Managing Urban Mobility Systems

MANAGING URBAN
MOBILITY SYSTEMS
ROSARIO MACARIO
Instituto Superior Technico, Lisbon, Portugal
United Kingdom North America Japan

India Malaysia China
Emerald Group Publishing Limited
Howard House, Wagon Lane, Bingley BD16 1WA, UK
First edition 2011
Copyright r 2011 Emerald Group Publishing Limited
Reprints and permission service

Contact: booksandseries@emeraldinsight.com
No part of this book may be reproduced, stored in a retrieval system, transmitted in any
form or by any means electronic, mechanical, photocopying, recording or otherwise
without either the prior written permission of the publisher or a licence permitting
restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by
The Copyright Clearance Center. No responsibility is accepted for the accuracy of
information contained in the text, illustrations or advertisements. The opinions expressed
in these chapters are not necessarily those of the Editor or the publisher.
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
ISBN: 978-0-85724-611-0
Emerald Group Publishing

Limited, Howard House,
Environmental Management
System has been certified by
ISOQAR to ISO 14001:2004
standards
Awarded in recognition of
Emeralds production
departments adherence to
quality systems and processes
when preparing scholarly
journals for print
Preface
Cities certainly are some of the most complex outcomes of human initiative and
activity, and they give us tremendous examples of diversity of physical shapes,
economic foundations, functional paradigms, attitudes to foreigners, and so on,
leading to what is frequently called the character of a city.
There is also a very strong diversity in the level and nature of planning that
occurred across cities, some with a very centralized concept and design, strictly
followed in implementation, others with a sequence of plans covering different parts
of its territory and adaptable modes of implementation, reecting in those modes the
evolving interpretations of priorities for the urban fabric, still others in which existing
plans are mostly an effort of rationalization of infrastructure following location
decisions already made by citizens and companies. Quite frequently these different
patterns co-exist side-by-side in the same city, as witnesses of different periods of its
past.
In all cities the issue of mobility is essential for it is mobility that allows the
interaction of people and the trading of goods, the two dening elements of
the reasons for the very existence of cities. And of course, mobility being so much at
the heart of the city, there are also very diverse examples of mobility patterns,
intensities, and shapes.
Also on the mobility infrastructure dimension of the city different levels and
natures of planning can be observed. In many, the very design of the city was
established on the basis of the street network, itself prepared with clear rational
foundations, in others we can see that it were just the main elements of that network
(the Avenues) that were pre-established, with the secondary and tertiary streets
showing a more haphazard layout, depending on the particular circumstances of
wealth and taste of those that were acquiring the successive land plots. Of course,
since the middle of the nineteenth century, many cities show clear elements of their
layout in adaptation to the railway lines and stations which became key players in
their connections to other cities and territories.
There is a very strong dialectic relation between the layout of the city and its
functions, and the mobility system that connects the different pieces. This mobility
system and its performance have become so important in the lives of people that it no
longer can be considered as an agent at the service of the city as its principal, and in
fact also has to be considered a fully assumed rst rank actor on stage, in par with the
xviii Preface
city itself. Each of them now shows its own objectives and their developments should
be harmonized, but the only hierarchy that is accepted is of them both to society at
large, and not of one to the other.
That is one of the leading reasons why the theme of this book is so important and
its timing so adequate.
The complexity of the problem and of the associated system have long been
recognized, and the quest for an integrated approach can also be found in many policy
and research documents for at least two decades, mostly in a European context.
However, these approaches have mostly followed two separate lines of endeavor: the
transportland use integration and the multimodal transport integration, where in the
former efforts are concentrated on urban forms and functional mixes that reduce the
propensity to move by private car, whereas in the latter those efforts are geared in
making the use of multiple modes of public transport more attractive, no matter
whether as legs of one journey or across the day or week in separate journeys.
The author clearly denes quality of the Urban Mobility System as the goal, and
produces a theoretical framework in the domain of Systems Dynamics to develop her
work. These options provide a much richer eld for a systematic exploration of
concepts across different levels of policy intervention.
A rather interesting part of the text is dedicated to the identication of the agents
in the system, the tensions that evolve among them as individual agents and as
collective entities, and to the different types of instruments available to organize and
manage their interactions.
This is followed by the main innovative contribution of this work, which is a
careful denition of Quality of the Urban Mobility System, for the system as a whole
and not as had already been done in several instances for any of its
components. This is a challenging step given the diversity of the agents and of their
interests, and it is where the theoretical framework proves its value, providing the
author and the reader with the foundations for a coherent approach to the challenge.
Careful review of the quality concept in Urban Mobility Systems and appraisal of
the effectiveness and sufciency of the traditional quality management models in this
eld are made, followed by analysis of what the direct and indirect quality factors are
and of their roles in the system.
Possible quality management issues in these systems are explored next, in full
recognition not only of its complexity but also of the dispersion of decision power over
the different subsystems and associated policies. The key elements for overcoming the
ensuing difculties are a clear separation between the strategic, tactical and operational
decision levels, and the full and explicit recognition and use of the numerous feedbacks
in the system, based on a careful denition of indicators. Naturally, Causal Loop
Diagrams come to the surface as the more capable instruments to represent these
interactions.
The proposed quality management model is nally reached after a careful
discussion of the different patterns of dialogue between institutions in different
cultural and political settings across the world, and indeed seems rather adaptable.
A key feature of that adaptability is the recommended separation between the
processes for quality planning, control, and improvement.
Preface xix
The proposed model seems of general applicability although the immense variety
of existing situations implies that its declination to the local circumstances must be
carried out by the reader. Not only this, but it is very likely that different agents in
the same urban area will produce different declinations from the same abstract model
proposed in this book. However, this should not be seen as a negative feature, as
those different declinations may well serve as starting points for the indispensable
discussions across the eld, with the advantage that there would be a reference
platform that could have been previously accepted by all parties.
That is indeed the major strength of this work, presenting an abstract model based
on substantial experience of the author in concrete cases. Publication in this form
should allow it to reach wider audiences and thus greatly increase its value for urban
communities across the world.
Jose M. Viegas
December 2010
Contents
Abstract vii
Acknowledgements ix
Quotation xi
List of Figures xiii
List of Tables xv
Preface xvii
1. Introduction 1
2. A Theoretical Framework for the Management of Urban Mobility

Systems 39
3. Simplifying Complexity in Urban Mobility Systems 79
4. Conguration of Quality Factors in Urban Mobility Systems 137
5. How to Set Up a Management Model for Urban Mobility Systems 197
6. The Need for a Strategic Approach to Urban Development 259
Bibliography 271
Annexes 287
Chapter 1
Introduction
1.1. Evolving urban societies
The concept of urban involves a series of interrelated dimensions, among which:

population size and density; spatial, economic and social organization; variety of
functions and institutional interactions; social values of population or degree of
civility (often also referred as urbanism), etc.
A review of existing denitions of urban area has been undertaken by the
Network for Urban Research in the European Community (NUREC) (Paddison,
2001, p. 31) and it concluded that the level of comparability between urban areas is
very low. This is mainly because the basic geographical units, and even the functional
building blocks, are substantially different and also because of the importance given
to economic and social functions existing in any urban area. As a consequence of this
diversity other factors result as essential for urban management, such as the
dimension of urban infrastructure and other supply systems.
In addition, the spread of inter-urban connectivity, that is the growing conurbation
effect,1 in the past decades called for a redenition of the concept, emphasizing
interactions and functional relations instead of geo-morphological criteria. As
reported by Hall (1969, pp. 408435) and Hart (2003, pp. 102123) much of the
movement that some decades ago was considered as regional is now incorporated in
urban agglomerations, sometimes even producing a cut across national boundaries,
such as the case of urban areas between Belgium and the Netherlands or between
France, Germany, and Switzerland.
Indeed, some authors have dened the city on the basis of a functional community
area, representing a self-contained labor and social market area characterized by
high frequencies of interaction (Frey & Speare, 1995, pp. 139190; Hawley, 1971,
pp. 149150)2. The concept of interaction and interrelation is present in the denition
of systems applied to many different sciences, from biology to management science, but
in urbanism and, consequently, in urban mobility these concepts gain an inter-
disciplinary character. The Athens Charter (1933) in its article 83, introduces an
1. Here understood as an aggregation or continuous network of urban communities often using common
supply services.
2. These authors considered only the labor market and ignore the existence of a social market, where
leisure activities occur, that is also fostering competition between cities and is one determinant factor of
mobility needs.
2 Managing Urban Mobility Systems
extended concept of conurbation by dening its limits according to the scope

of economic action, which implicitly shows that the interdisciplinary interactive
character of urban life was considered long ago. In a similar line of thought,3 although
more recently, the American Charter of the New Urbanism (Leccese and
McCormick, 1999, p. 15) rst discussed in 1993 and nally adopted in 1996, recognizes
that:
The last half-century has seen the rise of a social and commercial
geography that fuses town, city, and suburb into a new but unresolved
order the metropolitan region. Its becoming clear that the economic
building blocks of the global economy are regions not stations, states,
or cities. Its equally clear that many of our environmental challenges
are regional in scope
thus also admitting the need to go beyond the traditional geographical or admin-
istrative boundaries and suggesting a systemic approach.
A common argument around the systemic approach is that every system is
part of another system, that is the existence of formal hierarchies of systems, with
formal subordination of the lower level partition to the upper level partition of the
more global system. In urban dynamics, and especially in urban mobility, this formal
hierarchy is not so straightforward as it may be found in other disciplines. Moreover,
the interpretation and denition of these formal and informal hierarchies is a key
factor for how planning and control functions are designed and performed.
However, we leave this theoretical discussion for a later stage4 and dene the
urban mobility system, central object of this work, as an enabler of the urban system.
That is, a subsystem not only with great autonomy of organization but also with
strong symbiotic relations with the other subsystems of urban life (i.e., land use,
environment, telecommunications, security, education, etc.) as well as with the main
upper system, leading to causeeffect relationships between their performances and,
consequently, inuencing their evolutionary capacities.
Simons metaphor of the watchmaker, also known as the parable for biological
evolution (Simon, 1999, p. 188), provides a good illustration of the logic behind this
causeeffect performance link:
There once were two watchmakers, named Hora and Tempus, who
manufactured very ne watches. Both of them were highly regarded,
and the phones in their workshops rang frequently-new customers were
constantly calling them. However, Hora prospered, while Tempus
became poorer and poorer and nally lost his shop. What was the
reason?
3. Despite the differences (physical, functional, and cultural) between the European and American cities.
4. See Chapter 2 for discussion about the hierarchical and interactive character of urban mobility systems.
Introduction 3
The watches the men made consisted of about 1.000 parts each.
Tempus had so constructed his that if he had one partly assembled and
had to put it down to answer the phone, say it immediately fell to
pieces and had to be reassembled from the elements. The better the
customers liked his watches, the more they phoned him and the more
difcult it became for him to nd enough uninterrupted time to nish a
watch.
The watches that Hora made were no less complex than those of
Tempus. But he had designed them so that he could put together
subassemblies of about ten elements each. Ten of these subassemblies,
again, could be put together into a larger subassembly; and a system of
ten of the latter subassemblies constituted the whole watch. Hence,
when Hora had to put down a partly assembled watch to answer the
phone, he lost only a small part of his work, and he assembled his
watches in only a fraction of the man-hours it took Tempus
As referred by Simon (1999, p. 195), one of the sources of selectivity for systems
evolution is their capacity to create building blocks, which will constitute basic stable
congurations that will contribute to accelerate the evolutionary process.5 Urban
mobility, like all other referred subsystems, acts as one building block of urban life and,
contributes to its conguration through the interaction with land use, environment,
and other subsystems.
Urban areas together with their stakeholders members of urban societies can
be seen as complex6 systems with a wide span of control of several subsystems
(themselves, also of considerable complexity) (Figure 1.1). One of the main difculties
associated to this perspective is the identication of boundaries for interaction between
these subassemblies, and the understanding of the extent to which the interaction
between any two of these subsystems will also affect the others, through the simple
effect of contextual disturbance. We have thus three clear distinct dynamics within the
uppermost urban system: between upper and lower partitions within the same
disciplinary dimension and across different ones, designated as vertically nested
dynamics; between subsystems, designated as transversal dynamics; and the one within
each of the subsystems, designated as inner dynamics. The objective of this research
work was to develop a management framework for urban mobility systems,
5. Speed of evolution was discussed by H. Jacobson in Information, Reproduction and the Origin of
Life, in American Scientist, 43 (January 1955), pp. 119127, where he applied information theory to
estimate the time required for biological evolution. The essential idea of Jacobsons model is that the
expected time required for the system to reach a particular state is inversely proportional to the probability
of the state hence it increases exponentially with the amount of information (negentropy) of the state, in
Simon (1999), p. 189, footnote 4.
6. At this stage we follow Simons semantics and consider complex systems as the ones with a high number
of components, interaction and interdependencies, woven together in a logical whole.
Figure 1.1: Urban mobility dynamics (conceptual diagram). Source: Author.
considering the interactions and interdependencies accruing from this three-tier

dynamics, meant to foster the following attributes that should be understood as main
properties of an urban mobility system:
Robustness, meaning long-term stability and sustainability.

Adaptability, meaning the capacity to adapt services to evolutionary demands or
new technological opportunities, often resulting from exogenous changes, which
are typically initiated within the subsystems where urban mobility requirements
are generated, therefore not controlled by the mobility system).
Efciency, meaning high productivity, in the capacity to transform basic resources
into service outcomes, and these into consumption units, providing the best results
at the lowest possible cost.
Diversity, capacity to respond to the different demands of different market
segments in a dynamic match between supply and demand for urban mobility.
The evolution of information and communication technologies substantially changed

the conguration and processes of our societies. Business processes are becoming more
and more spatially dispersed, costs of knowledge will foster co-operation between
Introduction 5
business and institutions and more and more people will work with information as their
main productive resource, whereas workers of physical production are becoming a
minority. The economy is not only global but it is also transforming into a network
economy both at international, national, and regional scales, challenging all former
physical concepts associating space and time dimensions.
Business and social trends have consequences on the different spatial strata, the
reverse being also true. Network societies are built over the emergence of urban
networks where spaces are conceived according to the new social paradigm of a
highly differentiated and selective society. Some years ago cities used to be clearly
recognized by their hard boundaries or the simple landscape discontinuity. Today,
urban sprawl has multiplied centers, which compete to attract citizens, fostering
motorized zigzagging across distanced centers as one of the most appealing weekend
leisure activities for a good part of the population.
Current and future societies are characterized by high individualization and with
most developed countries showing a trend to early social emancipation, where the
individual becomes the basic reference unit, instead of the family or group of friends.
Consequently, individual freedom of choice is expected to increase as well as social
mobility, as transitional work increases and quick up and down turns in social
circumstances are seen as a normal characteristic of individuals evolutionary paths.
Co-existence of different social groups and heterogeneous areas, from the social and
spatial viewpoint, should thus be a consequence of the modern way of living and an
attribute of new societal congurations. As Ascher (2001) so clearly points out:
Les individus apparaissent ainsi comme socialement multi-appartenants,

socialement pluriels. Leur pratiques, leur syste`mes de valeurs, leur choix
individuels resultent de socialisations et de circonstances diversiees.
Par exemple, le sport pratique par un individu sera en correlation plus
forte avec son origine geographique, la musique quil ecoute se
rattachera plutot a` une classe dage, son travail dependra pour une part
signicative des origines socio-professionnelles de ses parents , ses
choix de vacances seront plus fortement determines par une negocia-
tion familiale, son vote politique dependra du lieu ou` il habite, etc.
La differenciation sociale semble ainsi peu a` peu pulveriser une societe
dans laquelle des individus plus differents et autonomes, ne partagent plus
que momentanement des valeurs et des experiences sociales. (Ascher,
2001, p. 34, emphasis is ours)
With this evolution, the use of urban spaces will be by and large dominated by a
context that is changing from providing unique and monofunctional options, for
each problem or need, to redundant multifunctional ones (Ascher, 2001, pp. 8283).
Enlargement of spatial implantation of cities, large conurbations in special, led to
a more restrictive offer of goods, services, and even social events within walking
distance. Theoretically this possibility still exists for people living near commercial or
business centers, but for the most part of the population access to those elements of
social life is increasingly dependent on motorized transport, particularly private car.
Moreover, concepts of quality of life and ways of living have changed substantially,
one of the most signicant changes being the valuation of urban places as a function
of the diversity of multiple options that they may offer.
This concentration of diversity can be found mostly in commercial areas, where
multiple functions can be performed at the cost of the same trip. Elsewhere it requires
moving at a distance that usually requires motorized means. These cases, as referred
by Viegas (2002a, 2002b, p. 36), lead us to the recognition that motorized mobility is
an individual right and a fundamental citizenship factor for social integration,
without which social exclusion might occur.
In this evolutionary context, with a growing accent on individual actions, the
steering role of local governments gains more prominence than ever before, with
management control systems being achieved through clear rule setting, close
supervision of planning, and implementation processes. The aim is avoidance of
misfunctionalities and inconsistencies between interacting subsystems, assessment of
systems (and subsystems) evolution and decisions of when and where to act to assure
good match with societal needs, together with an adequate balance between efciency
and effectiveness of the decision processes. Following the concepts presented in
(Friedman & Miller, 1965, pp. 192236, in Anthony & Govindarajan, 1998, pp. 13),7
local governments must assure the existence of the following ve elements, embedded
within their organizational and functional settings, as illustrated in Figure 1.2:
A detector or sensor, meant to identify what is really happening with the process
being controlled.
An assessor meant to determine the signicance of what is happening. This is
usually done by comparing information with some standard or previously dened
expectation.
An effector meant to accomplish the action, whenever the assessor indicates the
need to do so, as a result of a feed-back mechanism.
An information system, to translate what is happening into performance
assessment indicators.
A communication network, to transmit information between the detector and the
assessor and between the assessor and the effector.
As Ascher denes it, urban modernization is a process that has been developed
quite ahead of its own public awareness. It results from the interaction between three
socio-anthropologic dynamics, responsible for the current conguration of modern
societies. These are individualism, rationalization, and social differentiation (Ascher,
2001, pp. 1213).
7. Both Milner and Anthony ignored the need of the information system as a basic element.
Introduction 7
Figure 1.2: Elements of the control process. Source: Author, Adapted from
Anthony and Govindarajan (1998, p. 2).
Individualism can be considered as the self-view of the world, ltered by the

individual selsh interest, that is the stakeholdership8 role each citizen has been
taught to perform in his relation to community life. Rationalization relates with the
process of choice between different options and consequent acts, based on
accumulated information on empirical evidences, scientic knowledge, and utiliza-
tion of methods and techniques. Rationalization arose as opposed to tradition or
beliefs and (in the framework of community life) raised the individual ability of
contesting public decisions. Social differentiation in turn is a process of diversifying
the roles of individuals and groups within the society. The most fundamental
principle wrapping this triad is to accept change as a permanent process in the
interaction between these dynamics. That is, in modern societies stakeholdership is
a moving condition, with each individual playing different personalities in successive
moments and varying his position along time and circumstances, according to his
own capacities and conveniences.
Along time a number of milestones for the evolution of modern societies, and with
it also the concept of city, can be observed (Le Gates & Stout, 2000, pp. 1788).
8. That we dene as the attitude of standing for specic social and economic interests as a result of self-
perception of events and respective advantages and disadvantages.
A rst milestone concerns the period between 1434 and the mid-sixteenth century,
when European colonization and trade had a rich period and for the rst time
extremely different human societies entered into some degree of interactive evolution.
This period is marked by the Treaties of Alberti, which started in Florenze, as
reported by (Benevolo, 1993, pp. 104109) by then Europe represented 2.5 million
square kilometers concentrated around 130,000 towns, thus representing 2025% of
humanity that consumed 50 per cent of the available animal protein and had access
to ve times the per capita energy available in China. As he expresses (idem, p. 106),
Europe inhabited a continuous territory and from the top of any bell tower one
could see for or ve others.
The confrontation of paradigms between European settlements and others was a
source of creativity in the development of cities. The European paradigm was based
on a schematic and utilitarian rationale which dominated the centers of economic
and military power in the rst half of the sixteenth century of which we have good
examples in Bairro Alto, in Lisbon, and in the expansion of Antwerp. These models
were then reproduced overseas and became referenced as outcomes of a general
esprit de geometrie in which the persistent medieval tradition harmonized with the
humanist preference for regularity (Benevolo, 1993, p. 107). Later, after the Council
of Trento (1563), which contributed to stabilize the political and religious balance of
Europe, the architectural organization of cities evolved to a closer relationship
between city and geographical setting, later known as the perspective trend.
The industrial revolution caused a demographic explosion in the cities,
accompanied by a steep growth of spatial denition of urban areas, resulting in a
clear register of social differentiation in urban congurations. A new concept of city
has emerged ruled by the logic of the industrial revolution specialisation, that was
one of the remote origin of the zoning concept,9 reinforced by the Haussmanns
model of reorganization of cities, with a rigid separation of public and private spaces
(Benevolo, 1993, p. 177).
After World War II, and irrespective from the different urban forms adopted in
the several countries, the concept of Welfare State fostered everywhere in Europe
the development of public services and extended public intervention from the
economic and social dimensions also to urbanism. As we will see later, this concept is
still engraved in the current dynamics of provision of public services. It was also
during this period of urban evolution that urban planning was effectively empowered
as a visible component of city management. With the emergence of the concept of
Welfare State and the development of industrial capitalism, the rst needs were
adapting the conguration of cities to the requirements of production and
consumption, as well as of stock management.
The last decades observed a technological development that allowed interactions
to occur at such a spatial and time distance that there is a clear competition between
the physical urban area and the articial one enabling the development of a new
9. The term urbanism appears in different ways by the end of nineteenth century (Ascher, 2001, p. 16).
Introduction 9
concept of local area. Social and economic interaction tends to occur by way of
networks, where each individual can simultaneously belong to several networks. As
Castells concludes Networks constitute the new social morphology of our societies,
and the diffusion of networking logic substantially modies the operation and
outcomes in processes of production, experience, power, and culture (Castells, 1996,
pp. 468469).
In fact, this network structure allows the same economic agent to be active in
different countries and cities, whereas it enforces competition between cities, for the
most prominent positions in the regions. Given the exibility in the location of
activities (labor and social) cities will compete with one another to attract individuals
based on the quality of life they offer, which entails the level of accessibility to the
most relevant markets. This increases the governance difculties and also the
complexity of all urban subsystem (including mobility system), which have to serve a
much more differentiated set of clients, who are very often located out of the
administrative boundary of jurisdiction. The new logic will inevitably cause a
reordering of urban hierarchy and of economic and political links between places
(Fainstein & Campbell, 1996, pp. 67).
The composed effects of all these changing elements inevitably call for rethinking
the institutional architecture. As referred by (Ascher, 2001, p. 95, emphasis is ours)
Le gouvernement des villes fait ainsi place a` la gouvernance urbaine, que lon peut
denir comme un syste`me de dispositifs et de modes daction qui associe aux
institutions des representants de la societe civile, pour elaborer et mettre en oeuvre les
politiques et les decisions publiques. La gouvernance urbaine implique un enrichisse-
ment de la democratie representative par de nouvelles procedures deliberatives et
consultatives. Elle necessite a` la fois une relation plus directe avec les citoyens, et des
formes democratiques de representation a` lechelle a` laquelle doivent se prendre les
decisions urbaines structurantes et strategiques; and he still adds the concern that
Et la democratie de voisinage sans la democratie metapolitaine ne peut sufre pour
faire prendre conscience aux citadins que leur sorts sont lies (ibidem, pp. 9596).
The important point for any government is how well it will be able to mix
legislation, regulation, and persuasion to succeed in encouraging people to adopt the
right behavior in such a way as to minimize the societal and individual costs of keeping
the freedoms expected under the modern society paradigm. Given the perception that
both legislation and regulation are not in the leading edge of innovative promising
tools, we should expect to see in the near future urban governance science (and
forefront practitioners) investing in persuasion as a basic ingredient to develop new
ways of conceiving and implementing public decision.
Citizens participation in decision processes might well be an early example of the
new coalition style. Leadership skills will replace direct hierarchical lines of command,
shifting the emphasis from a social optimum determined by superior governance to
shared objectives largely built out of a community process of understanding and
validating main needs, made operational by local governments, as legitimate
representatives of urban communities.
The new logic called upon a new style of government that some authors prefer to
distinguish with the designation of governance and that in practical terms most
democratic political and administrative systems try to achieve, staying very often
somewhere in between (Table 1.1). Although evolving to this new political paradigm,
where institutional relationships imply a social objective (collective, i.e., society as a
whole), there is a collective benet that could not be obtained by governmental and
nongovernmental powers (actors and resources) acting separately (Stoker, 1995,
pp. 301315). Under this new framework for public intervention the steering
mechanisms and processes emerge as critical instruments for governance at good
performance levels.
The dynamics of the new societies, with network conguration and largely based
on communication facilities, also developed a sense of potential ubiquity in
performing part of our daily activities that contributes to changing our under-
standing of mobility, often seeing it as a potential corollary of that differentiation
effect. In the past decades the situation evolved from understanding mobility as a
vehicle-related concept to a recent, and only applied in a very reduced number of
cities, wider concept of mobility where for all modes and means specic missions are
attributed. The widening of the scope that forms the basis of the rationale behind
decision-making in these areas also widens the complexity of the problem regarding
the number of entities to be coordinated, the number of nested spatial domains to be
considered (EC, DGT, 1998, Cost 332, p. 23) and, consequently the potential for
conicts of interest and complexity of governance.
This change process between the new and old logic lying behind the rationale for
land use and transport planning was well understood in Cost 332 research, dedicated
to transport and land use policy, and is reproduced in Table 1.2.
City and local authorities have their decision-making process made more
complicated by the difculty to accurately evaluate the effects of their policies or
packages in the short, medium, and long term. The lack of systematic information on
these domains is even a limiting factor to develop mathematical simulation tools, as
some causeeffect relationships are not understood in all its extension. Besides, while
a single organization can evade goal controversy because it is only a part of the
system, within government institutional frameworks the intergovernmental dynamics
Table 1.1: Government versus governance.
Government Governance
Hierarchy Co-operation
Monopoly of public power and Vertical and horizontal relationships,
vertical relationships both intergovernmental and with the
private sector
Universality and permanence Sectorial variations in the networks of
of the network of relationships, which can be modied
relationships according to circumstance
Public intervention in all areas Auto-limitation of public intervention
Source: Adapted from Stoker (1995).
Introduction 11
Table 1.2: Logic elements behind the rationale for transport and land use.
Old logic Feature New logic
Expand Networks Manage and integrate

Predict and provide Forecasts Predict and manage
Hard-supply oriented Technology Soft-demand oriented
Engineers isolated Professional culture Managers open
Hermetic and sectorial Policy making Discourse and integration
Homogeneous Space Customized
Reduce travel time Time Niche, certainty
Disengaged Users Reengaged
Site-specic externality Environment Global emissions
Standardized, static Knowledge and information Tailored dynamic
Macro-extrapolation Modelling Microlevel responsive
Source: Adapted from EC (1998b, Cost 332, p. 47).
are representations of the whole systems and as such of the all polity (Christensen,
1999, pp. 2345).
Consequently, the need for innovative management tools is seriously felt by local
administrations. As Simon (1978a, 1978b, p. 5) stated in his Nobel Lecture The real
world in fact is perhaps the most fertile of all sources of good research questions
calling for basic scientic inquiry. Furthermore, some policy measures or packages
need a considerable time gap to produce effects, sometimes conicting with the time
gap of policy cycles, which often hinders continuity of objectives. The goals accruing
from the mandate of public organizations often result in multiple, conicting, and
vague operational objectives.
European research10 revealed that in most European cities the integration between
transport and land use is one of the driving factors for long-term sustainability.
Surprisingly, the awareness gained so far points out, as rst line target of public
concern, to issues like land scarcity, trafc congestion, and decreasing quality of life,
instead of local emissions and other environmental concerns. Despite the implicit
causeeffect relation, this evidence also raises the problem of dening sustainability
and, with it, local quality of life, as well as other concepts, which are up and
downstream of this complex and rather subjective denition of sustainability.
Later in this chapter, and also throughout the remainder of the book, we
demonstrate that both complexity and subjectivity lay in the need to trade-off between
10. TRANSPLUS Transport Planning, Land-Use and Sustainability, EC supported project under the
City of Tomorrow and Cultural Heritage key action, within the European Commissions Energy,
Environment and Sustainable Development Research Programme.
economic, social, and environmental sustainability while keeping an integrated

approach to development and quality of urban living. In this context integration arises
as a multidimensional task since its result is not only an attribute required at policy
level (often the weakest link) but also between the supporting organizational models,
monitoring indicators, and institutional structures. It is the structural integration as
a quality characteristic that imposes consistency in decoupling the sustainability
concept from a universal concept,11 as Brundtland Commision (1987) dened it the
development that meets the needs of the present without compromising the ability of
future generations to meet their own needs, down to the large diversity of its national,
regional, and local interpretations. The ne-tuning of sustainability concepts has been
an essential contributor to strengthen planning processes while providing in-depth
awareness on barriers to its operational interpretation, implementation, assessment,
and possibly damage control of inherent risks of nonaccomplishment. Evaluation and
monitoring of progress achieved along time are fundamental tools for these assessment
processes, although it must be evident that sustainability is a concept that should go far
beyond the clean environment objective and corresponding traditional indicators as
illustrated in the DPSIR framework12 developed by the European Environment
Agency (EEA) and illustrated in Figure 1.3.
Furthermore, sustainability aims induce urban planners and managers to cease the
traditional fragmented disciplinary thinking, by breaking down articial barriers,
which constitute inrmities to our contextual knowledge of factors inuencing the
dynamics of urban systems and even our ability to predict long-term evolution. The
prominence of this fragmented thinking has often hindered our capacity of establishing
relations between parts and wholes of subsystems and components interacting in urban
areas. This rationale limited our ability to understand and deal with chance and
uncertainty often caused by asymmetric information between agents. Dealing with the
unexpected is a common requirement in open systems, where the absorption of new
ows of information is continuous and where we have not only a considerable diversity
of stakeholders but also each individual plays several (sometimes, conicting)
personalities in different moments, according to the specic interests at stake.13
As highlighted by Morin (2001, p. 29), The universal problem for every citizen of
the new millennium is how to get access to information about the world and how to
acquire skills to articulate and organise that information, how to perceive and
conceive the context, the global (the whole/parts relation), the multidimensional, the
complex. This is also a main challenge in managing urban mobility systems, which
have their structural support on three main processes and respective feedback
11. Sustainable development concept rst introduced in 1987, with the Brundtland report, has been since
then objects of a wide diversity of operational interpretations and recommendations disseminated in a
comprehensive set of literature references.
12. DPSIR framework: Driving Forces, Pressures, State of the Environment, Impact and societal
Responses.
13. At least as an individual with its own moving interests, a member of society and a member of human
beings species (Morin, 2001).
Introduction 13
Figure 1.3: DPSIR framework. Source: TERM research project (EEA, TERM, p. 12).
knowledge: strategic denition, conguration, and delivery of supply and steering

system performance according to society needs.
Inevitably, the awareness of complexity of sustainability goals leads to a more
strategic thinking of urban planning, where the multidimensional character of society
is emphasized providing evidence on the interaction, cross feedback, and inter-
dependence between the different components that form the urban system. Although
this work has no objective of undertaking an in-depth discussion on the dynamics of
urban systems, it seems to be unavoidable that in particular urban planning has to be
addressed as a relevant contextual background for the management of urban mobility
systems. Land use and transport planning are key factors in this process for spatial
distribution of activities and as such potential originators of mobility ows and
ultimately of environmental externalities. As Fainstein and Campbell (1996) so well
observed Urban space gains its meaning as a consequence of the activities carried on
within it, the characteristics of the people who occupy it, the form given to its physical
structures, and the perception with which people regard it. Consequently such space
does not simply exist; it is instead a social creation (Fainstein & Campbell, 1996,
pp. 1011).
Land use and transport policies have been the two main streams of action to
inuence the spatial distribution of activities, often seen as alternatives to each other.
Direct intervention on land use policy or indirect inuence on land use patterns
through intervention on transport policy are common actions, reason why these
instruments should rather be seen as complementary to each other in the developing
and shaping of sustainable urban areas.
Irrespective from the selected process two main options14 gained prominence
in the past decades, the monocentric and the polycentric congurations of urban
areas. Both represent an empirical validation that no policy formulation, either
sectorial or global, can be implemented without considering a specic spatial
context that, in turn, will impose concerns on its structural characteristics.15 A third
trend that has also revealed in the past decades is the desorganized scattered urban
area typical of the illegal agglomerations but also of the suburban American small
city.
The recognition that decisions on land use and transport need to be co-ordinated
led to the notion of land use transport feed-back cycle, materialized in the
following sequence presented by Wegener (2002, p. 5): (i) The distribution of land-
uses, such as residential, industrial or commercial, over the urban area determines the
locations of human activities such as living, working, shopping, education or leisure;
(ii) The distribution of human activities in space requires spatial interactions or trips
in the transport system to overcome the distance between the location of activities;
(iii) The distribution of infrastructure in the transport system creates opportunities
for spatial interactions and can be measured as accessibility; (iv) The distribution of
accessibility in space co-determines location decisions and so results in changes in
land-use system.
A number of distinct aspects involve this problematique in terms of interdependence
between transport and land use policies that are both inside and outside the political
and administrative system. Outside the system we have mainly the interdependence
between territorial (or spatial) facts and processes associated with land use and
infrastructure use. Inside the system the interdependence focus is mainly between the
different approaches to planning and the need for co-ordination between the actors and
policies in the different elds.16
Shorey Petterson (unknown, p. 103),17 provides an extensive survey and
structured analysis and criticism on the multiple use that the co-ordination concept
14. Sometimes strategically dened but most often just a result of stakeholders movements and actions
under existing legal and economic constraints, in particular where public planning is not a cultural
tradition.
15. Simoes Lopes (1984), Desenvolvimento regional, problematica, teoria e modelos, Fundac- ao Calouste
Gulbenkian, p. 276.
16. Cost Transport, Transport and Land-use policies, Transport Research Cost 332, European
Commission.
17. Irregular form of reference because only a paper copy is available with date and identication of
publication unreadable.
Introduction 15
has among persons studying transport-related matters, and nally concludes for the
following diversied denition Co-ordination is the assignment, by whatever
means, of each facility to those transport tasks which it can perform better than other
facilities, under conditions which will insure its fullest development in the place so
found. Co-ordination is the creation, in any way at all, of effective joint services by
agencies which are directly complementary. Co-ordination is the attainment of such a
compromise between monopoly and competition as will insure the continuance of
essential agencies, maintain the maximum variety of service, eliminate undue waste
and preserve effective incentives to improvement. Co-ordination is the avoidance of
duplication through the subordination of rival agencies. These denitions are not
necessarily inconsistent, but they involve radical differences in emphasis, and suggest
divergent programs of execution.
For the sake of clarity and given the lack of unity of meaning that the concept of
co-ordination entails we nd it necessary to advance a denition. In this research
work co-ordination is understood as the function of steering two or more interacting
elements of a system (or subsystem), so that their joint performance attains previ-
ously dened goals.
Another schematic division for analysis of the problem, lies on the enterpreneur-
ship function regarding the policy denition which can be either land use focused or
transport focused, in both cases taking either a deductive approach (e.g., strategic
denition) or an inductive one (e.g., specic problem solving), the former usually
generated inside the political-administrative system, the latter generated in the users
domain of action leading to reactive policies or measures focusing a specic problem.
Choice of policy focus will of course determine the actor who should take the
initiative for dening urban planning priorities.
Today, a number of studies18 provide in-depth knowledge, based on empirical
evidence, on the difculties and barriers to the implementation of effectively integrated
sustainable policy strategies in urban areas. Although the type and number of obstacles
differ between any two communities, thorough observation leads to conclude that
some of the most common stumbling blocks are:
unclear and unt legal and regulatory frameworks;

land use and environment; counterproductive institutional designs and allocation
of roles;
unclear nancing and investment streamlines with inadequate pricing and scal
structures;
poor integration between transport and land use;
low quality or even nonexistent information systems;
18. Such as: PROGRESS and the accompanying measure CUPID; projects within the European
Commission CIVITAS initiative, such as MIRACLES, TELLUS, VIVALDI, TRENDSETTER, and the
accompanying measure METEOR and several ECMT and World Bank studies. The author was directly
involved in both CUPID and METEOR accompanying measures, with responsibilities for evaluation and
corresponding methodological developments.
unclear and/or unt application of subsidiarity and proportionality principles

within the governance mechanisms; and
lack of experience and knowledge in the adoption of community participation in
policy building processes and problem awareness.
Indeed, dening and implementing integrated policies also entails bringing together
national, regional and local levels of government, and also public and private entities,
business developers and citizens. Policy and institutional interaction between all these
types of stakeholders, where some of them have no obvious authority rights over the
others, requires an adequate managerial structure and mechanisms encompassing all
the mobility system and supporting the clarication of roles and missions of the
different agents.
Sustainability should have a denition of objectives in the economic, social, and
environmental dimensions, which needs to reect short-term actions to accomplish
long-term goals. That is, nding an adequate balance between the different vectors of
the problem, in such a way that assurance can be given to citizens for access to the
services and activities required for their daily lives, while minimizing negative
environmental, equity, economic, and health impacts of mobility. This means applying
strategic thinking in the local operationalization of the sustainability concept, having
the structure and growth of urban areas as crucial considerations to develop alternative
strategic options. In this work sustainability is understood as a quality characteristic
of urban systems, strategically dened at the local (micro) level, despite the need of
consistency with macro and meso19 objectives.
1.2. Policy and management problems in Urban Mobility

Systems (UMS)
Despite the awareness of the concerns behind the sustainability concept, the fact is that
for most cities and countries sustainable mobility strategies remain rather illusory. The
evolution of society in last decades led to profound changes in urban living conditions,
with an increasing residential dispersion, longer home-to-work distances, and more
complex mobility patterns. In conjunction with this, an increasing availability of
private motorized transport is a common trend almost everywhere, both through a
more diversied supply of vehicles and through easier nancial access with several
sources of solutions, such as leasing contracts or other type of credit incentives.
19. We adopt along this work the concepts of micro-, meso- and macrolevels as dened in TRANSPLUS
research (Deliverable 1.2, p. 7), where microlevel relates to the immediate surroundings of a specic
infrastructure, meso relates to the city under the management of one municipal authority, and macro a
metropolitan region including a core city and surrounding suburbs or satellite towns, under the rule of a
County or Metropolitan Authority.
Introduction 17
However, as urban space is a limited resource, the more private cars circulate
inside the cities the lower is the global level of accessibility (measured in travel time)
for all the population using the road space. Consequently, the quality in terms of
travel time and frequencies that surface public transport undertakings are able to
offer decreases in a direct relation with this global level of accessibility. From this
evolution three problems resulted which are a major cause of concern in urban
transport policy and can be considered as the Achilles heel of any Urban Mobility
System (UMS)20:
high congestion problems in urban areas;

high levels of public money involved in the operation of urban public transport;
and
public transport companies losing market share and accumulating decits, often
justied by the public service character of the service provided, are unable to
release funds to re-invest in the improvement of the service provided.
The combination of these factors clearly contributed to change environment where

urban transport develops. On the one hand, both the general public and the politicians,
have acquired a different perception on the importance of urban mobility, that is seen
today as one of the citizens basic needs and as such a political priority. On the other
hand, a growing awareness has developed through the years that one of the key factors
to solve the aforementioned problems is to shift part of mobility from private car to
collective modes, releasing the cities from congestion while increasing revenues to PT,
making it less subsidy dependent. However, there is also a growing evidence that this
shift of market share can only be achieved by making collective means more attractive
and responsive to citizens needs, that is more customer-orientated, or better t to use.
In addition, it is also worth to point out that general access to information and higher
levels of education have become more common in society, transforming citizens into
much more demanding clients.
To cope with these changes, some urban public transport (UPT) companies were
driven to look for the solution to their problems in the quality certication processes
that, as we will see later, were mostly seen as an effective marketing tool, and the
most advanced ones have even turned themselves to the introduction of Total
Quality Management (TQM) models. Signicant investment was dedicated to the
implementation of these models in operating companies, as well as in certication
processes. An even more important factor is that some urban authorities start seeing
20. In Europe, since in other parts of the world State Intervention is thought on different basis than in
European countries (where a strong historical public service concept exists), sometimes encrypted in the
constitution itself as we could observe in the survey done by the author in the study Examination of
Community law relating to the public service obligations and contracts in the eld of inland passenger
transport, EC-DGVII, 1997/1998.
it as a basic requirement in the selection of operators,21 aiming with it to provide

public evidence on their capacity to ne-tune with the developments of modern
societies, very often missing the deepest essence and advantages of the process as a
management control tool.
Despite these efforts, only marginal improvements were obtained in the
performance of the companies that have introduced these processes and management
models (e.g., reduction of maintenance costs through the organization of processes,
better assistance to clients) which in turn have proved to be successful in other
sectors. Moreover, none of these developments solved the main problems faced by
the urban mobility systems during the recent years, that is:
the trend for a decreasing market share of UPT is not being reversed;
congestion and nancing problems are becoming more and more acute; and
citizens needs are not being adequately satised with the current supply of UPT
and private car.
This lack of success called for an in-depth analysis of the reasons why tools so well
tested in other sectors were apparently deemed to fail in urban mobility (EC, 1998a,
p. 18).
As cities grow and consumption demands become more complex, mobility also
becomes an essential lever to undertake other economic and social activities, thus
affecting the overall society and leading to consider mobility as a public service.
Traditionally, State interventions in public transport has been partly justied by equity
considerations, namely to ensure that the transport network was available to all
citizens, and that no one should be deprived of its services by considerations of price.
The interpretation of this goal, which in itself is still valid nowadays in the implicit
concept of public service, led the authorities to increase the nancing of Urban
Transport through the use of concessionary fares and subsidies to cover companies
decits. The main factors leading to this essentially political attitude were the aim to
compensate the insufciency of revenue caused by a loss of patronage in favor of
private cars, together with a regulatory framework that restrained operators from
establishing competitive prices.
It is clear today that this practice was not an efcient way of intervention and
allocation of public money, once in most places all users beneted from the same
(subsidized) fares independent from their income levels, with an exception of conces-
sionary fares for elderly people, etc. Additionally, there is a growing awareness that to
achieve a sustainable balance between private and public means of mobility pricing
policies have to be able to send the correct signals to induce an adaptive behavior from
the users, which in turn will provide the system with a reliable feedback on the needs
for further investment and expansion of transport facilities.
21. Although in the rst years this was done in a rather unconscious way revealing severe lack of
knowledge from the authorities.
Introduction 19
This switch of perspective is also reected in the aims of the Citizens Network
green book (EC, 1996a, p. 25) where the Commission states that it wants to assure
that the needs of the citizens are put at the centre of decisions about transport
provision. The goal must be the achievement of networks of public passengers
systems. In addition public transport should ideally be a service open to all citizens in
terms of accessibility to vehicles and infrastructure, affordability in terms of fares
levels, and availability in terms of service coverage, suggesting that UPT should be
considered as a service of general interests.
Already in the previous year, in 1995, the European Commission launched another
Green Paper under the title Towards Fair and Efcient Pricing in Transport: policy
options for internalising the external costs of transport in the European Union
(European Conference of Ministers of Transport (ECMT) 1995, p. 12), advocating
that:
Pricing should be seen as a complement of regulatory and other market policies.

The main aim of a fairer and more efcient pricing policy is to use price signals to
curb congestion, accidents, and pollution.
Prices should reect underlying scarcities to ensure sustainable transport.
Appropriate infrastructure charging is needed to mobilize private capital and relief
the pressure on public budgets.
The operational interpretation of the previous statements should reect a price

structure clear to transport users; differentiated across time, space and modes,
nondiscriminatory between modes, and Member States. Recently, a European
Commissions White Paper on Fair Payment for Infrastructure use: a phased
approach to a common transport infrastructure charging framework in the EU (EC,
1998b, p. 16), though not directly applicable to Urban Transport, reinforces that:
In future, charging systems should be based on the user pays principle,

supported by a marginal social cost rationale.
To achieve it a phasing system is proposed entailing:
the adoption of a broadly compatible structure in the main modes of
transport until 2000, i.e. the rst phase;
a second phase dedicated to the harmonization of the charging systems, with the
implementation of a kilometer based charging system, differentiated on the basis
of vehicle and geographical characteristics; and
nally, the third and last phase envisages the update of the implemented
framework based on the experience gained in the previous phases.
From this evolution we can conclude that the EU transport policy goals can be
grouped into three main categories, which reect the fact that equity and efciency
are still the main criteria for policy appraisal as systematized by Gwilliam (1987,
p. 46), with the exception of sustainability that was not referred by this author:
allocative efciency of resources within the transport sector and between this
sector and other economic sectors;
meeting individual requirements at minimum resource costs, that is market and

productive efciency objectives; and
equitable distribution of benets and costs, that is equity objectives.
However, as we have seen in previous section, the denition of an Urban Transport

policy is a complex issue, very much related with the specic characteristics of the local
environment as well as with the respective political options, which may change between
localities within the same country, and even between neighbor communities served by
the same transport system. The diversity of variables involved causes a wide diversity of
approaches to Urban Transport policy that in turn are reected in the denition of a
number of elements of the system, among which pricing and nancing policies have a
special role.
During the past decade many initiatives were taken in this domain22 with the
following three main driving goals: increasing productive efciency; reducing the gap
between the price paid by the users and the real costs of providing the service; and
reducing the needs of public expenditure in the sector by introducing new ways to
involve private nance. Many pricing and nancing schemes have emerged all over
Europe during these years, though only some of them have successfully survived
the implementation process. These implementation difculties often arise from
the difculty of identifying winners and losers and consequently devising effective
schemes, and accompanying measures, to transfer the gains of the rst into compen-
sation to the latter. Besides fairness within comparable groups, there is also the need to
consider the longitudinal character of equity, that is everybodys goal of no reduction of
previously available benets (entitlements) that corresponds to one of the major
difculties in modern societies.23
The success of pricing and nancing schemes in Urban Transport Systems is also
strongly dependent on the regulatory and organizational framework of the system,
and in its potential to co-ordinate between the different policies with impact in
mobility demand patterns (e.g., land use, environment, etc.) as well as between the
push and pull measures developed by different agents of the system authorities and
operators. It is often forgotten that one of the remote causes of the external costs
caused by transport is the location of the economic and social activities, which create
the mobility needs. Furthermore, it is also the role of the transport pricing policy
to contribute to the control of these external costs, and this can only be achieved by the
combined use of market-based incentives together with control regulations. The
former should persuade users to adapt their behavior toward the policy aims, whereas
the latter is mostly meant to restraint practices leading to the growth of external costs.
Finally, it is worth to highlight at this stage that pricing and nancing policies
recommend the use of different mechanisms at the same time as demonstrated by
Vieira (2005, p. 116), that is single measures are seldom fully effective if applied in an
22. Reported in detail in the following European Research Projects: PETS, OPTIMA, FATIMA, FISCUS,
PATS, UNITE, developed under the 4th and 5th RTD Transport Programme.
23. Designated as longitudinal equity in PATS research project (EC, 2001b, p. 66).
Introduction 21
isolated way. The risks involved in the implementation of each measure, their
synergetic potential, as well as the assurance that the different measures involved in
one policy package do not produce contradictory effects, are important issues that
have to be included in the concerns of the decision-makers when choosing the most
appropriated policies for their local enveloping.
This complex institutional environment that forms Urban Mobility Systems can
be disaggregated into three levels of planning and control (EC, TIS, 1997c, p. 26), or
decision levels, being:
The strategic level, where the mobility policy, objectives, and means, are dened
reecting the needs of the citizens. The corresponding decision process should be
performed by political entities.
The tactical level, where the mobility system is conceived and the respective
policies are dened translating the strategic goals into operational specications,
assuring the effectiveness, and coherence of the system. Depending on several
parameters the functions that this level entails can be performed by different public
or private agents and contracts can also be allocated through competitive
procedures.
The operational level, where transport services are produced and consumed.
Depending on the regulatory option, public transport services can be performed
directly by the transport authority, in which case it accumulates also the design of
the system, or contracted out to an operator (private or public) by direct
negotiation or through a tendering procedure. It is worth referring that the
individual self-production modes and all the infrastructures are also a component
of the mobility system.
In the real world the division into these three levels is not so clear-cut as
described earlier. For most European cities (urban areas or conurbations served by
the same transport system) the boundaries between these levels are very often fuzzy
and the overlap between the strategic and tactical levels is common, with less clear
(or even nonexistent) strategic options made. In addition, a consistency gap is often
found in different interacting boundaries of the subsystems. As illustrated in Figure
1.4, this gap arises either between the denition of strategic options and the tactical
formulation to achieve those ends or between this set of strategic objectives and
tactical formulations and the monitoring system of operations that should provide
feedback for path adjustment, together with the good reading of the stakeholders
needs. Bridging this gap means gaining consistency between stakeholders needs,
strategic objectives of the mobility system (dening what to achieve), tactical
formulations (how to achieve it), and monitoring processes and outcomes (how have
we performed and what do we need to adjust).
The lack of a clear and well-structured regulatory and organizational framework
is a determinant factor that may hinder the successful denition and implementation
of a coherent mobility system, in particular if an effective interaction between the
different parts of the system is not properly assured through a sound institutional
conguration.
Figure 1.4: Decision levels-conceptual diagram. Source: Author.
The denition of objectives, and guidelines to achieve them, is a function

within the strategic level of the mobility system. A consensual strategic goal is to
achieve a conguration for the mobility system able to address concerns at the
following dimensions originally dened by (Ciufni, 1965, p. 73) and herewith
adapted:
Transport dimension adequate balance between modes and means of transport,

so that all those that give up the use of private car have available alternatives of
good quality, without any sort of social, geographical, or sectorial discrimination.
Environmental dimension the conguration of the urban mobility system should
result in a total sum of pollution below the endurance level.
Economic dimension the system should offer good value for money, induce an
adaptive behavior from the users, and be able to create new nancial resources to
support investment.
Social dimension the system should assure that citizens are provided with a
transport system adequate to their needs and that no exclusion through price, or
any other criteria, will be imposed on basis of economic or nancial goals.
Trade-offs among these four domains are conditioned by the socioeconomic

and cultural reality of each specic environment (urban area), and also by the
political options that result from the interaction between the local, regional, and
national levels of intervention. It is thus a function of the strategic level to assure
a denition of objectives that provides an adequate answer to the UMS
stakeholders requirements, considering the existence of two categories of interests:
the ones represented by the personalities of individual stakeholders, and the ones
of the society, usually best described through the public service and welfare
concepts.
Introduction 23
European policies aiming at an improved efciency and effectiveness of the

transport system are carefully pushing the revision of the regulatory and organizational
settings of this sector (revision of regulation 1191/69, proposed in July 2000 and
subsequently strongly conditioned by the Altmark case law, judged in July 2003),
through the introduction of competitive pressure in contract awarding, whereas
challenging old established monopolies to re-organize themselves.
In parallel, some of the main agents of this process (authorities and operators)
decided to start their changing process focusing on the regulatory and organizational
situation in their cities (EC, 2001a, p. 75) while others, following the general man-
agement trend, decided to become global and taking advantage of a large experience
initiated operations in several cities in Europe and beyond. However, this change
process is being hindered by the conict of interests raised by the confrontation
between the original statutory interests and mission of the institutions involved and
challenged by the process of change, and by the aims and ambitions of individual
decision-makers running those institutions.
In addition, the more numerous are the levels of government and diverse the
agents involved the higher are the coordinating costs, as well as the complexity of the
task, to assure consistency and coherence of action. Trade-offs will have thus to be
considered between dimension and diversities of the organizations involved in the
system. These interactive mechanisms affect both organizations (or agents) and
urban policies, and it is this dynamic characteristic that enables system exibility to
adapt to new demand patterns, creativity to conceive new services in order to
increase patronage and, consequently, organizational change to adapt agents to the
changing environment, whereas keeping congruence and consistency in their working
methods, processes, and organizational models.
Moreover, irrespective of the regulatory and organizational form, all urban
mobility systems need to have their quality performance monitored. However, if the
aim is to assess quality performance then we must gain awareness that in UMS, like in
any open system, the essential managerial characteristic is in the interaction of its parts.
Consequently, the individual improvement in the performance of its components taken
separately, although necessary and even positive, does not assure the overall
improvement of its performance.
An essential factor for system performance is how well the different parts of the
system t together, which is directly related with the interaction between the main
enablers and processes identied in the urban mobility system. That is, control through
co-ordination24 is required in all decision levels, based on existing interlinkages and
their impact in the way trans-organizational processes (i.e., processes managed across
different intervening organizations or agents) have to be managed both in a steady-
state regime and under a changing contextual environment.
24. Following our denition made at the outset of this work (see p. 17), co-ordination should be
understood as concerted decision-making, since no hierarchical dependencies exist between interacting
organizations.
1.3. Quality as a driver to enhance performance of UMS
For several decades quality concerns are becoming widespread across the economy,
with several important developments: from product quality control through a shift
of focus to quality of the process. A major breakthrough has been the transfer of
emphasis from a reactive to a pro-active approach, making it possible to bring service
industries to work with quality frames.
The next step led to a concept of quality concentrated in the human factor,
involving the entire organization with a shift of responsibility from a specic
department to the overall engagement of the company. Quality became a manage-
ment philosophy and the concept of Total Quality Management (TQM) was born.
Stahl, among others, dened it as a systems approach to management that aims to
continuously increase value to customers by designing and continuously improving
organizational processes and systems (Stahl, 1995, p. 35).
Increased acceptance and use of TQM principles became more intense and this
was mainly due to the following factors:
increasing domestic and global competition;

increasing complexity in the set of stakeholders interests; and
a need to achieve sustainable commitment of all organizational levels, to improve
processes, and achieve better use of available resources and reduce waste.
Figure 1.5 illustrates the evolution from quality control toward a TQM approach
in organizations.
Implementation of TQM is based on quality management models, such as ISO
(International Standard Organization) and EFQM (European Foundation for
Quality Management) models, the best known in Europe. However, the ISO-model
crowned by certication is often looked at as a system killing creativity and even
leading to bureaucracy. On the contrary, the EFQM model is based on very
important management values such as customer and staff satisfaction, and so far is
considered the most exible model for adaptation to several sectors. It allows
pointing out the weak and strong points of the organization providing a starter for
continuous improvement. Other frequently used models, in other parts of the world,
include the Malcolm Baldridge and the Deming (Prize) model, both of which are also
strongly based in sound consolidated management principles and theories.
Certication became more and more interesting and important to manufacturers
and service providers in nearly all kinds of business, leading to a tremendous increase
of the number of certied companies. However, this widespread dissemination and
implementation of certication processes based on the ISO norms does not mean that
the management principles underlying these norms were equally disseminated or
applied.
Certication methodologies gained prominence as tools that enforce the internal
review of processes, which in turn are a rst step to the implementation of TQM,
since it enables the organizations to identify their critical areas for improvement. It is
worth saying that a major pitfall of ISO processes lies in the fact that by wide
Introduction 25
Figure 1.5: Evolution of quality concept. Source: Adapted from EFQM, Economic
Aspects of Quality, p. 7, and QUATTRO.
spreading the need for quality certication this ended up by being understood by the
markets as a major marketing asset, instead of a sort of lesson-drawing tool for
management.
Only 25 years ago quality was nearly a nonissue for the public transport sector, in
Europe and elsewhere, but it became rather important at the end of the 1980s and it
is mostly in the beginning of the 1990s that quality management systems in public
transport have become popular. The reasons for this may vary from country to
country, but what follows can be found among the most common arguments to
undertake that management approach in this sector:
general trend toward more quality in the economy;

reduction of public subsidies with the consequent need to improve efciency and
increase patronage; and
introduction of competition in certain countries.
However, there is no model especially developed for this sector and, as already
referred, only the most innovative transport companies are implementing these
systems. This seems to be about to change, especially in countries where national
programs toward quality are being launched, and in which incentives (nancial or
training) are given to implement and/or certify quality management systems. Even the
European Committee for Normalization (CEN) is addressing the problem and decided
to set a working group (CEN TC 320 WG 5 Public Transport and Intermodality) to

deal with public services in general and public transport in particular.
Despite the general awareness of how important this could be for the sector the fact
is that no signicant development was achieved on these grounds. No harmonized
approach to public transport management between countries was achieved so far and
quality remains a rather vague concept in the sector. Although increasingly mentioned,
authorities have also failed to include it in service specication thus hindering its
potential to become an objective feature of the transport system. Most companies
willing to follow the approach are still faced with the absence of clear guidelines for
its application to public transport and have to do it according to their own
understanding on how to translate industry-born norms into the specic constraints
and peculiarities of transport services provided in urban areas.
Different background conditions contribute also for nonhomogeneous develop-
ments of quality approach all over Europe, challenging transferability attributes or,
as referred in (Rose, 2001, p. 5) the fungibility25 of quality practices. In general,
Scandinavia, Germany, France, and Great Britain seem to be leading, but there are
also remarkable achievements in some cities of the south, placing them among the
best in Europe. In other continents, even if public transport is generally less
structured than in Europe, there are also interesting developments in New Zealand,
Australia, Brazil, Japan, and Hong Kong, certainly presenting leading features in
some of their systems, in particular in the development of customer satisfaction
analysis and denition of quality concepts specic to the sector (Hensher & Prioni,
2002, pp. 97100).
There is no evidence to support the argument that certication should be
considered as a necessary condition to provide public transport services with high
quality. In fact, there are just a few European public transport companies certied
according to the ISO 9000 standards and even for the ones that are certied, this is
often only for specic departments of the company, such as maintenance (EC, OGM,
1998a, p. 57). The emphasis these standards have had in the harmonization of
outputs was certainly one of the major drawbacks of their mistness in a sector
where production and consumption occur simultaneously. This situation changed
substantially with the introduction of the new package ISO 2000, where the emphasis
in processes was reinforced.
However, a fact worth being referred is the existence of several operators
complying with the ISO standards without applying for certication (EC, OGM,
1998a, p. 66), which seems to reinforce the idea that the main added value of the
certication process is to act as a lesson-drawing tool by enforcing the internal
review of processes and attitudes for the implementation of TQM approaches, and not
for the certication itself. The learning process will certainly lead to a clarication of
25. As referred in Rose (2001, p. 5) fungibility, at the opposite of total blockage, of a policy as its ability
to travel from one context to another. We will use the concept with an explicit emphasis in the capacity
to overcome barriers to implementation of any quality policy and pursue spreading objectives.
Introduction 27
who is responsible, what is to be done, how and when to do it, why is it relevant for the
several nested processes existing in the critical structure of any service provision.
According to Juran (1998, p. 2.1), the main basic requirements for the
implementation of TQM systems are:
implementation of strategic quality management, including market segment

differentiation based on customer expectations;
communication of a culture of quality through the organization, implying the
involvement of everyone at all levels, and in special those at the lower end of the
organization, where contact with the customer normally takes place;
translation of the technical requirements into process specications; and
proposing and testing process improvements focused on customer expectations.
Quality management implies the increased focus on individual customer satisfac-

tion, which can only be achieved with the fragmentation of the global market into
homogeneous segments, receiving differentiated services, so that customer require-
ments can be met. The identication of customer requirements is thus the starting point
of a process, which requires the translation of these requirements into the organization
methods and processes. The main objective is to achieve customer satisfaction and
subsequent loyalty to the service(s) provided. The perception of this contextual reality
calls for a deep change in the organizational culture, through which the customer
becomes a borderline sensor of the company performance. Figure 1.6 illustrates the
dimension and impact of this change.
Public transport has specic characteristics that make quality implementation
even more difcult than in other sectors: rstly, as all other services, production, and
consumption occur simultaneously and the service is intangible but, on top of it, the
production of service by several transport modes is made in an environment (the
Figure 1.6: Customer as borderline line sensor of company performance.

Source: Adapted from EFQM (1994, p. 15).
public road space) in which multiple agents interfere with the transport operator
without any kind of subordination to its goals and very often entering into conict.
Moreover, because of its need to provide service to the society (i.e., collective of
citizens), it is almost inevitable that product segmentation leads to signicant
increases in unit costs. This was not a problem when most potential users had no
alternative, as the productive organization envisaging large demand was effectively
corresponded. However, this is no longer the case in many industrialized countries,
and even in developed countries the offer of possible alternatives to public transport
is increasing steadily, both through nancial solutions and through a very careful
adjustment to market segmentation undertaken by car industry, accompanied by
new freelance entrepreneurs offering personalized public services at low price.
These characteristics of public transport services have clearly contributed to a
growing gap between the stated political goals of the mobility system26 and the results
achieved at its operational level, thereby leading to a degradation of the terms of
competition of public transport versus the private car, whereas providing awareness on
the much more deeper nature of the problem, going beyond the traditional dichotomy
between public transport and private car.
Depending on the strategic aims dened for the mobility system, and the different
stakeholders interests, a different mix of quality aspects will result. These aspects
should be transferred into sets of measurable criteria to be used in the selection and
contracting of the operators, to assure a coherent relation between the strategic aims
and the performance of the system, as illustrated in Figure 1.7. This coherence has to
be assured through a continuous ne-tuning of the criteria used that should be in line
with the quality specication of the different services.
Most quality programs, despite referring to TQM methodologies, put their
emphasis only in the consumers perception at the end point of the production chain,
without addressing the problem of interaction between the different parts of the
system lying behind the operational output. This gap represents a constraint in the
potential for improvement of the system, once the decisions taken at both strategic
and tactical levels of decision are equally important, and very often critical, for the
quality of the nal set of services provided.
TQM methodologies developed so far are designed to apply to organizations,
dealing with it as a system, that is bringing to the organization the systemic approach.
As already said, UPT companies nowadays are putting signicant efforts in quality at
the operational level, mainly through the enforcement of those TQM programs.
However, there is little sense in these efforts if we forget the impacts accruing from the
denition of the strategic objectives of the mobility policy and from interaction with the
different services in the quality of the service, as perceived by the customer.
An UMS goes far beyond the provision of public transport and should entail all
services, infrastructure, and trafc management that in its whole enable citizens to
26. In this work mobility system is understood in its widest scope as the set of means and modes able to
provide citizens mobility.
Introduction 29
Figure 1.7: Conceptual dynamics for a quality approach to UMS.

Source: Viegas and Macario (1998a, 1998c, 1998d, 1998e, 1998f, 1998g, pp. 112119).
satisfy their mobility needs. Considering this wide scope denition, for an Urban
Mobility System to achieve a quality level that provides an appropriate and effective
answer to the needs of its clients (individual and collective, i.e. the society) three
different, yet complementary, quality dimensions have to be addressed:
Quality of the services provided.

Quality of the different organizations involved in the denition and production of
the services for which certication processes and TQM approach are within the
most used tools so far.
Quality of the overall UMS, where a number of policy and institutional integration
aspects have to be considered, to contemplate stakeholders perspectives and
society in general.
This research work deals with the last dimension by developing a management
model for urban mobility systems, where institutional and policy interactions are
considered, and performance outcomes assessed in line with the strategically dened
objectives. That is, quality is taken as the driver of a symbiotic process where a
concept of co-ordination without clear hierarchies is developed.
The objective of this work is thus to present a quality management model for
urban mobility systems that is congured by the following characteristics which, as
we will be able to proof in the course of the text, respect the previously referred
conditions of theoretical and empirical systems:
The focus of analysis (urban mobility system) is highly complex due to the number
and diversity of interacting components, nested processes, and institutional levels
of interventions.
Earlier digest of systematic knowledge about the holistic perspective of dynamics
of urban mobility systems is still rather limited.
Comparability between urban cases revealed to be rather low, given local
conditions and determined inuence of contextual variables.
There is a high degree of multidisciplinary interaction within urban mobility
systems and between this and other subsystems of urban context.
Key variables are not possible to experiment under controlled conditions, and
simulation of causeeffect relations within urban mobility dynamics has been
achieved only to a limited extent between some pairs of variables but is still far
from being reached for the whole system.
There is no experience on application of quality management methods to systems
of such complexity under a single hierarchical27 structure. All the available
knowledge is focused on its application to organizations.
Consequently, case studies offer a narrow and limited base for generalization.
We partially support the arguments listed by House, in particular in what concerns

the difculty to undertake transversal analysis between cases and even longitudinal
comparisons within the same case, as well as in the difculties to assess effects of a
specic policy implementation due to the time gap required for incubation or
maturation processes. However, experimental methods could have never been used
for the study of managerial problems at the system level due to difculties in
reproducing the real-world dynamics. As Ashby (1968, p. 95) formulates the real-
world gives the subset of what is and the product space represents the uncertainty of
the observer in its interpretation function, therefore the product space may change if
the observer changes, and the constraint is the relation between the observer and the
phenomenon being observed, so a substantial part of the theory of organization is
concerned with properties that are not intrinsic to the phenomenon but are relational
between observer and the phenomenon.
Indeed, the utility of experimental method for the study of managerial problems
had been questioned for a long time, even for research undertaken at the
27. We adopt Simon (1999, p. 185) concept of hierarchical system, which includes systems in which there is
no relation of subordination among subsystems. As Simon states By hierarchy I mean the partitioning in
conjunction with the relations that hold among the parts. That is much more wider than the etymological
meaning of the word that generally refers to a complex system in which each of the subsystems is
subordinated by an authority relation to the system it belongs to.
Introduction 31
organizational level (Weick, 1965, in Cummings and Harnett, 1968, p. 44), as the
method yields results that cannot be generalized beyond the conditions surrounding
the specic study in question. Therefore, practical knowledge should be obtained
through inductive methods based on selected case studies, despite the limitations of
these methods for transferability purposes.28
In support of the previous option, as reinforced by Yin (1994, p. 3) explanatory
case studies are in general the preferred strategy when how or why questions are at
stake, when the researcher has little control over events, and when the focus is on a
contemporary phenomenon within some real-life context. (y) The case study allows
an investigation to retain the holistic and meaningful characteristics of real life events
such as individual life cycles, organisational and managerial processes, neighbour-
hood change, international relations, and the maturation of industries.
The holistic dimension is one fundamental feature of the model developed in this
research work, but this does not deny the capacity of the experimental method to test
specic parts of this whole, as the capacity to obtain explanatory knowledge on the inter-
retroactive effects between elements of the systems, and their adequacy, lies on our
capacity to reduce this complex whole into more basic and elementary logic relations.
We started by building a virtuous management model for the UMS based on the
eld observation of several real cases, which enabled to establish the relational logical
links between the elements of the UMS. The observation cases were selected among
those analyzed within the following European projects29:
ISOTOPE Improved structure and Organisation of urban transport

operations of passengers in Europe, 4th RTD Program, Urban Transport,
EC-DGVII, 1995/ 1997 (EC, TIS, 1997b).
QUATTRO Quality Approach in Tendering/Contracting Urban Public
Transport Operations 4th RTD Program, Urban Transport, EC-DGVII, 1996/
1998 (EC, OGM, 1998a).
Study on Examination of community law relating to the public service
obligations and contracts in the eld of inland passenger transport (usually
known as PSO), Commissioned by the European Commission, DGVII, 1997/1998,
to support the revision of the Regulation 1191/69 (EC, NEA, 1998c).
FISCUS Cost Evaluation and Financing SChemes for Urban Transport
Systems, Fourth Framework Programme, Urban Transport, 4th RTD Programme,
DGVII, European Commission, EC-DGVII, 1999/2001 (EC, TIS, 2001b).
PATS Price Acceptability of Transport Systems, Transport Research, 4th
RTD Programme, Urban Transport, DGVII, European Commission, EC-DGVII,
1999/2001 (EC, TIS, 2001c).
28. Macario R., discussion paper Evolutionary Paths in Urban Mobility Systems: Structured assessment
of barriers and tools, July 2002 in MARETOPE research project, unpublished.
29. The author participated as a member of the co-ordination team on ISOTOPE, QUATTRO, FISCUS,
and PATS and was responsible, as leader of the work group, for the survey of all European countries in
PSO.
Study for The redenition of the public transport regulatory and organizational
framework in Ireland, Commissioned by the Transport Ministry of Ireland in
association with the Irish Transport Forum, 2001.
Study on Integration and Regulatory Structures in Public Transport (usually
known as Integration), Commissioned by the European Commission,
DGTREN, 2003.
The cases presented in Table 1.3 constitute a subsample of the empirical basis used
for the construction of the model based on the inductive method, which has been
done in a cumulative way as the number of cities observed was enlarged along the
surveys done in the different research and advice studies that support this research
work (annex 2). Simultaneously to the enlargement of the geographical knowledge
occurred also the thematic enlargement, since each research project encompassed one
specic theme of urban mobility.
The model was thus built in an incremental way with each city providing the author
with additional knowledge for its construction and simultaneously offering a new (or
renewed) scenario to test the development made until that moment. In several cases the
same city was analyzed more than once either for reasons of observation along the time
or for reasons of change in the perspective of analysis. This incremental process is
illustrated in Figure 1.8. In rigor, after the rst survey of state of the art, the
methodological combination presented in Figure 1.9, with the inductive and deductive
method, occurred simultaneously for several of the observed cities. The research work
that supported this work started in 2000 and the projects done until that moment
constituted its departure base.
Secondly, we undertook a transversal multidisciplinary theoretical approach, based
on the following building blocks and meant to assure consistency and tautological
verication of the management model:
System management theories.

Denition of the structural components of an urban mobility system.
Legal, regulatory, and organizational framework for urban transport.
Decision-making processes in local policy denition.
Analysis of quality elements and inuencing factors in urban mobility systems.
Identication of barriers to the implementation of total quality management
methods in the different regulatory and organizational frameworks.
Finally, a validation method based on deductive logic and largely built over the
falsiability hypothesis suggested by Popper (2002, p. 57) was applied. For the
falsiability tests we have again used real cases largely developed within the
framework of the following European research projects30:
30. The author participated in these research projects as a member of the leading team, in particular in
MARETOPE where the overall scientic co-ordination was under our responsibility.
Introduction 33
Table 1.3: Subsample of cities observed for development of inductive method.
Country Cities Country Cities
Australia Adelaide Italy Rome, Milan,

Bolzano
Austria Vienna, Graz Luxembourg Luxembourg
Belgium Brussels, Charleroi, The Netherlands Amsterdam, Arnhem,
Ghent, Bruges Maastricht
Brazil S. Paulo, Curitiba, New Zealand Wellington, Auckland
Belo Horizonte,
Recife
Canada Toronto Norway Oslo
Denmark Copenhagen, Aarhus Portugal Lisbon, Porto,
Coimbra, Setubal
France Paris, Lyon, South Africa Cape Town
Marseille,
Toulouse, Nancy,
Dijon, La Rochelle
Finland Helsinki Spain Madrid, Barcelona,
Valencia, Bilbao,
Germany Berlin, Hamburg, Sweden Stockholm,
Munich, Koln, Goteborg, Malmo
Bern, Hannover,
Karlsruhe,
Dusseldorf,
Dresden
Greece Athens, Thessalonica United Kingdom London, Bristol,
Leeds, Manchester,
Liverpool,
Leicester, Preston
Ireland Dublin, Cork, Belfast United States Chicago
MARETOPE Managing and Assessing Regulatory Evolution in urban

Transport Operations in Europe, 5th RTD Programme, Urban Transport,
EC-DGTREN, 2000/2003;
TRANSPLUS Transport Planning, Land Use and Sustainability, 5th RTD
Programme, City of Tomorrow, EC-DGTREN, 2000/2003;
CUPID Co-ordinating Urban Pricing Integrated Demonstrations, 5th RTD
Programme, Urban Transport, EC-DGTREN, 2000/2003 (accompanying measure
of the demonstration project PROGRESS); and
METEOR Monitoring and Evaluation of Transport and Energy Oriented
Radical Strategies for clean urban transport, 5th RTD Programme, Urban
Transport, EC-DGTREN, 2002/2006 (estimated) (accompanying measure of
Figure 1.8: Research and Advice projects relevant for this work. Source: Author.
CIVITAS program (http://www.civitas-initiative.org, 14-07-2005), that involves the

following demonstration projects: MIRACLES (http://www.miraclesproject.org,
14-07-2005), TRENDSETTER (http://www.trendsetter-europe.org, 14-07-2005),
VIVALDI (http://www.vivaldiproject.org, 14-07-2005), TELLUS (http://www.
tellus-cities.net, 14-07-2005)).
Theoretical validation of the model with the cities (annex 3) indicated in Table 1.4
proved the falsiability test, according to which the soundness of the model will be
contingent upon feasibility failure. The sample of cities selected for this validity test,
besides following the criteria of availability of information through the participation
in the European research projects, also respects the objective of diversity toward the
following attributes: dimension in population, variety of mobility modes, regulatory
Introduction 35
Figure 1.9: Methodological combination used in this research work. Source: Author.
regime for market access, interpretation of public service concept, socioeconomic

role in world urban networks,31 and variety of cultural backgrounds.
Moreover, the majority of the subset of European cities have been observed in the
period 19951998, and were included in the set of cities for application of the
inductive method. During the period 19982003 these cities have been subject to a
change either in the institutional conguration of their mobility system or in the legal
and regulatory framework for the provision of mobility services, which has been the
reason why they are again selected for the sample supporting the application of the
deductive method.
31. The so-called global, regional, and local cities concept dened according to their importance as
interacting nodes in the socioeconomic worlds.
Table 1.4: Cities used for theoretical validation in this research work.
Country Cities Country Cities
Australia Adelaide Ireland Dublin

Canada Toronto The Netherlands KAN region
Denmark Copenhagen Norway Oslo
Finland Helsinki Spain Barcelona
France Paris Sweden Stockholm
Germany Hannover, United Kingdom London
Greece Athens United Kingdom Bristol
Italy Rome United States Chicago
In addition to this theoretical validation two empirical works for application of

this model have been done, and are still on-going. In the metropolitan areas of
Lisbon and Porto, where due to a change of government the implementation work
was suspended. This work lasted the whole year of 2004 and entailed:
The diagnostic of the situation in the metropolitan area, considering the legal,
regulatory, organizational, economic, and nancial aspects.
The conceptual adaptation of the model to the reality of the two metropolitan
areas.
The development of a project law (assisted by a legal expert contracted by the
Metropolitan Authorities) incorporating the adjustment of the model to public
transport and considering the limitations imposed by the delimitation of power of
the Metropolitan Transport Authorities, which was already dened and
established at the time we have started the application.
Preparation of guidelines for public transport tendering and contractual relation
between authorities and operators.
Conception of a nancial solutions for the reform of the economic and nancial
reform of urban mobility sector (excluding the internal reform of operating
companies).
Creation of and harmonized accountancy norm for public transport operators to
enable monitoring of productive efciency and comparison between operators.
Denition of quality criteria for development of the public transport network.
Conception of an information system to support management of the whole urban
mobility system in line with the model proposed in this work.
Communication plan for discussion of the model with the main stakeholders.
The participatory process was left in the middle, but the model (together with
all material referred above) was discussed and approved by the following
entities:
Secretary of Transports of the Ministry of Public Works Transport and
Communication, and the technical body of this governmental unit;
Introduction 37
the two existing regulators for land transport (in 2005), the INTF Instituto
Nacional de Transporte Ferroviario (Rail Regulator) and the DGTT Direcc- ao
dos Transportes Terrestres (Land Transport Directorate); and
All operators of public transport of the Lisbon Metropolitan Area (in 2005).
A second case, providing a much wider experience of application of this model, is

being done by the author for the whole country of Brazil, under the authority of the
Federal Ministry of Cities, Secretary of Urban Mobility, and Transport. This work
started in November 2004 and the model herewith proposed was submitted to public
hearings and to the approval of Federal and State governments, as well as to the nal
approval of the parliament, embedded into a National Directive for the Reform of
the Urban Mobility Systems (annex 4). Main cities and metropolitan areas have
been formally engaged in the discussion, and the author undertook a total of 21
participatory discussions with selected groups of stakeholders (e.g., Council of
Cities, Operators, Public Transport Industry, Interministerial groups from Public
Administration, Academics, Representatives from political opponent parties, Unions,
Financial Institutions (BNDES, World Bank), Environmentalist, Consumers Associa-
tions, etc.) and 17 interviews with political key informants. For the implementation of
the model in the 27 Brazilian states, and given the fact that Brazil has a municipal
co-operative Federation the following stages were designed to achieve the complete
implementation until the end of the year 2005, as per the objectives dened by the
Ministry of Cities:
Production and discussion in several public sessions of a diagnostic of the current

situation in Brazil to achieve consensus over the current problem of urban mobility
systems.
Presentation and discussion by the author of the model proposed in this work in
several stakeholders sessions and private interviews to achieve consensus on the
tness of the proposed model to the Brazilian reality.
Drafting of the Directive (assisted by a legal expert from the Brazilian national
parliament) and respective presentation and public hearings in the Council of
Cities and in Regional sessions in all States.
Conception of a nancial program to stimulate faster adherence of municipalities
to the implementation process.
A capacitating program (pilot experience started in July 2005 with the main 11
cities that are capital of States plus the Federal District of Brasilia (Belo Horizonte,
Campinas, Curitiba, Fortaleza, Goiania, Manaus, Porto Alegre, Recife, Rio de
Janeiro, Salvador, and S. Paulo) to support Municipalities and Metropolitan
Regions in the implementation of the proposed model.
Some years are needed after implementation before any results and conclusion are
available regarding: causal effects among the elements of the model itself; on the
good tness for the various urban environments where it will be applied; and, on its
contribution to the improvement of urban mobility systems. At this stage we can
only consider theoretical validation and the reaction of practitioners (i.e., politicians,
decision-makers, city planners and managers, operators, users, etc.) regarding their
belief in the rationale supporting the proposed model, their understanding whether
it is correct or not and, consequently, their expectation on benets and disbenets it
can bring to their cities.
The results of the research that support this work are delivered in six chapters, of
which the current introductory chapter aims to present the set of domains involved in
the work, highlighting the main constraints and methodological options taken.
Chapter 2 is dedicated to the problematique of system management, covering from
the theoretical components related to general systems theories to the more practical
reality of the multi-institutional conguration of the urban mobility system and the
very specic type of co-ordination between its agents, where no vertical hierarchical
relations exist. In Chapter 3 the borders of the urban mobility system are dened as
well as its main structural components, such as several types of agents, setting of
political priorities and contractual relationships. Chapter 4 aims to identify the
quality elements and factors of an urban mobility system. In Chapter 5 the proposed
model is presented, and nally Chapter 6 draws conclusions for the near future and
provides suggestions for future research in systems management and decision-making
applied to urban mobility systems.
Chapter 2
A Theoretical Framework for the Management

of Urban Mobility Systems
2.1. General Concepts on System and Systems Dynamics
In the rst quarter of the last century the emergence of the system as a key concept in
scientic research could be observed, despite the fact that systems had been studied for
several centuries, with the oldest traceable references, as natural philosophy, being
found in Leibniz and Nicolas de Cusa (Bertalanffy 1968, p. 11). This new approach
was based on the study and observation of systems as an entity rather than as a
conglomeration of parts, which was consistent with the science evolutionary stage of
avoiding mechanical isolation of phenomena in narrowly conned contexts, accepting
instead open interactions for analysis of larger pieces of reality.
This trend toward more holistic methods of analysis, with mutual dependence of
relevant variables and, to a certain extent, in opposition to the specialized approach
of addressing scientic domains as an isolated matter, found application space in
many sciences, in particular in the ones where living entities interact and progressive
learning processes are required to keep dynamic interaction with the surrounding
environment, such as social domains, here addressed in its widest scope.1 As referred
in Ackoff (1959, pp. 145150), a practitioner of operational research and theorist of
organizational concepts, Under the banner of systems research (and its many
synonyms) we have also witnessed a convergence of many more specialized
contemporary scientic developments. (y) These research pursuits and many others
are being interwoven into a co-operative research effort involving an everwidening
spectrum of scientic and engineering disciplines. We are participating in what is
probably the most comprehensive effort to attain a synthesis of scientic knowledge
yet made.
In 1925, Lotka, seen by many as the precursor of this line of thought,2 provided
the scientic community with the basic formulation for the theory of open systems by
conceiving communities as systems while regarding individuals as a sum of cells. This
strange analogy has been justied by the scientic community with his statistician
1. We follow Bertalanffy (1968) broad denition of social domain entailing sociology, politics, economics,
etc. (p. 194).
2. Other preliminary works were reported by Bertalanffy (1968, pp. 1029), such as Kohlers physical
gestalten in 1924, but without dealing with the approach in full generality as the focus of this work was
restricted to biological and psychological phenomenon.
background highly focused on population problems, which is why the communities

were better perceived as systems. Almost in parallel with Lotka3 and Whitehead4
make statements advocating respectively in favor of an organismic conception of
biology and of organic mechanism in a very close formulation to what become
later known as a general system theory. The fact that similar ideas appeared
almost simultaneously and independently in America and Europe provides evidence
that at the time a gap in research methods was felt to address a number of life
phenomena.
In 1948, Cybernetics appeared. Cybernetics is a theory of systems control based on
communication (transfer of information) between system and environment and within
the system, and control (feedback) of the systems function in regard to environment.
In biology and other basic sciences, the cybernetic model is used to describe the formal
structure of regulatory mechanisms, e.g. by block and ow diagrams. Thus the
regulatory structure can be recognized, even when actual mechanisms remain
unknown and thus undescribed.
This theory on cybernetics, developed by Wiener, emerged together with Weavers
(1949) information theory and Von Neumann and Morgensterns (1947) game theory.
With Wieners theory the cybernetic, feedback, and information concepts where
introduced far beyond the traditional elds of technology and generalized in the
biological and social context. Despite the strong impact of Cybernetics, reported as
largely fostered by Wieners bold proclamation of the second industrial revolution,5
this theory had also its precursors in Canons concept of homeostasis and the less
well-known feedback models of physiological phenomena, developed by Richard
Wagner in the 1920s and by the Swiss Nobel prize winner W. R. Hess (Bertalanffy
1968, p. 16). Moreover, cybernetics is a particular case of systems showing self-
regulation, which is a fundamental function in systems of loose hierarchical
relationships.6
In 1948, L. Frank introducing a cybernetics conference,7 developed the following
considerations: The basic assumption of our traditions and the persistent implica-
tions of the language we use almost compel us to approach everything we study as
composed of separate, discrete parts or factors which we must try to isolate and
identify as potent causes. Hence, we derive our preoccupation with the study of the
relation of two variables. We are witnessing today a search for new approaches,
for new and more comprehensive concepts and for methods capable of dealing
3. This is highly contested in American literature. Main leading American biologists, such as Dubos (1964,
1967), Dobzhansky (1967), and Commoner (1961), address organismic biology without referring
Bertallanfys work in the early 1920s (cfr Bernal, 1957 and also recognized in Bertalanffy, 1968).
4. Whitehead A.N., Science and the Modern World, Lowell Lectures (1925), New York, The Macmillan
Company, in Bertalanffy (1968, pp. 12 and 278).
5. Kuhn (1962, p. 37) denes a scientic revolution by the appearance of new conceptual schemes or
paradigms of thought.
6. As it is the case with systems providing market initiated services.
7. About Teleological Mechanisms, in 1948 at the Academy of Science in New York.
Theoretical Framework for the Management of UMS 41
with the large wholes of organisms and personalities. y. The terms feedback,
servomechanisms, circular systems, and circular processes may be viewed as different
but equivalent expressions of much the same basic conception.
The term general system theory was rst introduced by Bertalanffy in a very
broad sense entailing three distinct, although not separable domains: systems science,
meaning the scientic exploration and theory of systems as a doctrine of principles
applied to all sciences8; systems technology, meaning hardware and software
development in modern technology; systems philosophy, meaning the development
of a new scientic paradigm of thought, which in itself divides into three parts:
systems ontology, systems epistemology, and systems value.
The system problems appear essentially as a consequence of the limitations of
analytical procedures in science since perception is a result of the interaction between
knower and known, which in turn depends on a multiplicity of factors. As
ascertained by Bertalanffy (1968, pp. 1819) Analytical procedure means that an
entity investigated be resolved into, and hence can be constituted or reconstituted
from, the parts put together, these procedures being understood both in their
material and conceptual sense (y)-principles of classical science, rst enunciated by
Galileo and Descartes-(y) Application of the analytical procedure depends on two
conditions. The rst is that interaction between parts be non-existent or weak
enough to be neglected for certain research purposes. Only under this condition can
the parts be worked out, actually, logically, and mathematically, and then be put
together. The second condition is that the relations describing the behavior of parts
be linear; only then is the condition of summativity given (y). These conditions are
not fullled in the entities called systems, i.e., consisting of parts in interaction. This
is later ratied by Rapoport (1966) and Simon (1965) both dening systems (or
organized complexity) as being characterized by strong nontrivial, i.e. nonlinear
interactions.
Indeed, a wide diversity of approaches has been developed to deal with these so-
called system problems, which go far beyond the object of this thesis but, for the sake
of completeness, we decided to briey refer them on an informational basis.
Excluding the approaches in applied system research, such as operational research,
linear and nonlinear programming and other specialized systems engineering
methods, Bertalanffy (1968) among other authors reports9 an extensive array of
approaches to investigate systems10 going from classical systems theory, applying
classical mathematics, to cybernetics, theory of control based on transfer of
8. Prompting the idea of unity of science, through which reality is seen as a hierarchy of organized
wholes and general system theory understood as the science of wholes or wholeness as Bertalanffy
(1968, preface, p. xx) pointed.
9. The analysis found are in general rather loose and lack rigor of comparisons in the sense that models and
mathematical techniques are brought together as if they constituted alternative methods in a set of similar
purposiveness.
10. Bertalanffy (1968, p. 23), detailed survey of cited alternatives could also be found in Rappaport and
Horvath (1959), Bayliss (1966), and Bernal (1957), Prigogine and Stengers (1985).
information11 and, information theory, exploring the expectation that information

can be used as a measure of organization12 and further to theory of deterministic
chaos based on the distinction between systems that are in equilibrium, system that
are near equilibrium and systems that are far from equilibrium.13 This last
paradigm of analysis sets that all systems contain subsystems that are in permanent
uctuations of variable intensity, that may even shatter the previously established
structure or organization, leading the system either to disintegration or to a new
higher level of organization. These are the so-called Prigoginian dissipative
structures, stimulated by the existence of time-dependent irreversible processes14
that constitute a main feature of the equilibrium of dynamic systems.
It is worth saying that the survey of approaches summarized earlier is rather loose,
and lacks rigor of comparison, in the sense that conceptual models and mathematical
techniques are brought together as if they constitute alternative methods in a set
with similar purpose of analysis. To a certain extent the common purpose of
this extended list of approaches is found in the challenge of understanding the
rationale behind the behavior of dynamic systems. The intensity of evolution
observed in system science during the 20th century, together with leaping of basic
thermodynamic concepts between disciplines in an attempt to look for structural
similarities in the different elds, largely justify the need to encompass, albeit at the
expense of less rigor, such a diversity in the comparative analysis of the available
alternatives.
Before analyzing the properties of a dynamic system in general, and of the UMS in
particular, we must understand the concept of system which has been approached in
different perspectives by the soft and hard system thinking streams of thought.
11. Maybe the widest scope alternative often wrongly taken as systems theory in general.
12. In the conventional Boltzmann Entropies (1965) sense order can be measured by negative entropy and
the so-called Prigogines Theorem stands for minimum entropy production as a criterion to dene steady
state in open systems, associating total entropy in a system with the reversible capacity of its processes, that
is in a system of fully reversible processes there would be no entropy and consequently no difference
between past and future. This interpretation, and further development, of the second law of
thermodynamics is consistent with Schultz (1951) ndings of nonrandom arrangements of amino acids
within a protein chain, which supported his concept of chain entropy, as reported by Bertalanffy (1968,
p. 151). However, it is worth referring that other authors follow different formulations, such as the
biophysicist Trincher (1965) who came to the conclusion that the state function entropy is not applicable to
living systems.
13. Within this spectrum the diversity of approaches developed along the years entails: computerization
and simulation, addressing nonlinear multivariables problems; compartment theory, dealing with the
critical boundary limits of decomposition of a system into subunits; set theory, axiomatical analysis in
terms of set theory; graph theory, addressing structural properties of systems; net theory, usually applied to
nervous networks; theory of abstract automata, based on simulation of processes through a nite number
of logical operations; game theory, entailing a system of antagonistic forces, where presumably rational
players undertake strategic behavior aiming to obtain maximal gain and minimum losses; decision theory,
mathematical theory concerned with choices among alternatives; queuing theory, concerning optimization
of arrangements under crowding constraints.
14. Prigogine and Kondepudi (1998) dene irreversible processes as having a dual role, destroyers of
order near equilibrium and as creators of order far from equilibrium.
As referred in Bertalanffy (1968, p. 56) A system can be dened as a set of elements

standing in interrelations. Interrelation means that elements, p, stand in relations, R, so
that the behaviour of an element p in R is different from its behaviour in another set of
relations R. If the behaviours in R and R are not different, there is no interaction, and
the elements behave independently with respect to the relations R and R.
For Simon (1969b, pp. 184186), in turn, complex systems are sets made of a large
number of parts that have many interactions and where the whole is more than the
sum of the parts in the weak but important pragmatic sense that, given the properties of
the parts and the laws of their interaction, it is not a trivial matter to infer the properties
of the whole. Furthermore he considers that the partitioning in conjunction with the
relations that hold among the parts form the hierarchy, which is the basic concept of
his thesis based on the near decomposability of the hierarchic structures15 that would
greatly simplify the understanding and analysis of the behavior of dynamic complex
systems. From the near decomposability postulate it would appear that those aspects
of the microstructure that control the slow developmental aspects of organismic
dynamics can be separated out from the aspects that control the more rapid cellular
metabolic processes (Simon, 1969b, p. 207, footnote 21 and p. 204). This fact has the
effect of separating the high-frequency dynamics of a hierarchy involving the internal
structure of the components from the low-frequency dynamics involving
interaction among components(Simon, 1969b, p. 204).
Understanding complex systems by way of a theory of hierarchy has no doubt the
strong appeal of simplication. However, the critical issue, also raised by Simon, is to
which extent interaction between two or more subsystems excludes interaction of
these subsystems with the others. This symbiotic effect is especially relevant in
systems where there are no formal relations of subordination between subsystems of
the different levels, which is the case of a UMS.
Checkland, in the soft systems approach, follows Simon hierarchic concept, and
denes system as a model of a whole entity; when applied to human activity, the
model is characterised fundamentally in terms of hierarchical structure, emergent
properties, communication and control(Checkland, 1972, p. 3).
Two major differences can be observed in these approaches: the rst is that soft
systems thinking takes the existence of a ill-dened problem as its departure point,
whereas hard systems thinking considers the existence of a well-dened problem; the
second is that the option in soft system thinking is the analysis of elements of
structure and process and their mutual relation, whereas hard system thinking draws
the analysis on the objective and its place in a hierarchy of systems. Despite these
different departure points and philosophies we believe, as already referred, the two
approaches are rather complementary for the management of real world complex
15. Simon gives the word hierarchy a wider meaning by including systems in which there is no relation of
subordination among subsystems. That is By hierarchic system, or hierarchy, I mean a system that is
composed of interrelated subsystems, each of the latter being in turn hierarchic in structure until we reach
some lowest level of elementary subsystem (Simon, 1969b, p. 185).
systems, and we will adopt a system thinking hybrid approach for the management of
UMS.
As Prigogine and Stengers (1984, p. 16) question The problem of irreversibility
still remains a subject of lively controversy (y) The world is far from homogeneous.
Therefore the question can be put in different terms: What is the specic structure of
dynamic systems that permits them to distinguish past and future? What is the
minimum complexity involved?
The answer to these questions will inevitably lead us to dene the contour lines for a
theory of change and organization of dynamic (living or open) systems, which can be
generically dened as complexes of elements standing in interaction. As formulated by
Bertalanffy (1968, pp. 3941) formulated this heuristic in a biological context and
observed that Every living organism is essentially an open system. It maintains itself
in a continuous inow and outow, a building up and breaking down of components,
never being, so long as it is alive, in a state of chemical and thermodynamic equilibrium
but maintained in a so-called steady state which is distinct from the latter. This is the
very essence of that fundamental phenomenon of life which is called metabolism, the
chemical processes within living cells.
As it happens with all heuristic formulations only mental tools to better support
the various steps of the discovery process, without providing further guidance on
where and to what extent those formulations should be used this rather generic
formulation shows structural similarities in the various scientic elds, from nature
to technological or social context, conrming that Abstractions and conceptual
models can be applied to different phenomena as long as similar key constructs and
equinality16 principles are used, contingent upon the so-called after-effects or
hereditary.17 Structure and function are two such constructs, the rst representing
the order of parts (or components) and the later the ow of processes.18
It is worth realizing at this stage that the concept of part (or component) is itself a
deep metaphor of science, as Weinberg so well observes so deep that we seldom
know that we are using it (Weinberg, 2001, p. 144), as it implicitly calls for the
denition of properties and boundaries so that the referred parts can be isolated and
gain their own identity. Based on this formulation and introducing human
16. Equinality is understood in this study as a consistency property. Following Drischel (1968)
formulations in his Formal Theory of Organisation, equinal systems are the ones which have the
property of reaching the same nal goals (or state) regardless of the initial state, input sequence or evolving
pathway. That is, consistency through equinality can be achieved even if initial state and selected
processes are different.
17. Dependence of a system on past conditions and previous course of evolution. These problems are
reported to be addressed in mathematical context by Picard E. and in an historical context by Volterra.
18. Functions as ow of processes is a concept recapped by Management Theories, after Whiteheads
(1925) philosophical approach derived in what is seen as his most important work Process and Reality
where he understands human experience as a process belonging to nature and states that no element of
nature is a permanent support for changing relations, each receives its identity from its relation with others.
In the process of its genesis each existent unies the multiplicity of the world, since it adds to this
multiplicity an extra set of relations (quoted in Prigogine and Stengers (1985, p. 95).
randomness in this compound we can easily understand the universe as being divided
into systems of organized complexity, i.e. with a hierarchic order linking its elements
(components or parts), and systems of unorganized complexity, i.e. without
possibility of establishing that order simply based on internal characteristic of the
components or typology of interactions.
According to the theory of change implied in Prigogines concept of dissipative
structures, chaos (unorganized complexity) emerges whenever uctuations force an
existing system into a far-from-equilibrium stage, threatening its structural stability
and pushing the system toward a critical point, dened as a bifurcation point, at
which it is not possible to pre-determine the next stage of organization of that system.
Chance drives the new subsystems (or what was left from previous bifurcation
moment) toward a new path of development, returning to a deterministic mode until
it reaches again another bifurcation point. Whenever the system reaches the new
bifurcation point, then the deterministic description breaks down and the process of
crossing the bifurcation develops with stochastic characteristics.
As Prigogine and Stengers (1985, pp. 177209) demonstrate chance and
determinism interchange to produce a synergetic effect in the evolution of an open
system nudging it from close-to-equilibrium to far-from-equilibrium stages, having
uctuations successive instabilities19 as the inuencing factor in the change
process. According to their postulates deterministic processes take over until a new
bifurcation occurs, but this moment is only determined by chance. Consequently,
systems evolve by joint action of deterministic and chance processes, thus revealing a
signicant synergetic effect.
A rst conclusion that can be taken from this brief recap of the evolution of
systems science is that the unifying principle of the scientic development of the past
century, largely centered on the universal principles of thermodynamics, is that the
need for organizational and integrated approaches (a consequence of the existence of
open systems) can be found at all systems levels and stages in the various scientic
streams. As one of the precursors of this rationale (Whitehead, 1925, in Buckley,
1968, p. 37) emphasized that an atom, a crystal or a molecule are organizations.
This assumption on this science unifying principle lead Alvin Toer, clearly blasted by
the potential overtone the Prigoginian model could achieve through analogical extension
into other scientic elds to state When we bring reversible time and irreversible time,
disorder and order, physics and biology, chance and necessity all into the same novel
frame, and stipulate their interrelationships, we have a grand statement arguable, no
doubt, but in this case both powerful and majestic.20
19. Bertalanffy (1968, p. 48), hypothesises that at organizational level this instability can be treated in terms
of Volterra theory, Volterras rst law being that of periodic cycles in populations of two species, one of
which feeds at the expense of the other. It is also worth referring at this stage that another analogy with
governance systems is made by several other authors, such as Douglas Yates (1977) who considered that
the political-administrative system in fragmented to the point of chaos (p. 34).
20. Alvin Toer comment in the foreword of Order out of Chaos (p. xxiii).
In a totally different context but embedded in a similar paradigm of thought

(Morin, 2001, p. 44), insists:
Let us not forget that the problem of determinism has changed over the
course of a centuryy In place of the idea of sovereign, anonymous,
permanent laws directing all things in nature [this] has been substituted
[by] the idea of law of interaction (y). There is more: the problem of
determinism has become that of the order of the universe. Order means
that there are other things besides laws: that there are constraints,
invariances, constancies, regularities in our universey In place of
homogenizing an anonymous view of the old determinism, [this] has
been substituted [by] a diversifying and evolutive view of determina-
tions.
In a reection about learning processes, Mather (1951)21 stated One of the

criticisms of general education is based upon the fact that it may easily degenerate into
the mere presentation of information picked up in as many elds of enquiry as there is
time to survey during a semester or a year (y) More important is the search for basic
concepts and underlying principles that may be valid throughout the entire body of
knowledgeyThus conceived, integrative studies would prove to be an essential part
of the quest for an understanding of reality (Bertalanffy, 1968, pp. 5051).
The more remote attempts to apply systems theory to organizational behavior date
back to Bouldings (1953) work The organisational Revolution, where he advances a
general model of organization in his so-called Iron Laws, presumably valid for all
organizations, such as the Malthusian law that the increase of a population is, in
general, always greater than that of its available resources, from which he elaborates
that the larger an organization grows the longer and more complex is the way of
communication, which is in line with Prigoginian postulates of indirect, second and nth
order of communication.22
In the organizational context Bertalanffy (1968, p. 48) refers that According to the
law of instability, many organisations are not in a stable equilibrium but show cyclic
uctuations which result from the interaction of subsystems and at an higher level
the important law of oligopoly states that, if there are competing organizations, the
instability of their relations and hence the danger of friction and conict increases with
the decrease of the number of those organisations. Thus, as long as they are relatively
21. Mather (1951), a professor, in a symposium of the Foundation for Integrated Education discussed
Integrative Studies for General Education (in Bertalanffy, 1968, pp. 5051), what is still today an issue
for discussion both in general and higher education.
22. Which allow a molecule or an organism to respond to signals which it cannot sense due to lack of
receptors, in which cases a third entity acts as a relay/converter of communication creating a new second
order channel where emitter and receptor adequately match their communication devices or codes. In this
context, Wiener (1961) stated that within any world with which we can communicate, the direction of
time is uniform. For Prigogine and Stengers (1985, pp. 295300), communication is seen as a rather
irreversible process.
small and numerous they muddle through in some way of coexistence. But if only a few
or a competing pair are left, as is the case with the colossal political blocks of the
present day, conicts become devastating to the point of mutual destruction. The key
question, not addressed by Bertalanffy, seems to be whether there exists at the whole
system level an adequate dissipative structure that can nudge back the system to a
close-to-equilibrium stage.
At this broad level of generalization there are two major approaches reecting the
two main categories of thinking about organizations. One was pioneered by the
sociologist Ferdinand Tonnies (1887) Gemeinschaft und Gesellschaft, which was
translated in 1955 by Loomis as Community and Association. Tonnies constructed
models of two types of society organizations. The natural living community into
which a person is born, the family or the tribe (Gemeinschaft) and the formally
created groupings (Gessellschaft), that a person typically joins in some contractual
sense. The orthodox view emphasizes the Gessellschaft nature of organizations as
they are conceived to do things collectively (achieve goals) that would be beyond the
reach of individuals. The alternative view stands for the interpretation that all social
groups have some Gemeinschaft principles within its organization. That is, an
organization is to some extent understood as having characteristics similar to a
family. Indeed, the detailed observation of organization provides the evidence that
most entities reveal characteristics of both models.
Despite this reality, there has been some dominance of the orthodox view reported
by some authors, such as Khandwallas (1979) concludes in a review of organization
theory, that transfer of the open systems view into the organizational world was the
most powerful orientation in organisation theory today (Khandwalla, 1979, p. 251).
Another not less important stream is the one relating the organizational structures
with their core tasks and embedding them in a context of interaction with the
surrounding environment. Despite the pioneer character of Bouldings aforemen-
tioned work Reed (1985) reports in its survey that systems theorists y had
dominated organisational analysis since the 1930s (p. 35).
Another challenging view of the orthodox stream of thought is given by Silverman
(1970), who contrasts the systems view from the 1930s with what he called action
frame of reference in which action is driven by the meanings that members of
organizations attribute to their own and each others acts. As observed by (Checkland
(1994, p. 77) this view transforms organizational life in a set of collective processes of
attribution of meaning, supporting the importance that soft system thinking gives to
the Weltanschauung concept, which was developed by Dilthey and rst published in
1931, and known as the theory of the structure of Weltansschauungen (Checkland,
1999, pp. 276279) as a structured method to understand society.
Diltheys method comprises a circular process of discovery called the
hermeneutical circle, that is a means of perceiving social wholes simultaneously
as whole and parts. He assumes an iterative cycle with no xed or absolute starting
points that gradually leads to an increased understanding of social reality. For
Diltheys a Weltanschauung is a whole compounded of three elements: our cognitive
representation of the world, our evaluation of life, and our ideals concerning the
conduct of life. Based on this structure and in the philosophical interpretations of the
concept he develops the idea that there are three basic world images: naturalism,
embodying the view that man, relying on his senses to understand nature, adopts a
goal of manipulating nature for his own ends; subjective idealism, in which the
holder projects moral ideals that transcend given reality; and objective idealism in
which the world is neither manipulated scientically nor dominated by the assertion
of moral will, but is viewed as an object to be appreciated aesthetically the world as
a universal harmony (Checkland, 1999, p. 276).
After World War II arduous efforts have been made to capitalize knowledge
acquired during wartime, built on operational research and logistics, into industrial
companies, government agencies and civil supporting systems in general. This was
largely the main lever for the boost of systems thinking. Systems were then dened
as goal seeking and cybernetics was put at the service of system control. This trend
was strongly aligned with the aforementioned orthodox view on organization theory,
and so management was conceptualized as being problem-solving and decision-
taking in pursuit of predened goals.
This paradigm is further elaborated by Ackoff (1957, pp. 214218) by stating that
problems ultimately reduce to the evaluation of the efciency of alternative means for
a designated set of objectives. Checkland and Scholes (1999), in line with this
problem-solving orientation, establish a fundamental difference between hard and
soft systems that remains as one of the pillars of Checklands soft system thinking
approach: What we think of as engineering begins when a need is established; and
the engineers task is to provide something which meets the need, whether in the form
of a physical object or a procedure or both. (y) Systems engineering looks at How
to do it when what to do is already dened (y) Being a manager, y, is to be a
decider of what to do as well as how to do it. This means that naming a system to
meet a need and dening its objectives precisely the starting point of systems
engineering is the occasional special case (p. 17).
This was also the eld where Simon (1960) made remarkable contributions through
his work on The New Science of Management Decision (1960), where he developed
a science of administrative behavior and executive decisions aiming, as Zannetos
(1984) observed, to provide a theory of problem solving, programs and processes for
developing intelligent machines, and approaches to the design of organisational
structures for managing complexity. Together with the orthodox organization
theory, Simons theory forms the main stream of hard systems thinking of the 1950s
and 1960s, although Simon slightly23 abandoned the optimization paradigm and
replaced it by a more dynamic concept of satisfaction, meaning that the main idea of
managers and administrators should be to search for solutions that are good enough,
given the perceived circumstances and constraints.
23. We consider that the optimization paradigm has not been fully abandoned by managers and
administrators as suggested by Simon, instead there was a shift of emphasis for a multiperspective analysis
in the search and identication of problems, based in the existence of gaps between performance and goals,
where trade-offs in the satisfaction of several groups of stakeholders is often required.
As Checkland (1983) differentiates, hard systems thinking assumes that the world
can be organized into sets of systems that can be engineered to achieve objectives,
whereas in the I shifted to the process of inquiry into the problematic situations
instead of being assumed as a structural characteristic of the world. Vickers (1965)
development of appreciative systems complements this view and provides the
operational frame for soft systems methodology24 by introducing the diagnosis by
way of what he called an appreciative function,25 that justies intervention in human
interaction.
Vickers rejects both the optimization and the cybernetics paradigms and proposes
a self-nurturing system, driven by an interacting ow of events and ideas unfolding
through time, where appreciation, created by our ability to select and choose, leads
to intervening action that contributes itself to the ow, recursively generating the
next appreciation moment and so forth. Appreciation is limited to the selective
perception of reality and making judgments about it, which stem from standards of
both fact and value given by the accumulated history of the system, that is the
memory of previous appreciations. It is worth noting that standards are themselves
products of use and the very fact that we use them may lead to their modication.
Vickers, in a letter to Checkland, expressed the postulates behind his notion of
appreciative system:
It seems to me in retrospect that for the last twenty years I have been
contributing to the general debate the following neglected ideas:
(1) In describing human activity, institutional or personal, the goal-

seeking paradigm is inadequate. Regulatory activity in government,
management in private life consists in attaining or maintaining
desired relationships through time or in changing and eluding
undesired ones
(2) But the cybernetic paradigm is equally inadequate, because the
helmsman has a single course given from outside the system, whilst
the human regulator, personal or collective, controls a system
which generates multiple and mutually inconsistent courses. The
function of the regulator is to choose and realise one of many
possible mixes, none fully attainable. In doing so it also becomes a
major inuence in the process of generating courses.
(3) From 1 and 2 ows a body of analysis which examines the course-
generating function, distinguishes between metabolic and
functional relations, the rst being those which serve the stability
24. Checkland denies any precedence and refers that only by coincidence Soft Systems Methodologies
(SSM) emerged in an action research program at Lancaster University and only later it was discovered that
there was considerable similitude with Vickers model (Checkland, 1994, p. 86).
25. For Checkland (1985) systemicity is the capacity of being systemic and the appreciative function is
made of two main moments: perception and judgment.
of the system (e.g. budgeting to preserve solvency and liquidity), the

second being those which serve to bring the achievements of the
system into line with its multiple and changing standards of success.
This leads me to explore the nature and origin of these standards of
success and thus to distinguish between norms and standards,
usually tacit and known by the mismatch signals which they generate
in specic situations, and values, those explicit general concepts of
what is humanly good and bad which we invoke in the debate about
standards, a debate which changes both.26
The two recursive mechanisms, the intervention loop and the standard raising
loop, keep the system continually changing through a self nurturing process that
Varela (1984) called natural drift. Indeed, this natural drifting is a major cause
of managerial complexity, which is characterized by conicting appreciative settings
and norms, largely conditioned by a social system that is permanently changing the
interactions between three elements: roles, norms, and values.
Roles are the social position recognized as relevant by the other participants of the
problem situation. As Checkland and Scholes state this position can be institutionally
or behaviorally dened (Checkland, 1999, p. 49). A role is characterized by its expected
or assigned behavior, inuenced by pre-established norms (current roles will inuence
future norms, adding the third recursive mechanism not identied by Vickers nor by
Checkland to Vickers concepts), and its performance is judged against local
standards, or values. For Vickers, managing relationships is the natural reason for
human action, goal seeking is left to be only an occasional special case. This is quite
evident in his argument of differentiation of his own stance from the one made by
(Simon, 1957) in his work on Administrative Behaviour, about which (Vickers, 1965,
p. 22) coherently comments:
The most interesting differences between the classic analysis of this

book and my own seem to be the following:
(1) I adopt a more explicitly dynamic conceptual model of an organi-

sation and of relations, internal and external, of which it consists, a
model which applies equally to all its constituent sub-systems and to
the larger systems of which it is itself a part.
(2) This model enables me to represent its policy makers as
regulators, setting and resetting courses or standards, rather than
objectives, and thus in my view to simplify some of the difculties
inherent in descriptions in terms of means and ends.
(3) I lay more emphasis on the necessary mutual inconsistency of the
norms seeking realisation in every deliberation and at every level of
26. Vickers G. (personal communication) with Checkland (1974) (in Checkland, 1994, p. 80).
organisation and hence on the ubiquitous interaction of priority,

value and cost.
(4) In my psychological analysis linking judgments of fact and value
by the concept of appreciation, I stress the importance of the
underlying appreciative system in determining how situations will
be seen and valued. I therefore reject weighting (an energy
concept) as an adequate description of the way criteria are
compared and insist on the reality of a prior and equally important
process of matching (an information concept).
(5) I am particularly concerned with the reciprocal process by which
the setting of the appreciative system is itself changed by every
exercise of appreciative judgment
Despite the correctness of the comparison with Simon postulates, we cannot

agree with the rejection of the energy concept. Moreover, point 5 of Vickers
argumentation holds some inconsistency of analysis since the reciprocal process is
assured through the effects of action and not through judgment which is upstream in
the process of intervention. In addition, each intervention entails perturbation of
system equilibrium, irrespective of whether the system is referred to one organization
or to a set of organizations, and so it may simultaneously impose a transformation
process upon several agents, although the effect over these agents may be
different according to the nature and intensity of their relationship with the problem
situation.
Vickers analysis concentrates on appreciative judgment and misses the action
moment (raison detre of the appreciation function, to our judgment) the
intervention on the natural drift of the system largely ruled by the mechanics of
energy ows between agents (sources of energy in the system). Besides, the
appreciation gives place to a dual transformation process in the whole system
and in the agent carrying out the intervention, which is itself a subsystem. Figure 2.1,
inspired in Vickers concepts and Checkland respective pictorial interpretations,
illustrates these interventive mechanisms where appreciation plays a fundamental
role as a lter for selective actions on the system. This action, in turn, will condition
the future path of evolutions through the reaction of the agents being its victims or
beneciaries (feedback effect), that is the path of the system evolves as a second order
effect of the interventive mechanism.
Contrary to the cybernetics rationale, this tri-loop mechanism is embedded in
the structure of the system, the regulatory mechanism being the capacity to
control natural drifting through appreciative interventions. The repeatability of
the process provides accumulated knowledge by enabling formulation of new
hypotheses in forecasting future system evolution that, in turn, will reshape norms
and values. As Checkland puts it, while discussing the soft systems method, The
three characteristics which dene the pattern of activity are reductionism,
repeatability, and refutation. We may reduce the complexity of the variety of
the real world in experiments whose results are validated by their repeatability,
Figure 2.1: Conceptual illustration of a system intervention based on Vickers

appreciation concept. Source: Adapted from Checkland (1999, p. A-51).
and we may build knowledge by the refutation of hypotheses (Checkland, 1999,

p. 51).
Soft Systems Thinking, as dened by Checkland, is thus a learning process entailing
four major actions for problem-solving: perceiving; predicting; comparing; and
deciding on action. However, we consider the method misses two major subsequent
stages of a complete system thinking cycle, which are: fth action-interpreting
feedback signs and information; sixth action-incorporating feedback learning into
next wave of system perception. The method lies on the concept of human activity
system and meaningfulness of attributions regarding a particular image of the world,
Weltanschauung, already referred (Checkland, 1999, p. 18).
System thinking is founded over the following two pairs of concepts-emergence

and hierarchy; communication and control herewith described:
Emergence and hierarchy (Checkland, 1999, p. 314):

Emergence, Emergent Properties The principle that whole entities exhibit
properties which are meaningful only when attributed to the whole, not to its
parts. Every model of a human activity system exhibits properties as a whole
entity which derive from its component activities and their structure, but cannot
be reduced to them.
Hierarchy The principle according to which entities meaningfully treated as
wholes are built up of smaller entities which are themselves wholes y and so on.
In a hierarchy, emergent properties denote the levels.
Communication and Control (Checkland, 1999, p. 313):
Communication The transfer of information.
Control The process by means of which a whole entity retains its identity and/or
performance under changing circumstances. In the formal system model the
decision-taking process ensures that control action is taken in the light of the
systems purpose or mission and the observed level of the measure of performance.
However, and irrespective of our reserve in the universality of the transcribed

descriptions,27 these concepts may also have different operational interpretations,
depending on the nature of the system we are observing. A particular distinction lies
with multi-organizational systems, which are not addressed by Checkland, and in
which case the agents (parts or components) of the system encroach on each other
(Lindblom, 2001, pp. 2122), creating a very strong interdependency that does not
preclude the separate analysis of some effects, but leaves the more extended
comprehension of the effect to be achieved only when the whole system is studied and
the collateral effects of interdependencies are understood.
Indeed, the interdependencies28 characterize the system need for co-ordina-
tion29 on a comprehensive basis, although primary co-ordination, entailing a
direct line of command, can only be implemented if hierarchical relationships exist
between these entities. Confronted with this limitation of absence of formal
27. The concepts of hierarchy and control which are proposed by Checkland are focused in a single entity
environment and, as such, only with a very open interpretation of his denitions it is possible to consider
their application in a multi-organizational system (e.g., UMS). Besides, the denition of control used by
Checkland better ts the concept of self-control then the function of controlling the action of other entities,
which is also the case in the management of UMS.
28. Following Lindblom, we dene interdependency as the capacity of one organization to cause effects on
another one by way of its own actions, or as he, enounces Within the set, each decision-maker is in such a
relation to each other decision-maker that unless he deliberately avoids doing so (which may or may not be
possible), he interferes with or contributes to the goal achievement of each other decision-maker, either by
direct impact or through a chain of effects that reach any given decision-maker only through effects on
others (Lindblom, 2001, pp. 2122).
29. See endnote 26 of Chapter 1 for the denition of co-ordination.
hierarchy, primary co-ordination tends to fail and unorganized systems emerge,

with each organization pursuing its own selsh goals, unless reasons for
cooperation exists, in which case benets from co-ordinated actions should
overcome the costs of undertaking these efforts. In the last situation co-ordination
is then assumed has a more sophisticated steering function.
In an interactive system antagonism or even incompatible goals may exist, as well
as the risk of encouragement of irresponsibility (through misallocation of
responsibility) and redundancies in the application and exploitation of resources,
that is, as Landau designated a stage of multi-organizational sub-optimality, often
found in public sector (Landau, Chisholm, & Webber, 1980; pp. 35).
This state is especially acute whenever there is capacity of appropriation of vast
resources and there is no strong set of objectives shared by these entities.
Interdependence increases uncertainty, as causeeffect relations between the
dependent entities may not be well understood due to insufcient knowledge. Each
organization seeks to reduce or eliminate this uncertainty while operating under the
rule of rationality. This is done in several ways mergers, partnerships, government
regulation, etc. although the public sector central control (implying line of
command of the upper level entity over the lower level entity) and vertical integration
have very often been prescribed as a panacea for this typical public sector disease.
However, evidence exists in Europe and elsewhere that structural reform and
institutional conguration are to be considered in the suboptimality discussion to
cope with the low decomposability that characterizes hierarchical systems. As Simon
(1969a) said, intracomponent linkages are generally stronger than intercomponent
linkages and higher levels of interdependence usually require higher levels of
connectivity30 in the formal structures.
From the pitfalls identied we can understand that a multi-organizational system
needs to use elements of soft and hard systems thinking to dene adequate identity
and managing mechanisms. Just like chance and determinism have a synergetic effect
in the development of open systems, so it seems that also hard and soft systems
paradigms can convey to dene social systems of unorganized complexity, made of
interactive wholes, which are themselves subsystems of organized complexity,
representing a community of purpose (equinality) with functional interdependence,
that is precisely what happens in an urban mobility system.
Also Sterman (2000, p. 27) after some experimental studies, concludes that the
observed dysfunction in dynamically complex settings arises from misperceptions of
feedback. The mental models people use to guide their decisions are dynamically
decient, which in turn is caused by the nonlinear character of the feedback system
of the real world, where the decisions of any agent represent but one of the many
feedback-loops that operate in any given system (Forrester, 1958, pp. 3940).
30. We adopt the connectivity concept presented by Checkland (1999, p. 313), that is the property which
enables effects to be transmitted through the system. The connectivity may have a physical embodiment (as
in an order processing system) or may be a ow of energy information or inuence.
Indeed a good part of systems dynamics insight lies in understanding and interpreting
feedback loops. The learning function itself is made operational through the feedback
process. This has been recognized by Dewey, who described learning as an iterative
cycle of invention, observation, reection, and action (Schon, 1992, pp. 119139).
Also Deming, as one of the main precursors of Total Quality Management, developed
his PDCA cycle as the main method for an iterative learning process incorporating
feedback obtained from process control. PDCA stands for Plan (meaning, establish
the objectives and processes necessary to deliver results in accordance with
customer requirements and organizations policies), Do (meaning, implement the
processes), Check (monitor and measure processes against policies, objectives and
requirements), and Act (take action to continually improve process performance).
2.2. Urban Mobility System Components and Dynamics of Interaction
At the light of systems thinking concepts and having observed the realities of several
cities, we can identify the following characteristics as common to all Urban Mobility
Systems (UMS):
Symbiotic relations between agents: The systems behave as a whole, the changes in
every element produce effects on all the others. This wholeness characteristic
occurs not only within the urban mobility system but also between this system and
the other subsystems of the urban whole. This raises the problem of competing
and/or conicting objectives and the possible interdependence of objectives at
different decision (or planning) levels. The multi-organizational dimension of the
system constitutes by itself a constraint for truly optimal solutions, since priority
must be given to upper level objectives (system objectives, above individual
organizations) to assure system consistency.
Nonadditivity of change effects, i.e. the change of the whole does not correspond
to the sum of the change of the parts. This characteristic largely occurs as a
consequence of the previous.
The two previous aspects produce high degree of uncertainty regarding the
feedback path followed by the affected agent.
Consequently, interactions that we understand as a special case of interdepen-
dence, where the action of one entity causes the reaction (feedback) of the other,
are also affected by uncertainty in the behavior of agents.
Nonlinear rationale between causes and effects, largely due to the multiloop effects already
referred and to the synergetic character of the interactions observed within the system.
The nal product mobility results from a sort of productive chain where
several agents (authorities, operators, and users) intervene at different stages of the
mobility chain (and also at different decision levels) to pursue the nal objective
that is to access a number of urban functions. In line with this perspective we
understand urban mobility as a process-orientated system.
Consequently, most management decisions at system level given are originated and
give origin to feedback ow with the several affected agents. As production and
consumption occur simultaneously, system reformulation is largely based on

interpretation of results achieved in previous iterations. This is also one of the
reasons why a considerable number of decisions are irreversible.
Evolutive reconguration of the agents, which is a characteristic that intensies
with the degree of existing competition between operators, i.e. survival of the
ttest. System evolution leads to selection and development of some agents
whereas others are either extinguished or recongured. That can happen to
operators for reasons of competition or to authorities for reasons of political or
administrative behaviour. This often occurs embedded in a reshape of business
concepts and values and norms (specially where open markets have been
introduced), and may imply a situation of progressive segregation along time that
will lead to an increasing complexity of the whole system as the number of agents
may increase. Progressive segregation reveals as the opposite to globalization of
production chain. Both strategies can be observed in urban transport in Europe
and indeed these practices enforce the change of norms and values as referred by
Vickers in his appreciative model of intervention. This segregation effects are
based in a change pattern of organization that closely follows the principle of
punctuate equilibrium developed by Stephen Jay Gould in 1972 for biological
evolution of the species, according to which species do not gradually evolve into
new species (Sterelny, 2001, p. 68). According to Gould, adaptation is a less
relevant characteristic of an organism since the relevant attribute deals only with
adaptation to a specic environment. The degree of adaptation is largely
dependent on the regulatory background of the local environment under analysis,
with heavy regulated environments showing agents with low degree of adaptative
capacity, given the structure and weight acquired.
Effectiveness dependent on stakeholders acceptability. Success is largely dened in
terms of public acceptability of outcomes making it a trade-off-dependent system
given different time-delays in feedback cycles of the various types of intervention.
This is especially relevant when short-term effects of an intervention are compared
with its long-term effects since it is often found that long-term policies generate
worse-before-better answers and behaviors. A good illustration is the interdiction
of motorized trafc in specic urban zones for the creation of pedestrian areas.
As we have seen in the previous section, in soft systems thinking a system is

perceived as a model of a whole entity, which is characterized mainly in terms of
hierarchical structure, emergent properties, communication, and control. The
decision-making process is the procedure by which means the system organizes
itself, reacts to any disturbance, and pursues its purposes and, as Forrester (1958,
p. 66) said all decisions (including learning) take place in the context of feedback
loops (y) decisions are the result of applying a decision rule or policy to information
about the world as we perceive it. The policies are themselves conditioned by
institutional structures, organizational strategies and cultural norms. These, in turn,
are governed by our mental models.
Taking the decision-making process as the organizational departure point, as we
illustrate in Figure 2.2, we can envisage feedback as the main instrument for system
management, with complexity arising from the interaction (i.e., feedback process
itself) between elements of the systems (inner interaction) and between these elements
and the surrounding environment (outer interaction) but not from the intrinsic
characteristics of the elements themselves.
Dynamics of a complex system arise thus from two types of feedback loops: the
positive loop which is self-reinforcing, that is generating growth and development;
and negative loops which represent counteraction and opposition to evolution (or
change) in the direction previously followed. Together, they enable any system to
create processes that will seek growth and equilibrium, that is to have some degree of
self-regulation. A good illustration of these effects using the urban reality is provided
by Sterman (2000, p. 12): The more attractive a neighbourhood or city, the greater
the immigration from surrounding areas will be, increasing unemployment, housing
prices, crowding in the schools, and trafc congestion until it is no more attractive
than other places where people might live.
Figure 2.2: Interactions of decision-making process. Source: Author.

It is worth to highlight some features of the diagram in Figure 2.2 with our
interpretation of the structure of decision ows:
The rst is that just like Vickers anticipated, each decision an intervention
process leading to transformation of the system from one stage to the other
alters the overall environment, leading to an extensive chain of decisions and
triggering side-effects.
The second is that decision-making is a response to the gap between objectives set
for the system and a perceived insufcient attainment of those goals. The
distinction between real and perceived achievements is often a consequence of the
existence of multiple communication channels, which in turn result from an
institutional setting not devoid of interdependencies, interests, bias, etc. Despite
the difculty that might occur in dening precise goals and accurately measuring
system performance against those objectives, the goal-setting process is present at
all system levels and for all entities acting within and across those levels. A major
difculty is to assure consistency in the process of decoupling those objectives and
induce purposive behavior in all interacting agents. For the designer of the
management control system a key issue is of course how to identify and deal with
these multiple decision loops (positive and negative), and how to decouple
objectives (often entailing conicting objectives and the need to obtain trade-offs
from the agents holding those conicting interests), in such a way to develop
policies and procedures that will tie up all entities involved in a shared
commitment toward the strategic goals of the system.
A third feature is that the transformation of decisions into results takes place
through complex processes that involve institutional settings and channels,
organizational structures, and market relations between agents, that is, the real
world of interactions where each agent has also its own systemic dimension to
consider. Very often this structure is only apparent sometimes due to random
behaviors or, in other occasions, due to numerous sources of noise. As Roberts
(1978, p. 392) emphasizes, the cause of emergence of these structures can also be
attributed to the time delay often occurring between cause and effect, which may
lead to unanticipated results and ineffective policies.
The fourth is that, just like in corporate management, we nd also in systems
management a critical (and continuous) cycle of decisionresultsmeasurement
evaluationdecision.
Finally, it is still worth to emphasize that in reality there is no such thing as side-
effects, all are simply effects, the main difference being that the ones we have
anticipated in our conceptual mental model31 are the intended effects, whereas the
31. We adopt the systems dynamics concept of the mental model presented by Sterman (2000, pp. 1416),
as including our beliefs about the network of causes and effects that describe how a system operates, along
with the boundary of the model and the time horizon considered. For the boundaries of the system, we
follow Checkland (1999, p. 312) and consider the area within which the decision-taking process of the
system has power to make things happen, or prevent from happening.
nonexpected ones (often hindering the system) are the so-called side-effects and so
the nonintended effects.
Decisions act as ow regulators for the rate of change of (decision or planning)

levels from which the ows originate and to which they are destined. As Forrester
(1958, p. 40) envisaged:
A feedback control system exists whenever the environment causes a

decision which in turn affects the original environment y the regen-
erative process is continuous, and new results lead to new decisions
which keep the system in continuous motion. The study of feedback
system deals with the way information is used for the purpose of
control. It helps us to understand how the amount of corrective action
and the time delays in interconnected systems can lead to unstable
uctuation. Driving an automobile provides a good example : The
information and control loop extends from steering wheel, to auto, to
street, to eye, to hand, and back to steering wheel. Suppose the driver
were blindfolded and drove only by instructions from his front seat
companion. The resulting information delay and distortion would
cause erratic driving. If the blindfolded driver could get instructions
only on where he had been from a companion who could see only
through the rear window, his driving would be even more erratic.
Despite being a formulation developed for application in organizational business

context, some elements of Industrial Dynamics nd a good t when applied to
systems where no formal hierarchies exist between institutions. This is mostly due to
the fact that this philosophy, as Roberts (1978, p. 395) noticed:
recognises a common systems base in the ow structure of all social-

economic-industrial-political organisations. This perspective ties the
segmented functional aspects of formal organisations into an integrated
structure of varying rates of ow and responsively changing levels of
accumulation. The ow paths involve all facets of organizational
resources men, money, materials, orders, and capital equipment and
the information and decision-making network that links the other ows.
Our proposed approach lies on the cybernetics principles of control and

communications, which have also been extensively used by Jay W. Forrester in his
Industrial Dynamics approach, where decision and feedback processes are seen as
main controllers of the system, the rst acting on a pro-active basis and the second
on a reactive basis.
In any eld, and in particular in mobility, the dimension and diversity of elements
existing in the system, calls for the need of a structure if we are to interrelate and
interpret empirical observations. Without this structure all the data collected will only
represent a fragmented set with limited meaning. As Forrester (1968a, 1968b, pp. 13)
emphasized without structured information it is hardly possible to learn from

experience and so to transfer knowledge from past to future, since information is the
basis for the decisions that control the action stream. The main rules and postulates
supporting Forresters models of dynamic systems32 are (Forrester, 1968a, 1968b
pp. 41 to 417):
Systems are the cause of dynamic behavior. As such, the focus of analysis is the
interaction within the system. Any specied behavior must be produced within a
boundary that denes and encloses the system.
In concept a feedback system is a closed system. The dynamic behavior arises
within its internal structure.
The feedback loop is a path coupling decision, action, level, and information, with
the path returning to the decision point.
Every decision is made within a feedback loop and a decision process can be part
of more than one feedback loop.
The feedback loop is the basic structural element in systems.
A feedback loop consists of two distinctly different types of variables the levels
(states) and the rates (actions).
The levels integrate (or accumulate) the results of action in a system. The levels
create system continuity between points in time.
System dynamics theory is not devoid of criticism, despite the large spectrum of
possible application, from biology to social, economical, political, and technological
systems. Being a problem-oriented approach the key issues at the denition stage
involve setting the boundary of the system, its level of aggregation and the models
temporal horizon. The setting of the boundary includes the choice of scope33
and aggregation levels, which is in itself a major area for criticism. As an example,
the urbansuburban denitions are a typical case for unclear boundaries, and
Forrester has been strongly criticized for not including these relations in his
urban model (Ansoff & Slevin, 1968, 1968a, and also admitted in Forrester, 1969,
pp. 412).
Indeed, suburbs compete with cities and vice versa in several domains, such as
attraction of inhabitants, property value, educational quality, conditions for business
establishment, leisure offer, etc. Urban and interurban communities interact in such
an intense way that very often the mobility services generated by one go beyond the
administrative limits of the circumscription to which they belong.
Another relevant vulnerability common to all systems thinking method is strongly
related with the problem and boundary denition phase, that is, the replicability
32. Based on the feedback theory, Forrester developed three dynamic systems approaches: Urban
Dynamics, Industrial Dynamics, and World Dynamics.
33. The number of distinct spheres of interest covered by the model. Typically these include the
technological, economical, demographic, sociological, and political spheres of activity.
of the model built. The method used to identify factors and relationships which
are essential to any problem situation consists in gathering empirical and theoretical
information and using interviews with decision-makers (key informants) to identify
the rules and rationale they use to reach a decision.
The vulnerability to subjective inuences lies on the risk that the inquired
decision-makers do not correctly identify all factors, or do not fully understand the
complexity of all feedback processes affecting the systems or even, as Ansoff and
Slevin (1968, 1968a pp. 7 and 24) notice, that the rules and rationale used by key
informants may change from one set of circumstances to another or from one sector
of the economy to the other.
Naturally any model can be improved through accumulated knowledge of proven
experiences but this vulnerability remains with all system thinking models. Moreover,
a general limitation of any modeling experience is that there is no assurance that two
persons with the same quantity and quality of information and even using the same
logic principles would separately arrive at the same model formulation for a given
problem.
In the different scientic domains there is a large spectrum of system types ranging
from what is usually designated as hard systems, in an analogy to their physical
characteristics, to conceptual systems with no physical representation and explained
in terms of logic relations between their main elements and between these and the
surrounding environment.
Taking these principles and discussion as our departure base, we follow the advice
in Platt (1964, p. 351) that logical constructs are the strengths of any mathematical
formulation:
Equations and measurements are useful when and only when they are
related to proof; but proof or disproof comes rst and is in fact
strongest when it is absolutely convincing without any quantitative
measurement. Or to say it another way, you can catch phenomena in a
logical box or in a mathematical box. The logical box is coarse but
strong. The mathematical box is ne grained but imsy. The
mathematical box is a beautiful way of wrapping up a problem, but
it will not hold a phenomena unless they have been caught in a logical
box to begin with.
Hence, we consider that the root denition of a given problem34 covers the
full conceptual construction of the system, which entails problem recognition,
system conceptualization, and model representation. These three phases in
their application to quality management of UMS constitute the object of this
research.
34. We extend the original concept of Checkland (1999, p. 317).

2.3. Multi-Organization Systems
2.3.1. Network Structures and Attributes
In a market each agent (individuals and institutions) tends to pursue its own
objectives. As we have already seen, in a system with several agents acting in the same
market (e.g. urban mobility market) it is the interaction of its elements that leads to
an outcome, largely driven by the market dynamics whenever no hierarchy exists to
enable a formal coordination.
Joint management implies bringing into a relationship otherwise independent
activities or events, which presupposes a predetermined expected outcome or purpose
that typically constraint the action of agents. In addition, as Mitchell (1991, p. 105)
points out, hierarchy presupposes a stratication of authority and the following of
rules. To this respect Weber (1968, pp. 217226) states that the effectiveness of legal
authority rests on the acceptance of the validity of the following mutually
interdependent ideas:
Any legal norm may be established by agreement or by imposition on grounds of

expediency or rational values or both, with a claim of obedience35 at least by a part
of the group affected by the norms. This is usually extended to the sphere of
authority or power, which in territorial bodies is represented by the territorial area.
Every body of law should constitute a consistent system of abstract rules and
administration of law is held to consist of the application of these rules to
particular cases, which implies setting administrative processes in the rational
pursuit of the interests specied in the order governing and bound by the principles
approved by the same ordering group.
That the typical person in authority is the representative of the institution in all
actions associated with his status.
That obedience toward authority is not universal, that it occurs only within the
sphere of a determined stakeholdership domain, so the state of obedience happens
only toward a principle or law.
Obedience occurs in an impersonal order, not toward individuals.
The practical application of these interdependent ideas results in what Weber

(1968, p. 219) designated by categories of rational legal authority, being:
A continuous organization of ofcial functions bound by rules.

A specied sphere of competence. This involves sphere of obligation to perform
functions; provision of the incumbent with necessary authority to carry out these
35. Field G., in Monteiro (2003, p. 609) also states in his postulates of political theory that in any political
situation the key issue is not the identication of who will have the power but of who will obey, making
thus obedience as the main factor for the exercise of power in social communities.
functions; and means of compulsion clearly dened and their use subject to explicit
conditions.
Organization of ofces36 following a hierarchy of principles, that each lower
ofce is under the control and supervision of a higher one, with conduct regulated
by technical rules or norms.
A complete separation from administrative function and ownership of means, and
a complete absence of appropriation of positions by incumbents.
Webers characteristics reect the mechanism of the bureaucratic organizations.

However, most human activities are performed by different social agents, operating
with different purposes and objectives, forming overlapping systems with relatively
precise boundaries where no hierarchical characteristics exist neither within nor
across the systems. As Castells points out, modern society is characterized by
networks: Most people in our advanced societies, but also in others, are building
their projects as individuals, in the family, in the economy, in everything. Even in the
economy, people train themselves with the idea of having individual portfolios,
which you can negotiate with different people. So we are in a world of individuals
and the Internet actually is very good for that, because rather than creating virtual
communities that practically dont exist, what exists is networks of individuals which
provides the basis for increasing, not decreasing, our sociability, but our sociability
as individuals.37
This networkwise view of the world is becoming more and more common and this is
not independent from a general movement of regulation toward market stimulation
that provides each agent with much more freedom to act following its own interests and
advantages, within a space bounded by a set of rules and principles for the functioning
of the markets and for actors behavior. Moreover, it is also rather consensual that
networks tend to spread through technology, but the key question is whether there is an
underlying architecture based on common organizational principles for the structure
and properties of complex networks equivalent to our understanding of the functioning
of hierarchical systems.
In a recent research on development of network systems (Barabasi & Bonabeau,
2003, pp. 5259) observed the structure and evolutionary behavior of a set of
diversied networks entailing among others: cellular metabolism, intern, research
collaboration, world wide web, transport, etc., which revealed that some networks,
the so-called scale-free networks, contain nodes (hubs) with a high number of links
and other with just a few connections. This research conrmed that the characteristic
of these networks is that the node linkage follows a power law. According to this
distribution a network with a low number of nodes would tend to have an unlimited
number of links.
36. Weber refers to ofces in the sense of administrative organs or institutions.

37. Castells Interview in Berkeley: Conversations with History; Institute of International Studies, UC
Berkeley, 9 May.
This same research also reveals that the scale-free networks grow incrementally
and a process of preferential attachment exists, that is the choice where to establish
a link is usually conditioned by the number of connections each available node
already has. These two processes-growing and attachment largely justify the hub
phenomenon since the more connections a node has the more it is preferred for
attachment by the new formed nodes. At the limit, as Barabasi and Bonabeau (2003,
p. 58) conrmed in their simulations tests, if connecting speed increases substantially
the network will tend toward a star topology with a central hub, in which case
connectivity will be very high whenever this central node is used and very low
elsewhere in the network. The limit being the capacity to satisfy demand and
maintain attractiveness of the preferred hub node. From that saturation point
onward the preferential attachment relation will tend to form new subhubs that will
start their evolution cycle until saturation is reached.
Transferring this analytical framework to a transport network means that scale-
free network will have high service level in the central hub in detriment of the rest
of the network, what to a certain extent has been observed in the deregulation
processes both in Europe and the United States. Looking only to human activities
systems we can nd in social science and organizational theory several networks
examples and in all of them the key characteristic is that network interactions
tend to be less formal, more cooperative, nonhierarchical and with more
reciprocity, whenever legal and regulatory environment allows this free association
movements.
A network is a set of specic types of effective and potential relations linking a
dened set of persons, objects, or events. These elements (the nodes) are identied by
Knoke and Kuklinski (1982, pp. 1214) as possessing some attribute(s) that identify
them as members of the same equivalence class for purposes of determining the
network of relations among them. Network structure is thus revealed through the
description of classes of nodes and effective and potential links between those nodes.
In an urban mobility network system, actors represent the nodes and some actors
have more intense relations than others.
As Mitchell (1969, p. 14) noticed that the importance of network analysis does not
resume to the denition of the framework of relations the structure of relations
among actors and the location of individual actors in the network have important
behavioral, perceptual and attitudinal consequences both for the individual units and
for the system as a whole, which occurs due to the synergetic characteristic of the
network system that we have already discussed.
The structural analysis of any system implies the identication of the most
relevant positions within a given network of relations that connect the system actors
(i.e., the nodes) which implies dening the four main structural elements of a
network: units, complexity, form of relation, and relational contents.
The network units identify the social form represented in the network, that is
individual, formal and informal groups, formal institutions, classes and strata,
communities, regions, nation-states, etc., which are the actors of the interactive
processes represented by the links or connections, that is the relations.
The complexity of the networks can be dened by the level of interactions observed.
The simplest level is the egocentric network, formed by each individual node and the
ones with which this virtually centered node has relations, with each actor (or node)
being described by the number, magnitude, proportion of reciprocated linkages,
density of links to the egos rst zone, and other characteristics that may be relevant for
the denition of the link. This egocentric zone is responsible by the attractiveness of
each node when being selected by others as a preferred attachment node.
The next level of complexity is the dyad, formed by a pair of nodes, the analysis of
which is usually focused on the variation of joint characteristics of the pair, such as
similarities between attribute proles. The key question to dene a dyad is whether a
direct link exists between the two nodes, or whether the connection is made with indirect
links through a third node. The following level of complexity is obviously the triad,
where we add to the dyad complexity the determination of transitivity relations, that is
if A prefers B and B prefers C can we deduce that A will also tend to prefer C, or not.
According to Knoke and Kuklinsky (1982, in Thompson, Frances, Levacic, &
Mitchell, 1998, p. 178) experience in network analysis, Beyond the triadic level, the
most important level of analysis is that of the complete network, or system. In these
analyses, a researcher uses the complete information about patterning of ties between
all actors to ascertain the existence of distinct positions or roles within the system and
to describe the nature of relations among these positions.
Systems exist in small dimensions, such is the case of a classroom for example, or
in very large dimensions, such is the case of world wide systems of e-bay clients that
establish commercial relations between them, but what provides insight into its
structure is the identication of signicant positions within the network of relations
that links the system actors. Positions, or roles, are subgroups within a network
dened by the pattern of relations, some of which are observable behaviors, but all
bearing special rights and duties.
For the identication of positions in a complete network, Burt (1978) proposes
two alternative criteria. The rst is social cohesion, where actors are aggregated
according to their direct connections by cohesive bonds. The second criterion is
structural equivalence (Lorrain, & White, 1971, p. 63) where aggregation is made
according to their common set of linkages to other actors of the system (e.g., in a
mobility system this will be authorities, operators, suppliers, clients, etc.). The key
assumption for the operationalization of the concept being the distance between a
pair of actors, which is measurable in terms of degree of similarity in their patterns of
relations with other system actors.
In another work, Burt (1982, p. 22) also dened the relations between system actors
as being constituted by contents and form. The former refers to the substantive type or
relation represented in the connection, such as monitoring, planning, etc., and the
later refers to the complexity of the connections between actors (properties of ego,
dyads, or triads) which are independent from the respective content. As Burt (1982,
pp. 1516) further species that two basic aspects of relational forms are: the intensity
or strength of the link between two actors, and; the level of joint involvement in the
same activities.
For the relational content of networks Knoke and Kuklinski (1982, pp. 1516)
identify several types from which we select the following as being more relevant for
the study of UMS:
Transaction relations: actors exchange control over physical or symbolic media,

such as economic sales and purchases.
Communication relations: links between actors are channels of communications
within the system.
Boundary penetration relations: ties between actors are subcomponents of the
system held in common.
Instrumental relations: actors contact each others in efforts to secure valuable
goods, services, or information.
Authority/power relations: identify rights and obligations of actors to issue and
obey commands.
The social structure of the system is thus given by the regular pattern of relations
between the concrete actors and positions, or social roles, which are subgroups within a
network dened by the pattern of relations that is, real observable behaviors. This
means that each actor by the simple fact of occupying a position in a network structure
will consequently have links to other actors of that network, who also occupy other
structural positions. New positions will be created whenever new unique set of ties to
pre-existing positions are created (e.g., the case of rolling stock leasing companies).
2.3.2. Management of Change
No single theoretical approach prevails to explain organizational change. Indeed we can

observe a wide diversity of approaches and a major criteria to identify a subset is certainly
to look at the approaches which follow an open system view of change. That is, the ones
that recognize that institutions are embedded within larger environments and that the
precise denition for the institutionenvironment boundary is subject to several
variations over time developing from the a wide range of forces (social, economic, and
political) and also from the mutual inuence between institution and environment.
So, understanding the change is rst and foremost a matter of understanding
interactions between the entities that form a specic organizational eld38 and its
environment. The continuous interplay between theoretical and conceptual approaches
and empirical evidence will progressively contribute to improve our capacity to better
identify forces driving organizational change within a system, and to improve our ability to
dene the transferability limits of propositions to induce change in a desired direction.
Five organizational theories dealing with change in open systems have been
identied in Knoke (2001, pp. 3773): organizational ecology theory; institutional
38. As Knoke (2001, p. 38), dened it an heterogeneous set of functionally interconnected organizations,
in our case all the institutions, interest groups, government bodies and agencies, etc. that deal with
mobility-related issues.
theory; resource dependence theory; transaction cost economics; and organizational

network analysis. Each theory raises substantive questions and each addresses a
distinctive dimension of organizational structures and processes. Knoke provides a
comparative table that we reproduce, where it is obvious that these theories do not
propose alternative explanation of the same phenomenon, instead they provide
complementary insights on the dynamics of organizational change.
Organizational network analysis is the least developed of the ve theoretical
perspectives, but it is the one that holds better potential for tying together the many
dimensions of organizational behavior that we can nd within a UMS. Three basic
assumptions underlie this perspective:
The social structure of any complex system consists of stable patterns of repeated
interactions between actors.
These relations are the primary explanatory unit of analysis, rather than the
intrinsic attributes and characteristics of the individual actors.
The perceptions, attitudes, and actions of organizational actors are shaped by the
larger structural networks within which they are embedded and, in turn, their
behavior can also change the network structure.
Political networks and communities work on basis of inuence, domination and

also a considerable degree of implicit and voluntary trust with unofcial rules of
appropriateness. Trusting behaviors can be dened following (Lorenz, 1989, in
Gambetta, 1989, pp. 194210) who settled these as a sort of behavior that consists
in action that (1) increases ones vulnerability to another whose behavior is not under
ones control and (2) takes place in a situation where the penalty suffered if the trust
is abused would lead one to regret the action.
There are implications to this denition which are worth highlighting for their
relevance in change processes, such as:
First, this concept of trust implicitly entails decision-making in a situation of

uncertainty, and so, in a situation where risk is perceived; risk, in turn, is attributable
to the strategic behavior of others or to the possibility that they behave
opportunistically, or still to the existence of random variables that might even not
inuenced by others or self.
Second, both action and risks are not fully avoidable. Although institutions can to
a certain extent avoid engagement in trade with one another (although this implies
renouncing the potential benets of trade), the fact of being part of the system
limits this degree of freedom. So, the conservative fundamental strategy of
avoiding the relations, which implicitly means avoid risk of capture, has a limited
application once in a system some relations are inescapable and others may be
pursued or changed at varying opportunity costs.
Third, we have to distinguish between risk associated with the behaviour of others
and the risks related with acts of Nature or unpredictable changes in markets. In
these last cases trust is not related with opportunistic behavior or violation of
commitment and so the Lorenz postulate is not applicable.
The observation of institutional behavior in different sectors of activity led39 to

ascertain that change is more likely to occur when institutions possess no rules binding
their usual behavior or dening the appropriateness of their behavior (Table 2.1). The
no rules arena, just like it is often used in the policy participatory processes is a
powerful instrument to boost change processes. In the analysis of policy change in
United Kingdom Dudley and Richardson (2000, p. 20), observed that institutional
processes are dependent variables of the dynamic of ideas y.Similarly, policy
networks, as exemplied by (in the case of roads) a hegemonic policy community, or
(in the case of rail) a hollow core, are important in determining the trajectory of
ideas, but are also, ultimately, dependent variables of those ideas.
In mobility systems the symbiotic stage of interdependence between actors forces
the system to have structural consistency, which entails a horizontal and a vertical
dimension. Horizontally consistency is given through the equinality of action
between the different agents mobility providers acting at different moments and
with different roles in the supply chain. Vertically the consistency is achieved by way
of three attributes: coherence, efciency, and accountability.
Coherence is given by the alignment of decoupled objectives down through the
different decision levels assuring that the objectives settled at the strategic level will
be well derived into adequate goals for the tactical and operational levels. Efciency
represents the capacity to best transfer the basic resources (that is the means that
were allocated through strategic decision) into service outcomes, assuring this
property at the tactical level, and further down into consumption units, revealing
then the efcient performance also at the operational level. At last, accountability
constitutes a main instrument for management control, which enables a feedback
loop between eld results and strategic decisions.
The cross effects between horizontal and vertical specialization in a mobility
system lead to a complex network of institutions with different degrees and forms of
interaction, but all should be linked by a set of quality performance objectives. Given
this tight net of interactions, tness of purpose and action is important, indeed a truly
indispensable attribute that can only be assured by a continuous adjustment of
institutional design to policy and regulatory changes, that can be materialized in
changes in goals, instruments, or settings.
To assure this tness of purpose a clear distinction between policy-making and
executive decisions is required. The rst aims to provide direction, coherence, and
continuity to the course of actions for which this body is responsible for, whereas
executive decisions are aimed at giving effect to policies. In this perspective, as
Vickers (1965, p. 39) refers every decision-making body is to be regarded as a
regulator of the dynamic system of which it forms part, but its scope for regulation
and hence the meaning of that term is much more complex than at the simpler levels
at which our concepts of regulators and regulation are commonly formed. This
complexity resides chiey in the presence of policy-making as a constituent of
regulation.
39. Such as Dudley G., Richardson J., Vickers G., Knoke D. (1990), and McCool D.C.
Table 2.1: Comparison of ve open-system theories of organizational change.
Theory Primary level Focus of change Main underlying

of intervention process concepts
Organizational Population Organizations that t Organization form,

ecology better to their population, niche,
environments are selection,
more likely to be foundings, failures,
selected for survival, growth, and density
while ill-suited dependence
organizations are
more likely to perish
the change process
Institutionalism Organization Organizations Isomorphism,
eld conform to a legitimacy, symbols,
common form that is taken-for-granted,
legitimated by higher norms, and values
level institutions
Resource Organization Organizations and Resources, exchange,
dependence subunit subunits exchange uncertainty, power
resources to dependence, and
maximize power, autonomy
and avoid
dependence
Transaction Transaction Organizations decide Transactions,
cost decision whether to make or contracts, market,
economics buy goods and hierachy, hybrid,
services depending opportunism,
on transactions efciency, and
costs, including bounded rationality
administering
contracts
Organizational Multilevel Organizational Relations, centrality,
networks structures and cohesion, clique,
actions are both structural
causes and equivalence,
consequences of position, exchange,
multiplex relations and social distance
between and within
organizations
Source: Adapted from Knoke (2001, p. 43).
The complexity accruing from policy-making lies in the fact that a basic problem
must be sorted out before dening the course of intervention, that is how to
assimilate change with the greatest net gain (or smallest net loss) of value, whereas
preserving the balance of the system. The analysis made by Buchanan (1963, in
Vickers, 1965, pp. 6471) on the impact of increased demand for trafc movements in
United Kingdom towns, provides a good example. The terms of reference for the
Buchanan committee were to study the long-term development or road and trafc in
urban areas and their inuence on the urban environment. The report starts by
showing that the relation between trafc and other human related activities is not
self-regulated for any acceptable level giving thus room to search for more ambitious
regulation. The next phase was to identify the manifold relations underlying the
problem and challenging the coherence rule for any adopted solution.
Four propositions were advanced to embody the optimizing-balancing problem
involved in policy-making, for what was considered at the time a master piece of public
education for public reality judgment, stressing relations between amounts that can be
usefully invested and their effects, for value judgment, by betting the value of good
environment at the opposite of exclusive value for accessibility and, nally, for instrumental
judgment, by setting the problem before the public, revealing its complexity and showing
efcient, veriable and accountable solutions, that is pointing possible executive courses of
action. The four propositions used to build the construct reduce the conict to the
interaction of two variables (Buchanan, 1963, in Vickers, 1965, p. 65):
Trafc is an aggregate of individual journeys, (y). Their density is a function of

the density of buildings; their pattern a function of the spatial relation of these
buildings to each other;
All buildings need some degree of accessibility. All buildings need a number of
environmental characteristics and motorised trafc is inimical to some of them.
The conict involved in reconciling trafc needs and other needs can be expressed
as the conict between accessibility and good environment.
Urban trafc movements take place through streets that still serve the multiple
needs they served in medieval times (y.) This is an historical legacy, not a law of
nature(y). The use of space between buildings, no less than within buildings,
could and should be differentiated so as to minimise the conict between
accessibility and environment
The conict, even if minimised will remain. A given minimum of environmental
value implies an upper limit to the amount of accessibility. How high this limit may
be depends on the tolerable upper limit of a third variable, cost.
The Buchanan example provides evidence40 that in a system change can be observed
from two perspectives: the narrow focus on service or production, that is the
operational decision or planning level; and the wider whole system spectrum covering
40. The Committee clearly stated that existing institutions could not do the job on the required scale and
recommended the creation of executive agencies.
the three planning levels, strategic, tactic, and operational. The rst perspective is
largely dominated by the effects caused by different forms of market access, whereas
the later focuses on overall system performance that is, how the mobility system
contributes to the economic, social, and environmental objectives and how the needs of
urban performance are reected and implemented at the three decision levels. In both
cases change is a permanent transition process that adjusts institutions to the dynamic
environment in which they are embedded, that can be materialized in changes in goals,
instruments, or settings.
However, this transition process is only possible within certain limitations, which
can be divided into three main categories:
Logical limitations: the policy-maker or the executive decision-maker must be

able to demonstrate sound comparison of the current situation with the existing
norm and also with some predictive scenario that justies the reason for change.
Means for devising, selecting, and later communicating, between possible
courses of action must be available before any change proposal can be put
forward.
Functional limitations or minimum ability to produce solutions and alternatives in
domain of problem, which are largely inuenced by the following factors:
understanding the process that requires prediction and maybe change
intervention to adjust future path, in particular understanding interactions
between agents and possible impacts from change;
capacity to collect, store and process relevant information, so that minimum
control and awareness on the outcome of the change process can be assured; and
theoretical predictability of the process itself.
Organic limitations related to the structure and function of the institution within
which the change process is driven as well as its relations with other institutions,
which affect their performance and even their standards of success. Moreover,
structure and function of the several interacting institutions, and their mutual
relations in a change process are not static. On the contrary they tend to evolve in a
rapid change, which is partly inuenced by the policy-making process itself.
Ideas at the genesis of any change process, although related with existing political
forces and institutions, have a logic of their own that serves to shake existing
balances and be the remote driver of change. The evidence from the large majority of
mobility systems observed was that most change processes have been boosted by
nancial problems threatening to stop services, although the ve challenges for
reform, as systematized by Oster and Strong (2000), remain also valid as secondary
reasons for change:
re-estructuring to promote competition;

keeping competitors behavior;
maintaining small markets services and access;
managing concessions; and
safety and environment preservation.
To the previous we add change in pricing regimes which are strongly conditioned
by public acceptability, and technological changes either at the level of infrastructure
or superstructure, or change of energy sources, or even introduction of new concepts
of mobility. In all these cases, change management is required and processes
engaging all intervening agents must be organized.
Regulation, being a major change instrument in UMS, is often thought of as an
activity that restricts behavior and prevents the occurrence of certain undesirable
activities that is, a sort of red light concept. But the inuence of regulation may
also be enabling or facilitative, that is as a green light permission to move. Like in
many other sectors, also in UMS no universal receipt for regulatory intervention has
been found, despite exhaustive analysis and discussions within the transport research
community, and evidence exists that the optimum of regulatory intervention is
achieved through a mix of push and pull measures. Reinforcing this idea (Viegas,
2003a, p. 9) reports the preferable nature and level of regulation at a certain point in
time as being mainly dependent on:
the dimension of the market and potential number of suppliers;

the existence of alternative (and acceptability similar) forms of transport;
the sophistication of the services being offered (higher sophistication constitutes an
instrument of differentiation and thus a barrier to destructive competition); and
the level of technical competence and sophistication of the regulatory agency,
which should be in correspondence with the complexity of the supply require-
ments.
As systematized in MARETOPE research, the narrow denition of regulation

denes strictly the so-called regulatory measures as corresponding to the above
referred red light concept. The broader denition corresponds to the concept of
regulatory framework, where the state/public authorities play a central role and
regulation is achieved often through a mix of the above referred red and green
light measures. Finally, we dened also the concept of regulatory context which
includes not only the framework of rules and actions which stem from the public
authorities, but also other sources which contribute to regulate system behavior (e.g.,
markets, nongovernmental organizations, etc.), that is the broader delimitation of
the green light environment.
From the previous arguments we can understand that change processes force a
continuous adjustment of institutional and regulatory frameworks. However, several
institutions with responsibility for mobility services have developed as if they were
social services institutions with management decisions being taken in the framework
of politically appointed boards of administrators causing institutional rigidity and
consequently a mismatch to those adaptive evolving needs, whereas raising several
barriers to the change process very often caused by a state of strong entropy. In other
cases it is the simple balance between gains and losses as an effect of the change
process that originates stakeholders negative reactions.
These reactions can be divided into the following categories which simultaneously
express the intensity of opposition and target the eld where intervention is needed to
overcome opposition:
Asymmetry of information (i.e., stakeholders do not understand change):

Stakeholders do not have enough or the correct information to perceive the
reform process. Consequently they are unable to assess the effect or even to
understand how to accommodate the objectives and the change process in its
organizational subsystem.
Capability problems (i.e., stakeholders are unable to support change):
These may concern the lack of instruments, personnel, skills, experience, and
competencies for the different stakeholders to implement the change process.
Means have to be provided for stakeholders to deal with the change process.
Conicting interests (stakeholders dont want to support the change):
The reform determines advantages and disadvantages for the stakeholders that
react in positive or negative way according to the impact suffered, in which case
a mitigation process should be entailed in parallel to the change process to
minimize the negative impact of change.
A main division within theories of conict is between those that consider conict
as a pathology accruing from interaction between agents who have competing
interests, and those that consider conict as a sort of contest in which participants
select their strategies aiming to win (i.e., game theory) and so the latter is seen and
analyzed as a natural behavior of agents.
The study of these behaviors in a eld of study designated as strategy of conict
was consecrated by Schelling (1960, p. 15), largely inspired in game theory, which
distinguishes three types of games: games of skill; games of change; and, games of
strategy. Like in the eld of conict related issues also here the term is intended to
transmit the interdependence of the adversaries decisions and their expectations
about the others behavior. For Schelling most conict situations are essentially
bargaining processes, which may be explicit or only tacit manoeuvre.
This denition has the implicit consideration that in addition to the divergence of
interest over the variables under dispute there is also a common interest for the
agents in negotiation to achieve an outcome in which the destructive impact on each
others values should be contained. An aspect worth highlighting is that the theory of
strategy of conict is nondiscriminatory between conict and common interest. As
Schelling (1960, p. 18) explains The theory degenerates at one extreme if there is no
scope for mutual coalition, no common interest at all even in avoiding mutual
disaster; it degenerates at the other extreme if there is no conict at all and no
problem in identifying and reaching common goals.
Negotiation is a discussion between two or more parties with the aim of resolving
divergence of interests and so escaping social conict situations. According to
conict management theories developed in several organizational elds of analysis,
the parties can be individuals, groups, organizations, or political or governing units

or bodies. From the literature we can distinguish ve broad strategies that are
commonly used in negotiation processes (Pruitt & Carnevale, 1993, pp. 313):
Concession-making: that is reducing own goals, demands, or offers.

Contending: trying to persuade the other party to concede or trying to resist similar
efforts by other parties. There are a number of diversied tactics to undertake this
strategy that go from threats, that is messages indicating that the one who fails to
conform will be punished, to positional commitments, that is messages assuring that
one of the parties will not move from a determined position.
Problem-solving: trying to adopt compromise options that satisfy both parties
goals. A wide diversity of problem-solving tactics is available.
Inaction: doing nothing or as little as possible, for example calling off meetings,
talking around the issues and avoid tackling the real problem, etc.
Withdrawal: dropping out the negotiation, in which case the paradox of the
prisoners dilemma might take place, according to which if all parties41 in dispute
fail to cooperate they will be worse off than if they cooperate.
Concessions involve a reduction in the goals underlying demands that is, changing
ones original proposal so that it provides less benet to oneself and more benet (or
less disbenet) to the other party. Three main ndings are reported by Pruitt and
Carnevale (1993, p. 27) concerning the impact of concessions on the outcomes of
negotiation result from an extensive observation done by several authors. The rst is
that, if agreement is reached, negotiators with high goals, who make large initial
demands and resist negotiation making, will usually achieve larger outcomes. The
second is that rmness contributes to lengthen negotiation and risk noneffective
agreement to be achieved. The third nding is a derivation from the previous two and
it says If reaching an agreement has any value to a negotiator, there is likely to be an
inverted U-shaped relationship between rmness and negotiation outcome (Pruitt &
Carnevale, 1993, pp. 28 and 29). Negotiators who demand too much will fail to reach
agreement and will have a poor performance and the ones who demand too little will
reach agreement but at the cost of low prots. So, the most successful negotiators will
be the ones between the two extremes, that is the moderates.
Contending can surge in a wide variety of forms, usually designated as tactics,
among which we can nd threats, harassment, positional commitments, and persuasive
arguments, all having the aim to persuade the other part to make concessions, the major
difference being the existence of punishment in the rst two, whereas on the latter the
tactic lies on statements of determination that lead the opponent party to reconsider its
own strategy. Problem-solving in turn involves an effort from all parties to nd
41. In its simplest version the prisoners dilemma involves only two parties, each with two options (Pruit
and Kimmel, 1977; Rappaport and Chammah, 1965). The prisoners dilemma falls in the technical
category of games of moves, in contrast to negotiation, which is a game of agreement.
mutually acceptable agreements. Inaction and withdrawal are substantially different

form the other strategies as they do not move the negotiation toward an agreement.
To accompany these strategies we can nd three broad classes of procedures:
Joint decision-making: including negotiation and mediation, involving a third

party.
Third party decision-making: which includes adjudication, arbitration and
decision-making by a legitimate authority within a legally binding institution.
This procedure can be quite expensive, and besides third parties decision-makers
often do not have enough knowledge on the parties interest to successfully place
winwin agreements.
Separate action: in which the parties make independent decisions. Three types of
separate action are:
Retreat, in which one of the parties yields to the others requirements. Similar to
concession making in negotiation.
Struggle, which is identical to contending in negotiation.
Tacit coordination, in which the parties accommodate to each other without a
discussion.
As expressed by Sabatier and Jenkins-Smith (1993, p. 49) levels of conict can be

understood as a reection of the degree of incompatibility of basic beliefs between
competing coalitions, threatening the core values and precepts. The higher the level
of conict the greater the incentives of both subsystem coalitions to commit resources
in defending the core values.
Consequently, identifying structural relations is essential to understand the
sources of power in the organizational eld as the dominant coalition will constitute
the power core of the organizational system. Power is an important resource which
intensity and clarity of appropriation by the different parties varies substantially. The
way that use or application of intensity of power varies can be described and
controlled by a process.
A barrier to change is thus something that is causing hindrance, preventing progress
or movement. In an evolutionary process a barrier exists all the time, they are part of a
natural causeeffect dynamic and represents always a negative effect over the
evolutionary process where they are acting. They can be visible or not limiting the
decision-makers awareness and consequent action. Barriers can also be material or
immaterial and within the later it can still have a visible or hidden representation.
Besides, the classication of barriers cannot ignore the stage of development of the
change process itself. A barrier can be more or less severe depending on the moment in
which it is raised and the potential damage (political or otherwise) that can result.
In fact, there is a great diversity of ways to classify the different kinds of barriers.
Moreover, within the diverse array of barriers that can be recognized it is possible to
verify that they can overlap each other, be interrelated or even nested within each
other. From our observation we conclude that barriers can be classied according to
their end-object of incidence, that is:
Resource related: relates with the lack of nancial, material, or human resources to
implement the change.
Process related: related with the change process itself, including coalition games
between stakeholders.
Framework related: related with the overall environment where the change process
is embedded, including the aspects related with the rules of change.
This interpretation seems to be well-adjusted to reality if we complement the

analysis of barriers by further desegregating it into consensus building levels, where
intervention occurs, taking into account the concept of level of social analysis
originally proposed by Williamson (2000, p. 608), where each level determines
the scope of intervention of the respective actors. This distinction gains importance
due to the fact that instruments and measures cannot be freely established within the
institutional framework, so the identication of the governance level and scope of
intervention is extremely important to enable the correct allocation of measures to
overcome barriers. Within this perspective we can observe barriers at the following
levels (EC, TIS, 2002b):
L1: cultural and social regime: entailing the cultural, ideological, political, and
social orientations.
L2: regulatory context: entailing the laws to which the different mobility services
and actors are submitted, as dened earlier.
L3: regulatory framework: entailing the general rules that are decided for the
organizational eld of transport, within the scope of the law.
L4: organizational forms for governance: entailing the choice of organizational form
by authorities and operators within the scope of the existing laws and regulations.
L5: contractual relationships: entailing the choice of incentives (contractual
relations) between actors, within the scope of laws, regulations, and organizational
forms.
L6: allocation of resources: decisions to be taken concerning conicting allocation
of budget and/or resources within the same institution.
A hierarchy exists between these levels. They are classied with decreasing level of
temporal inertia for change to produce effects (e.g., traditions generally have a longer
change process than laws, which generally also take longer to change than regulation
rules, and so forth until the allocation of resources). Thus, from L1 to L6 each level
largely determines the scope and conguration of the next one. Besides, each of the
hierarchical levels has a specic context in which not all measures and instruments are
equally effective. Consequently, each governance level has a possible set of instruments
and each instrument optimizes its effectiveness in a determined governance level.
Moreover, each level has different degrees of intervention in the planning levels of the
system, with the higher levels having more presence in the strategic decision levels and
Figure 2.3: Distribution of stated barriers in MARETOPE case studies (EC,

TIS, 2002c).
Figure 2.4: Boomerang effect of barriers and tools (EC, TIS, 2002b).
the lower ones at the operational level. Figures 2.3 and 2.4, illustrates the distribution of
barriers found within these levels in the cities observed.
The role of each agent is thus dependent and inuenced by the level where its
intervention takes place, that is to the degrees of freedom to act within the system
and to the power that agents has over the other. Consequently the potential to raise
barriers is also affected by that delimitation. From our observation of the change
cases in UMS we conclude that the process of raising and overcoming a barrier is
bound by three basic elements: actors/institutions, territories, and instruments (EC,
2003a, TRANSPLUS, D4.2, p. 35).
A barrier is frequently a consequence of an actors attitude regarding his position
in the territory he perceives to be under his area of inuence and of the instruments
available to develop effective opposition. Like with a boomerang effect, the barrier
life cycle gives place to the development of instruments that aim to mitigate the
barrier but, as this occurs, a new potential is also raised that another stakeholder
group might be affected by the impacts caused by the intervention with the new
instrument. To some extent we can then say that there are cascade effects in this
evolutionary path.
Managing the changing process implies introducing a disturbing element that will
act has a trigger for the readjustment of the course of evolution of the system,
overcoming barriers generally raised by incumbent interests, nding mitigation
effects to compensate losses and assure fairness between agents and, nally,
re-establish again a new equilibrium in the system until new change needs are
detected, which should be done through feedback channels.
Chapter 3
Simplifying Complexity in Urban

Mobility Systems
3.1. Interacting Agents

3.1.1. Nature and Missions
Most organizational and institutional planning tends to pursue the objective of work
division through horizontal specialization, assigning existing or foreseen tasks and
functions to different bodies. As referred by Gulick (1937, pp. 34), work division is
the foundation of organization; indeed, the reason for organization. Since then, the
theory of organization evolved, and today, there is a general awareness that careful
analysis and interlinkage of functional processes, describing each step of the different
activities and providing a consistent base for analytical and interactive tasks, is a
fundamental tool to understand the dynamics of both organizations and complex
systems.
In an urban mobility system, besides the horizontal specialization, we have to
consider also the vertical specialization of the system that, inspired in the concept of
Simon (1997, p. 7), we dene as the division of decision-making duties across the
strategic, tactical, and operational levels, already referred in Chapter 1. The main
reasons supporting the need for vertical specialization are mostly related with the
consistency of the system. Although horizontally this is achieved through equinality
of mission for the different agents and complementarities between modes, vertically,
the consistency is achieved by way of three attributes: coherence, efciency, and
accountability.
Coherence is given by the alignment of decoupled objectives down through the
different decision levels assuring that the objectives settled at the strategic level will
be well derived into adequate goals for the tactical and operational levels. Efciency
represents the capacity to best transform the basic resources (i.e., the means that were
allocated through strategic decision) into service outcomes, assuring this property at
the tactical level, and further down into consumption units, revealing then the
efcient performance also at the operational level. Finally, accountability constitutes
a main instrument for management control, which enables a feedback loop between
eld results and strategic and tactical decisions. The cross-effects between horizontal
and vertical specialization in a system result in a complex network of different types
of institutions with different degrees and forms of interaction, but all must be linked
by a robust chain of quality performance objectives and associated performance
evaluation mechanisms.
Before any discussion on formal or informal interactions between institutions, it is

worth to understand what the meaning of interdependence is1 and where we should
look for it while observing the dynamics of an urban mobility system. A basic rule
can be to look at all sorts of ows exchanged between institutions, decision-making
agents, which have specic roles (i.e., specic work assignments for decision-making
duties) and missions (i.e., purpose or the reason why institutions exist and the
philosophy guiding their strategic choices) within the system and interact in such a
way that each organization depends on the other for pursuing its own goals,
although not necessarily bounded by subordination status.
This state of symbiotic relationships is characterized by symmetry and potential
reciprocity between the parties of such a relationship, implying that all changes that may
hinder the interest of one such agent must be solved through negotiation or often through
unilateral inducement methods such as the application of principleagent theories of
incentives developed by Laffont and Martimort (1947). Besides, as Lindblom (1965,
pp. 2122) clearly identied, these relations and resulting effects can be direct or indirect:
Within the set, each decision-maker is in such a relation to each other

decision-maker that unless he deliberately avoids doing so (which may
or may not be possible), he interferes with or contributes to the goal
achievement of each other decision-maker, either by direct impact or
through a chain of effects that reach any given decision-maker only
through effects on others.
To identify the nature and scope of interaction between two or more institutions,
it is essential to characterize their mission and their role in the system. For this, we
continue to follow the ISOTOPE approach where three decision or planning levels
have been identied, as described in Chapter 1: strategic, tactic, and operational.
From the observation of several cities, referred in Chapter 1, we have identied that
interdependence between agents is present not only within each of these decision or
planning levels but also across these levels.
The complexity of any mobility system is thus largely dependent on the number of
entities in each level, which condition the adequacy of mechanisms meant to induce a
concerted action among these agents. The diversity and disparity of agents is also
essential for the characterization of interdependencies that can be typied along the
following categories:2
Bilateral interdependence exists whenever the object of relation affects two parties
in any interaction process.
1. We distinguish between interdependence and interaction. The latter representing only the coherent
transfer of information between any two basic elements of the system. See also Section 2.2 in Chapter 2.
2. In his analysis of the San Francisco Bay area, Chislom (1989, pp. 4063) denes also bilateral and
multilateral interdependencies but ignores the existence of hierarchical and longitudinal ones, which seems
to be a strange conclusion, even considering the American urban mobility framework.
Simplifying Complexity in Urban Mobility Systems 81
Multilateral (or transversal) interdependence exists whenever any agent taking part
in the system creates more than one bilateral dependence with other entities, as
long as there is no relation of subordination between the parties.
Vertical interdependence exists whenever the relation between entities is bounded
by a subordinate status of one regarding the other, both entities having specic
roles within the same domain of intervention.
Oblique interdependence is the equivalent of vertical interdependence in the case when
the entities bounded by subordination status belong to a different domain of inter-
vention. It is worth referring that one of the most difcult aspects in managing urban
mobility systems lies in the conicts originated by these types of interdependencies.
Longitudinal interdependence exists whenever dependency is irreversibly based on
previously taken decisions, that is, a sequential relation exists between decisions
taken in different time moments. These time-related dependencies might be due to
decisions of one or more entities.
In Figure 3.1 a pictorial representation3 highlights the typology of institutional

interdependencies. From the case studies observed, we concluded that these
interdependencies can be found in different circumstances, such as the follows:
between agents of the same nature (e.g., between operators) or

between agents of different natures (e.g., between authorities and operators) or
even between agents with different roles within the same system decision level (e.g.,
land use and trafc authorities) or
between agents in different levels (e.g., central government and municipal
government).
Furthermore, we have seen that interdependence can be established by formal or

informal instruments: the former can vary from simple partnerships up to contractual
frameworks and other legally established obligations, whereas the latter may even be
more complex because interdependence can exist without any direct relation. For
indirect relations, the informal institutional link can be established through third
parties or even by simply sharing a common resource without any protocol. Examples
of these interdependencies without formal direct relationship are wage revisions in one
entity that may induce workers of another agent to have analogous behavior or a
municipal decision of revision of trafc space assigned for parking, without any
consultation to the public transport company and so on.
Several authors relate degree of interdependency with complexity of coordination
mechanisms and the corresponding framework. As suggested by Scott (1981,
pp. 407422), higher levels of interdependence need more extensive and complicated
coordination mechanisms. As a corollary to this note, Chislom (1989, pp. 5657)
3. Inspired in the graphic codication developed in the research project TRANSPLUS, D4, Barriers,
Solutions and Transferability, chapter 2 Conceptual Framework (pp. 720). The author of this book
was one of the coauthors and Work Package Leader in the research project.
Figure 3.1: Conceptual identication of the typology for institutional

interdependencies. Source: Author.
suggests that by increasing the requirements for successful co-ordination, the level
of interdependence among the components of the system may be articially increased
as an unintended consequence. Under these conditions, co-ordination requires
greater agreement across a broader range of values and the solution of cognitively
more complex problems.
Indeed, from our empirical observation, we concluded that unless some degree of
vertical (or oblique) interdependency exists, the key function to bring a complex
system to good performance levels is concertation of decisions,4 which is a form of
4. Coordination entails direct hierarchical command, which in reality does not exist in most relations
between agents.
Figure 3.2: Chain of key decisions across planning levels. Source: Author.
leadership mostly based on inducement of actions and decisions instead of direct

hierarchical command.5
By systematic observation of European cities, we have found a chain of decisions
taken by different actors, which, despite being taken before the moment of production
and consumption of the services, is an integral part of the process of conceiving,
developing, and implementing mobility services, that is, they are part of the mobility
supply chain. As described later, those decisions can be allocated to the three planning
levels that were briey presented in Chapter 1 (Figure 3.2):
At the strategic level, the main concerns relate with long-term decisions, such as
denition of mobility policy, market shares, level of cost coverage by revenues,
denition of areas of intervention for the mobility system, dening the levels of
accessibility to be provided to different areas at different times, delimiting the public
service character of the services to be provided, and the means to be allocated to the
production of those services, establishing the degree of intermodality provided by the
system.
At the tactical level, the main concerns are medium-term decisions largely related
to the conguration of system supply, individual service denition to match the
different market segments, and detailed specications such as type of vehicle,
routes, timetables, different fares, additional services, denition of performance
standards, and denition of contractual basis for engagement of service providers.
At the operational level, the concerns are mostly short term and related to
management of services and resources. This is the action level where the service is
carried out but also where the performance monitoring is undertaken. Production
scheduling can extend from infrastructure management until vehicle and staff
rostering. Depending on the degree of integration of activities, all these functions
can be allocated, through different ways as dened in the regulatory framework, to
one or several entities.
5. We dene command as the capacity to exercise authority without allowing stakeholders in subordinate
position the participation in the decision process. Authority, in turn, represents the capacity of one entity
to guide the decisions of another, without giving the latter the opportunity to independently assess the
merits of those decisions.
To undertake these activities, a wide spectrum of institutional settings can be

found across Europe. In some countries and cities, it is for the authorities to develop
all functions, whereas in others, private agents carry out signicant parts. Whatever
the solutions adopted, it is through the denition of their function in the whole
mobility system that these entities have their departure point to move into a certain
direction and to dene purpose and values to guide their characteristic actions and
reactions, that is, to establish the patterns of their institutional behavior.
As several authors refer (Piercy & Morgan, 1994, p. 16; David, 1989, pp. 9097),
in the institutional mapping, the mission of the different entities fullls two types of
information needs: direction, by dening boundaries for strategic choices, because
the denition of mission sets the institution on a particular heading; legitimization,
by dening the mission, internal and external stakeholders are entitled to check
whether the institution is pursuing its values in a proper way.
The type of mission is also one of the characteristics that dene the nature of the
interacting entities. From the cities observed, in the different research projects, as
referred in Chapter 1, we have identied a diversity of entities that can be organized
according to their natures as presented in Tables 3.1 and 3.2. Irrespective of the
economic interests of each agent, the relation between these entities is characterized
by transactions, which can typically take the form of exchange of goods, services,
nancial means, information, or any other asset with value for the utility function of
the parts (Simon, 1991, pp. 2544). Once this utility argument is accepted, then
agents become nodes in a network of transactions.
Table 3.1: Nature and roles of entities interacting in an urban mobility system
(part I).
Nature and roles of entities Type of entity
Political authorities National government

Regional government
Local government
Regulating authorities Transport authorities (e.g., PTA,

passenger transport authorities)
Economic authorities
Fiscal authorities
Technical authorities Transport authority (e.g.,

passenger transport executive)
Trafc authorities
Land use authorities
Environmental authorities
Safety authorities
Security authorities
Source: Author.
Table 3.2: Nature and roles of entities interacting in an urban mobility system
(part II).
Nature and roles of entities Type of entity
Operators Transport operator

(e.g., train operating company)
Developers (i.e., land use operators)
Other service operators
Emergency services
Suppliers of productive Vehicle industry

resources Staff (e.g., man power rms for crew
members and public attendance)
Management information systems
(e.g., radio and TV)
Consumables
Etc.
Clients Transport user/consumer group (e.g., cyclist

association and commuter group)
Specic local residents (as property owners,
threatened parties, etc.)
Specic local businesses (as employers,
as threatened parties, etc)
Other interested parties Nongovernment organizations

Community group (i.e., meaning here a group
based on a particular local area)
Other interest group/ activist group (e.g., green
lobby; pro-roads lobby)
At this stage, we must recall Coases (1937, pp. 386405) argument that
transaction costs are themselves the reason of existence of the rm, meaning that
if information and enforcement were costless, it would be hard to envision a
signicant role for organizations (North, 1990, p. 63).
Enforcement costs arise mainly due to lack of information. If there is no
information on the attributes of a good or service or all the characteristics of the
performance of agents, then problems of acquiring information and measuring accrue
and enforcement are made necessary with a full cost structure behind it.
But there are also some economies of scale in the provision of enforcement and
corresponding laws that largely depend on the necessity of developing agents and
hierarchical structures to monitor, measure, and collect information across the whole
mobility system. The purposive economic and social society in which agents are
embedded leads to strategic behaviors aiming to (formally or informally) inuence the
evolution of the system in favor of its own goals; therefore, principalagent
mechanisms are at the center of the interactive behavior between all agents, irre-
spective of their nature and roles in the system.
Regulatory frameworks, contracts and prices, are the main instruments to steer
these behaviors in favor of system performance. As Hayek (1945, p. 528) notes, In a
system where the knowledge of the relevant facts is dispersed among many people,
prices can act to co-ordinate the separate actions of different people in the same way as
subjective values help the individual to co-ordinate the parts of his plan.
Agents own utility function plays an essential role in the principalagent
mechanisms, in which the agent should pursue the interests of a principal. In urban
mobility systems, there are several levels of nested principalagent relations, the
topmost level being the one between the government (principal) and the regulator
(agent), and the second level the one between the regulator (principal) and the
organizing authorities (agent), and a third level between each organizing authority
(principal) and the several agents acting in the operational level, in which the private
citizen, who has the capacity to use common mobility facilities on self-service basis, is
also one of those agents. Moreover, even the operator of mobility services can play the
role of principal whenever part or the totality of productive services is subcontracted.
Indeed, we can observe in the different cases supporting this research study that
agents are often confronted with playing a double role, where they have to act both
as principal and as agent. Being so, it is important to come up with a systematic
denition for the structure of these institutions, which is dened along ve main
aspects (DONIR):
Domain of intervention: Set of values, principles, axioms, and rules, based on

which intervention is made for system ruling and control, at different decision or
planning levels. In an urban mobility system, we can categorize the existing
domains of intervention in three main types framework related, process related,
and resource related, which entail:
Framework-related domain, entails the denition of the underlying structure of
a UMS, entailing policy denition in the several related elds of intervention, such
as land use, trafc, and environment, but also the identication of the type of actors
involved in the system. Most of the strategic decisions are located in this domain;
Process-related domain, entails the denition of the management algorithm6
leading to the production of services in the most efcient and effective way; and
Resources-related domain, related with supply of resources.
Ownership: Capital ownership that formally means the nancial resource to
establish the agent as an economic or social unit.
Nature: Set of attributes that characterize institutional behavior allowing an
institution to be recognized by their actions and reactions to external stimulus.
Typical natures in a mobility systems are indicated in Table 3.1
6. The algorithm concept is here referred in its widest meaning, that is, the precise rule (or an ordered set of
rules) specifying how to solve some problem and not in the sense of mathematical algorithm.
Institutional (Legal) statute: The statutory form adopted by the agent that
determines some of the legal procedures followed in the management of the entity
and to a certain extent its degrees of freedom as a collective entity.
Roles: Functions and mission developed by an agent, which is best represented by
the set of activities performed. Typical roles are regulator, organizing authority,
auditors, planner, operator, infrastructure manager, and supplier of services or
materials.
The synergetic combination of these categories of aspects results that each agent
will be in a unique institutional setting that she will try to alter and adjust to the most
adequate conguration in the process of pursuing its own objectives, occasionally
leading to progressive segregation, as explained in Chapter 2. This diversity
and evolutionary process provides uniqueness7 to each urban mobility system that
cannot be directly reproduced to t another system without going through careful
transferability analysis.
This evidence of uniqueness is in line with the rationale followed by Richard
Lewotin (2000) that living organisms are not genetically determined only by its genes,
but instead are a unique outcome of an ontogenetic process contingent on the sequence
of environments in which it lives. Lewotin supports his analysis in the experiment of
Jens Clausen, David Keck, and William Heisey in 1957 on cloning plants in different
environments,8 later conrmed with several other experiences. These experiments led
to the emergence of one of the basic concepts of modern biology coevolution or
tness to landscape according to which all living organisms (i.e., dynamic entities)
result from the interaction between its genotype (structure) with its phenotype
(environment/circumstances to which it has been exposed).
3.1.2. Industrial and Market Structures
John Bain (1968, p. 25) dened industrial organization as the market decomposition
into three parts structure, conduct, and performance. Market structure means the
way sellers interact between them, with buyers, and with potential new competing
entrants. Market conduct means the way rms behave in a given market structure,
that is, how rms determine their strategies, their policies. Market performance
7. Often raising serious difculties in comparability exercises, as experienced in ISOTOPE, MARETOPE,

TRANSPLUS, and others.
8. The experiment was done (published in 1958) with three cut pieces of the plant Achillea millefolium. One
piece was planted at a low elevation, 30 meters above sea level, the other at 1.400 meters, and the third at
3.050 meters. The result was different growth patterns and no conclusion could be taken on the way of
predicting growth order or even which genotype caused the best growth (Lewotin, 2000, pp. 2023). This
reection was launched in TRANSPLUS in the process of developing a framework to access transferability
of policy measures. Within the accompanying measure METEOR, the author developed a transferability
framework to support transfer of good practices within the cities engaged in CIVITAS initiative promoted
by the European Commission, which is currently being tested by those cities.
means the welfare outcome of market interaction, that is, to assess performance,
we must measure whether the market interaction leads to a desired outcome or whether
a failure occurs in which case regulatory intervention is needed. The legal structure that
rules and supervises market behavior is usually designated as antitrust laws.
Edward Mason (1939, p. 69) interpreted the core problem of industrial analysis in
the following way: The problem, as I see it, is to reduce the voluminous data
concerning industrial organization to some sort of order through a classication of
market structures. Differences in market structures are ultimately explicable in terms
of technological factors. The economic problem, however, is to explain, through an
examination of the structure of markets and the organization of rms, differences in
competitive practices including price, production, and investment policies.
The answer to the typical questions raised by economic analysis is conditioned by
a number of factors, which we may also understand as specications for the
denition of market behavior, which can be divided in two groups:
Static parameters, describing the characteristics of the market agents:

Which type of agents (i.e., productive units) can we nd in a market.
Theoretically, rms are not the only organization with economic activity; on
the contrary, in most markets, we can nd nonprot organizations, associations,
and governmental agencies. In urban mobility, we still have individuals
transporting themselves and interacting with all those types of organizations in
the same market.
How many agents interact in the market, is it a xed number, or is entry of new
rms allowed.
The scale of geographical presence of agents allowed to enter the market, that
is, whether the market is open for local, regional, national, or global agents or
restraint for agents with any of these characteristics.
Dynamic parameters, representing bounding rules for agents interaction:
Which actions are available to each rm, that is, for example, choosing a price,
setting quantities produced, and setting production capacity or location.
Agents expectation about available actions and how competing agents will
answer to each agent action.
Firms expectation about market contestability and number of competitors and
potential entry.
Therefore, the denition of market structure represents the specication of the

rules of the game that will be played by agents in the course of their short- and long-
term relationships within a market.9 Setting these rules means describing agents
entitled to play it, a nite action set available to each player, as well as specifying the
elements that dene the end of the game (e.g., time period, outcomes, effects, and
results). Many authors have dedicated their works to the analysis of market
9. Market is herewith understood as the virtual arena where producers place their products or services to
attract the same set of consumers.
structures. We follow Shys (2001, p. 59) approach, illustrated in Figure 3.3 where the
main existing market structures are presented.
There are two main categories in market structures: perfectly and imperfectly
competitive. The strategies dened by each agent (i.e., market conduct) in their
interactions are conditioned by the possibilities offered by the market structure. The
perfectly competitive market structure assumes that each agent takes the market
price as given and determined by the intersection of the market demand curve and
the producers aggregate supply curve. Consequently, the set of actions for each agent
is limited to its decisions on quantities of production. These structures are usually
explained in economic theory either by assuming a xed number of rms (referred as
short-term equilibrium) or by free entry (referred as long-term equilibrium). In
imperfectly competitive market structures, we can nd three types of structures:
monopolies, duopolies, and oligopolies.
Monopolies are characterized by having only one seller who is theoretically entitled
to choose any pricequantity combination on the consumers demand curve, the choice
falling either on price or on quantity. Monopolies are prone to practice higher prices
given the lack of competitive threat and consequent consumer capture. To this respect,
monopolies can be considered as discriminating, when products are sold to different
consumers at different prices, as opposed to nondiscriminating monopolies, where the
Figure 3.3: Market structures. Source: Adapted from Shy (2001, p. 61).
same price is practice for all segments of consumers. Monopolies can still be subdivided
into static and dynamic, the rst sells its product only once, and the latter sells it over
more than a single act or period.
Duopolies (two sellers) and oligopolies (more than two sellers) can be either
cooperative or noncooperative. The rst is characterized by collusion of producers by
setting the productive level at a prot maximizing level or by agreeing on a monopoly
price. The latter is explained by situation in which agents choose their strategies once
and at the same or in sequential moves (i.e., leader-follower), in both cases choosing
either prices or quantities produced.
Market contestability is conditioned by these structural frameworks that model
short- and long-term relationships between productive agents and regulatory and
planning agencies.
In urban mobility systems, it is the regulatory and organizational framework that
provides the legal and institutional background congurations where markets develop,
and the market supervision goes beyond the antitrust function because there are many rules
dened by authorities that limit the action of operating companies, such as rules related
to network stability (even in the United Kingdom) and to administrative price setting.
In the past decades, we have observed substantial changes in market structure for
urban mobility. In urban mobility systems, the trend is to move from production
based on public capital monopolies to a limited competition situation where
competitive pressure is ensured through contracts and other instruments.
In most countries, the movement of change transformed markets where typically
one local monopolistic company operated into a more competitive environment
where global companies are able to operate either directly (in United Kingdom) or
entering the market through capital ownership of local operators. These changes
raise a number of questions that are usually approached in studies of industrial
organization and that should be upstream of the decision on which regulatory regime
should be selected for a given service in a determined city:
What is the markets optimal dimension to divide the network in such a way that
minimizes costs?
Does the market produce a socially optimal number of services adequate to
consumers preferences and diversity?
Are companies dynamically efcient, that is, are investments done to ensure the
adequate amount of resources to develop new technologies for current and future
generations?
3.2. Understanding the Main Tensions in the System
3.2.1. Territorial Denition of the Urban Mobility Systems
Nowadays, as the size and shape of urban areas developed and spread across peri-
urban areas, forcing the mobility network conguration to loose its original radial
shape and to extend beyond the administrative borders of the city, the need to extend
the scope of intervention of the mobility authority to all communities with a direct
stake in the mobility system becomes more obvious.
Despite the easy understanding of the rationale behind the functional enlargement
of the scope of intervention for the urban mobility systems, the concrete denition of
the specic territorial limits, or the respective nancial autonomies of the organizing
authorities, is still a problem difcult to solve, given the diversity of current legal,
administrative, and nancial frameworks existing in Europe, despite the harmoniza-
tion of concepts made by Eurostat, which dened urban area and central area for
statistic purposes, but with no relation to mobility needs.
As noted by Hagerstrand and Clark (1998, p. 21), The critical link between
human society and the terrain with its living content is constituted by the parcelling
of land and water in spatial domains of various size and shape. (y) Boundaries tend
to be very stable over time. Some units may persist for generations (y) These
primary domains form together a mosaic which is placed directly upon the
geophysical landscape.
For many years, developed societies chose to organize in such a way that these
domains are administratively nested in hierarchical order, from the state level down
to the private ownership level. The higher level acting only on land and spatial
domains of the lower level if there are properties rights involved; otherwise, the
geographically dened boundaries correspond also to the limits of administrative
competencies.
As also referred by Hagerstrand and Clark (1998, p. 23), all these domain borders
were typically dened on basis of social agreements and almost by an implicit logic;
upper level institutions have integrative roles, whereas lower level ones have sectorial
roles. Notwithstanding the robustness of this logic, today, this type of organization is
crisscrossed by a critical element, which is the urban mobility system. In fact, this
system reects citizens needs, and as such, the base logistics for the functioning of
any society. But its territorial insertion goes often beyond the institutional limits
dened by the traditional rationale of integrative versus sectorial roles, without
having a dedicated institution to enable the good matching between functional and
operational delimitation of the system and its legal and administrative correspon-
dence within the urban governance structure.
As observed in our set of cities, and also reported by several other research
projects (e.g., SESAME research project, COST 332 research action, both promoted
by the European Commission and covering EU countries, and World Bank, 1996,
report, among others), those boundaries represent an important constraint factor in
the denition of pricing and nancing policies for the mobility system and also in
enabling policy concentration between transport, land use, environment, and scal
incentives. The main reason for this pitfall is the need to satisfy two levels of
intergovernmental (i.e., between adjacent urban communities or between the city and
the suburban communities) and intragovernmental (i.e., between policy sectors,
such as land use, environment, energy, and employment, usually under different
institutional authorities) concertation. These evidences point to a general problem
concerning the jurisdictional structure of governance institutions in spatial-related
policies, that is, traditional administrative boundaries are now dysfunctional and cause
discontinuity of decision-making.
As pointed out by Raffestin (1980, p. 129), territory is a constantly changing
reality, territory is generated out of space; whatever the level, it is the result of the
actions of a syntagmatic player (i.e. a player implementing a programme). By
appropriating a space in either a substantive or abstract form (for instance by
making a representation of it), the player territorialises space.
Functionality lies thus in a virtual reality a relational territorial dimension
dened by economic and social interactions and lying on a concept of continuity of
interdependencies, materially represented by the services provided to communities and
by their collective interests. Yet, local legitimacy forces to neglect these interdepen-
dencies with different arguments. A rst argument often used is that local governments
respond to demands of local population and are presumably expected to neglect the
benets their citizens obtain from other communities, even if costs are imposed to those
communities. A second argument is that a consequent cream-skimming, related to
competition in households and industrial location, could lead to a lower efciency in
the application of public money and risks of overinvestment in transport infrastructure.
In the second chapter, we have expressed the boundary of a system as the area
limit within which the decision-taking process of the system has power to make
things happen, which involves the choice of scope and aggregation levels. Although
we consider this denition to be correct from the point of view of systems dynamics,
we admit that it is an abstract and subjective denition way to provide a structured
territorial denition of the urban mobility system.
The system boundary is indeed a uid denition, given its dependence both on
the judgment of the observer on what she takes as being the system and on two kinds
of constraints, legal and technological competencies, that is, respectively, the actors
territorial and spatial competence, and the actors technical capabilities. Besides, in
reality, what we have is two sets of decision-makers, one set in the supply side that
even is the case of a good organizational framework, only rarely addresses the
decision at system level, and another one in the demand, which is characterized by
hundreds of disorganized decisions.
As recognized by Viegas (2002a, 2002b, pp. 12), the denition of the
contemporary city is ambiguous and complex, the former because it lies on relations
of belonging, regarding territory, people, and even institutions. Complexity, in
turn, grows from the spread to peri-urban areas, which imposes a diversity of spatial
relations, with each citizen very often relating with two urban areas or having
stronger links to other cities than the ones where he formally (i.e., administratively)
has his residence.
The territorial denition of the urban mobility system is indispensable to dene
the boundaries within which the power of institutions that are in charge of its
governance is dened. Three issues must be considered for an efcient territorial
denition: the systemic reality, the need to compare and transfer of solutions, and the
nancial manageability of the system. The concept of system by itself is required to
ensure that the relevant analysis considers the reality observed as part of a set in
which the whole must be referred.
Urban mobility systems are nested in an upper system, which is the respective urban
agglomeration, as dened by their land use pattern and jurisdiction of its political-
administrative institutions. As the size of an urban agglomeration is a consequence of
its economic organization and of the opportunities for social relations, we may
conclude that there is a relation between the dimension of the urban area and the factors
inuencing the need for mobility.
The need for comparability and assessment of potential for transferability of
solutions is a serious handicap felt by both the research and the practitioners
communities. In fact, we can observe in our set of cities that comparability and any
further performance assessment of mobility systems are hindered, and sometimes
even result unfeasible, due to a wide diversity of base concepts for the administrative
denition of urban boundaries, thus lacking a common reference unit.
For mobility management purposes, the basic concept should be the urban region or
urban basin served by the same mobility system. However, the concept is missing in the
administrative organization of most countries, what precludes effective management of
any nancial process due to absence of institutions with a jurisdictional capacity
adjusted to the territorial insertion of the mobility services.
Along the above-dened relational perspective of the territorial dimension, we
also have a formal institutional territory, complementary to the previous one, and
represented by the land occupancy by countries, regions, and municipalities. Some
authors (e.g., Sack, 1986, p. 160) consider institutional territory as no more than an
imperfect arrangement due to the following main reasons:
rigidity in which it is organized;

difculties in establishing optimum scales (e.g., nancing services and infra-
structures shared by a number of communes, cantons, or even countries, or the
opposite, i.e., one citizen belonging to several jurisdictional units10); and
the fuzziness in dening institutional and legal engagements for tackling cross-
border problems.
These arguments support the conclusion that the static concept of institutional
territory is necessary but clearly insufcient to deal with the dynamics of modern
societies. Relational territorial dimension is required to deal with the networkwise
dynamics of urban communities. In brief, the territorial denition of the urban
mobility system requires the use of variable institutional geometry so that the
interaction between land use and mobility can be effectively achieved. Institutional
design should thus be guided by a network logic providing service-related (and
associated decision-making) continuity in the administrative and jurisdictional setting
of the institution holding responsibility for the territorial management of urban
mobility.
10. Cost 332 research reports that the Swiss national census carried out in 1990 shows that one person in every
two works in a different commune from that of his or her residence, besides the possible additional mobility
element related to the consumption of goods and services (EC, DGT, 1998, Cost 332, p. 52).
3.2.2. Collective Interest
One of the pillars of public service practice lies in the postulate that governments and
their agencies are in a good position to interpret the (more relevant) collective needs
of society, and therefore, their action is always driven by what they think is best for
the citizens. However, this is not always the case as formulated in the Theory of
Public Choice (Truman, 1995, pp. 3240). According to this theory, there are
failures from the state level that reect mostly on the offer side. These result from the
deviation of state agencies from those public service objectives to satisfy objectives of
their own agendas. Contrary to what is usually formulated by more radical
approaches to public service concept, we believe that systems organized to provide a
public service can have specic steering mechanisms so that agents involved in the
provision of public service may simultaneously stand for their own (corporate)
interests and for the ones of the communities they serve.
In his theory of justice, Rawls (1999, p. 10) observes that those who engage in
social cooperation choose together, in one joint act, the principles which are to
assign basic rights and duties and to determine the division of social benets.
According to Rawls, political institutions have the capacity to generate a just
society, but there is no just society generating just institutions (Rothstein, 1996,
pp. 133166).
The main problem of this logic of distributive justice is thus the choice of the social
system itself and of its background institutions that determine how the different social
precepts are balanced. Institutions are enduring entities, they cannot be changed at
once, and some of them do get an almost sacred statute, such as constitutions where the
most important and basic rules of a society are bound together.
As stated by Rawls (1999, p. 15), the relevant construct is that certain precepts
tend to be associated with specic institutions political institutions. In the narrow
sense, these can be dened as formal arrangements for aggregating individuals and
regulating their behaviour through the use of explicit rules and decision processes
enforced by an actor or set of actors formally recognised as possessing such power
as (Levy, 1990, p. 405) dened it. In parallel, Rothstein (1996, p. 157) identies four
basic types of political institutions:
The rule making type that makes collectively binding decisions about how to
regulate common interests;
The rule applying type that implements previous decisions;
The rule adjudicating type that takes care of individual disputes about how to
interpret the rules laid down by the rst type institution; and
The rule enforcing type that takes care of and punishes insiders and outsiders
who disrespect the rules.
Much has been discussed about the different ways and intensities for the existence
and intervention of the public sector in the markets. Irrespective of the opinions
taken regarding direct state intervention in the provision of services or products,
public policy is generally recognized as needed to guide, correct, and inuence
market behaviors mostly due to the following circumstances that are general market
requisites:
The claim that market mechanisms lead to efcient use of resources is based on the
condition that competitive factors and product markets do exist. This logic entails
the implicit assumptions that market entry is easy and both consumers and
producers have full market knowledge. These conditions fall in the policybusiness
interaction area but can only be secured by policy regulation.
Similar assurance is required whenever decreasing costs patterns give room to
inefcient competition.
A regulatory and legal structure is needed to protect and enforce contractual
arrangements and exchanges required by market operations. This structure can
only be secured by political institutions.
Even if all the previous conditions are met, there are still problems arising from
externalities caused by the production and consumption of some goods and
services and requiring compensatory mechanisms that can only be activated
through public sector action.
In addition, there are other social and economic objectives such as employment
and rate of economic growth, which cannot be assured by market systems, and
again, public policy is required to intervene in securing these objectives.
Four government branches can be envisaged to answer these needs, each of them
consisting of various agencies, or departments or activities (depending on the
government structure) charged with preserving certain social and economic conditions.
These branches are to be understood as different functions, such as allocative,
distributive, stabilization, and nally the concertation, coordination, and conict
management between the previous functions (Musgrave & Musgrave, 1984, pp. 318).
Although these divisions have no match with the government organizational setting, it
is desirable to consider the essence of these functions in the conguration of the
network of policy institutions.
The allocation branch has the responsibility to keep price mechanisms within the
principles of competition and to prevent upsurge of unreasonable market powers. As
Musgrave (1984, p. 8) observes, the benets from social goods are not vested in the
property rights of certain individuals, and the market can not function, that is, there
are cases where the signaling system between producers and consumers does not
work properly, not only because social exclusion is undesirable but also because it is
frequently impossible or very expensive to apply. Therefore, this branch is
responsible for identication and correction of efciency deviations.
The distribution branch is responsible to preserve justice in the distributive shares by
means of taxation and often also through adjustments in property rights. Within this
stream of action, levies and regulation are applied with twofold objectives: the rst is to
correct the distribution of wealth envisaging the provision of fair opportunities, and the
second is to withhold revenues that will be later applied as nancing sources to secure
provision of public goods and services. Taxation techniques provide a wide diversity
of solutions, which will lead to differentiated effects that are addressed in questions
of political judgment and analysis that will not be discussed here as they largely
overcome the objectives of this work.
As observed by Rawls (1999, p. 247), The two parts of the distribution branch
derive from the two principles of justice. The taxation of inheritance and income at
progressive rates (when necessary), and the legal denition of property rights, are
to secure the institutions of equal liberty in a property-owning democracy and the
fair values of the rights they establish. Proportional expenditure (or income) taxes
are to provide revenue for public goods, the transfer branch and the establishment
of fair equality of opportunity in education, and the like, so as to carry out the
second principle. To achieve these objectives, Musgrave (1984, p. 13) postulates
that efciency in the broad term must allow for the two following concerns: (1)
any given distributional change should be accomplished at the least efciency
costs, and that (2) a need exists for balancing conicting equity and efciency
objectives.
The stabilization branch is meant to solve some of the additional problems left
from market mechanisms, such as balanced employment and steady growth rates.
This branch acts always in synchronized partnership with all the others. Together
with the allocation branch, they maintain the general efciency of market economy.
Together with the distributive branch, a transfer function is developed where social
needs are taken into account and priorities are assigned with respect to other
competing claims. In these, attention must be paid not only to active claims from
different sectors of economic and social activity but also to time-related effects of
implemented policies, in particular, issues of justice between generations.
Finally, it is through policy concertation and coordination of planned actions of
the different economic sectors that priorities are dened and the concept of public
service is made operational for the different sectors. Despite the autonomy of these
sectors, each country tends to adopt a common philosophical approach to the
concept of public service, transversal to all economic sectors. To this respect, we can
observe two main political streams in what concerns the public service character of
transport sector (Hensher & Macario, 2002, p. 351), these are the so-called Code
Napoleon and the Anglo-Saxon philosophy.
The Code Napoleon approach, best exemplied by the French reality, which
takes transport as an input into a wider socioeconomic and political framework, in
which case the sector should have a strong state intervention and thus the full
application of public service obligation, usually materialized in the obligation to
operate, the obligation to carry, and the tariff obligation.
In all countries observed and where this philosophy stands (EC, NEA, 1998c,
p. 97), it was possible to conclude in general terms that these obligations are meant to
ensure an obligation of providing a service that is expected to satisfy xed standards
in terms of continuity, regularity, and capacity, as well as to accept and carry
passengers, and in some cases also goods (e.g., transport to Islands, like the case of
the Portuguese archipelagos of Azores and Madeira, or the Spanish archipelago of
Canaries), as well as to respect predened limitation regarding fares, sometimes with
price levels politically set. The base rationale behind this approach is that public
service obligations guarantee the satisfaction of the population mobility needs that
otherwise would not be properly provided by the market.
The Anglo-Saxon philosophy, best represented by the UK experiences, takes the

opposite rationale, that is, transport is like any other sector in the economy and as
such should be provided as efciently as possible; thus, private participation and
commercial approaches are the dominant orientation, and markets are thought to
provide better performance to the transport systems than the one obtained though
government intervention. In practical terms, most countries found a mixed solution
in between these two philosophical streams, and we can nd several different ways of
applying the public service concept to the transport sector.
3.2.3. Symbiotic Characteristics in Agents Decision-Making Processes
As stated by North (1991, p. 6), a major role of institutions in a society is to reduce

uncertainty by establishing a stable (not necessarily efcient) structure to human
interaction. However, it is also this interaction structure that enables the creation of
constraints in the relation between actors, largely materialized by codes of conduct,
norms for social and institutional behavior, and conventions. Literature usually
divides these rules into formal and informal constraints. Formal rules include
political (and legal) rules, economic rules, and contracts (North, 1991, p. 46).
Several studies provide empirical validation of the phenomenon of spreading of
informal constraints that, as Sugden (1986, p. 54) dened, these (y) are rules that have
never been consciously designed and that it is in everyones interest to keep. Three
categories of informal constraints are identied by Ellickson (1987, pp. 67100 in North,
1991, p. 40), as a consequence of the need to repeatedly coordinate human interaction.
These are (i) extensions, elaborations, and modications of formal rules; (ii) socially
sanctioned norms of behavior; and (iii) internally enforced codes of conduct.
Formal rules are often used to complement and improve the effectiveness of
informal rules. They may reduce the need for information, monitoring, and even
enforcement cost as well as contribute to clarify the legitimacy of some actions or
decisions. As demonstrated by Weingast (Weingast & Marshall, 1988, pp. 132163 in
North, 1991, p. 40), the power of US congressional committees, which is not
explained by the formal rules, is a result of a set of informal unwritten constraints
that have evolved in the context of repeated interaction (exchange) among the
players. These constraints evolved from the formal rules to deal with specic
problems of exchanges and became established as recognized institutional constraints
even though they were never made a part of the formal rules.
One decade earlier, Ostrom (1971, p. 57) showed that the constitutional structure
devised in 1787 was conceived not only to facilitate exchange but also to raise
transaction costs to that type of exchange, thus promoting the interest of factions.
Therefore, costs of exchange and enforcement have long ago been used to foster
institutional interactions or, at the opposite, to raise articial barriers.
Interaction and exchange between institutions are based on different sorts
of social or economic agreements, which provide the framework that enables
organized (or precodied) and stable interaction, given the interests of the parties.
Principalagent theory is at the core of the process to structure this interactive
behavior of agents, although it is worth discussing other complementary formulations.
An argument developed by Margolis (1982, p. 154) was that individuals possess

not one but two utility functions, one governed by the self-interest and another
purely social. In a similar line of thought, Viegas (2002a, 2002b, pp. 23), as already
referred in the rst chapter of this book, refers to a societal evolution leading to the
permanent simultaneous involvement of each individual in several groups, with
different agendas and interests. Parallel to this, Sugden (1986, p. 173) stands for the
argument that a convention acquires moral force when almost everyone in the
community follows it, which he denes as a morality of co-operation. The dialectic
strains between formal and informal rules can be softened by a mission statement
that is clearly understood by all internal and external stakeholders. By providing
direction and legitimization, it also motivates behaviors according to the stated
purpose, irrespective of being developed through informal or formal channels, thus
minimizing the deviation from the institutional goals and containing conformity to
the main values and strategies. In addition, direction and legitimization also
contribute to a more efcient institutional communication.
In a different approach, North (1981, p. 7) abandoned the concept of efciency of
institutions as he considered It was possible to explain the existence of inefcient
institutions, but why wouldnt competitive pressures lead to their elimination (y).
Following this reection, he rmly states that Institutions, together with the
standard constraints of economic theory, determine the opportunities in a society.
Observation of reality conrms the existence of several institutions of recognized
inefciency, which have been legally or administratively attribute a mission and are
not subject to any type of competitive pressure. These institutions are often kept due
to the high cost of their dissolution.
It is also known that collective wealth maximizing behavior usually requires high
cooperation between players whenever they possess good levels of information about
the other players past performances and the number of players is relatively small.
Whenever rivalry between agents is absent, the more interaction exists between players
the more prone they are to cooperative behavior as each player gains knowledge about
the others through repeated interaction, and therefore, its perception about the
uncertainty of behavior of others is reduced. In cases where rivalry exists, then the
good levels of information are used to better understand the strategic movement of the
other so that predatory strategies can be more effective.
Therefore, cooperation seems to be a strategic game of knowledge because each
individual is expected not only to know the preferences of others but also to be aware of
how much the others know of his own preferences. As Schoeld (1985, pp. 1213) states,
the theoretical problem underlying cooperation and the analysis of community lies in the
following formulation: what is the minimal amount that one agent must know in a given
milieu about the beliefs and wants of other agents to be able to form coherent notions
about their behaviour and for this knowledge to be communicable to the others?.
To the previous idea, Hardin (1982, p. 57) adds that the difculties of collective
action depend not just on the size of the groups but also on the costbenet balance
that results from the analysis made by each agent by comparing its current position
and potential with the future, that is, his own end state after cooperation. By
bringing costbenet balance to the discussion, we are driven to the rationality of
game theory where different individual strategies alter the pay-off of the disputing
players and determine their behaviors, although no objective knowledge exists on

how transaction costs affect strategic behavior and how, in turn, those costs are
affected by institutional structures.
Returning to the principalagent construct, and considering our adoption of the
Lewotin posit, we must still address the inuence of context over individuals acting
as principals, which we will be addressing with more detail in Section 3.3.1. In a
rather simplistic, or even na ve, formulation of the behavior of agents, and so almost
opposite to game theory, Mills (1972, p. 484 in Stigler & Becker, 1977, p. 81) declares
that It is not true that the actions even of the average rulers are wholly, or anything
approaching to wholly, determined by their personal interest, or even by their own
opinion of their personal interest (y) I insist only on what is true of all rulers, viz.,
that the character and course of their actions is largely inuenced (independently
of personal calculations) by the habitual sentiments and feelings, the general modes
of thinking and acting, which prevail throughout the community of which they are
members; as well as by the feelings, habits and modes of thought which characterise
the particular class in that community to which they themselves belong (y) They
are also much inuenced by the maxims and traditions which have descended to
them from other rulers, their predecessors: which maxims and traditions have been
known to retain an ascendancy during long periods, even in opposition to the private
interests of the rulers for the time being.
The analysis by Mills lies very much on the behavior of the ruler within the agent,
which brings to the discussion the interests of the agent (as collective) versus the
interests of the individual (decision-maker within the agent). To this respect, it is
commonly accepted in both social and economic sciences that people understand the
environment in which they are involved through the lter of preexisting mental
constructs, which are their main tools for problem formulation but also their main
obstacle to a natural congeniality (in the sense of cross-fertilization) that should
simply accrue from contact with new information. Hereafter, we will consider
individual behavior bounded by the following attributes:
information, dened by the level of knowledge on self and other players;

degree of complexity of a given problem, largely dened by the number of
interacting variables;
contextual inuence, for which the determinant factors are socioeconomic
behavior rules and institutional interaction;
class afnities, the afnity attribute being an indispensable factor to induce11
cluster formation; and
time frame, as reected on inherited mental constructs.
Along the years, a number of researchers brought together their work aiming to
explore afnities between the underlying features of genetic survival, evolutionary
11. In this work, induction means stimulation that produces a determined empathy calling up a particular
class of behavior or line of thought.
models among animals and human beings, and organized human behavior. A
renowned scientist in this eld (Dawkins, 1989, pp. 78107) made the parallel with
the genetic pool, where good and bad genes are part of a team and evolution is seen
as the process through which some genes replicate in a faster and more numerous
way than others. That is, evolution is a stage process and develops through the
differential survival of genes within the pool.
Almost simultaneously, Hirshleifer (1897, p. 221) compares biological evolutionary
models with socioeconomic ones and states that (y) the evolution of a single nations
economy is the result of changing relations among populations of individuals, trading
units, and the like. Evolutionary models represent a combination of constancy
(inheritance) and variation. There must be an unchanging as well as changing element,
and even the changing element itself must be heritable if a system can be said to evolve.
In biological evolution, the emphasis is upon differential survival and reproduction of
organismic types or characters from one generation to the next. Here the constancy is
due to Mendelian inheritance of permanent patterns of coded genetic instructions
(genes). Variation stems from a number of forces, including internal mutations of these
instructions (genetic copying errors), recombination of genes in sexual reproduction,
and the external pressure of natural selection. Socioeconomic evolution mainly
concerns the differential growth and survival of patterns of social organization. The
main inheritance element is the deadweight of social inertia, supported by intentionally
taught tradition. As for variation, there are analogues to mutations (copying errors as
we learn traditions). Also, natural selection is still effective. Finally, imitation and
rational thought constitute additional non-genetic sources of economic variation.
Institutions can be attracted for cooperation either when a positive benet will
accrue for both parties or in two other ways, as explained by some authors, who have
explored the potential of principalagent theory applied to punishment incentives
(Milgrom, North, & Weingast, 1990, p. 23), that is, through a communication
mechanism that enables to know when punishment is needed and through the
provision of incentives for individuals that carry out punishment when called to do so.
The critical question seems to be who should be in charge of this enforcement. North
(1981, p. 58) states that while the transaction costs of voluntary agreements would be
enormous, there are important economies of scale in policing and enforcing agreements
by a polity that acts as a third party and uses coercion to enforce agreements.
However, in a diversied context such as the one involved in urban mobility
systems, largely characterized by problems of organized complexity,12 doubts can be
raised whether this entity can be set up only with third parties without any state
interference or alternatively should be assumed by a public body, lying above the
12. As dened by Weaver (1948, pp. 538539), problems which involve dealing simultaneously with a
sizable number of factors which are interrelated into an organic whole. They are all, in the language here
proposed, problems of organized complexity, at the opposite of problems of disorganized complexity,
which are the ones in which the number of variables is very large, and one in which each of the many
variables has a behavior which is individually erratic, or perhaps totally unknown. However, in spite of this
helter-skelter, or unknown, behavior of all the individual variables, the system as a whole possesses certain
orderly and analyzable average properties.
economic and social agents with a direct stake in the mobility system, and so acting
as a sort of dissipative structure between the UMS and the surrounding environment
where mobility needs are born.
As North also points out, some uncertainty can accrue regarding the capacity to
have state entities behaving as impartial third parties and being able to assure that
fairness conditions between players are not compromised. This is particularly
relevant whenever dealing with public services because citizens, although very
sensitive to the stability and performance of these services, simultaneously play the
voter role, that is, that of the principal in the relation with state institutions.
The existence of a system implies that agents are functionally related, and therefore,
any change in an agent will affect all the other agents in the same network of
interdependency to a considerable degree, this being the symbiotic condition of
the system. As Chisholm (1992, p. 7) puts it, there is a disjuncture between the
organizational character and the functional properties of the system. Although the rst
seems to transmit an autonomous self-sustained structure, the latter imposes
the symbiotic attribute in the interactions between agents. The lack of clear hierarchical
structures between entities related by these symbiotic effects weakens the capacity of any
coordination, as traditionally understood in the sense of unity of guidance, resulting that
the guidance (or leadership) of these entities has to be done through concertation efforts
instead, which corresponds to a more sophisticated form of institutional leadership.
Concertation13 means placing elements in an adequate position relative to each
other, that is, bringing together the parts of a whole in some kind of order, not
necessarily a hierarchical formal structure. Therefore, in brief, we dene concertation
as a combination of a number of elements, actions, and processes. Following Simon
(1997, pp. 116139) representation,14 there are three main elements to consider to
achieve a well-succeeded concertation process:
obtain information on goals and respective causeeffect relationships and

devise and communicate a plan of action for the parties making the resulting effect
clear for the whole; assure acceptance by the parties, which is only achievable
through community of purposes, or as Barnard (1971, p. 27) designated
community of interest. These communities are better developed through informal
mechanisms than through formal ones, although formal mechanism can be an
important instrument to stimulate and support the upsurge of these communities.
Table 3.3 summarizes the difculties and failures of the concertation process, which
are strongly related with information ows and levels detained by the parties involved.
13. A number of authors (such as Simon and Chisholm) dene this as coordination. However, we recall
footnote 26 in Chapter 1, where we have dened coordination as a steering process above action and
processes. Therefore, in this study, concertation stands for combined action and processes as dened in the
main text.
14. Note that we disagree from Simon who uses the concept of coordination instead of concertation as we
understand it.
Table 3.3: Types of failure in the concertation processes.
Element Permanent failure Occasional failure
Development of plan No provision for

of action building coalitions
No provision for
building community of
interest
Communication of Mismatched channels Slow channels
the plan of action Incomplete channels
Acceptance of plan of No provision for Inability to expand zone
action building coalitions of acceptance
Collection of essential Mismatched channels Slow channels
information Inability to collect Mismatched channels
sensitive information
Formal structure
incomplete
Source: Adapted from Chislom (1989).
As Hayek (1945, pp. 521522) concludes, The peculiar character of the problem of a
rational economic order is determined precisely by the fact that the knowledge of the
circumstances of which we must make use never exists in concentrated or integrated
form, but solely as the dispersed bits of incomplete and frequently contradictory
knowledge which all the separate individuals possess. The economic problem of society
is thus not merely a problem of how to allocate given resources (y) It is rather a
problem of how to secure the best use of resources known to any member of society, for
ends whose relative importance only these individuals know. Or, to put it briey, it is a
problem of the utilisation of knowledge not given to anyone in its totality.
3.3. Articulation between Agents: Balancing Policy Consistency

and Stakeholders Requirements and Expectations
3.3.1. Policy Processes with Stakeholders Participation
For many years, the implementation of mobility policies followed a logically structured
cycle of allocative decisions based on a rational approach that is usually illustrated with
four main steps: specication of objectives; development of alternatives by which the
objectives may be accomplished; evaluation of the consequences of each alternative;
and, nally, selection of the action that maximizes the net benet according to decision-
makers criteria and valuations.
This conceptual approach assumes a unitary decision-making or considers a group
acting as a unit and ignores situations of conict, which arise whenever social activities
are at stake and the different interests of societal groups are confronted. However,
policy denition is a much more demanding process as it requires the ability to dene a
problem lying on a multiperspective environment, to draw arguments from a diversity
of sources, to adapt the argument to the audience (also a multistakeholder audience),
and nally to educate public opinion to achieve consensus around policy objectives and
design (Majone, 1989, p. 87).
Policy denition thus requires understanding the diversity of personalities repre-
sented by the stakeholders, the interests they represent, and the interdependence
those interests impose in the different stages of policy denition, given the subjective
nature of the citizens decisions, in particular, when called to express their public
opinions, for example, through voting mechanisms (Lau and Redlawsk in Kuklinski,
Paul, Jennifer, & Robert, 2001, p. 143).
Despite the criticism surrounding the staged approach to the policy process, that
is, the policy cycle, this is a model for policy analysis not devoid of advantages,
in particular, for its ability to provide a departure framework of thought, which
represents an illustration of reality, reducing its complexity to a manageable
dimension, and perfectly suitable for the case of the denition of the conguration of
an urban mobility system. Notwithstanding, the argument of many authors can be
easily accepted that the real world is far more complicated and not composed of tidy,
net steps, phases, or cycles (Parsons, 1995, p. 243). Less consensual are the criticisms
that the model fails to consider interaction between actors, in particular, between
stakeholders and the different stages of policy life cycle.
This policy cycle reveals a good level of adaptation when applied to decision
processes related to urban mobility systems.15 In particular, the incorporation of
interaction with stakeholders seems to be possible from the practical point of view16
as well as theoretically robust. The several stages are composed of the following
steps:
Stage 1: Denition of problems and objectives.

Perception of problems and dynamics of environment (by decision-makers);
Disseminate perception of problem so that stakeholders gain awareness;
Denition of objectives; and
Gaining acceptance from stakeholders in setting the goals based on problem
perceptions (including feedback and possible redenition of objectives).
Stage 2: Policy design
Identication of alternative solutions/responses to the problem;
Planning concepts/future scenarios, including determination of favorable and
unfavorable patterns;
15. The author presented a similar cycle applied to decisions on transport pricing issues in the workshop of
the thematic network TRANSTALK, included in the 5th RTD program of the European Commission,
which was later object of publication (Viegas & Macario, 2003a).
16. The author followed this cycle in the application of the model proposed in this study to the reality of
Brazil, and no difculties or major dysfunctionalities were found.
Selection of policy instruments;

Gaining acceptance from stakeholders for the selection of instruments, which
can reach the preferred dened patterns, assuming that each policy/instrument
has its own goals and effects; and
Assessment of expected impacts and stakeholders reactions (including possible
feedback for redenition of policy instruments).
Stage 3: Policy implementation
Deployment of policy instruments including detailed design of legal, organiza-
tional, nancial, and operational instruments for the effective implementation of
the policy;
Identifying and bringing together implementers, that is, all individual and
institutions involved in the setup of policy organization, including gaining
acceptance from stakeholders for that organization; and
Denition of an implementation plan, setting up resources, time, and space
scales.
Stage 4: Policy evaluation
Checking stakeholders reactions;
Monitoring and evaluation of implementation;
Evaluation of policy impacts, side effects, and overall outcome, by measuring
the impact and policy success; and
If necessary return to stage 2 for ne-tuning (feedback mechanism) and
reconsideration of initial stages, whenever evaluation results in low effectiveness.
Despite its inclusion of an interaction process, this staged approach can still be
criticized by oversimplifying the possible involvement of multiple levels of
government in the decision process, which can result in the existence of interacting
cycles to consider the negotiation and bargaining process between those institutions.
In addition, this approach is also unable to reect policy motivation to move from
one stage to the next, neither to assess potential negative political balance that might
lead to policy disruption in the middle of the process.
The reection over these aws will certainly lead us to the conclusion that the best
analytical model would be a multiframed one, although we must also recognize that
its main disadvantage would be the degree of complexity and the strong growth of
interacting cycles that would considerably reduce our capacity of understanding the
decision process entailed in each of its steps.
When applying this cycle in the real world, what happens is that we execute the
stakeholder consultation step entailed in each stage as many times as the number of
group of stakeholders we have, which represents the multiframe image in the
theoretical modeling. While moving from one group to the other, we enter into an
incremental learning process on public acceptability of the proposals being discussed,
and, quite often, in between those incremental steps, we make adjustments based on
trade-offs between the different groups of stakeholders.
In what concerns evaluation, we recall Parsons (2000, p. 5) concept that evaluation
is fundamentally a learning process, and different frameworks of valuing inevitably
generate different ways of thinking about the problem, that is, the multiperspective
reality of policy building and decision-making in public domains. Therefore,
evaluation is focused on the task of integrating knowledge (i.e., designing methods,

processes, and institutions, which can best serve to clarify competing arguments)
aiming to answer the question Who gets what knowledge when and how, and,
ultimately, whose values get to dominate.
A number of evaluation frames can be selected to accompany the staged process
dened earlier, although it must be stressed that the selection of the frame must be in
line with the objectives of the different stages, which is why no best evaluation
framework seems to exist for the overall policy cycle that drives urban mobility, but
instead, a more appropriated frame should do the job in each stage.
A multiframe method for evaluation will set the grounds for having these tools as
self-learning instruments for policy denition and ne-tuning. Evaluation as
calculating and distributing costs and maximizing benets and social welfare can be
used in stage 2, whereas evaluation as the measurement and monitoring of performance
is to be preferred for stage 4.
Whenever stakeholders reactions are to be assessed (in the different stages of the
model), then a pragmatic interpretivism frame seems to be more effective, because it
adopts evaluation as education and empowerment of all stakeholders through
dialogue and deliberative democracy, rejecting the positivist and objective claims
of other frames in favor of an argumentative approach. This multiframe path applied
to the denition of the different components of the urban mobility policy (e.g.,
quality levels, choice of technologies, and pricing) should facilitate the nal and
aggregate policy measure of evaluating through economic valuation (e.g., value for
money).
Therefore, evaluation should allow markets to facilitate experimentalism,
learning, and self-organizing in conditions of uncertainty and complexity, just like
in the real world. While the traditional model of rational choice assumes that people
attempt to optimize their decisions within given rules, and that the underlying
principle of maximization of expected utility only guarantees that the choice is
consistent with the decision-makers valuation of the probabilities and utility of the
consequences of the various alternatives at stake. As suggested by Majone (1989,
p. 125), the policy process becomes much more understandable if one assumes that
the actors (in their exercise of strategic behavior) view the rules of the policy game as
possible targets of political action, striving for changing those rules in their favor. In
designing and implementing urban mobility policy, the more society evolves, the
more we must understand citizens as active participants and with strategic behavior,
whenever represented by collective organizations.
Being so, assessment of stakeholders reactions cannot possibly be considered only
at the stage of policy implementation; on the contrary, the potential winners and
losers status change along time. That is, as we already said, while introducing a policy,
we have to consider not only the absolute effect over each stakeholder personality at a
given moment but also the marginal effects on the social and economic statute of those
personalities along time, meaning that the impact of change in the status of the receptor
of the policy effects must be continuously assessed in the short and long term assuming
a dynamic evolution of that status.
This perspective allocates a dominant importance to the argumentative capacity at
all stages of the political cycle, and even to the public acceptability phenomenon
itself, because as Dewey (1927, in Majone, 1989, p. 119) once observed, the most
important thing about popular voting and majority rules is less the current outcome
of the voter choice than the fact that the electoral process compels prior recourse to
methods of discussion, consultation, persuasion, and the resulting modication of
views to accommodate the opinion of the minority.
3.2.2. Market Access Regimes
Satisfying mobility needs in an urban area involves the provision of several different
complementary services that result in a chain able to ensure safe and uid movement
of citizens between the several possible origins and destinations. This means that
mobility represents the outcome of an articulated effort that intertwines pedestrian
movements, movements with private individual motorized transports, and move-
ments with public collective and individual transport. That is, the supply side of the
mobility systems gathers mobility services provided directly or indirectly by the state,
by the market, and also the ones consumed on self-provision, for all of which there
is the need to ensure the adequate infrastructure. The balance between the utilization
(or consumption) of these different means represents a key element of the sustain-
ability of urban areas and as such must be subject to careful thought.
From the four networks that form the urban mobility system walking, cycling,
private motorized transport, and public transport, with associated infrastructures
the public transport network is one of the components of the system where the
market access regimes represent an instrument of articulation between agents and
where regulatory issues gain prominence. Another element where market access
issues can be raised is in the access to infrastructure (e.g., limits to circulation of some
vehicles in certain roads), although here the constraints to market access are not
related with reason of competition or market contestability.
Depending on the regulatory option, public transport services can be performed
directly by the transport authority, contracted out to an operator (private or public)
by direct negotiation or through a tendering procedure, or directly in the market by
an operator in deregulated regime. The system design (i.e., the planning) is also a
service on its own and can be outsourced with or without tendering, although
it is usually seen as a separate market from the one on provision of transport
services.
In ISOTOPE research, a global classication of regulatory and organizational
frameworks for public transport was consolidated and further rened in
MARETOPE. Figure 3.4 is based on the ndings and formulations done in these
research projects and illustrates the global classication of these regimes in what
concern supply of public transport services. The main distinction is given by the
entrepreneurship variable, that is, the dichotomy between authority-initiated or
market-initiated regimes. In the rst, authorities have the exclusive right to initiate
services, and any direct attempt of market entry is deemed to be illegal according to
the legal framework that rules the way services are allowed to be provided.
Figure 3.4: Regulatory and organizational congurations. Source: Adapted from

Van de Velde (1999, p. 149).
Consequently, in these cases, all service supply results from a planning initiative
from the authorities, which is the current situation17 with urban public transport in
France, Belgium, and also formally in Portugal. In the latter, that is, market-initiated
regimes, service supply is based on spontaneous market entry of operators, which
results from a normal market process subject to some regulatory evaluations at
the moment of entry, such as technical, legal, and economic ability to perform
the operation, based on legally preestablished principles and criteria. This is the
current situation in Great Britain, Germany, and Netherlands, although each of
these countries has opted for different categories and processes for verication of
professional and social capability of service provision. In these cases, authorities are
still entitled to develop planning functions, but operators are legally allowed to enter
the market with new services as long as the relevant criteria for acceptance (by the
authority) are fullled.
17. End 2003.

In both cases, with the exception of the pure open market regime, it is possible to
maintain competitive tendering for parts of the whole of the services because, as Van
de Velde (1999, p. 150) stresses, competitive tendering is merely a selection mechanism
in the context of outsourcing. Indeed, it is simply a method of organization of
production available to any service supplier whatever the regulatory regime and
organizational form used.18 Therefore, the alternative regimes to consider are public
production systems, limited access (also known as controlled competition), and
open entry (subject only to demonstration of professional capability), as represented in
Figure 3.4.
This sequence corresponds also to a spectrum of market openness, which is
embedded with an increased competitive pressure to enter the market that can be
observed when we move from public control regimes to open entry ones. However,
market access is not the only way of imposing competitive pressure as this can also
be achieved in all regimes through contracts and monitoring dispositions. To this
respect, we also observe that when moving from public to private capital
ownership, the competitive pressure also increases by way of those management
tools.
Within the authority-initiated regimes, it is possible to distinguish two categories of
regimes, the publicly owned regime and the concession to a private or public company.
Publicly owned regimes can still be divided into two forms, public management where
vehicles and other installations are directly run by the public administration, which is
common in small cities, such as the case of Carcassone in France and Barreiro in
Portugal, and delegated management, where the authority makes the assets available to
a private operator company to whom the management of the network is delegated
through contractual arrangements that can be very diversied according to the way
commercial and operational risks are shared, and to the degrees of freedom that can be
given in the service design as discussed in Section 3.3.4.4. As we have observed in
ISOTOPE, QUATTRO, and MARETOPE researches, France is the country where
more of these types of contracts have been identied.
The alternative category for authority-initiated systems is the concession. Here,
the authority selects a private company to operate public transport services in a
route, area, or in a network, and the concessionaire is normally responsible for the
vehicles and installations through ownership or leasing. Depending on the legal
framework in place, this company is selected either directly (not any longer allowed
in the EU, but still a practice is other places) or through negotiations after a short-
listing procedure or through competitive tendering.
Market-initiated regimes are largely divided into two categories, open entry
regimes where the so-called on-street competition exists, of which the best example is
the urban transport in Great Britain outside London (very often incorrectly
18. There is even evidence of contracting out experiences in free competition environment, as reported by
Van de Velde (1999, p. 150); such is the case of services general interest not offered by the market as
established in Southern Vectis, House of Commons (1995, p. 227).
designated by deregulated regimes), and restricted authorization regimes19 where

licensed operators are given the right to apply for an authorization to exploit a
certain service with some degree of exclusivity, which varies enormously from one
country to the other and even within the same country from one city to another or
between services of different type, although in all of them the operator is, for the
period of time of the authorization, at least partly protected from competition. One
of the main drawbacks of this regime that has been pointed out by some authors
(e.g., Van de Velde, 1999, p. 151) is the risk that this protection, if extensively
applied, will eliminate market contestability and develop regulatory capture.
Open entry regimes in turn have been associated with some degradation of service
quality, despite their efciency advantage (European Commission (EC), TIS.PT,
1998b, p. 194). The need to address this problem was at the root of the development
of quality partnerships in Great Britain. Through these agreements, seen as light
touch regulation (Carr, 1997, p. 6), a number of rules committing both authorities
and operators can be established, such as provisions for service coordination,
integrated ticketing, obligations to use vehicles accessible for handicapped, and bus
priority lanes. However, these agreements have also been condemned by the advocates
of free competition regimes, given their potential to raise barriers to new entrants,
whenever the quality partnership is restricted to existing operators. Indeed, risks exist
that collusion might build up if the partnership is allowed to be closed to new entrants.
The organizational forms and respective regulatory regimes described represent
the pure conceptual existing options. In practical terms, what we nd in the real
world are mostly intermediate forms and also coexistence of different regimes for the
multiple services provided in an area or network. All regimes and forms present
advantages and disadvantages, and no universally best solution can be identied.
Market-initiated regimes enable an active participation of the operator in the service
design, providing the stimulus for the improvement of the service and consequently a
stronger willingness to share the planning and revenue risks with the authorities. The
latter is mostly related to patronage and fares, and the former is highly inuenced by
the quality and appropriateness of the service to the customer needs, reason why the
involvement of the operator to the design of the services is so important.
Where the creation of the services is left to the authorities, that is, authority initiative
systems, the compliance with requirements established in accordance with the strategic
goals can be, at least theoretically, more easily achieved, and consequently, enforce-
ment should be made at a lower cost than in other regimes. The main advantage of these
regimes is that they give structural priority to integration and stability of supply, while
seeking cost-efciency through other instruments.
Despite the consensual awareness of the positive and negative aspects inherent to
each pure regulatory option, the mix of regimes for each specic mobility system has
19. It is worth referring that the concept of license and authorization is often used with different meanings.
In this work, we consider, license as referring to the professional qualication (including
creditworthiness and reliability) and authorization as the right to the commercial exploitation of the
service.
a considerable variation from one country to the other. However, we could identify
the following key variables as being present in all decisions regarding the structural
conguration of the system:
Variables external to the mobility system:

Political-administrative organization of the country/region.
Regulation emanating at a higher level but applicable to the urban mobility
system in question (e.g., case of EU or Federal States like Brazil).
Variables internal to the mobility system:
Legal possibility of having a plurality of initiatives on the market (i.e., degrees of
freedom) and entrepreneurship for those initiatives (i.e., who takes the
initiative).
Degree of competitive pressure and incentives in the system.
Level of technical competence of the interacting agents for planning complex
networks.
A number of different solutions have been implemented across Europe and

worldwide, and there is strong evidence that the trend has been to replicate the
political-administrative division of the countries (i.e., national, regional, and local
division) into the organizational framework of the transport system. One of the main
reasons for this almost systematic option is the need to match scal and nancial
autonomies with the organizational responsibilities to facilitate the handling of the
funding sources and mechanisms that support the management of the urban mobility
system.
Additionally, the varieties of organizational forms that can be identied are the
result of crossing the functions to be performed at the different planning and decision
levels with the regulatory regime. This way we could nd the following organizational
solutions in the surveys done for ISOTOPE research and later conrmed in
MARETOPE:20
Central planning and tendering of the operations the transport authority

determines the policy goals that represent the planning framework for the planning
department or agency who in turn is obliged to contract out part or the whole of the
planned services to private operators under competitive tendering. This organiza-
tional option is known as Scandinavian model, which we could nd in
Copenhagen, among other places, where a set of local governments cooperate to
form a transport authority (which is the political body of HT), which, in turn, has a
planning body (HT Hovedstadsomradets Trakselskab). An alternative to this
form is to have a management contract between the transport authority and the
planning company that obliges the latter to contract out the realization of all or part
of the planned services to a private company through tendering procedures.
20. Also reported in Van de Velde and Sleuwagen (1997, p. 154).

This organizational form was used in Malmoe region,21 although it is no longer in

force.
Similar to the previous but allowing the operator to redesign the services in their areas
of operation as an incentive, although this freedom is limited to preserve system
integration. The planning company sets fares, bears the revenue risk, and induces the
operators choices by setting minimum standards and paying a passenger.kilometer-
based fee, that is, sharing the commercial risk with the operator. This organizational
form was known as the Adelaide model because the real case that inspired this
specication was Adelaide in South Australia (Radbone, 1997, pp. 118).
Tendering planning and operations normally designated by concession, where the
authority can also establish the minimum standards and set the public service
obligations. It is worth clarifying this is also the organizational setting adopted in the
British Rail and to our opinion wrongly designated as franchising, because
the objective was to give more freedom to the operator for service denition and
respective marketing, which is precisely the opposite to what happens with the
commercial franchises in the retail sector (product specication is rigid and marketing
is centralized by the franchising entity). Concessions are common in many European
countries, namely, Portugal and France. A derivation of this form can be the tendering
of the planning function at a different procurement process than the one used for
the operations. This option was conceptually recommend in the conclusions of
ISOTOPE report (EC, TIS.PT, 1997c, p. 83), and currently, its discussion is gaining
prominence within practitioners, although no effective implementation was yet
reported.
Finally, the free competition form, with or without light touch regulation, already
explained earlier.
3.3.3. Regulatory Framework
Theoretically, several methods exist for the selection of agents, and these are conditioned
by the regulatory frameworks in place that determines the available options for market
access as we have seen in the Chapter 2. Selection of agents is thus a function of the way
the tactical level of planning and control is organized whenever outsourcing of the whole
or parts of service production is done. However, it should be made clear that outsourcing
is simply a method of production available to any initiator of services irrespective
of the organizational form or the regulatory regime (Van de Velde, 1999, p. 150).
As already referred, competition is seen as an important mechanism to improve
service performance, in particular, efciency, quality being imposed through
regulation. In a more economic perspective, competition is considered by many
authors as the most important mechanism for maximization of consumer benets and
to limit monopolistic power with the underlying intent of improving industry
performance by increasing the role of market forces. The main pillars of this rationale
21. Reported in ISOTOPE research (EC, TIS.PT, 1996b, D153).

are rivalry and freedom to enter the market, and the most relevant game instrument
between actors is the competitive threat from potential as well as existing competitors.
As noticed by Beesley (1997, p. 41), about British Airways, the simple
announcement of the intention to privatize the company helped to increase efciency,
thus providing another evidence that the issue at stake is the ability to keep competitive
pressure over the market agents as we have also concluded in ISOTOPE research
project (EC, TIS.PT, 1997c, p. 83), regarding urban public transport.
In fact, a number of studies have been dedicated to the analysis of the efciency
effects accruing from the introduction of competition. In ISOTOPE, based on
information obtained from 207 operators from 108 cities for the period between
1993 and 1996, we have concluded that Deregulated markets have theoretical
and empirical advantages in terms of efciency of production. Regulated markets
have theoretical and empirical advantages in terms of efciency in consumption.
Limited competition markets may have advantages of both (EC, TIS.PT, 1997c,
p. 104).
In a more recent study commissioned by the European Commission to Colin
Buchanan and Partners for the Study on Good Practices in Contracts (Buchanan
et al., 2002, pp. 26) using information from a sample of 43 cities with a population of at
least 500,000 inhabitants (29 with no competition, 10 with controlled competition,
and 4 with deregulated market access), all of them with at least 6 years of data available
and for a similar period, between 1991 and 1996, found that
in cities with no competition, there was a small average annual decrease in

passenger numbers of 0.2%, for cities with limited competition there was an
annual increase of 1.7%, and in the cities with free competition (the so-called
deregulation) a drop of 2.6% per year is reported;
the fare coverage ratio has increased at a rate of 1.2% for cities with limited competition,
whereas in the ones with no competition this rate was 0.5%. For free competition
cities, a low increase of 0.1% was found for this ratio, but in these cases, fares tend
to cover most, if not all, the operating costs, because they are not regulated; and
the number of employees in public transport companies reveals an average annual
decrease of 2.1% in cities with no competition and an increase of 1.2% in cities
with limited competition. In cities with free competition, this data was not
available for the period covered.
Despite the impact caused by the above-reported gures, none of these studies was
able to prove a causal relation between the regulatory and contractual practices and the
productive efciency of the urban mobility system in general or the public transport
operation in particular. Looking at the last 15 years of studies done on regulatory and
organizational issues and contractual analysis,22 all the evidence collected only allows
22. The main studies in this eld in Europe in the domain of regulatory and organizational issues in Urban
Mobility were Tyson (1994), EC, TIS (1997b), EC, OGM (1998a), EC, NEA (1998c), Buchanan (2001),
NEA, (Integration, 2003), and EC, TIS.PT (2002b).
us to reinforce the statement that competitive pressure contributes to the enhancement

of efciency and other elements of performance. In addition, the study by Buchanan
and Partners lacks rigor in the constitution of the sample of cities because it brings
together a group of cities that cannot be considered in the same cluster, such as
Paris, Porto, Lille, and Rotterdam, to name a few examples only, and applies
statistic averages over it on basis of the assumed similarity of the regulatory system.
The clustering pitfall is mostly related with the fact that in the group of cities selected,23
there are several cases of noncomparability in the domains of city dimension, mix
of modes offered, rate of change in the process of introduction of competition, degree
of participation of the state in operating companies, and so on. It is only fair to refer that
the lack of consistent series of data was a major problem in all the studies referred,
which is also largely recognized by Buchanan and Partners (2002, pp. 210 and 211).
3.3.4. Contractual Relationships within the Public Transport System
3.3.4.1. Selection of agents The principles reected in the studies presented in the
previous sections have been transferred to the relation between authorities and
operators through the use of competitive tendering for the selection of efcient
operators and contractual formalization. In Europe, in particular, the European
Commission introduced in July 2000 a proposal for regulation of public service
requirements in public transport that requires public service contracts to be concluded
by authorities whenever they wish to award and exclusive right or an operating subsidy
to an operator. The award of these contracts should follow the rules set by the EU
Directives (92/50/EEC, 93/36/EEC, 93/37/EEC, and 93/38/EEC), related to the
coordination of procedures for the award of public service contracts.
The underlying theory behind this competitive tendering procedure is the economic
theory of agency that, as Ross (1973, p. 134) notices, provides one of the oldest and
commonest codied modes of social interaction. An agency relationship exists
between two parties when one, designated as the agent, acts on behalf of or
representing the other, or simply representing the interests of the other, designated as
principal, in a particular context of decision and problem-solving. Examples of agency
relationship are very diversied and present in all sectors of society, and they are
mostly reected in contracts. In essence, any contractual arrangement has elements of
the economic theory of agency.
The rationale supporting this theory is that both parties in the relation have a state
independent von Neumann-Morgenstern utility function (Ross, 1973, p. 136), and
they act as to maximize their utility function. This can be briey described as an
23. Amsterdam, Barcelona, Berlin, Birmingham, Bochum, Bologna, Bordeaux, Brussels, Cologne,
Copenhagen, Dortmund, Dresden, Dublin, Dusseldorf, Essen, Frankfurt, Genoa, Gothenburg, The
Hague, Hamburg, Hanover, Helsinki, Lille, Lisbon, Liverpool, London, Lyon, Stockholm, Manchester,
Marseille, Milan, Munich, Nancy, Newcastle, Nuremberg, Paris, Porto, Rotterdam, Strasbourg, Stuttgart,
Valencia, Venice, and Vienna.
agency problem involving choice under uncertainty. The agent and the principal
agree upon a fee (or some other kind of remuneration) to be paid by the principal to
the agent for her services. The fee is in general dependent on two variables, the state
of the world and the action to be performed by the agent. The choice of the fee is the
outcome of a bargaining process between the principal and the agent, or, in larger
games, it is the outcome of a market process in which asymmetries of information are
a determinant factor to inuence the strategic behavior of both agent and principal.
As the tender procedures are usually complex due to the number of criteria
to consider, with respective trade-offs, and the contracting periods involved, a
prequalication procedure is advisable. Prequalication should be entirely based on
the potential of prospective bidders to perform the particular contract satisfactorily,
taking into account aspects such as experience in public transport services,
professional competence, and nancial capacity, besides the minimum technical
qualication to hold a technical license to perform the job.
These prequalication criteria should however be carefully applied to avoid in one
hand closing the market to new comers and in the other the risk of leaving the
authorities without degrees of freedom to cancel a contract whenever the operator
does not fulll the expectation laid down in the contract. In addition, a balance is
required so that candidates who do not have the predened expertise and experience
are not systematically confronted with this barrier to new entrants into the sector.
Legislation can overcome this problem by preventing the use of criteria that will bias
the selection in detriment of new entrants and favoring incumbents. In Germany, for
example, previous experience is not permitted as selection criteria (EC, NEA, 1998c,
pp. 6780).
The tender document is simultaneously the starting point for the quality assurance
and for the bidding process that should aim to be fair, transparent, and
nondiscriminatory. The support document, which is the invitation document, must
contain clear and objective rules for the different aspects that are at stake when
selecting a contractor and should also reect the authoritys objective in the future
contractual relationship. Besides, in the invitation to tender, the rules of the
tendering procedure should be very clear regarding the list of criteria that will be used
to select the candidate to the contract, the different stages of the negotiation process
for the nalization of the contract, and the criteria through which the future
contractor will be monitored.
The decision process to select the agents will always be the result of an assessment
based in more than one criterion, even if no formal competitive tendering procedure
is followed, to cover the different aspects of performance divided into industrial and
commercial performance. Ex ante evaluation is required to assess the potential of the
agent to achieve the desired performance, that is, assessing the risk of contracting a
specic agent before doing it. Ex post evaluation is equally required but with the
purpose of providing feedback for the continuous learning process of the tendering
authority and for future adjustments.
The two performance concepts have a different coverage. Industrial performance
is concerned with productive efciency, that is, transformation of resources into
services, whereas commercial performance is concerned with the transformation
from results (service provision) to impacts. As Viegas (2001, p. 11) recommends for
the supply of public transport services, commercial performance should be analyzed
against three levels of impact:
First, the impact on citizens use of public transport measured by passenger.kil-

meters in public transport.
Second, the impact on trafc congestion, measured through market share of public
transport.
Third, the impact on the nancial situation of the operators and authorities
(reduction of subsidy needs), measured through the revenue obtained.
The selection of the operator should enable the development of a learning process
for the evaluation method underlying the selection procedures, so that accumulated
information will enhance the ability to better select the contractor based on the
statistical relations between characteristics of an operator and its ability to develop
good performance levels in service provision. This knowledge will enable the
authorities to ne-tune the criteria for evaluation of mobility providers taking into
account the constraints imposed by internal (e.g., changes in the local regulatory
framework) and external changes (e.g., for Europe, revision of the Directive ruling
the procedure for contract awards). It is this recurrent cycle of renement of criteria
that forms the learning process illustrated in Figure 3.5.
The evaluation of tenders can take several forms, but from the observation
of several tendering procedures (EC, TIS.PT, 1997c; EC, OGM, 1998a; EC, NEA,
1998c; EC, TIS.PT, 2002b), there are a restricted number of alternatives on which the
bid should be based, which are as follows:
only price/minimum subsidy in these cases, full-service specication is required;

combination of price/minimum subsidy and quality of service in these cases, it is
common to have a minimum service specication for the services that have to be
offered under the contract that will be awarded plus the possibility for the tender
to offer higher levels or quality of service in addition to what is specied in the
invitation to tender. In this alternative, it is indispensable that the tender authority
should indicate in advance the score mark for these options.
only service quality depending on the type of contract a clear statement of
receivable fees (for gross cost contracts) and receivable subsidies (for net cost
contract). In any case, the scoring system on which basis the tender will be
evaluated has to be informed to the bidders.
Very often, authorities request bidders to present a combination of price/

minimum subsidy and quality of service, so that they can negotiate a best deal with
the rst ranked candidate. If no success is obtained with the negotiation, the
authority still retains the right to start negotiation with the second ranked candidate,
and so forth. In this iterative process, the number of possibilities for a best business
deal is maximized without having to tender again, and so with a low marginal cost
increase for the authority.
Figure 3.5: Learning process for qualication and selection of operators (Macario,
1999, p. 8).
It is worth referring that the EUs public procurement directives only allow post-tender
negotiations in some limited circumstances. To ensure conditions of fair competition is an
important principle in tendering procedures, but this concept can have several
interpretations. In general terms, fairness depends on the consistency with past practices
and rules to which people and organizations have become habituated. However, two
different interpretations can be found in the literature (Baumol, Ordover, & Willig,1996)
for this concept: the procedural notion of fairness based in the equity of opportunities
offered and the type of access provided to an economic process, and the end-state fairness
based on the assessment of the state in which individuals or institutions emerge from an
economic process, that is, the impact of the economic process.
Fairness theories appear associated with the games of fair division (EC, OGM,
1997a, p. 15), which in complex problems are achieved on basis of the interest of the
persons or entities involved. The individual efciency of this process is extremely
dependent on the level of information that one part has on the preferences of the
other. The concept of fairness is based in the value of the utility function that each
individual attributes to a specic good (or service). If two alternatives are to be
compared, a distribution is considered as fair if individual A, to whom good (a) was
distributed, considers the utility function of good (a) as superior than the utility
function of good (b), which was distributed to individual B. This denition highlights
the fact that fairness concept assumes the existence of asymmetry of information and
that the individual judgment is done disregarding the utility valuation of the other
parties involved (Baumol, Panzar, & Willig, 1988, pp. 425429).
One problem of the fairness analysis is to select one most equitable solution
among the available existing possibilities given the individual utility functions of the
parties involved and the misinformation that each part has on the preferences of the
other. The determination of the optimal most equitable solution is a difcult task
given all uncertainties involved in the process. Despite this, an alternative that has
been defended by some authors is to look for fairness regions bounded by
acceptability constraints instead of looking for an optimum of the fairness function.
When discussing fairness issues, we are always seeking for tools that can be able to
determine the level of equity of a specic proposal or institutional arrangement. The
main question for decision-makers is whether a specic option is fair or fairer than
other alternatives (EC, OGM, 1997a, p. 16), and this largely depends on the
assessment of impacts each alternative has over the respective winner and losers.
Transferring the above concepts to the UMS drives us to the underlying principles
of the contractual relationship, the already referred principalagent problem,
according to which the principal aim is to establish incentives that will motivate
an agent to act in such a way that contributes to the principals objectives. This
process is highly affected by the level of information that each agent controls and by
the agents perception on the end-state fairness of the contractual relation in general
and the incentives in particular. Asymmetries of information between authorities and
operators or between bidders are a potential factor for end-state unfairness or for a
perception of procedural unfairness.
This perception can by itself hinder the operator availability to react to incentives
as expected, as it is the case in tenders for net cost contracts whenever an incumbent
has advantage of market information or other, possibly leading to the dissuasion of
potential competitors and thus jeopardizing the benets of tendering as an
instrument to foster competition.
The preceding issues were raised presuming the case for an open tendering, but in
fact, all these principles and concerns are valid for situations of closed tender, that is,
the authority invites a selected set of operators to tender, the so-called short list, as
well as cases of negotiated contract arrangements, where the authority deals with one
operator only. In addition, there are still some soft elements that can also be used to
enhance the likelihood of success of a contract. One such case is the prespecication
of a negotiation period and the later management of the contract, which is mostly
related with monitoring and contacts between the authority and the operator to
overcome all the situations that were possibly left out of the contract detailed terms.
Contract monitoring can be either formal or informal. Informal relations are
usually a positive factor to bring forth a relationship of trust, which in turn enables a
mutual understanding of the problems existing in both parties and even a better
understanding of the potential to improve performance. Formal monitoring in turn
should cover all issues contained in the contract, but a particular emphasis is usually
placed in aspects such as safety, efcient performance, perceived quality of service,
and nancial issues.
One of the key aspects in monitoring relates with the issue on how to obtain data.
Basically, we can nd three possible ways of doing it: the operators, in which case the
contract must specify their obligations to provide the information required by the
authority or have relevant information audited by the authority or a third party;
collected by the authority using its own staff and resources or outsourcing it to third
parties. In any case, it is within the role of the authority to analyze what type of data
it requires so that its management control function can be duly fullled. This
information should be part of the contract, detailing what data is to be collected and
by what form and frequency.
Data collection can be a delicate issue whenever it is provided by operators;
therefore, contract dispositions should clearly state which part of the data can be
made public and which part should be kept to the authorities only. In cases where the
authority has contracts with more than one operator, some of the data collected can
also be used by each operator and even by the public, for the sake of transparency
in the management of the mobility system to benchmark the performance of each
operator against the others. If the authority is the one to collect the data, then the
results should be made available to both operators and public. Given the commercial
sensitiveness of some data, two options might be considered, either to clearly identify
the operators and their performance on a name basis, which happens in Spain and
British train companies, for example, or otherwise to present the data and the
performance assessment in an anonymous way as it is done with London Bus
contracts (EC, OGM, 1996a, p. 119).
3.3.4.2. Contractual forms A characteristic of network activities is the collection

of goods or services from a (possibly large) number of producers to a large number of
users. Both producers and users are often scattered in a wide geographical area, and
competition typically starts with the unbundling of various stages of the delivery
process. As Crampes and Estache (1997, p. 2) note, the type and degree of
competition is usually justied by the classical trade-off between the relative choice of
inputs by rms (internal efciency) and their sales and pricing policies (external
efciency). If economies of scale are not too strong with respect to the size of the
market, the risk of both types of inefciency is weak.
In cases where goods or services need a continuous material infrastructure, then
standard competition is unlikely to offer a best solution because duplication of
infrastructure costs offsets the potential social benet resulting from the presence of a
competitor. But the obvious solution, a monopoly, does not offer a good solution
either, especially in terms of allocation of resources. An alternative option is then to
keep property rights over productive assets in the hands of authorities and concede
management or operation of services to a private rm. The effectiveness of this is
much better and more appropriate to the infrastructure management; however, it can
also be implemented with success only for operations, as it is the case with the
implementation of management contracts, which is very common when the aim is to
introduce competitive pressure on incumbent state-owned public transport operators
(e.g., STIB in Brussels, EC, OGM, 1997a, p. 67), because it allows to use competitive
mechanisms in the selection of service providers, without risking uncontrolled
maximizing prot behavior hindering the public provision of service without
exclusion through price discrimination, which is a major concern within the
provision of mobility services.
This way the monopolistic provider of the service will have a prot-driven incentive
to minimize operating cost, and it is for the authority to enforce quality standards in
the service provision and design further incentives to stimulate the service provider to
increase quality above minimum standards as a driver for increased demand.

Distributive objectives in turn are only achieved if the authority collects efcient access
fees for the exclusive rights of operation entailed in the concession. However, this
economic rationale is not always followed because authorities often have other
priorities to satisfy through the concession strategy, such as social and scal objectives.
According to Crampes and Estache (1997, p. 4), these political priorities impose
some constraints in the specic design of the concession system, the regulatory
regime, the tariff regime, and even the way the concession is awarded. From the
economic theoretical viewpoint, all the decision process should be subordinated
to efciency concerns, but in practice, the political rationale very often considers
efciency as a second-order concern at the benet of income distribution, budgetary
concerns, or even employment levels. Although the standard approach in public
economics is to maximize social welfare subject to constraints of several orders, such
as legal, scal, technological, and informational, the political approach is to balance
between four main groups of interest, the so-called stakeholders:
users, who are normally concerned with prices, service quality, and reliability;
rms, which give priority to prots, risk assessment, and market power;
governments, who tend to give priority to the reduction of scal burden, generation
of nancing resources through the reform process, dealing with unions on the issues
of labor redundancies resulting from the reform process, environmental preserva-
tion, and distributional issues, besides their own political survival; and
nally, the regulatory and organizational bodies, who in their key role as
facilitators of the reforms process are mainly concerned with their own reputation
with the other stakeholders and with the effectiveness of implementation of the
reform process itself.
On top of all these interests, contracts should be designed in such a way that they
constitute an effective pressure on the service suppliers toward the desired quality
levels, while simultaneously, they should also be the main instrument to assure
continuity and stability of services. In parallel, as Viegas (2004a, 2004b, p. 10) notes,
public bodies must keep their ability to procure effectively and efciently for the
provision of services. In this respect, contractual design is indeed a delicate and
complex issue with a number of difculties related with the following elements:
adjustment conicts between contract duration and policy evolution during the
lifetime of the contract;
information on the service being provided;
risk-sharing between the parties;
skills to manage and monitor the contract; and
level of decoupling of service provision.
In the study by Colin Buchanan and partners from the 49 contracts analyzed,
related not only to urban mobility, three major risks are reported: revenue risk,
whenever demand is below expectation, service quality is poor, and so on; productive
(or operating) risk, that is, maintenance costs, fuel bills, and so on; and capital risk,
related with purchasing and replacing assets (Buchanan et al., 2002, pp. 26).
Net cost contracts with investment, usually with xed periods, are also a variation of
net cost contracts that can be found in Europe (UK) and also in South America (Santos &
Aragao, 2000, p. 276)) and Australia (Cox & Duthion, 2001, p. 25). In these contracts,
the operator provides, in addition to the public transport service, the required rolling
stock (which is more common) and a xed infrastructure facility (which is less
common). For bus contracts, depots and bus stops are usual assets to include; for rail
contracts, combinations of track, depots, or station are the most common elements.
The number of operators in each market is not very high, and these are causing
reasons for concern within the authorities, especially in the bus sector where a
number of the so-called global operators exist, because of the potential reduction of
competitive pressure that might accrue. In several cities, authorities have taken a
number of measures to encourage new entrants, namely, as follows:
Moving to gross cost contracts, although developing incentives and penalties

associated to performance, to avoid quality degradation. The reasons that
are usually pointed out to choose this type of contract are the fact that they reduce
bidders risk and cost and also the costs for authorities as the need for expensive
customer survey to allocate revenues among operators is reduced. In addition,
because fares are set by authorities, operators have a very reduced scope to increase
patronage and revenue, even when incentives are included in the contract, so market
concentration is more unlikely.
Choosing route rather than area contracts to encourage smaller companies to bid.
Requiring bidders to subcontract a proportion of their services whenever there are
risks of market concentration.
Maintaining ownership of local depots and specialized equipment (e.g., ticketing),
as already referred, as lack of access to these equipments often is an entry barrier,
in particular, where quality partnerships exist between incumbents.
It is worth referring that these trends from the authority side are not left without
answer from the operators. For example, in Sweden, smaller operators have grouped
themselves into cooperatives to enhance their capacity to bid for larger contracts. In
practical terms, this means concentration of the market structure and with it a
signicant reduction of market contestability.
Contract theory identies several contractual categories considering the object
they are expected to rule and the following attributes:
Whether they describe a bilateral or multilateral relation;

Whether they are static or dynamic; and
Whether they are complete or incomplete.
On the basis of the structural attributes dened in contract theory, there is a wide
diversity of practical contracts in urban mobility systems although all of them provide
a quite similar structure rst the basic type of contract according to risk-sharing
options, second by the set of incentives. The basic types of contract that we have found
in urban mobility systems are the following ones:
Management contracts
Gross cost contracts
Net cost contracts
In the management contracts, the authority usually retains the ownership and
control of depots and rolling stocks as well as all revenues and pays for all capital
and current expenses. In addition, there are also cases where the authority chooses to
be the employer of the staff engaged in the provision of the services. In fact, the
management contract represents a form of delegation from the authority to the
operator who is conned to the professional management of the operations on behalf
of the authority. The degree of delegation and of engagement of the contracted
manager in any risk-taking is decided on a case by case basis, but in all circumstances,
the contract is negotiated for a xed period of time and agreed price, usually entailing
rather strict performance standards. These contracts are also adequate for transition
periods from noncompetitive to competitive environment as they enable the
progressive adaptation of the incumbent operator.
In gross cost contracts, the authority releases the control of the productive means
vehicles/rolling stock, depots/other infrastructure, and so on to the operator, often
setting also certain specic standard for quality of service, required eets, and so on,
together with the agreed price for the production of the service. Under these contracts,
all fare revenues are transferred to the authority with the respective commercial and the
operator is left with the risks conned to the productive aspects of the operation. Very
often contract length has to be associated to the life cycle duration of material assets
involved in production; this is a common situation with railways companies. However,
more recent evolutions enable to have contract length almost independent from life
cycle of material assets through operational leasing with heavy maintenance being
performed by the suppliers of the assets, or subcontracted under their responsibility,
and still with disposal provisions at the end of the contract ensured by the supplier.
In net cost contracts, both the productive and the commercial risks are born by the
Operators, and ownership of assets may be retained by authorities or transferred
to a separate entity, which is also a possibility in gross cost contracts. The authority
also provides specied standards for quality of service, required eet of vehicles/
rolling stock, and so on, as well as for an agreed subsidy, premium, or compensation
for fares levels below the economically viable level. In these contracts, the operator is
normally entitled to retain all fare revenue and bears all the risks (productive and
commercial), which typically may involve trafc disturbances, uctuations in
revenues for diverse reasons, and changes in regulatory regimes. Special risk-sharing
dispositions may be negotiated to limit the operator exposure to these risks. These
contracts can also involve specication for investment in xed or moveable assets, in
which case, they are designated as net cost contracts with investments.
Apparently, net cost could be thought as the best answer from the authority
viewpoint, because the operator bears all the risk and receives only a xed subsidy.
However, experience reveals the opposite. For the operators, these type of contracts
imply a number of additional costs and associated risks such as costs of permanently
steering the market; investments that have to be done to foster demand with a lead-in
time that often goes beyond contractual agreements; and much higher costs for
preparing the bids. For authorities, there are some externalities in reduction of
market contestability as a consequence of lower number of bids. In addition, all
changes in transport policy or trafc regulation may affect the commercial side of the
mobility system and consequently give room for claiming new nancial compensa-
tions to operators.
However, net cost contract with investment can be the most appropriate for
situations where there were no previous services and new infrastructures are required.
This is often the case for implementation of light rail operations. An example of such
case is the Croydon light rail, where the operating contract is made for 5 years while
the contract supporting the building activity is for 99 years (Buchanan et al., 2002,
pp. 422), which is far longer than the more common contracts usually for 1530
years. These contracts provide the advantage of engaging the operator during the
construction stage what in principle should enable to ensure the system can be
operated efciently. In Buchanan and Partners (2002, pp. 422), we can nd a
number of examples where the operator is unable to provide the level of service it
desires due to cost-cutting measures during construction that in the long term results
in reduced revenues or higher operating costs.
The choice of contract type is not a linear decision and depends on a number of
factors that inuence the level of risk and also the entity best positioned to control
those risks. Consequently, the solution adopted by one city cannot be directly
transferred to another city without thorough consideration of the following factors:
local policy; planning of services; level of fare integration; geographical area to be
covered by the contract; quality of service issues; budgetary constraints; pricing;
entities able to control and mitigate the risk; number of expected bidders; external
constraints; and so on.
3.3.4.3. Negotiation Negotiation can be dened as a discussion between two or

more parties aimed at converging to a mutually acceptable option or sets of options,
possible not optimal to any of the negotiation parties. The noncoincidence of goals
gives origin to conicts, and negotiation is the main instrument to escape conict.
The dynamics of negotiation are made out of the reactions of the different parties
that can be described as matching and mismatching; the rst means corresponding to
the others position, that is, demanding more if the others demand is enlarging and
conceding faster whenever the others make rapid concessions, while the last means
taking the symmetric position, that is, demanding more when the other demands are
small or conceding more rapidly whenever the other concessions are slower.
Matching and mismatching are intertwined in the negotiation cycle and may appear
at the beginning, middle, or end of a negotiation process.
Several authors (Liebert, Smith, Hill, & Keiffer, 1968; Pruitt & Syna, 1985; Yukl,
1974; Esser & Komorita, 1975; Smith, Pruitt, & Carnevale, 1982; Benton, Kelley, &
Liebling, 1972) observed the negotiation cycles in different sectors of activity and
concluded that mismatching is common at the beginning of a negotiation as a reaction

to the other partys initial offer. This rst reaction is commonly designated as
tracking, that is, an effort to place ones goal and demand at a reasonable distance
from the best offer that can be expected from the other (Pruitt & Carnevale, 1993,
p. 64). The offer made by the other party is seen as valuable information about its
objectives. However, in reality, this assumption is very often not conrmed because
the behavior of the other party might be conducted through a more complex paradigm-
like blufng. The experience of the parties in negotiation and the knowledge each
party has on others is essential for the behavior adopted in the different phases.
As Keer (1986, pp. 1618) explains, we need to distinguish between matching the
others concession and matching the others nonconcession. The rst we designate as
positive matching and the latter as negative matching. Positive matching is often
motivated by the social norm of reciprocity or the desire to reinforce the others
cooperative behavior. Negative matching in turn reects also the norm of reciprocity
but added from the fear of tangible loss or image loss, depending on the political
weight of the negotiation at stake. This matching phenomenon occurs at the different
stages of the negotiation cycle, the longer one being the middle stage of the
negotiation. At this stage, the undermatching phenomenon also exists, that is,
conceding considerably less than the other had just conceded.
The higher the tendency to undermatch, the longer the negotiation period is
expected to be.24 Consequently, the determination of a time limit for the negotiation
period might help in reducing the frequency of these phenomena along the
negotiation cycle, although it will increase substantially the pressure at the end stage
of negotiation as when deadlines are looming. Some studies (Bartos, 1974; Benton
et al., 1972) report the fact that many negotiators engage in mismatch at the end
stage of the negotiation cycle to force the other party to ll the gap. The success of
this strategy depends very much on the level of information on the other capacity or
willingness to concede; otherwise, there is a strong risk of rupture if the mismatching
party holds rm at a point, that is, beyond the others limit, the so-called crunching
point of the negotiation, when no decision is often by itself an irreversible decision. In
addition, there is still a disadvantage that whenever negotiation restarts after rupture,
it will create a departure point where both parties have a substantial knowledge on
values and behavioral strategies of the other and so less bargaining room is available.
According to Pruitt and Carnevale (1993, pp. 34), there are ve broad strategies
that can be used in negotiation:
Concession making which implies one party to reduce goals, demands, or offers.
Contending which implies having one party persuading the other to concede.
There are many ways of implementing this strategy, of which the most well known
are as follows:
24. This situation is often observed in negotiations with Unions, in particular, where there is a scenario of
reform and likelihood of loosing some statutory or contractual benets.
Threats that is, messages indicating that one party will punish the other if that
other fails to conform.
Positional commitments that is, messages indicating that one party will not
move from a certain position.
Problem-solving trying to converge in options that will satisfy both contending
parties through a vast number of possible techniques.
Inaction doing nothing to let time lower the pressure between parts in conict.
Withdrawal dropping out negotiation, usually followed by a later reformulation
of the negotiation process.
As Schelling (1980, p. 5) states, most conict situations are essentially bargaining

situations, that is, situations where the ability of one participant to achieve his goals
is dependent on the choices or decisions that the other participants will make.
3.3.4.4. Conicts and risks In urban mobility, conicts can arise from several
different sources. They can be in the framework of industrial relations, that is,
between operators; or in the framework of contractual relations, that is, between
authorities and operators; or in the framework of political relation, that is, between
authorities; and nally between clients and the agents of the system. In ISOTOPE
research (EC, TIS.PT, 1997c, p. 51), we have identied six types of conicts: roles and
responsibilities, denition of objectives, nancial related, fare policy, and performance.
In Table 3.4, we relate those types of conict with the source framework where they are
more likely to occur. The result of this qualitative assessment reveals the highest
potential for conicts to arise within the contractual framework.
Table 3.4: Allocation of conicts.
Allocation of conicts Source framework

Political Contractual Industrial Client
Conict types Roles and responsibilities
Denition of objectives
Financial
Fare policy
Performance
Source: Author.
Contracts usually include provisions for conict resolution, namely, conciliation

or arbitration. In the rst, decisions are left to the contending parties, whereas in the
latter, decision is transferred to a third party. The possible conciliatory paths include
recourse to the contract and its interpretation, bilateral negotiation followed by
agreement; recourse to another (higher hierarchic level) authority; and recourse to
third parties. Contract design is also a learning process where experience managing
contracts provides extremely valuable feedback for the subsequent opportunity to
redesign it. In particular, in what concerns conicting situation where risk-sharing is
involved, as very often at the time of writing a contract, a good part of potential risks
are not even known. In fact, there are three different types of risk, according to their
likelihood of occurrence that can potentially degenerate in conict situation,
justifying different types of contractual provisions (EC, TIS.PT, 1997c, p. 51):
The relatively common risks, for which the contract should try to provide rules as
clear as possible.
The risks that can be identied, but which arising circumstances are very difcult
to predict, in which cases, a structure for conciliation and possible arbitration
should be dened.
The risks that cannot be identied at the outset, for which the adequate provision
is the incompleteness of the contract that is discussed in Section 3.3.4.5.
In general, risks should be allocated to the party who is in a best position to avoid
its occurrence or mitigate its consequences. This risk allocation should normally
be reected in the contractual provision, although there are always some degree
of nonidentied risks that usually give origin to some conicts at the moment of
proceeding with the ex post allocation of its consequences. Despite this permanent
element of uncertainty, it is possible to identify the following risks in urban public
transport (EC, TIS.PT, 1997c, pp. 4149):
Production risk
Revenue risk
Financial risk
Planning risk
Environmental risk
The production risk is an area that has been traditionally allocated to operators
because it is associated with the ability of doing the activities that contribute to the
transformation of productive inputs into vehicle.kilometers. This risk usually
subdivides into two main categories: loss of service provision and cost overrun.
In the last case, cost overrun may be caused by action or inaction of the authority in
which case some form of compensation has to be considered in the contracts.
Production risks can still be further divided into the following categories:
Infrastructure related, entailing construction costs and delays, maintenance and

upgrading costs, congestion, and temporary loss of access.
Rolling stock and technical equipment and consumables related with the operation,
entailing delivery times, maintenance direct costs, and availability of equipments,
loss of operationality, and so on. Long-standing guarantees are usually asked by
transport operators to their suppliers to transfer to them the penalty for any possible
nonperformance that will reect on a loss of quality near the user.
Labor related (agreements and regulations), entailing wages and salaries, loss of
net productivity, and industrial disputes.
The revenue risk is mostly related to patronage and fares, and in some cases, there
are other types of activities such as publicity in vehicles or stations that also
contribute for revenues. There is a wide variety of factors inuencing patronage,
namely as follows:
Mobility patterns, which are inuenced by lifestyle, location of activities, citizens

purchasing power, demographics, schedule of economic and social activities, labor
regulations, and so on.
Market share, inuenced by perceived quality of urban public transport, level of
private individual motorization, and push measure inducing the shift toward
public transport.
Fare structure and levels, entailing convenience of fare structure, tariff integration,
concessionary fares, and transactions costs (technology).
Change of urban mobility policy, which effectiveness is conditioned by the level of
integration between policies interacting with mobility, such as land use, energy,
environment, scal, and nancing.
Through the different types of contracts, revenue risks can be shifted either to
authorities or to operators. The latter providing the incentive to improve service to
customer. A key concern in this option is to preserve market knowledge and
customer needs by the organizing authorities; otherwise, capture by the operator
might well occur in only a few years.
Financial risk means the risk attached to the costs of money availability and
associated opportunity costs and advantages of retention of revenues accruing from
operation. This type of risk is quite peculiar from situations where installation of new
systems occurs, and consequently, there is exposure to new nancing schemes, or
currency exchange rates, or simply whenever companies nance their capital assets
with instruments available in the nancial market, which have a risk of their own.
Financial risks can thus be divided in two subtypes, the rst usually taken by
purchasers of capital assets and the second by the operators whenever net cost
contracts are established:
Risks of nancing costs on purchases, entailing interest rates, and currency

exchange rates.
Risks related with the remuneration of current short-term surpluses, entailing sales
distributions across different traveling titles and nancial remuneration of short-
term surpluses.
Planning risks result from the effects on production and revenue risks accruing
from changes in any demand generator, which usually occur with some time lag
between the moment of change in planning and the consequent effect on demand.
These risks result from three different sources:
Town planning, entailing ofces/factories, homes, shops/leisure, Urban Public

Transport (UPT) priorities, parking policy, and so on.
Road planning, entailing ring roads, and pedestrianization.
Public transport planning, entailing location of access points and routes.
Interaction with regional, national, and European links to which urban areas are
the nodal representation of these wider networks and largely the main generators
of mobility needs.
Finally, environmental risks, which in the last years have become an increasingly
important issue in urban planning and urban transport operators, because the risk of
having to pay penalties for environmental unfriendly vehicles is increasing leading to
the inducement of a faster technological transfer. This risk is very likely to occur in
developing countries where aged vehicle eets are more common and depends very
much on changes in public perception and consequently public opinion. The more
informed societies are, the more these concerns gain priority in the political agenda.
The environmental risks are usually reected in the following elements, all of them
with incidence on the production side of the mobility system: noise pollution, air
pollution, visual intrusion, and land use.
From the analysis of these ve risk categories, we can easily understand that their
incidence is always divided between the production and the consumption side of the
mobility system. In fact, this is the understanding reected in the contractual forms
used between authorities and operators where, as we will see later, risk-sharing is
usually identied with only these two categories. However, in a signicant number of
European research projects (e.g., ISOTOPE, QUATTRO, LEDA, MARETOPE,
and METEOR, the accompanying measure of the CIVITAS program25), we have
observed that despite the conrmation that all the ve categories of risk referred
earlier can be contractually resumed according to their incidence, there is in fact an
advantage of keeping it decoupled as the institutions with capacity to inuence the
degree of risks are in fact different and the need for institutional interaction is by
itself a major risk factor.
3.3.4.5. Contract completeness and renegotiation Today, there is a strong awareness

that contractual completeness is impossible to achieve as this implies that at the outset
of the relationship, all possible contingencies that may affect the agreement are
25. Where the author is responsible for developing a methodology to improve the transference of
experiences between European cities. This work was done in the years of 2003 and 2004 with a sample of 30
cities indicated in Annex 2.
foreseen. As said by Hart (1995, p. 1), at the root of contract formulation, we must
understand that in an ideal world, contracts would not be needed because partners
simply trust each other and rely on everybody behaving fairly with no aim to jeopardize
the other parts in the relationship. The less trust exists, the more we need contracts.
Also, when one of the parties is a state agency, a contract is needed, not least because of
the inherent rotation of the persons occupying the position, as it is the case in urban
mobility systems. As Loasby puts it, rms provide contracts for future options,
whereas markets provide options for future contracts (Loasby, 1994, in Foss, 1996,
p. 2).
The study of incomplete contracts in urban mobility systems is still at a very early
stage, and we could nd several authors, in the last years, have engaged in the
observation and analysis of contract completeness in several sectors developing
different frameworks of analysis. These research streams still use nonharmonized
concepts, which make these frameworks largely uncomparable.
Notwithstanding, in this eld of research, the work of Guasch (in Estache &
Quesada, 2001, p. 2) is especially relevant for our purpose, where a detailed analysis
of 1000 contracts was developed, for utilities and transport signed during the 1990s
and found out that renegotiation happened in around 50% of the cases, suggesting
that the likelihood of renegotiation is highest when the auction criteria is driven
by objectives of minimization of average prices paid by the users of the services
submitted to bidding procedures. In his study, one of the main pitfalls identied
was the low effort put in the assessment of the factors inuencing uctuation of
demand.
Preceding Guasch studies Tirole (1999, p. 779) presented a set of theoretical
reasons that can justify contract incompleteness, such as the follows:
Unforeseen contingencies that may arise during the execution of the contract.
Even if there is prospective capacity to anticipate all possible contingencies the cost
of doing it would be a reason not to proceed.
Contracts can only be contingent upon variables that can be veried by a third
party, otherwise the contract is nonenforceable.
For this author, these reasons justify the preference for an incomplete contract as
opposed to a complete one with high probability of being renegotiated in the near
future. This preference lies on the fact that renegotiation of contracts is not without
risk because neither the parties are symmetrically informed nor the contingencies are
fully foreseeable. In addition, the loss of bargaining exibility by all partners and the
increasing potential for collusion or corruption between the rm and the public
administration in charge of the renegotiations are also among the factors that
contribute to increase the risks associated with renegotiations. Therefore, reduction
of the probability of renegotiation may well be a variable to consider when selecting a
rm in a competitive procedure.
As observed by Hart (2003, pp. C70C71), both approaches encompassed in the
theories of the rm and of privatization are concerned with whether it is better to
regulate a principalagent relationship through an arms-length contract or through
a transfer of ownership.26 However, these two theories did not develop along similar
lines. In the theory of the rm, focus on vertical integration leads to formulate the
relation-specic investment between two rms along two main options: the existence
of an arms-length contract with the two rms remaining as independent; or the two
rms will merge and carry out the transaction within a single rm. In the theory of
privatization, the same question receives a different treatment. Let us assume that
rm A represents the government, who has to deliver a service to society, and rm B
represents a rm supplying the government with that service, in our case, mobility
services, or more specically public transport. The options open for this relationship
are the two entities have a contract, with the supplier (agent) remaining as a private
rm; or the government (acting as principal) can buy (nationalize) the supplying rm.
The two situations are of course different in some of the implicate elements. If we
assume rm B to be a mobility provider, this entity will have a direct relation with
clients, irrespective of the option taken for the relation with government, and also with
other service providers, because mobility is a chain service provision. Consequently,
the contract with the government will represent an attempt to regulate the companys
relation with its own costumer and with its own partners and suppliers, which does not
happen in the case of a strictly economic decision such as vertical integration, as
opposed to the previous one that is a political decision.
In much literature on these issues, reported by Laffont and Martimort (1947/
2002), Holmstrom and Milgrom (1991), Hart (1995), Schmidt (1996a, 1996b),
Shleifer (1998), and King and Pitchford (2001), among others, the supporters of the
theory of the rm take the incomplete contract perspective but consider that
inefciencies arise because it is difcult (and in some cases even impossible) to foresee
some of the contract variables, while supporters of the privatization stream take the
complete contract perspective and consider that the reason for imperfection lies on
moral hazard of asymmetry of information, which implicitly leads to the conclusion
that organizational forms (that entails ownership and rm boundaries) is not
relevant. In this perspective, the owner has no special rights once everything that can
ever be specied is included in the contract. At the opposite, ownership is of utmost
importance when contracts are incomplete, so that the owner of an asset or rm can
take all decisions that are not included in the original contract.
In Hart (1995, pp. 88), several similar reasons why a contract should have a
reduced number of variables are discussed. These are as follows:
Contracting parties are forced by their inherent bounded rationality, even when
abstracting from the costs of negotiating, writing, and legalizing the contract, to
neglect some variables whose effect on the relationship is difcult to evaluate.
26. Hart uses car manufacturer and electricity supply to illustrate his reasoning, but the rational and
arguments are largely valid for urban mobility systems, which enable us to establish the parallel between
Hart formulation and our urban mobility problem.
Very often the cost of considering an unlikely contingency outweighs the benets
of specifying it in a specic clause.
It is sometimes cumbersome to assign a probability to some relevant events and
thus to condition on such circumstances, which is again originating in our bounded
rationality.
The main idea brought forward by Hart (1995, p. 4) was that contractual
incompleteness and power can be used to understand a number of economic
institutions and arrangements. In fact, several theories have addressed the problem of
contracts before but never associating it to power as this is not a standard feature of
economic theory. For example, in general equilibrium theory, it is assumed that every
agent abides by the terms of any transaction she enters, making power an irrelevant
aspect, while in game theory, market power concept is put forward. However, this
concept says nothing about how to allocate power within the relationship, and we
should emphasize that an institutional arrangement is always an instrument designated
to allocate power among the agents. However, it is only the theory of the rm that raises
the discussion on power, based on the idea that rm boundaries are chosen to allocate
power optimally among the various parties of a transaction.
According to Hart (1995, p. 29), ownership is a source of power when contracts
are incomplete. In fact, ownership gives privileged information, especially through
the access to the accounting system. The owner chooses which depreciation method
he wants to use and has the capacity to distort transfer prices. For some authors
(Grossman & Hart, 1986, p. 705), this privileged access to information is not the type
of right that can be contracted to some outsider, it is rather a residual right of
control, which is tied together with ownership. It is worth referring that these authors
ignore the effects of information access inside the rms and the impact of the internal
structures of power in that process, which stands in both sides of the relation that is
internal to the agent and internal to the principal, and may have signicant
interference in the external relations of the rm.27
As an exception, Schmidt (1996a, 1996b p. 573) reports his work on cost and
benets of privatization, where he slightly addresses this issue distinguishing the
behavior of the employee manager from the owner-manager. In his argument, the
manager of the rm is an empire builder who derives benets from higher production
levels and rm expansion. That is, the principalagent problem is placed at the rms
inner context. This is an issue also extremely evident in public capital rms where
managers are subject to political nomination. Also argued by Schmidt (2002, pp. 55),
as a corollary to Hart conclusions, that if comprehensive contracts are being written,
as supported by the principalagent theory, then the organizational structure will not
play a role and reinforces that Any allocation that can be implemented through a
given organizational structure could be implemented within any other organizational
27. This problem is extensively developed in the eld of psychosociology of organizations but falls beyond
the borders we have dened for this work, the reason why we limit ourselves to point out the importance of
enclosing it within the factors affecting the contractual dynamics between agents.
structure through the appropriate choice of a comprehensive contract. Hence, parties

can only write incomplete contracts ex ante that have to be completed and renegotiated
as they go along.
Observation of reality in urban mobility systems (EC, TIS.PT, 2001d,
MARETOPE, D3) provides evidence that interaction between agents conicts,
alliances, negotiation, cooperation, and even exercise of power occurs indeed within
the constraints and opportunities offered by existing structures, but it is simulta-
neously the interaction between agents that forms the movement of restructuring and
change of the system, as well as external factors, outside the control or even inuence
of any of the agents in the system. This means that structures are expected to have a
dual role, as they constrain some actions but should also enable those evolutionary
processes. Thus, the evidence of a permanent coevolution between agents, structures,
and contracts is a major reason to write incomplete contracts.
An incomplete contract will have gaps, missing provisions, or ambiguities that will
create situations in which some aspects related with the use of nonhuman assets will
not be specied in the contract. According to Hart (1995, pp. 2325), missing usages
will fall into the logic of property rights approach, according to which it is the owner
of the asset who has residual control rights over that asset, that is, residual control
rights are themselves the virtual denition of ownership.
The rst formal model of incomplete contracts is due to Simon (1951, pp. 293305),
and despite being developed for a completely different environment (employment
relationship), it raised a lot of questions, which are valid for the specic relation
between an authority and an operator and at large for any principalagent
relationship. Among the most important, we select the following ones: Why do the
buyer and the seller have to agree to a trade before the uncertainty is realised? or
Why cant the buyer and seller agree to a contract specifying a stage-contingent
delivery plan rather than an acceptance set?.
Kartacheva and Quesada (2000, pp. 7587) show through a government auction
of the concession of some public utility that the possibility of renegotiation provides
incentives for rms to announce lower bids than they would if the renegotiation
possibility would not exist. This is based on the argument that rms have the
capacity, through their market knowledge, to anticipate a contract renegotiation if
demand is low. In this case, results are very likely negative and governments will
prefer to renegotiate instead of stopping the provision of the service, and at that
stage, there will be no competition.
From the evidence provided in several studies reported in the literature (Estache &
Quesada, 2001; Crampes & Estache, 1997), we can conclude that the efciency level is
far from being the unique relevant variable driving the selection of a rm to operate a
concession; equity may also be considered and even prevail to efciency. In addition, an
authority (principal) cannot force a rm (agent) to charge prices close to their average
cost because the rm will react according to its bargaining power, and therefore, the
authority will not be able to extract all the rents from it. This constraint, raised by the
agent behavior and bargaining capacity, leads to the conclusion that if fear of
renegotiation exists, then efciency cannot be the main decision variable when selecting
a rm to operate a concession. Indeed, the authority (supposedly acting on behalf of the
consumers well-being) may prefer to award the concession to a less-efcient rm at

exchange of contract stability and lower renegotiation risks.
As suggested by Estache and Quesada (2001, p. 23) model, the benevolent, welfare
maximizing government should make every possible effort to balance the bargaining
power of its regulators with that of the operators to ensure a fair treatment of all users.
Consequently, a main precaution to take is to minimize the potential of collusion or
corruption between the rm and the authority units in charge of renegotiation, which
may not have such a degree of benevolence.
Despite all evidence on the advantages of incomplete contracts, we should also
convey that a number of potential inefciencies can accrue from this formulation for
several reasons, as pointed out by Schmidt (2002, pp. 56), such as the follows:
Haggling in renegotiation yields delay and inefcient decisions.

Asymmetric information may arise during a relationship, which prevents the
implementation of an ex post efcient allocation.
These costs would not matter very much, if the parties could easily switch to
alternative contracting parties.
In many situations, the parties are locked in with each other and a breakup would
be very inefcient. In these situations, the holdup problem arises.
Today, given the experience with contract rigidity, both in services and in
infrastructures, there is an advantage in specifying in the contract the principles that
should be followed whenever there is the need to deal with unexpected issues not
foreseen in the original negotiation.
In organizational economics, incompleteness means there is room for alignment of
incentives. In this eld, adaptation to unforeseen contingencies is a key issue, and
contingencies are seen as a possible problem outcome. Consequently, an open-ended
structure has the virtue of exibility to adjust to a problem-solving framework and
the learning that accrues from the multiple iterative ows between successful and
unsuccessful approach to problems. Therefore, problem-solving and consequent
learning are activities of contingent character that cannot cope with the rigidity of a
closed (complete) contract that leaves no other option than entering into a
renegotiation process to face any unforeseen occurrence or simply the natural
strategic drifting of the different agents and consequent misalignment regarding a
previous stage of stability.
Transaction cost theory, which started with Ronald Coase (1937), already placed
the emphasis on the cost of writing and negotiating contracts. However, many
authors extended the interpretations of transactions costs to various other sources
(Williamson, 1975; Klein, Crawford, & Alchian, 1978), and more recently, Hart,
1995); some of them associating it to a special type of production costs that
materializes after the product rather than before or associating it to knowledge costs,
that is, the costs of being asymmetrically and imperfectly informed (Foss, 1996, p. 4).
But as Foss refers, the transaction costs associated with these problems are of an ex
ante decision nature, that is, whenever an agent is confronted with taking an option
on an ex ante basis.
Agency theory, which already incorporates some contracting costs, assigns all these
costs to the costs of observing variables. That is, if a variable is observable by both
parties, then the theory assumes that it can be contracted with no marginal cost. In
practical terms, this means that in a standard principalagent theory, setting the optimal
contract cannot be the rst best, but instead, it should specify all parties obligations in all
future states of the world, to the fullest possible extent (Hart, 1995, pp. 2122).
In fact, in the typical contractual relation within the urban mobility system,
irrespective of the type of tender and the contract chosen, the authority (principal)
faces the problem that (even with an experienced and known operator) it does not
know with certainty the capabilities of the other candidate operators, but it knows
with certainty that no institutional design and structure will possibly enable the
constant monitoring of agents actions, given the fact that there are too many
stochastic variables that will affect agents performance.
The prevailing problem will then be to obtain an optimal solution for the contract
design, contemplating adequate incentives, with the information available to the
principal at moment zero, which continues to be optimal throughout the contract
execution, without the need to adaptation. In this perspective, transaction costs are
associated to quantity of effort put in the adaptation processes, and contracts should
be designed as problem-solving structures and, as such, open-ended structures, that
is, incomplete.
The advantages of contract incompleteness must be thought for the two parties in
negotiation. That is, incompleteness is expected to protect not only authorities
interests but also the ones of the operators.
As Coase (1937, p. 87) noted while discussing long-term contracts, owing to
the difculty of forecasting, the longer the period for the contract is for the supply
of the commodity or service, the less possible, and indeed, the less desirable it is for
the person purchasing to specify what the other contracting party is expected to
do. It may well be a matter of indifference to the person supplying the service or
commodity which of several courses of action is taken, but not to the purchaser of
that service or commodity. But the purchaser will not know which of these several
courses he will want the supplier to take. Therefore, the service which is being
provided is expressed in general terms, the exact details being left until a later date.
However, legal and economic perspectives on contract renegotiation are extremely
divergent. As synthesized by Schwartz and Watson (2000, pp. 13), there are a
number of effects accruing from contract renegotiation:
The renegotiated contract is likely to reect better the interests and intentions of
the parties in the agreement as it is based on more current circumstances and not
only on a ex ante analysis done by the parties at the beginning of the relationship.
Some authors have identied potential for ex post efciency as renegotiation will
allow parties to realize gains that under the initial contract would have been lost,
although economists counterargue that even with the advantage of ex post creation
of value renegotiation of contract tends to distort ex ante incentives, since even
the probability of renegotiation by itself generates an opportunistic behavior of the
parties and hinders the potential impact of incentive mechanisms.
Other authors point out the advantage of exibility gained through contract
renegotiation whenever contractual terms are tailored to individual contingencies.
This exibility is seen as an advantage to help managing the risk while maintaining
incentives in a principalagent relationship. Flexibility is also associated with
economies in contracting costs to the extent that parties can wait to see what
contingencies arise before specifying the terms.
Finally, renegotiation can also benet from ex post bargaining power to motivate
ex ante-efcient investments.
As we have seen in this chapter, problem-solving is essentially a learning process

instead of a command-execution activity. In the principalagent relation, this enables
each actor to adapt and exploit unanticipated learning. However, as Foss (1996, p. 9)
notes, problem-solving is by denition an activity with hierarchical complementary
features, that is, problems are usually solved by dividing it into subproblems. In
other words, the problem-solving approach starts from generic features and goes
down in hierarchy to detail, decomposing the original problem along an iterative
trial-and-error process.
In a complex system like urban mobility, the existence of a concertation agent, as
liaison between the several problem-solving processes, enables superior exibility,
which is largely facilitated by the stock of accumulated knowledge that each
problem-solving process provides to the others, progressively reducing coordination
and information costs as new problems arise, while it eases the interpretation,
interagent communication, and even the handling of unforeseen contingencies.
Moreover, there are economies of scale in knowledge acquisition and institutional
learning accruing from the complementarity between problem-solving processes with
several agents. However, a delicate issue is the concertation between all the problem-
solving processes, which is in itself the pragmatic justication for the existence of an
agency acting as the principal for all the urban mobility system and ensuring the
good design of the contractual incentives that will have a strong complementarity
and, consequently, promote interaction between agents.
Given all the arguments expressed earlier, we conclude that, rather than being a
decient instrument for the management of urban mobility systems, incomplete
contracts do represent an essential adjustment instrument as it provides room to
develop a dynamic learning process for systemic interactions, while accumulating
essential knowledge to devise more effective incentives. That is, in urban mobility
systems, it is precisely the incompleteness of contracts that allows the market,
with several levels of interactive processes between agents, to function as an adaptive,
cognitive system, as long as feedback mechanisms are established to enable creation of
an effective information system that will feed all agents in their process of development
of knowledge and innovation. As Tushman and Romanelli (1985, p. 177) explain,
agents interacting in a market create webs of interdependent relationships with
buyers, suppliers, and nancial backers(y) and patterns of culture, norms and
ideology, which are at the root of their resistance to change.
The main limitation of long-term relationships is the existence of external
opportunities, which are forcing two partners to remain attached to each other
without obvious gains, which leads to a situation in which breaching may be desired
even if punished with high penalties. Therefore, instruments have to be thought to
nd an optimal trade-off between exibility and the prevention of opportunism.
However, as noted by Tirole (1988, p. 27), another hazard related with long-term
relations is the tendency for collusion between the partners, a long time horizon
gives them time to reciprocate favors and to become condent that collusion is
sustainable, which in turn creates inefciencies. At the opposite, he argues that
short-term relationships are seen as advantageous by parties who believe they have
good future opportunities, and as these opportunities are supposedly related with the
ability to perform well, that party should see in short-term contracts an incentive.
The rst situation is rather common in urban mobility systems (both in operators
and in infrastructures), whereas the latter is extremely rare in this sector.
Following this rationale of pros and cons of long- versus short-term relationships,
Williamson (1985, p. 347) inspired in the US context considers that regulation may
be described contractually as a highly incomplete form of long term contracting in
which (1) the regulatee is assured an overall fair rate of return, in exchange for which
(2) adaptations to changing circumstances are successively introduced without the
costly haggling that attends such changes when parties to contract enjoy greater
autonomy. In complement, Newbery, (2000, p. 55, citing Spiller, 1993) states that to
ensure superiority of privately owned network utilities over publicly owned ones, the
regulatory agency needs to be restrained and have its discretion limited, supported by
governments committed to ensure stability of the regulatory framework. However,
this equilibrium will only be credible if appropriate institutions ensure enforcement
of the basic requirements.
In the cases observed, we noted that in Europe, there is a clear trend to use shorter
contracts for system exploitation, typically of 5 years for buses taking advantage of
the existence of market for these actives (i.e., the rolling stock market) but with the
risk of sacricing investments in the soft side of the service and in market knowledge,
which will only be possible with longer contracts. The use of short-term contracts
implicitly assumes that know-how from exploitation is transferable (which is far from
being consensual given its implication is several corporate aspects, namely, labor
relation), and therefore, a stronger emphasis is put on competitive pressure between
operators and the possibility of redening the authority requirements.
For North (1991, pp. 97112), institutions are sets of rules, formal and informal,
that organize and limit human interaction, which are supported by a heavy and
complex set of rules, norms, expectations, sanctions, largely responsible by their
inertia, making their evolution typically incremental, and history related. North
notes also that if players are few and interact in repeated games with full information,
cooperation will be a natural outcome of this interaction, but whenever these
conditions are absent (which happen in most if not all cases), additional structures
are needed to restrain opportunism and secure efcient outcomes.
The recognition of these problems is further consolidated by Williamson (1985,
p. 359), who takes the new institutional economic approach, where information is
considered as incomplete and transactions costly. Transactions are considered as the
basic unit for institutional analysis, and, consequently, the institutional ability to
organize transactions economizing on bounded rationality, warding against

opportunism, is the key determinant of a successful evolution of institutions. As
Newberry (2000, p. 53) explains, this perspective of institutional design builds on
three following main features:
Bounded rationality, or the costs of acquiring and processing information;

Opportunism, or the use of guile and deceit to distort outcomes to the benet of
the agent (also known as moral hazard);
Asset specicity, or the extent to which the original value of assets is hard to
realize after they have been committed to their chosen purposes.
It is the relative importance of these features that determines what is the

appropriate institutional setting for mediation of these transactions, that is, what we
usually designate as the regulatory framework. As Newberry observes, if there is
bounded rationality and opportunism but no asset specicity, competitive outcomes
are efcient (e.g., contestable airlines with easy assets relocation); if there were no
problems of bounded rationality, both opportunism and asset specicity could be
handled in contracts with detailed anticipation of contingencies; if there were no
opportunism, transactions could be handled by simple contracts, or even just
promises, without need of further monitoring or enforcement.
However, if all the three features are present, then governance structures are
needed to specify tasks ex ante as well mechanisms to monitor and enforce outcomes
ex post. This is what we usually designate as the organizational frameworks where
processes are depicted and responsibilities allocated to the various agents. Together,
regulatory and organizational frameworks set the scene for agents interaction
and system management, therefore constraining the evolution of market structures,
and contracts represent the main instrument to maintain system exibility and
equilibrium.
Chapter 4
Conguration of Quality Factors in Urban

Mobility Systems
4.1. The Quality Concept in UMS
Given the concepts advanced by soft systems thinking and further elaborated in the
previous chapters, the root denition of a dynamic human activity system with
diversity and disparity of agents,1 such as an urban mobility system, should be a
concise structured description of what the system is, its main elements, the processes
which determine the way in which steering of activities should be organized, and the
properties or attributes which describe its main characteristics.
These concepts can be organized into a logical body with ve basic
constructions that form the key building blocks of the mobility system:
Finality, that is system purpose.

Entities, that is agents with different roles and capacity to inuence the
performance and reconguration of the system. These can be decision agents or
simply reactive agents in the sense that these reactions are able to modify the level
of resources left for the next system iteration.
Boundaries of decision, following Checkland (1999, p. 312) are limits up to where
the decision-taking process has intervention power, that is power to make things
happen or to prevent or delay them from happening. In our interpretation
boundaries determine the place of variables in relation to the system as
endogenous, exogenous, or irrelevant for the decision process.
Decision processes, being sets of activities that enable the system to move from one
stage to another, that is the means for its organization and reaction to external
stimulus or phenomena that may impact on the capacity to pursue its purpose. In
this process the system continuously strolls between near to equilibrium and far
from equilibrium situations. We can envisage three domains for developing
decision processes, which represent the underlying normative structure for the
system functions and also for the decoupling of objectives in a consistent way:
purpose-related processes (entailing strategic decisions);
organizational (or institutional framework)-related processes (entailing tactic
decisions); and
1. We distinguish diversity, meaning number of types of agents, from disparity, meaning number of basic
types of organizations to which the agents belong.
production-related (entailing operational decisions).

Each decision process should in turn be identied with the following
components:
Goal, that is the driver.
Decision-taker (s), that is an agent of the system with capacity to directly
inuence its evolutionary path.

Discrepancy (gap) between goal and apparent condition, that is the decision
booster or the result of the appreciative function we have referred in Chapter 2.

The apparent condition against which the goal is compared (stocks2), is the
characteristics of the departure situation over which transformation is
expected.
Resulting transformation action, that is interactions caused and outcome
obtained and the respective distribution of impacts by the agents, in the three
relevant time horizons, that is short, medium, and long term.
Feedback loops, basic mechanism for the dynamics of the system, through
interaction among agents of the system and also between these agents and any
entity in the surrounding environment. These loops should be identied in terms of
the following elements:
decision level from where the ow of information or action is coming;
decision level where the ow of information or action is heading; and
the loop with the identication of positive and negative feedback, that is the
causal link. A major difculty in the complete description of these loops relates
with the type of ows (action) that is induced in each agent by the behavior of
the others.
Following Sterman (2000, p. 102), stocks and ows emphasize the physical
structure of the feedback mechanism. Stocks characterize the state of the system and
generate the information upon which decisions are taken. Rates are the ow of
increase or decrease in stocks. Decisions, in turn, are the instrument to introduce
variation in the rates of ow.
As Forrester (1964, p. 33) observed, dening the system boundaries and the
degree of aggregation are two of the most difcult steps in successful modeling.
Notwithstanding the importance of the structured description of the Urban
Mobility Systems it is also indispensable to understand the meanings of the term
quality.
Of the many meanings found in the literature, we take as starting point of our
discussion the meaning adopted by Juran, as being of critical importance for
managing quality. These are Quality means those features of products which meet
customer needs and thereby provide customer satisfaction (i.e., income perspective)
2. In systems dynamic terminology the stocks characterize the state of the system and generate the
information on which decisions are based. Stocks and ows are the physical structure underlying the
feedback structure of a system (Sterman, 2000, p. 102).
Conguration of Quality Factors in Urban Mobility Systems 139
and Quality means freedom from deciencies freedom from errors that require
doing the work over again (rework) or that results in eld failures, customer
dissatisfaction, customer claims, and so on (i.e., cost perspective) in the sense of
tness to use (Juran, 2000, pp. 2.12.2).
These denitions are far away from the long-standing denition of quality as
conformance to specication, which implicitly assumed the full knowledge of the
customer needs and its correct translation into product and service specication.
Indeed, as already pointed out in chapter 1, the 80s witnessed a wide broadening of
the quality umbrella, encompassing upper management functions, and substantially
changing the meaning of some terms. Table 4.1 below adapted from Juran (1990,
pp. 110) and Juran (2000, p. 2.4) highlights the main differences of that movement
of increasing scope.
The broadening of the quality concept was also accompanied by a more realistic
perception that customers were not so predictable and controllable as implicitly
assumed during the era of conformance to specication. With this evidence, the
falsication of the previous statement in which quality approaches were largely based
was also done, that is the condence that products conforming to specication would
also meet customer needs ceased to exist, and in some systems (such as the urban
mobility systems) evidence was found that the remote causes for customer
unsatisfaction could be placed at the strategic and tactical levels of decision, that
is a quality gap was found to exist at the specication stage. These ndings
contributed to the genesis of what we call today CRM Customer Relationship
Management, by creating the awareness that there is a need to know better and
continuously who our client is and what she needs.
As Malhotra and Birks (2003, p. 110) state The principles of CRM are simple.
Businesses gather accurate information about customers and prospects. Having
identied the customers that account for the highest prots, they devise marketing
strategies that differentiate between different groups. Greater resources are
focused on higher value customers. Every opportunity is used to amass additional
information about each client to personalise sales messages and build a closer
relationship.
The opening of the quality approach to services reinforced the need to deepen the
knowledge on the differences of customer cognition of quality for products and
services, and on the discernment that customer needs include many things that
cannot be found in product or service specication. These are features such as service
explanation, condentiality, freedom from burdensome paperwork, convenience,
etc., some of which can only be enclosed in the quality frame of a given service
through mechanisms of analytical scaling and packaging at the system level.
A direct consequence of adopting this holistic quality doctrine for the manage-
ment of the Urban Mobility System is thus on the denition of some related concepts
required for the identication of the quality elements, which so far have not been
subject to the standardization of their meanings. These are:
Mobility service: The output of any process that enables the spatial displacement
of persons or goods.
Table 4.1: Evolution of quality related concepts.
Concept designation Restricted meaning (o1980s) Broad meaning (W1980s)
Products Manufactured goods All products, goods,

and services. Even
the ones not for sale
Processes The ones directly All processes, including
related to manufacture manufacturing support,
of goods business, etc.
Industries Manufacturing All industries, manufacturing,
services, government, etc.,
whether for prot or not
Quality is Technological problem Business problem
viewed as
Customer Client who buys All who are affected,
the product externally and internally
How to think Based on culture of Based on universal trilogy:
about quality functional departments quality planning, quality
control, and quality
improvement
Quality goals Among factory goals Within companies
are included business plans
Cost of Costs associated with All costs that would
poor quality decient manufactured disappear if everything
goods were prefect (e.g., service
disruptions, complaints,
indemnities, service
guarantees, etc.)
Quality evaluation Conformance to factory Responsiveness to
based mainly on specications, procedures, customer needs
and standards
Improvement is Departmental performance Company performance
directed at
Training in managing Concentrated in the Companywide
for quality is quality department
Coordination The quality manager A quality council of
is by upper managers
Source: Author.
Mobility infrastructure service: The use of any mobility infrastructure by service

providers or by the nal user on self-service regime (air, rail, road, waterborne,
walking way, or cycling ways).
Ancillary mobility service: All services that contribute for the provision of essential
mobility services and inuence the users perception of quality (e.g., information,
police, emergency services, signaling, etc).
Service feature: An attribute of the service that is intended to directly or indirectly
meet customer needs;
Customer: Anyone affected (so a stakeholder) by the essential or ancillary services
or by the process used to produce those services. This denition includes the
individual user and also society in general.
Quality criteria: Aspects valued by the customers in their use of the mobility
system. These criteria have to be considered as variable according to customer
needs and expectations raised on system attributes. Customer needs can vary
between different segments of clients but even for the same client when exposed to
different circumstances.
Direct quality factors: Observable elements of the mobility system that inuence
appreciation (observation and judgment) of customers on the capacity of the system
to fulll their real needs and expectations. These elements are the valuation features
of the mobility systems over which the customers assess their perceived quality,
ranked through the criteria referred above, and presented in Table 4.2.
Indirect quality factors: Elements not visible by the customers that inuence the
planned and delivered quality and, consequently, have an indirect inuence on
customer perception.
In this perspective we consider the integrated urban mobility system to be composed

by a set of means and services, linked through interchanges, and classied along a
functional spectrum according to the type of use (i.e., services provided or self-service),
the availability to users, and the freedom of movement offered (in terms of time and
space).
The physical networks of the urban mobility system are then composed of: the
traditional public transport (also designated as mass transit, mainly in the United
States); intermediate services, also designated as paratransit, that includes the jitney
(small van with xed route and fare but exible stops very common in South America
and South Africa), the collective taxi (very used in the Netherlands and also in Brazil,
although here it is still illegal), the individual taxi (with car), the moto-taxi (currently
proliferating in South America), and all types of call mobility services; the private
services, such as walking, cycling, motorcycles, and private cars.
Customers can be internal and external to the system. The former group includes any
entity that is part of the production process, and so internal to the mobility system, who
has to rely on the good performance of the suppliers of the system components placed
upstream and downstream of its own contribution. This interdependence highlights the
process orientation of the functional organization of the mobility systems.
Entities which are not part of the production process are considered as external
customers, just like the users who are the raison detre of the mobility system. An
Table 4.2: Valuation features in urban mobility systems.
Hard feature Soft feature

Territorial insertion of the different Price-quality relation of
networks (i.e., the role of the UMS all elements perceived
as a rope interlinking all economic by the customer
and social activities, ensuring citizens Information services
are not excluded of access to those activities) (to agents, customers
Network of available services (conguration and society)
and hierarchies) Customer handling
Service characteristics (e.g., frequency, services (including handling
stability, etc.), diversity and disparity of complaints)
Interchanges, stations, and terminal Human agents and the
buildings way they perform
Prioritization of public transport for Trafc uidity
road and rail infrastructure Security perception
Vehicles (availability of staff and
Security (hard) devices (e.g., camaras, devices providing
etc.) in vehicles and infrastructures perception of security)
Infrastructure (moving and parking) Safety perception in services
Signaling and infrastructures
Auxiliary and emergency facilities
Roadways
Railways
Parking
Pathways (walking and cycling)
Police
Source: Author.
exception to this denition occurs when either the traveling process or the
technological devices require the user of the mobility system to be involved in the
productive process through any sort of self-service action (e.g., self-ticketing,
paying the parking station, etc.) in which the user takes an active role usually in the
last part of the service. Notwithstanding, in general we consider the individual and
collective user of the mobility system as external customers. An exception to the
previous is done for the driver of the individual motorized transport and other
services in self-regime where the user is simultaneously internal and external client.
Internal customers can be further disaggregated into the main roles an entity (agent)
can play within a mobility system: policy maker; regulator; organizer (or planner),
supplier, processor, etc. Very often the same entity plays multiple roles, such as
supplier, processor, and customer, at different stages of the process of setting and
running the services. That is, an entity can receive a part of the service from another
entity, process that part within its own production process and pass it to a third party.
Information provides a good illustration for that multiple role character that
substantially increases the complexity of the identication of internal customer needs.
Besides, it is worth referring that, given the organization of the mobility system, the
effectiveness in meeting the needs of internal customers is critical to the provision of
services of high quality to the external customer, since it affects performance of processes.
Indeed, quality denitions may differ substantially when switching from the internal
to the external customer perspective. Within the work developed in the research project
QUATTRO (EC, OGM, 1998a, p. 99), also reected in the norm issued by the European
Standardization Committee (TC320/WG5/13816-2002EN),3 we have decoupled the
denition of quality into four main concepts that we now adapt to the system level, as
illustrated in Figure 4.1:
Expected quality (Q1) This is the level of quality which implicitly or explicitly is
required by the customer. The level of quality is understood as a composition of a
number of criteria. Qualitative analysis on customer proles and preferences can
assess the contributions of these criteria.
Targeted quality (Q2) This is the level of quality which the service provider or
manager of mobility system is aiming to provide to the customers as a consequence of
his understanding of the customer expectations and of the capabilities of the
productive side of the system. Targeted quality must be set in an objective way and
decoupled through the different services available within the mobility system. For this,
there is the need to identify for each perceived quality element which variables are
meant to be kept constant across the all system (e.g., cost coverage for public
transport), which ones are depending on geographical incidence (e.g., average vehicle
age; emissions per passenger.km, etc.), and also which is the decision level responsible
for each variable, so that the corresponding decision-maker can be identied.
Delivered quality (Q3) This is the level of quality effectively achieved in the
provision of mobility services by the different components of the system, although
not necessarily a coincident image to what is visible by the customers. Delivered
quality must be measured also from the customer viewpoint and not only from the
supply side perspective that is, it should be assessed against the clients criteria.
Perceived quality (Q4) This is the level of quality perceived by the user-customer.
This is inuenced by several factors, such as their personal experience of the service
or from associated or similar services, the information received about the service,
from the provider or other sources, the nonservice elements (e.g., convenience,
etc.), or even the personal environment and needs. In the recently explored
domains of marketing (in its different approaches: mass, one-to-one, relational,
and afuent) this concept is very close the one of customer experience.
3. The author was a member of the co-ordination team of QUATTRO research and was also involved in
the development of the European Standard which was done based on the work of a mixed team CEN-
QUATTRO.
Figure 4.1: Decoupling of quality denition. Source: Author.
The operationalization of these concepts differs from one city to another and even
within the same city, whenever assessed under different circumstances or for different
types of mobility services. However, we can group these concepts in two different
categories: one represents the production perspective-targeted and delivered quality;
and another representing the consumption perspective expected and perceived
quality.
The relation between these four concepts is of utmost importance to adjust the
service both to the stated and to the real (revealed) needs of the external customer.
These relations can also be observed at the service and organizational scale, but their
complexity increases substantially when we consider the urban mobility system as a
whole, as the quality gaps identied in Figure 4.2 result from the interaction of
several agents and processes at the different decision (or planning) levels.
The difference between expected quality (QE) and targeted quality (QT) reveals the
existence of deciencies in the process of identifying needs of the external customers
and sometimes to distinguish between stated needs and real needs, that is, difculties
in reading market signs. These deciencies can be caused by problems at the
observation or at the decision levels: in the former case, this means lack or poor
Figure 4.2: Quality gaps. Source: Author.
effectiveness of the mechanisms for observation and study of customers perceptions

and needs; in the latter it will be either the malfunction of the strategic or tactical
level of decision or the nonexistence of one of these levels, which is a common
situation in Europe. Together these decision levels are responsible for the service
denition, irrespective of the number of agents involved in the process or on who is
the agent taking the entrepreneur role in the creation of services, which in turn is
dependent on the regulatory regime for market access.
Deviations between targeted quality and delivered quality can be caused by
several reasons that might be related either with the service design or production, that
is any situation of under performance related with the provision of the services. This
performance gap is either a measure of the effectiveness of one (or more) of the
several service providers in achieving their own targets or of the effectiveness in
decomposing targeted quality through the different service components.
Perceived quality often holds little similitude to delivered quality. This disturbance
has several causes, it can result from customers accumulated knowledge about
service delivered and of personal or reported experiences with the service under
assessment or with similar ones, and of personal background and environment,
which create an expectation on the service provided. Consequently, perceived quality
is assessed having an expectations scale as a lter for this assessment.
A particular case of application of these four quality concepts to urban mobility
system is the case of infrastructures that are used by customers in self-service base.
The best-known example is the case of road infrastructure where the authority still
has some intervention in the specifying of the quality levels of the offer, namely in the
type of infrastructure, in the circulation rules and parking, and in the management of
trafc control systems and respective time ows, whereas the rest of the service
specication is made by the customer itself during the use of the infrastructure. This
case is of particular difculty in terms of quality management since the perceived
quality is very much inuenced by the good use that all customers make of the
infrastructure. That is, is these cases customers education is fundamental for the
overall quality perception of the infrastructure.
4.2. Direct Quality Factors
4.2.1. Quality Elements
There is no universally prescribed set of quality criteria for urban mobility systems as
this should result from the interpretation of customer needs and aspirations in each
agglomeration. These market readings and criteria setting processes should be done
through an iterative process with the following main steps, largely corresponding to
an application of the principles entailed in the Deming cycle (ISO/TC 176/SC 2, 04/
Dec/2003, p. 14) to the setting of quality targets.
The rst main step is the explicit and implicit analyze of customers expectations,
that is observe and understand current state of needs and future trends, with
adequate instruments such as data collection in mobility patterns, attitudinal surveys,
key informant interviews, focus groups, participant observer techniques, etc. (Clifton &
Handy, 2001, pp. 410). It is worth referring here the work done by Poulenez-Donovan
and Ulberg (1994, p. 3) where they observe that The world of the survey, however, is
bounded by the perspectives and goals of the survey writers. The survey restricts not
only the question frame but also the answer frame, anticipating the important issues
and questions of the responses. The implication is that surveys are not by themselves
fully suitable for exploratory work where the research looks for unidentied issues.
Other instruments should be considered in complement.
Next there is the need to assure adequate frameworks for the provision of services
and, considering those expectations, set minimum performance thresholds for all
components of the mobility system. After this performance, assessment of the several
services should be done, considering adequate deployment of quality criteria to
all components and minimization of performance gaps. This process of quality
deployment should entail relevant attributes for internal and external customers, since
there are interdependencies within the different processes extending across
the different spheres of customer types. Finally, the last step consists in the assessment
of customer satisfaction, analyze of results and consequent readjust of quality targets.
This systematic process has to be repeated for all quality elements of the mobility
system to give an operational dimension to the quality criteria used when setting the
performance thresholds for a specic mobility system. From the observation of the
different urban cases that support this research we conclude that the operationaliza-
tion of these criteria is specic to each system.
For this operationalization a three-tier approach to the analysis of the evolution
of demand is required, complementing usual quantitative studies that support
forecasting models with qualitative studies, for a better understanding of individual
choices and attitudes and, nally, with studies on evolution of spatial distribution of
the social and economic activities inuencing demand. In fact, greater use has to be
made of research work being carried in the domain of activity-based travel modeling,
where the daily agendas are considered and the most convenient mobility chains
identied, possibly using multiple modes of transport.
Despite the uniqueness of each city regarding the operationalization of quality
criteria, it is possible to identify the following list of aspects that underlie the
denition of those criteria, representing aspects valued by the external and internal
customers in their relation with the mobility system:
Observable attributes valued by the external customer of the urban mobility

systems:
For the individual customer:
availability of alternatives for mobility needs (space and time coverage of
mobility network, that encompasses all modes);

physical and nancial accessibility to public transport;
reliability of public transport services (schedules and routes);
comfort in public transport and respective interchanges (seating, ride comfort,
personal space, cleanliness, light and ambiance);

good facilities of interchange between publicpublic and publicprivate

options of the mobility chain covering physical, logical, and price aspects
(e.g., with the same card paying private parking and public transport);
conditions of trafc ow in infrastructures for individual motorized
transport;
availability of parking and good qualityprice relation for individual
motorized transport;
reliable auxiliary services (emergency, police, etc.);
safety and security in all mobility services;
customer care in services and infrastructures;
human agents and the performance of their contact with users of services and
infrastructures;
concise and precise information where/when needed; and
time spent in door-to-door mobility.

For the collective customer (i.e., society):
overall economic efciency in production, meaning efcient management of
resources in production, and conception of services and infrastructures, given

the objectives set for the mobility system;
overall economic efciency in consumption, reected in the capacity to induce
adaptive behavior of the user to reduce the social costs borne by society as a
whole;
cost coverage of public expenditure, or economic sufciency, as the ability to
nance implementation, operation, maintenance, and expansion of services

and related facilities;
distributional effects, as seen in a holistic perspective of welfare impacts,
considering the discriminatory potential the mobility system might bring in

the geographical, social, and sectorial dimensions in an equity perspective;
a balanced mobility system enabling the exercise of citizens free choice but
ensuring simultaneously ecological sustainability, that is limiting the
externalities to an acceptable endurance level.
Attributes valued by internal customers, largely invisible to the external customer,
although indispensable for a good overall system performance:
a functional and vocational approach to the interaction of the different modes
with the various urban demand proles enabling to provide each mode and
service with the most adequate operating conditions;
based on the previous aspect, performance valuation of the widest concept
of network design (entailing all modes) submitted to the urban development
plans;
clear objectives and allocation of means for the whole mobility system (services
and infrastructures) with objective description of intermediate targets for the
different agents and services;
good provision of a stable basic structure for the mobility system underlying
institutional interaction within the system;
clear allocation of responsibilities and roles in the different processes;
management structure that provide the urban mobility system with sufcient
robustness so that the several private and public agents can engage in long-term
commitments for system improvement;
fair and efcient rules, monitoring and enforcement mechanisms;
fair and transparent rules and mechanisms for procurement and management of
contracts for public services (operations and infrastructures); and
Nondiscriminatory treatment of suppliers of services and materials.
Despite the long, although not exhaustive, list it is not sufcient to identify the
quality aspects. As already referred we need to understand which decision level is
responsible and able to manage and control each of these aspects and through which
processes.
4.2.2. Sufciency of Traditional TQM-Based Models for UMS Management
From the identication of the quality elements in the previous section evidence
accrues that managing the urban mobility system requires addressing an environment
with high diversity of interacting agents, that confront a not less diverse set of clients
perceptions and aspirations, reected in different market segments.
Therefore, as we have referred in Chapter 1 of this book, to ensure high quality level
of the urban mobility system three distinct universes have to be subject to the Total
Quality Management (TQM) paradigm: the service, the organizations and the whole
system. For each of these universes different quality approaches have to be used.
Indeed, quality management models refer not only to frameworks with their own
unique ways for categorising and organising information but also to a process and
approach to performance management (EFQM, 1999, p. 2). Current models raise
several difculties when we try to apply them to urban mobility systems, due to the
fact that they have been conceived for implementation in organizations with
comprehensive control of their strategic options, specication of outputs, relations
with customers and utilization of means.
Current models are context independent. They assess an organizations processes
against a set of criteria which are neither industry, nor sector, nor strategy specic.
Urban mobility systems are fully embedded in the environment and reveal a strong
interdependence with the evolution of urban environment. The same system can
perform very well in one city and very badly in another, because the tness between the
system and its environment varies and with it the clients requirements and
expectations. In addition, the denition of what is best largely depends on how the
city approaches the different problems since several alternative solutions exist for each.
Besides, the different perspectives will condition from the outset the identication of
the problem.
The decision-making context, which acts as the backstage for the strategic
denition of the UMS, is complex, for several reasons, among which the fact that
decision-makers are inuenced by neighboring cities, regional, national, and
international policies. The participation of private entities in the process of providing

transport reduces the possibilities for governments to decide by themselves: Instead,
as we have seen in Chapter 3, the number of problems which require participatory
approaches is substantially increased.
Another difculty is the fact that most of the criteria used in TQM models assume
the existence of a clear hierarchical line of command within the organizational
context in which the model is applied. This exists at the level of organizations
providing the services, but it is very difcult to adopt at the service level, as this
represents a process with multiple agents intervention and without a clear line of
command.
To illustrate these difculties Tables 4.34.6 provide the analysis of the
applicability to the different quality dimensions of the Urban Mobility Systems of
each of the assessment criteria proposed by EFQM model, which is the most used of
all TQM-based models for excellence worldwide.
This model is based on nine predescribed criteria (presented in Tables 4.34.6)
against which an organization current processes are self-assessed. Its main
characteristics are responsible for its universality. Nevertheless, and paradoxically,
this largely represents the main reasons why some criteria used by the model are not
directly applicable to the urban mobility system as described also in Tables 4.34.6.
These characteristics are:
It is context independent, with no specic orientation either to industry or sector.

It is organization focused and its formulation with enablers and results has an
implicit cause-effect logic assuming the existence of an hierarchical chain of
management control and a quite deterministic relation, which does not reveal in
the case of urban mobility phenomenon.
It is to enable worldwide benchmarking the criteria used are the same for any
organization.
Management activities are not prioritized. The model gives only a comprehensive
description and assessment of how processes across an organization are managed
and deployed but doesnot provide any indication on where to start neither on
sequences of action for improvement.
Continuous improvement is given as a principle in line with the theoretical
principles underlying TQM approach once the model provides only a snapshot of
the current state of the organization, without prescribing which activities should be
focused for improvement.
From the current analysis we can perceive the existence of several limitations for
the applicability of this TQM-based model to the management of UMS, as presented
in Tables 4.34.6. The multi-agent context that characterizes the UMS forces the
deployment of objectives through the several agents involved. This function is better
done by an executive entity internal to the system, the management authority, an
entity that is responsible for the functions at the tactical level of decision. The
deployment internal to the agents organization is left to be done by their own inner
structures.
Table 4.3: Applicability of the EFQM excellence model to the quality management
of Urban Mobility Systems (Part 1 of 4).
Key aspects Applicability of the criteria in the different quality domains

of the Urban Mobility System
What should be Organizational Service System
addresseda
General EFQM approach

Aspirations and Overall philosophical focus is applicable
benets sought:
Performance
improvement.
Identication
of strengths
and
weaknesses for
improvement
across
processes
Core values: The four core Partially Not applicable:
Organizations values are fully applicable: The
following applicable, but Services are management
these models government provided on a focus is at the
have four intervention in multi-agent holistic
underlying the transport environment dimension of
core values: sector gives so business the system.
origin to a fth focus takes the Business focus
implicit core partnership entails the
Business value Policy approach and system
process focus focus people focus is sustainability
Customer managed taking into
focus autonomously account public
People focus by each interest.
Learning organization. Customer
focus In addition, focus entails
the services two types of
consumed on clients
self-service individual
basis are customers and
reduced to the society at
customer and large. People
learning focus focus is left to
the
Table 4.3: (Continued )

Key aspects Applicability of the criteria in the different quality domains
of the Urban Mobility System
What should be Organizational Service System
addresseda
autonomous
management
of each
organization.
Learning focus
has a similar
philosophical
approach but
it has to cope
additional
dimensions
(system and
customers in
self-service
mode)
Development Applicable: Partially Partially
approach: applicable: applicable:
Process driven But with sector Set of objectives Management
Self-assessment specic and focus at the
fact gathering, adjustments. measurements holistic
data Set of are unique to dimension of
collection, objectives and every service the system
scoring based, measures are provided and should be
detail unique to the must also be strategy
oriented, and transport dened for driven, and so
present sector self-services long-term
focused focused. The
Continuous incremental
improvement learning
process has to
cope with all
stakeholders
needs and
expectations
Source: Author.
a
The synthesis on the general EFQM approach is based in Lamotte, G. (1999) and Hardjono, ten Have,
and ten Have (1997).
Table 4.4: Applicability of the EFQM excellence model to the quality management of Urban Mobility Systems (Part 2 of 4).
Key aspects Applicability of the criteria in the different quality domains of the mobility system
What should be addressed Organizational Service System
Assessment criteria
(1) Leadership: Applicable Partially applicable: Partially applicable:
Conguration of Quality Factors in Urban Mobility Systems

Develop mission, vision, and values. Service purpose and objectives Long-term strategy denition
Personal involvement in have to be predened and must be developed considering
management. Be a role model for a deployed to all agents governance levels (with no
culture of excellence. Motivate, (companies) taking part of the explicit hierarchy between them)
support, and recognize the service supply process. So, in and all sectors interacting with
organizations people what concerns service urban mobility. It is thus a
development, leadership is a consensus building approach
function divided between the steered through the system
managing authority (who leads management by persuasion for
toward system objectives) and alignment of the objectives of
the company running the service the different agents
(who leads toward it owns
objectives
(2) Policy and Strategy: Applicable Not applicable: Partially applicable:
Based on present and future needs, Key processes go across more Policy and strategy imply
and stakeholders expectations. than one organization, so clairvoyance regarding future
Based on information from deployment, communication, evolution of the system,
performance measurement, and implementation occur concertation with stakeholders
research, learning, and creativity within a multidimensional and other sectors and nally
related activities organizational framework very rm steering of
Should be developed, reviewed, implementation. Processes are
updated, deployed through a necessarily multi-organizational
framework of key processes for
communication and
153
implementation
Source: Author.
154
Managing Urban Mobility Systems

Assessment criteria
(3) People: Applicable Not applicable:
People are involved and empowered, People are managed within the organizations, neither at service
rewarded, and recognized for nor system dimension given the multi-agent character of
competencies and responsibilities, service specication and provision, often involving general
and cared for interest constraints in service denition. As soon as higher
Knowledge and competencies are levels of integration are ensured on the production side it is
identied, developed and sustained. very likely that Unions will try to harmonize (or reduce
Resources are planned, managed and differences) in working conditions. However, even if similar
improved agreements are made development of human resources will be
made within an organization or, alternatively, trends might
develop to use outsourced man power type of agreements
(4) Partnerships and resources: Applicable Not directly applicable: Not applicable:
External partnerships and internal Resources are shared Resources needed are
resources are planned, managed and within partnerships constrained by
assessed against measurable internal to the arrangements decided
objectives that should reect strategic mobility system but with partners external
choices made external to the to the UMS
organization
(5) Processes Applicable:
Processes are systematically designed, Overall philosophical focus is applicable. The only difference lies in the fact that processes are
managed, measured for performance, organized in a multi-organizational environment
and continuously improved
Processes are improved as needed using
innovation to fully satisfy and
generate increasing value for
customers and stakeholders
Source: Author.
Assessment criteria
(6) Customer results: Applicable
Conguration of Quality Factors in Urban Mobility Systems

External customers are
central to the development
of the organization. Entails
measures of perception and
improvement of external
costumer satisfaction on
quality of the product or
services provided
(7) People results: Applicable Not applicable:
People are recognized as key Given the multi-agent character of the service and system, the focus should be on
stakeholders. Entails agents results considering the objectives of the principal in the relationship. As
measures of perception and referred in the previous item people it is the agent who has the one-to-one relation
improvement of peoples with people. This relation enables to recognize merits, develop careers, and achieve
satisfaction, motivation commitments, even if motivation can be partially reached through the diffusion of
and commitment integrated results on system performance. Consequently at system level only agents
results should be addressed
(8) Society results: Applicable
Entails measures of societys
perception of overall social
responsibility. Performance
indicators should be
devised to monitor,
understand and predict,
and improve societys
perception
155
156

(9) Key Performance results: Partially applicable: Partially applicable: Applicable:

Entails key performance Service specication is Denition of service top priorities is Strategic denition of the UMS
indicators for operational often not under the dened at the system level. However, priorities entails political decision-
measuring in order to full control and there is the need to decouple the key making and concertation processes
monitor, understand, decision of the performance indicators dened at system external to the UMS. Key
predict, and improve key service provider level to service provision level and also performance results must reect the
performance outcomes to self-service regime consistency of the UMS along the
Key performance outcomes three decision levels
are used to provide focus
on top priorities and set
main objectives
Source: Author.
Within this network of relationships communication of a quality culture is

necessarily done through mechanisms of induction between the entity responsible for
the UMS management, acting as principal in the relation, and the different
operational agents (operating companies, infrastructure managers, trafc control,
and user in self-service modes). These limitations, that concentrate mostly in the
so-called enablers of the model (criteria 15) , constitute the main justication for the
development of a UMS-specic model, also based on the principles of the TQM
approach but adjusted to a multi-agent system.
4.3. Relation between Direct and Indirect Quality Factors
As referred in previous sections there is an implicit causeeffect logic underlying the

TQM approach. In a context-independent interpretation this means that there
should be a permanent understanding between the cause and effect, i.e. between
the transformation process and the desired outcomes. This is clearly reected in the
decomposition of the quality concept into the productive and consumption side as
well as in the relation between direct and indirect quality factors.
Indirect quality factors are elements not directly observable by the UMS customer
that are capable of pulling the system in such a way that improves the perception of
customers on the observable quality elements. These can be isolated elements (e.g.,
new energy programs) or processes, that is a logically organized chain of different
elements that may entail organization of people, materials, energy, equipments,
information, etc., into work activities designed to produce a pre-dened end result
(e.g., new organization of clients handling services enabling reduced response time to
satisfy clients complains) or, even more upstream in the supply of urban mobility, the
process of concertation between policies with inuence on the urban mobility systems
(e.g., reorganization of the decision processes between two agglomerations to
improve the response time in adjustment of the mobility system to changes in
land use).
From the previous discussion we can accrue that to improve direct quality factors
(normally observable at the operational decision level) we need to understand the
causeeffect relation between these and the indirect quality factors (spread along the
tactical and strategic decision levels), so that performance control of the later can
contribute to improve the quality perceived in the former. So the processes that relate
direct and indirect quality factors are critical to assure a consistent pattern of
decisions and activities, as described in Chapter 3, which is a precondition for good
performance of a UMS.
According to Riley (in Juran, 2000, p. 6.1) there are three dimensions for
measuring process quality: effectiveness, efciency, and adaptability. A process is
effective if the output meets its customer needs. It is efcient when it is effective at the
least cost. The process is adaptable when it remains effective and efcient in the face
of the many changes that occur over time. To be managed a process must fulll
certain minimum conditions, such as:
It should have an owner, for an easier accountability.

It should be fully dened (inputs, resources to be used, activities to be developed,
outputs, and objectives to be achieved) to enable its control.
Its management infrastructure should be in place.
Its measurements and control points should be dened.
It should be able to demonstrate stable, predictable, and repeatable performance.
To dene which aspects of a process should be measured knowledge on the process

mission and customer (internal and external) needs is required. For example, process
measures based on customer needs are suitable to measure process effectiveness,
whereas process measures based on cost, cycle time, labor productivity, etc., are more
adequate to measure process efciency. Simultaneous maximization of process
effectiveness and efciency, if successful, will very likely lead to service production of
higher quality at lower cost, that is it will increase value for customer and owner.
However, performance measurement is closely related with the direct quality
factors selected and with the structural organization of the urban mobility system.
Indeed, when designing the organizational structure of a UMS two things should be
considered, the process ow, patterns of which are achieved by understanding the
way activities are structured and agents are engaged on it, and how the process
interacts with each decision levels. The later is largely conditioned by the design of
the institutional network where agents interact, the span of control and account-
ability of each decision level and of each agent.
Span of control is dened by the number and diversity of processes that an agent
has under his direct control. Span of accountability is dened by the number and
diversity of functions that each agent is expected to develop and the performance
standards that it is expected to meet. Processes and span of control and
accountability converge to dene span of attention, which is referred to the number
of activities that are within the two previous. As Simons notes (2000, p. 54) in his
analysis about organizational performance, span of attention is a concept with a
different nature since it is dened by the agent itself, while span of accountability and
span of control are always dened by upper decision level, that is they are top down
dened concepts, while the previous is set by the agent as a tool to enable his capacity
to better correspond with the responsibilities received from the upper decision levels.
The good performance of the UMS largely depends on the capacity to control the
critical quality criteria. For this it is fundamental to understand the (rather
stochastic) causeeffect relation in the chain of activities (or processes) that leads to
those quality outputs, in order to know what should be controlled, when to do it and
which performance measures should be used.
As quality criteria differ between cities, since they result from a local
interpretation of stakeholders needs and expectations, no universal set of measures
can be recommended. However, it is possible to identify some categories of
performance measures that should accompany the development of city-specic
performance goals according to the quality criteria selected and the respective cause
effect relation with indirect quality factors.
Performance goals, if well communicated and appropriately deployed, possibly

also through incentive mechanisms, set the guidance for each agent to contribute to
the achievement of the strategic objectives of the system.
For goal deployment a sound measurement mechanism is needed, aligned with the
consistency requirements illustrated in Figure 4.3, where a goal represents a formal
general aspiration that denes purpose or expected levels of achievement (e.g.,
improve productive efciency), and objectives are measurements standards and time
frames to gauge progress and/or simple success (e.g., reduce the waste and scrap in
maintenance department by 5% each semester over next year) that are made
operational through targets, that is concrete values attributed to objectives. As
Simon also refers (2000) goals and objectives can be made actionable only when a
measurement is attached to any set of aspirations (p. 231).
The following categories of the largely complementary performance measures
have been systematized by Harbour (1997, pp. 919) through the formulation of the
following questions that represent a logical sequence of analysis for a given
performance goal/objective:
What is the current performance level?

Baseline measures establish the current performance and form the basis for all
subsequent measures. Collecting initial data to feed baseline measures usually
requires intensive working. The basic rule of thumb is that absence of baseline
measures means nonexisting performance measurement system and conse-
quently what is not measurable can not be improved.
How is a specic performance level changing over time?
Trend measures show the evolution of a performance indicator along time and
enable to devise preventive actions whenever progress evaluation results negative
or too slow, when compared with what was planned.
Figure 4.3: Strategic nature of performance goals. Source: Author.

Is performance staying within some predetermined boundary or tolerance level?

Control measures are feedback measures and like trend measures provide early
warnings when some activity or process is straying from a predetermined
performance level.
What is causing a specic problem and where is the problem area located?
Diagnostic measures provide orientation to a specic problem area and have a
preventive character, as they can avoid the rst occurrence of a problem.
Given past and current performance levels, what levels should be planned for in the
future?
Current and past performance can and should be used to support the
development of future scenarios and as feeders of future strategies and
subsequent plans.
As already referred, quality criteria changes with cities, not only because of the
importance given to the different quality aspects but also because of the different
availability of each city to invest in its measurement. To exemplify it we present in
Tables 4.7 and 4.8 the quality criteria adopted for the same purpose by the CEN, the
European Standardization body, (EN 13816:2002), meant to be the operational
translation of the quality aspects for public transport as viewed by the customer, and
the quality aspects adopted by the Public Transport Operator of Oslo, where a
quality guarantee was implemented since 1994.
It is worth referring that the CEN norm is goal and context independent and
ignores the existence of an urban mobility system that goes beyond the public
transport subsystem, which is itself a part of the rst. Consequently this norm fails to
provide a consistent relation between direct and indirect factors of quality. Contrary
to this standardization approach the Oslo Travel Guarantee, despite being also
exclusive to public transport, reects a consistent relation between strategic goals,
quality objectives, and travel guarantee.
4.4. Indirect Quality Factors
4.4.1. Market Segmentation
As already said, the quality management process implies an increased focus on

individual customer satisfaction, which despite the public service component of
mobility services can only be done through the fragmentation of the market into
homogeneous segments with differentiated preferences, so that market oriented
services can be designed to match customers requirements.
The identication of customer needs is thus the start of the process, the major
challenge being rst to understand needs, second to obtain their satisfaction through
good service specication, and third to enable stable provision of the services within
those satisfaction thresholds. The mobility system has specic characteristics that
make quality implementation more difcult. As all other services, production and
Table 4.7: CEN quality aspects for public transport.
Level 1 Level 2
Availability Modes
Network
Operation
Suitability
Dependability
Accessibility External interface
Internal interface
Ticketing availability
Information General information
Travel information
normal conditions
Travel information
abnormal conditions
Time Length of trip time
Adherence to a scheduled
Customer care Commitment
Customer interface
Staff
Assistance
Ticketing options
Comfort Usability of passenger facilities
Seating and personal space
Ride comfort
Ambient conditions
Complementary facilities
Ergonomy
Security Freedom for crime
Freedom from accident
Emergency management
Environmental impact Pollution
Natural resources
Infrastructure
Source: Adapted from EN 13816: 2002.
consumption occur simultaneously and the services are intangible. Additionally, the
production of services with several transport modes is made in an environment
(the public urban space) in which multiple agents have to act on a time and space
sharing basis.
Moreover, in what concerns public transport (the main mobility service provided
on a professional basis), because of its need to provide service to masses of citizens,
Table 4.8: Oslo Sporveir quality guarantee.
Strategic goals
Ensure service quality and Provide a ready channel to Put pressure on the
the customer right to customer complaints company to improve
compensation
Quality aspect Travel guarantee (commitment)
You can rely on our schedules All services will be provided according to current
schedules, unless otherwise announced in
advertisements, posters, and the like
We depart on time not early Departures will not be earlier than scheduled
We always announce the next stop The name of the next stop will be announced on a
vehicles public address system and show on the
electronics display of vehicle tted with them
Destination signs and route number are The destination and when applicable, the route
clear number of a vehicle will be clearly and correctly
displayed
Information at stops is complete Each stop will be clearly marked with its name
and the numbers and destinations of routes it
serves, along with a current table departures
times
Information on board is complete All vehicles will carry information on schedules,
fares and fare regulation. Underground
carriages will also have a map of the
underground
Our service stall will answer your All passenger service staff, at stations and on
questions board vehicles, will answer your questions on
stop and transfer points to bus, tram, and
underground lines. Please do not disturb the
driver when a vehicle is motion
Delays will be announced All delays will be announced as soon as they
occur and periodically thereafter, over the
public address systems of vehicles and stations
Vehicles will be clean and pleasant The interiors of all vehicles will be clean, so you
do not risk injury or soiling your clothing
We will answer your queries We will post a written reply within two weeks of
receiving your query
We will listen to you The complaints, suggestions, and queries we
receive, along with our markets surveys, help us
to improve our service
We will refund taxi fare if you are If a fault in our service results in your being
unduly delayed delays by more than 20 min, we will refund up
to NOK 300 in taxi fare, provided you send us
a receipt for the fare and state the line involved
and the time and place of the delay. The only
exception is when it is clearly obvious that we
are not responsible for the delay
Source: Adapted from UITP-PTI 2/2000, p. 24.

product segmentation is difcult to achieve without signicant increases in unit costs.

This was not a problem when most potential users had no other alternative than
public transport, but this is no longer the case in industrialized countries, in which a
large proportion of mobile citizens have (or could have) access to a private car for
their mobility. Consequently, there is the need to know the different customer
proles so that the bundle of available services may be congured in such a way that
suits the needs of the different segments.
In the past decades new perspectives have been developed in this domain and even
the operators view on the role of marketing evolved substantially and today it is
clearly seen as one of the main driving forces for good corporate results. Already in
2000, the marketing managers of two well-known European operators HT, from
Copenhagen and RATP, from France reported ve ways, largely based on
increasing service awareness to the customer, through which marketing could
increase revenues (Oostergard & Vautier, 2000, pp. 1012):
Increasing the number of customers, especially among the new residents of the
cities;
Increasing the market share per customer, i.e. making existing customers travel
more;
Increasing retail prices;
Increasing the number of paying customers, i.e. avoiding fraud;
Increasing the time period that customers stay with the public transport company
In addition, they conrm that it is no longer sufcient certainly not cost

efcient to direct the service to an anonymous mass of clients, the only exception
being brand building, which creates general knowledge and sympathy for the
services. Relationship marketing is thus elected as a primary tool for systematization
of dialogue with customers to make communication individualized, targeted, and
cost efcient.
This dialogue should be based on customers individual need for trafc guidance,
for example each time there is a change in, e.g. timetables, routes, etc., it should be
announced to the customer directly (on what is called today the one to one
relationship), by means of email messages, mobile phones or ordinary letters, for
which a Customer Relationship Management system (CRM) must be developed to
support adequate market segmentation.
For Zeithaml and Bitner (2003, pp. 158159), focusing on service markets,
relationship marketing (or relationship management) is a philosophy of doing
business, a strategic orientation, that focuses on keeping and improving current
customers rather than on acquiring new customers (y) The primary goal of
relationship marketing is to build and maintain a base of committed customers who
are protable for the organization. To achieve this goal the rm will focus on the
attraction, retention, and enhancement of customer relationship.
In public transport, like in most goods and services of collective offer, quality is
simultaneously simple and very demanding. As these practitioners refer (ibid. p. 12),
basic quality means transporting the customer from point A to point B on time and
with good standards for vehicles, stops and stations, information, acceptable
frequency and geographical coverage, and this is the basic quality level the customer
expects. If not fullled it can lead to dissatisfaction, but even if fullled will not
generate extra customer satisfaction, as the perception of extra quality is only
achieved when the level delivered is beyond expectations, provided it generates
goodwill.
In an urban mobility system, quality is even more demanding as citizens expect to
have available alternatives (that are very often represented by an organized chain of
mobility means that enable them smooth and seamless mobility, at a fair price and
with a low level of organizational, logical and physical effort, whereas undertaking
their professional, social and leisure activities. That is, the challenge of relationship
marketing is to move from the relation between one organization and the customer
to a much more complex interaction at system level and so between a set of
interrelated services and the customer.
Since 1990, Cushman & Wakeeld Healey & Baker, real estate service providers
and analysts, publish the European Cities Monitor evaluation on an yearly basis,
using quality of urban mobility as one of the assessment items to measure
competitive advantage of cities both to attract major investments and in providing
quality of life to corporate senior executives. This recognition of the importance of
urban mobility as a whole illustrates the systemic nature of the problem and points to
the potential that the urban mobility system has to create added value to the city,
whereas suggesting also the need for an integrated approach beyond public
transport.
As Viegas (2003a, p. 90) argues, it makes little sense to impair individual transport
conditions without improving the quality levels of public transport, and creating
good conditions for integration with pedestrian solutions. The unied management
of the urban mobility system should thus take and integrated service provider
perspective of the system, decoupling the quality denition and the continuous
monitoring of the performance of all components as well as the gaps identied in
Figures 3.1 and 3.2 of Chapter 3, aiming to achieve at least the basic quality at system
level as the citizen/customer expects, which is expressed by the level of uidity at each
hierarchical level of the mobility network and by the level of accessibility of the
different areas and for the different attraction poles.
The gaps identied in Figure 4.2 allow positioning the marketing strategies and
decisions as key tools to contribute to close the central gap that is, the satisfaction
or customer gap. Table 4.9 below, identies the main factors at the root of these
quality gaps according to our observation of urban mobility cases and readings from
other public sectors (e.g., health, education, and telecommunications).
From the analysis of the market we can understand that knowing the costumer
is one of the main foundations of relationship marketing which is achieved
through market segmentation. Most of the aspects related to segmentation and
targeting for services are similar to the ones used for industrial or commercial
goods. Table 4.10 (Kotler & Armstrong, 1991, pp. 216249) presents the basic
principles used for market segmentation and targeting and the main steps of this
procedure.
Table 4.9: Contribution of marketing tools to close quality gaps.
Quality gaps Factors contributing to quality gaps
Market (mis) reading gap Lack of process management to focus on customer

requirements
Lack of systematic service development processes
Lack of adequate process to set service quality goals
for basic and extra quality levels
Poor service design (individual services and
networks)
Failure to link service design to customers perception
of value for money
Performance gap Not matching performance with planning
Inadequate coordination between operations and
service design
Inadequate deployment of quality performance
objectives across the system
Delivery gap Difference in policies and procedures of the different
service providers
Inadequate operational procedures
Inadequate feedback for improvement of procedures
Poor service standards denition
Failure to achieve good match between supply and
demand
Satisfaction (customer) gap Lack of knowledge on customer expectations
Wrong service design and standards setting
Inadequate integration or nonevidence on
complementarity between services
Problem in communication to the customer.
Mismatch between performance and previous
promises (building expectation)
Source: Author.
The most relevant difference between marketing for goods and services regards
the need for compatibility of market segments because services are often performed
in the presence of the customer. Consequently, whenever there is a collective offer,
the service provider (or organizer) must be certain that the customers are compatible
with each other. Segmentation requires thus understanding how clients will use the
service and whether segments addressed with different service variations will be
compatible. Air transport explores this attribute in the yield management strategies
where, to avoid incompatibilities, market segmentation is based in the nonvisible
service attributes, such as date of return, number of days staying in destination,
advance notice period for reservations, etc., although recently we can already nd
examples of segmentation based on discrete visible attributes (e.g., seat pitch).
Table 4.10: Principles for market segmentation and market targeting.
Main steps in market segmentation and targeting for services

Step 1 Step 2 Step 3 Step 4 Step 5
Identify bases for Develop Develop Select target Ensure that
segmenting the proles of measures of segments segments are
market resulting segment compatible
segments attractiveness
Bases for market segmentation
Demographic Geographic Psychographics Behavioral

segmentation segmentation segmentation segmentation
Requirements for effective segmentation
Measurability Accessibility Substantiality Action ability
Criteria for evaluating market segments for market targeting
Segment size and Segment structural Company objectives

growth attractiveness and resources
Source: Author based in Kotler and Armstrong (1991, pp. 216249).
The second main difference between goods and services is, as Zeithaml (2003, p. 166)
refers that service providers have a far greater ability to customize service offerings in real
time than manufacturing producers. For this the second step of market segmentation
process is critical, since it is at this stage that the analyst must understand how and
whether the proles of the different segments differ from each other and, consequently,
which service attributes must be used to attract them.
In 1960, Theodore Levitt in his renowned article Marketing Myopia, largely
inspired in the evolution of the petroleum industry, clearly anticipated the role of
strategic and relationship marketing in transportation:
The failure is at the top. The executives responsible for it, in the last
analysis, are those who deal with broad aims and policies. Thus: The
railroads did not stop growing because the need for passenger and
freight transportation declined. That grew.(y) The reason they
dened their industry incorrectly was that they were railroad oriented
instead of transportation oriented; they were product oriented instead
of customer oriented. (Levitt, 1960, reprint HBR 2004, p. 138)
In fact, public transport felt into some pitfalls of what Levitt called conditions for
self-deceiving cycle (idem p. 140), being:
The belief that growth is assured by an expanding and more afuent population.
The belief that there is no competitive substitute for the industrys major product.
Too much faith in mass production and in the advantages of rapidly declining unit
costs as output rises.
Accordingly marketing effort should be viewed as a product developer, not as a

product consequence or a post-sale tool as many people still see it today, and if
needed take it to the limit of what he so well dened as creative destruction toward
customer-satisfying logic. For a better understanding we reproduce hereunder
extracts of Levitt illustration of this marketing concept:
(y) It can be shown that motorists strongly dislike the bother, delay, and
experience of buying gasoline. People actually do not buy gasoline. They
cannot see it, taste it, feel it, appreciate it, or really test it. What they buy is
the right to continue driving their cars. The gas station is like a tax collector
to whom people are compelled to pay a periodic toll as the price of using
their cars. This makes the gas station a basically unpopular instituti-
on.(y) Hence, companies that are working on exotic fuels substitutes that
will eliminate the need for frequent refuelling are heading directly into the
outstretched arms of the irritated motorists (y) Once the petroleum
companies recognise the customer satisfying logic of what another power
system can do, they will see that they have no more choice about working
on an efcient, long lasting fuel (or some way of delivering present fuels
without bothering the motorist) than the big food chains had a choice
about going into the supermarket business of the vacuum tube companies
had a choice about making semiconductors. For their own good, the oil
rms will have to destroy their own protable assets. No amount of
wishful thinking can save them from the necessity of engaging in this form
of creative destruction. (Levitt, 1960, reprint HBR 2004, p. 146)
Following Levitt rationale public transport companies should also stop looking at
transport as their core business and learn from the loss of patronage that occurred
everywhere in the past years, recognizing the need to adopt a customer-satisfying logic and
refocus their businesses toward mobility needs. Today most public transport companies,
and urban mobility agencies (i.e., authorities), have already initiated this movement of
change, however only a few have reected this evolution in their monitoring processes.
Most monitoring systems are still focused on productive performance neglecting
the importance of commercial and network performance at the system level, where
the integration between modes and services takes a prime role, making possible the
continuous assessment of the match between supply and demand of mobility services,
covering all modes and types of services, and enabling the early identication of the
need to develop new intermediate services.
The value of these services is more and more recognized in a bridging function
between more traditional modes, with clear mass transit vocation. As opposed to the
vocation of the latter, intermediate services are market segment orientated and
largely due to their high exibility (physical and organizational) they play the role of
missing link in the design of customer-orientated services as an extension of the
traditional mass transit services.
Good examples of these market segmented services are the train-taxi service
offered in the Netherlands in low demand periods, or the Taxi-Lotac- ao from
Porto Alegre (Brasil), one of the oldest intermediary regulated services in South
America (operating since 1974) (Costa et al., 1998 in Brasileiro, Henry, & Turma,
1998, pp. 338370) initially used where the regular bus had limited physical access or
not sufcient demand and also currently competing in segments of higher income. It
is worth referring that the two examples given conrm the effectiveness of market-
segmentation tools in very different cultural, political, and operational environments.
Many authors have developed on the importance of marketing in transportation.
Hibbs (2003, p. 151) provides a good illustration on the importance of the marketing
management function in managing urban mobility. He states that In transport,
marketing means making sure that what we produce we can sell and we sell what we
produce, at a price that satises both user and seller. According to this, the
following management activities are entailed in the marketing function:
Identication of customers, their needs and wishes, and various market segments.
The conception, production, and delivery of products and services which will
satisfy those customer needs and wants.
Setting the correct prices for the different clients segments and to fulll the
objectives of the mobility system and of the organizations providing the services.
Communicating to the customer the services offered.
Ensure that all activities entailed in customer relations meet customer needs.
In order to fulll these activities the marketing function has to be divided into the
following main areas:
Market intelligence consisting of gathering and interpreting information

concerning the market, covering the following areas, to enable regular monitoring
and also future-orientated decisions:
Modal and intermodal competition;
Actual performance of the rms;
Actual performance of the services offered in the market;
Behavior of the different customer segments;
Evolution of mobility needs as a consequence of evolving spatial planning and
effective occupation;
Changes in the state of the economy;
Social and cultural trends;
Evolution in government and politics;
Technological development.
Service development consisting of developing customer-orientated services

that can also suit the purposes of the service providers and of the all system. This
entails the marketing strategy associated with the placement of the service in the
market. As highlighted by Gubbins (1988, p. 120) the denition of a marketing
strategy implies that all the interrelationships in the marketing mix product,
price, place, promotion, people, physical evidence, and process between the
various elements must be understood and used to enhance the attractiveness of
the service.
Marketing communication and promotion consisting of spreading awareness
about the services offered to existing, new and prospective customers, ensuring
that service development outcomes are well placed.
In a marketing perspective the provision of urban mobility services should begin

by understanding the clients needs and developing the service backwards. First, by
focusing on the material and immaterial attributes of customer satisfaction. Then
moving back and conceiving those element-attributes (e.g., comfortable seats, nice
sightseeing in the walking path, real time information on trafc congestion) that lead
to achieve that satisfaction. Again, moving back still further, looking for ways to
improve the essence of the attributes (e.g., material used in the construction of
the seats or in the cycling lanes) that enable to develop mobility services, and so
forth.
Marketing is thus an instrument to create value in the conception, production,
and delivery of urban mobility services, that should not be mistaken for a selling
technique, and as such is an important part for the development of the conguration
of urban mobility systems. Indeed an important marketing function is precisely to
identify the good balance between the market share of the different modes and
services in the different urban areas, recognizing that in some areas the absence of
adequate public transport, and lack of capacity to promote that type of service in the
short term in areas of very low demand, gives the authority an effective interest in
organizing the mobility system of those areas with a stronger component of
individual transport.
4.4.2. Integration
Integration must be seen as an opportunity to improve quality of the services

provided by the mobility system and with it increase public transport market share
and reduce trafc congestion. It is also consensual that integration, when existing
also at the tactical level, enables to obtain additional advantages on cost efciency
for the whole system since it contributes to avoid duplication of services when no
added value exists for the client. Consequently, intermodality should not be
considered as an objective in itself but instead as a fundamental instrument to
achieve system integration.
Despite the general recognition of importance of integration its operational
translation and optimum level can differ substantially in the views of the different
stakeholders involved in public and private transport. By talking with different

professional key informants we arrived to denitions as diverse as this: optimal
integration is taking into account all relevant aspects related with the service; or, it
is an effective and efcient allocation of resources (EC, NEA, 2003c, pp. 18 and 38,
respectively).
From this eld contact a major lesson obtained was that integration and
intermodality are understood as an important positive element for a good
conguration of the urban mobility service, although at this stage the comparison
between cities is still very difcult and of limited effectiveness, and with it also the
capacity to dene how much integration is needed to achieve a pre-specied quality
effect, not only considering that integration should exist within the public transport
but also between the different forms of public and individual transport and, in the
limit, also within two different options of individual transport.
In the overview of literature related to integration we could nd little development
beyond the descriptive cases, however it was possible to identify four main theoretical
perspectives and respective designation that can be used for the analysis of this issue:
The engineering perspective on integration, based either on evidence of what is

perceived to be a good practice or on efcient use of resources.
The economic perspective largely based on market failure and suggesting
interventions aiming to achieve welfare optimization.
The public management (policy oriented) perspective based on institutional
frameworks and focused on behaviour of actors.
The recent, institutional, or evolutionary economics, driven by the institutions and
their evolution.
In its productive perspective of efcient use of resources the engineering

approach often establishes the optimum with a strong focus in elements such as
eet management, vehicle, and staff roster and planning of service bundling,
without considering the effects of the interaction on agents behavior and the
potential for emergence of optimized congurations of integration out of
market dynamics.
A main weakness of this approach is the reduced focus on the cost-revenue
balance and the implicit assumption that the solutions accruing from resource
optimization correspond to a good reading of passengers perception of quality of
services as an answer to their needs. Under this approach service denition and
integration tends to be done under rather rigid parameters and not as a reaction to
the behavior of demand. Another limitation is the fact that it does not consider the
effects of incentives to integration. So, under this approach integration is largely
established through an optimal mix of different public transport services/modes
having each service characterized through the hard aspects of service design, such as
speed, frequency, and distance between stops, and giving a second level priority to
clients perception, in particular soft elements such as vehicle comfort, service
convenience, information, fares, etc.
The economics perspective is essentially focused on market functioning and takes

as departure base the functioning of markets under the implicit assumption that
deregulation leading to perfect competition will contribute for optimal efciency
outcomes. Suppliers of services are seen as agents pursuing their own selsh
objectives of prot maximization but it is the market through the interaction of
several competitors (in an open market environment) that enables an automatic
control of any disruptive behavior.
In this perspective market failures are seen as the major cause that hinders the
achievement of welfare maximization and these can originate from different sources,
such as:
existence of externalities (that is value effects which are not absorbed by the
market pricing mechanism);
existence of economies of scale and the correspondent development of natural
monopolies;
existence of asymmetric information between participants;
existence of public goods, dened as goods where consumption is not rival and
nonexcludable which is not the case for public transport; and
economies of scope, better known as network effects which are fundamental for
the analysis of integration in mobility services and which are largely under
researched in public transport, at the opposite to what happens in air transport,
although this knowledge cannot be directly transferable given the difference of
complexity between these two realities of services/modes.
It is worth to highlight that according to the ndings of some authors

(Economides, 1996, pp. 681683; Shy 2001, p. 3) a main reason for the appearance
of network externalities is the compatibility and complementarity between
components of the network, represented by the services provided and by the
interchanges where the services come across. To obtain this complementarity
between network elements they should have compatible attributes that is, in practical
terms, feasibility of integration in a strict sense.
In urban mobility systems there may be various levels of compatibility of
integration, as the transfer between subnetworks is always possible, although with
different levels of cost and inconvenience. This creates some competitive pressure
between the different chains of service options offered by the network. Relevant
elements of this cost and consequently also relevant for potential compatibility, are
the information costs on complementary services, costs associated with acquisition of
tickets, opportunity costs (or disutility, i.e., cost of time) of the excess time spent in
transfer as a lack of synchronization between services (or subnetworks), etc.
In practical terms, the concept of network effect is materialized in the quality
improvement achieved through the reduction of the generalized cost for the network
user as the number of transfer options increases, as well as the number of service
options. This improvement can thus be achieved either through the addition of more
services to the main network or through the creation of intermediary services that
will enable the consolidation of the net tissue and which normally also have a
demand complementary to one of the core network.
Indeed, this direct network effect, which occurs as a consequence of the
complementarity between main and intermediary services, results largely from the
fact that improvement in one segment (through cost reduction) leads to an easier
access of the segments used in combination with the rst and thus might result in
subsequent increase of demand. This effect provides evidence on the clear benet
accruing from the use of intermediary services as feeders of the main services, which
is a common practice in South America and South Africa with the van services which
have emerged from market initiative (although illegal at the beginning) as a natural
mechanism of response to market opportunities. However, we should not ignore that
for the operator the costbenet balance resulting from service compatibility depends
on a set of factors related with the way common costs are shared and even on their
dimension.
Besides the direct effects there are still the indirect ones that cannot be ignored,
which is the case of the Mohring effect in public transport (Mohring, 1972, p. 599),
according to which the increase in the demand of one segment will lead to the
increase of frequencies in that segment, which in turn will render it more attractive
and foster additional increases in demand as a response.
The public management perspective on integration is another stream of thought,
where the emphasis lies on agents behavior and barriers to implementation. This
approach typically suggests interinstitutional collaborations, however, tends to ignore
the dynamic characteristics of the organizational and regulatory frameworks. Another
weak point of this approach is the fact that it ignores the spectrum of different possible
commitments between agents that can go from the simple partnership agreement to the
integration of services or, even further, to the common planning.
Finally, institutional economics formulates solutions of integration in relation to
the evolutionary stage of institutions. However, it provides no help for optimization
of institutional design neither for overcoming barriers to achieve system congura-
tion leading to optimal integration. Another aspect that hinders the utilization of
this approach is the reduced transferability of institutional conguration between
cities (Macario, 2003a, p. 548 ).
In brief, each approach has its own perspective on strengths and weaknesses of
integration, and the emphasis between the different aspects of concerns varies
considerably between the different approaches. However, they are also quite
complementary in the contributions given to understand the genesis of the
integration concept and associated processes, which entails providing answers to
the following questions:
Do benets of integration overcome the costs of the process? From the different
studies analyzed the engineering and economic approach allows to conclude that
some forms of integration produce benets but no indication was found on
optimum levels of integration.
If there is support for the argument in favor of the existence of clear benets of
integration, why does optimal integration not appear spontaneously in the market?
So far the main conclusion we have obtained in the already referred study
commissioned by the European Commission (EC, NEA, 2003c), is that emerging
costs and benets are supported by different groups, which suggests the need
to develop redistributive mechanisms, at least to assure that all operators have
some net benet from integration. However, the experience obtained in the
restricted analysis done to integration examples also suggests that an in-depth
analysis of this problem requires addressing it at the light of principles of network
economies.
How to assess the current level of integration and how to decide whether to search
for more integration? Considering the previously presented rationale, the denition
of the appropriate level of integration should be part of the marketing mix
choices for the mobility system. However, while in a corporate business this market
analysis corresponds to a continuous process of sequential phases of planning,
action, check, learn and feed-back, corrective action, etc., in an urban mobility
system this process is undertaken by several agents (authorities and operators)
which makes it much more complex and less prone in producing clear results.
Indeed the concept of integration applied to urban mobility systems is complex

and embodies various scales, as some authors have already pointed out. Potter &
Skinner (2002, p. 282) constructed a nested denition for integrated transport based
on diversied scales of meaning, encompassing aspects such as functional and
planning, environmental, economic, social and political, etc. Viegas and Macario
(2003a, pp. 219222) also emphasized the social dimension in the political decision-
making processes that leads to the denition of the mobility system as well as the
holistic concept of integrated decision-making that should be a key instrument for
sustainability. Earlier, Davidson and Lindeld (1996, p. 34), taking a public
management perspective while addressing the problem of infrastructure development
in India, dened integration as a formalized decision-making procedure to enable a
system to work efciently and effectively, that is to bring its pieces together into one.
Notwithstanding the many reasons that support integrated decision-making for
the benet of the user, there is a major difculty related with the allocation of the
additional costs that are associated to integration of the various mobility agents.
These are, as Viegas (2004a, 2004b p. 4) reports:
network planning, related with the smooth transition from one system component
to the other;
preparation of contractual agreements between the mobility agents involved in the
integrated system;
information systems, to support management both at agents and system level;
contingency planning, to face potential disruption incidents; and
operational staff to ensure clients transference.
Besides the previous list we must also consider the additional equipment required
in the interchanges for clients transference across modes. In fact, as Macario (2004a,
2004b, 2004c, 2004d, pp. 1415) reports, the complexity of mobility patterns that
characterize current societies results in organized chains of unimodal trips, using

preferential nodes of transference that ensure the smooth transition between services
of a multimodal mobility system.
4.4.3. Intermodality and Role of Interchanges
As we have already referred in Chapter 1, the development of cities is usually

accompanied by increasingly complex mobility needs, which is largely caused by the
multifunctional way of living of current societies. This evolution of the logistic
organization of our societies places new demands on the mobility system requiring it
to fulll a number of attributes, such as:
good levels of spatial, temporal, and economic accessibility (i.e., element of

efciency in consumption);
reliability of the services offered (i.e., element of robustness); and
organization of services in such a way that allow the user to improve its range of
reachable destinations.
To achieve this level of quality the transport network and subnetworks formed
of services (arcs) and interchanges (nodes) supporting the mobility system have to
assure the complementarity and compatibility of services. We reinforce here that the
design of the mobility system, and the articulation of the different modal
subnetworks to create an integrated urban network, is within the main functions
performed at the tactical level of decision (or planning level). A poor performance at
this level results in a fragmented urban mobility system, characterized by a prot-
maximizing behavior by all parties, with little or no concern with the performance of
all the system and, consequently, without offering a network that effectively provides
reasonable quality of service for all users with overall acceptable production costs.
The relatively reduced capacity of each mode to serve by itself all mobility needs
associated with a scarce urban space calls for the need to build the network based on
service complementarity and compatibility, that is a synergetic approach to the
mobility system, where each mode has the mission to serve specic market needs
according to its best vocation. Within this structured network the several modes and
services will feed each other allowing different perceived congurations for the
possible mobility chains (or service mix), according to the different intensities of
demand (peak, off-peak, night services, week-ends, etc.).
The door to door mobility chain is thus obtained through complementarity
between all mobility services previously identied. Besides, the service offer provided
by the different modes and services runs in competition between these services,
provide the system with some degree of redundancy and thus making it less
vulnerable to the underperformance of any of his elements.
An important factor for the success of this integrated approach is that customers
must perceive the existence of effective alternatives to the complete journey based on
individual motorized mobility. That is, the dichotomy between private car and public
transport should be replaced by the competitive perspective offered with the concept
of alternative mobility chains where the private car is also one of the elements. To
compete with the attributes of the private motorized mobility the alternatives
involving public transport must be seen as an equally available, easy to use, exible,
and comfortable option, with low perceived costs, that is offering a good relation
between quality and price (the so-called good value for money).
In this structure public transport plays a key role as the main mass transportation
option but it is far from being the unique mobility service that will meet citizens
needs. However, it is indispensable that an effort is made by public transport to
simplify what is originally a complex product, the use of which requires the ability to
manage considerable amounts of information in the advanced knowledge of
timetables, routes, connections, etc. So to improve its ease of use urban public
transport needs to offer (Morrison, 1996, p. 256; Viegas & Macario, 1998b, pp. 35):
Good information on the available options:

Coverage of the operations of all modes and operators, allowing the user to
perceive the network as a whole.
Possibility of offering tailored information limiting the complexity and amount
of information to the minimum required to satisfy user needs.
From any point of the network offer information on the possible competing
connections for a specic destination.
Stability of perception of service (long validity periods for timetables).
Network integration, i.e., the above referred structured interrelatedness of services
enabling:
easy changing of route;
easy use of different operators in the same trip;
availability of interchanges to enhance uidity of transfer ows between modes
and services;
schedule co-ordination and minimization of transfer times; and
fare integration for both frequent and occasional users.
In fact, for a mobility system to provide a set of attractive services, offering

economies of scale to the users, the following dimensions of integration should be
considered:
Three dimensions perceived by the customers (EC, TIS, ISOTOPE, 1997b, p. 55):
Physical integration in space, time, and technology, reected in network
design, existence of well-designed interchange stations, and schedule co-
ordination. That is intermodality strictu sensu, which concentrates in reducing
the users time and effort spent in the transfer process from one network link to
the other.
Tariff integration including fares and tariff integration and the consequent
revenue sharing between operators where applied. As Viegas (2004a, 2004b, p. 3)
expresses in a well integrated system, inasmuch as transport prices are a
function of distances traveled, prices of composite services (two or more links
used in connection) should reect what would have been the price of a direct
service, not the sum of the prices of the various components.
Logical integration entailing a global information system where all modes
and all operators are brought together into a unique system, also envisaging the
possibility of tailored information and of good perception of reliability of
connections based on real time information. This communication interface
between the client and the system is also an important tool for system
monitoring, as we will see later.
One underlying dimension, not visible to the customer but indispensable to ensure
visibility of the other three dimensions of integration, is the organizational or
managerial integration entailing decision processes, allocation of clear responsi-
bilities between different authorities (i.e., land use, transport and environment),
between transport authorities and operators (i.e., through contracts), between
operators from different modes (i.e., through agreements and partnerships). An
important element to consider here is the scope of authoritys competency, which
may either improve or jeopardize the dynamics of the relationship between
interacting institutions. The existence of an authority that unies management of
all mobility vectors, land use and urban environment for that same territorial area
allows the consideration of interrelations among these types of decisions and the
development of coherent policies with a clear advantage to all economic agents
interacting in the system. However, this naturally implies a heavier administrative
structure for this entity, besides a possible need to redene mission and
competencies in a number of other bodies of the public administration.
From the different studies done we can conclude that there are fundamental
requirements for the successful implementation and monitoring of integration
measures. These requirements can be divided into three categories:
Within the framework conditions:

Clarity of purpose from the authority side, who should decide on the level of
integration, commitment, and political priority of the other public entities
toward integration so that the necessary means to achieve it are made available.
A clear allocation of integration responsibilities in the institutional framework,
considering the different stages where decision and actions are needed, such as:
entrepreneurship, implementation, nancing, and monitoring. The objective
should be to improve causal relations between decision taking and quality
performance of the mobility systems.
A exible regulatory and organizational regime so that market initiative can
provide better answers to customer needs, possibly with the creation of new
services, complementary, and compatible with the existing ones.
Within the implementation process:
The introduction of integration measures requires the involvement of the three
decision levels:
Strategic to decide how much integration is wanted and what means will be
made available to support costs of integration.
Tactical to decide what type of integration is needed to achieve the dened

goals.
Operational to decide who should be responsible for which processes.
Competent authorities able to effectively monitor, evaluate, and steer the
mobility system, setting integration standards and quality objectives for the
agents involved in the provision of mobility services at planning and operation
stages.
Utilization of integration criteria in contract awarding and supervision for
service operation, giving some room to initiate new services beyond the
minimum established, but always ensuring these services will not cream-off the
market or hinder in any the network performance formed by the existing
services.
Within the resource management area:
Clear rules to allocate common costs for infrastructure use and
Compatibility of (information) technologies used by the various services.
As we have referred todays society is characterized by mobility needs that result

in the use of logical chains through the interlinkage with transfer points between the
available subnetworks. Once these chains are organized we have the physical and
functional representation of a network offering a set of intermodal points linking a
multimodal mobility system. From the strict physical viewpoint interchanges are
structural transfer points between the various subsystems, but from the functional
viewpoint they represent horizontal and vertical interlinkages. The rst between
modes and services within the same territorial scale, the latter between different
territorial and administrative scales, such as local, regional, national, and
international. Network exibility is thus given by the possibility offered to the
clients to organize their own (sub)network conguration out of the total offer.
Integration should thus be seen as an organizational process through which the
different elements of the mobility system are conducted to a closer interaction, with
higher overall productive efciency and enhancement of quality of services, resulting
also in improvement of efciency in the utilization (i.e., consumption) of the mobility
system.
As already referred it is consensual that integration has numerous advantages and
provides a global positive benet for societies, an argument which has been
recurrently used in the European Union documents related with sustainable mobility.
However, integration also generates additional costs, although their magnitude and
distribution are sometimes less clear. For instance, intermodality, as a main
instrument of the integration process, is often associated with some of the following
negative effects:
losses of time in the less-efcient connections caused by the reduction of direct

origin-destination connections as a trade-off with wider economies of scope;
increase in risks of lack of system security and reliability; and
passenger discomfort.
These disadvantages cause different impacts on the agents of the system where, in
general, all stakeholders aim to reduce their costs. However, if these aims are strictly
followed it will lead to a strong trend to reduce transfers between subsystems. In the
particular case of the citizen client of the mobility system her perception of
optimization of cost reduction is materialized through the utilization of direct trips
for all destinations, and the one that better fulls her quality requirements is the one
of individual motorized transport.
Indeed, several authors have identied interchange nodes as the more visible
expression of organizational complexity in multimodal transport networks, as for
each and every agent (i.e., authorities, tactical planners, operators, and clients) the
involvement in intermodal chains (physically made through interchanges) is
associated to additional efforts in the physical, logical, organizational, and nancial
domain, that cannot be ignored. Costs of the integration process have to be identied
and fully understood not only to identify the impact of gains and losses for the
system as a whole but also upon each agent. The following Tables 4.114.14 present
the main advantages and disadvantages accruing from some dimensions of
integration that are worth to highlight at this stage (Viegas and Macario, 1998a,
1998c, 1998d, 1998e, 1998f, 1998g, pp. 36)
In what concerns organizational integration, there are still other elements
to be considered, the scope of competencies of authorities and the responsi-
bilities of the different agents (mobility providers) regarding initiatives toward
Table 4.11: Physical integration.
Physical integration
Main elements Advantages Disadvantages
Network design High intermodal standard Direct routes customers

of service levels might be penalized by a
route design favoring
Economies of scale and
connections to other
scope
routes
Network co-ordination
Interchange stations Same as previous Might raise difculties in
cost and risk sharing
Waiting and transfer
related with interchanges
conditions for customers
investments and running
clearly improved
expenditure
Timetable co-ordination Increases condence (and Difcult to implement in
ease to use) in public periods of heavy trafc
transport supply
Source: Adapted from Viegas and Macario (1998a, 1998c, 1998d, 1998e, 1998f, 1998g, p. 3).
Table 4.12: Logical integration.
Logical integration
External information Ease of use Difculties in allocating the

covering various costs among operators
modes and operators Increases condence in PT Difculties in keeping the
information updated
Favors market share
growth
One operator Facilitates planning and Higher costs for each
information covering execution of multi- operator
connections to other operator trips Difcult to assure
modes and other permanent update of
operators information of others
operators supply
Table 4.13: Tariff integration.
Tariff integration
Tickets for Incentive to rst experience Increases overhead costs

Infrequent in public transport related with revenue sharing
travelers Lowers barriers to Distribution coefcients are
infrequent users more expensive to estimate
due to the higher variance of
the population
Tickets for Increases customer loyalty Same as above with less
frequent travelers expensive estimation of
distribution coefcients
Revenue sharing Allows better intermodality, Increases overhead costs
with stable network related with revenue sharing
design and relationships Agreement covering all cases
between operators may be difcult to reach
system integration. This division of responsibilities between agents entails the

following alternatives for the entities that can be involved in the integration
initiatives, that is, transport authorities, other authorities, operators, and users
representatives.
Table 4.14: Responsibilities in integration initiatives.
Responsibilities in integration initiatives

Main agents Advantages Disadvantages
Transport Political and social objectives Economic efciency of the

authorities are better achieved operators might be left as a
second priority
Other authorities Better integration with other Economic efciency of the
urban policies transport system might be
considered as a second
priority
Allows for an integrated Integration benets must be
approach with other merit careful assessed to justify
goods adhesion by operators
Risk of lack of knowledge of
real transport costs may lead
to cross subsidization with
other activities
Operators Higher level of cost efciency May represent attempts to
tend to be achieved gain information about
other operators markets or
Represents the understanding
to create collusion of
that integration brings
incumbents against new
positive results
entrants
Associations of (same than with operators)
Operators and
Better distribution of
integration costs given the
good knowledge on real
costs
Users Social objectives and user Risk of lack of knowledge of
representatives needs will be highly real transport costs, ignoring
considered in both network that integration should be
design and productive matched by additional
organization customers
Integration can still be reinforced if complementary measures are taken that not
only guarantee the effectiveness of an interchange as an element of the urban mobility
system, but also increase the value associated to the use of public transport, such as:
Development of several city functions with diversied density in areas at a

nonmotorized (walking, cycling, etc.) distance of the interchange, so that it can be
easily reached and have its utility value increased at the eyes of the citizens, who
are the potential clients of this infrastructure.
Management of accessibility facilities for walking and cycling, including conditions for
mobility impaired access, as well as private motorized access so that integration between
modes can be processed the smoothest and seamless possible way. This implies a careful
management of parking facilities using price as an instrument of demand management.
In this context park and ride facilities are a fundamental element to ensure the overall
integrated logic in the management of an urban mobility system.
Provide a friendly walking environment around the interchange.
Besides location and functional integration with the surrounding urban area, a
major problem that limits good performance of interchanges is its internal functional
design. Both internal and external design of these urban facilities are often the result of
visual architectural competition, leaving for a secondary role the effective functionality
of the equipment, such as poorly located ticket ofces and/or information desk or
simply information placed not in the optimal decision point for users,4 unfriendly
electronic devises, lack of lift and ramps, and lack of waiting areas. In addition we can
still observe contrasting situation, which inuence the mobility ow internal to the
interchange, between high-quality hallways and poor platforms or waiting areas or
poorly kept toilets, creating a disruption in the quality standards perceived by users.
The European research project GUIDES, surveyed 14 site-specic case studies5, and
made a synthesis of the main barriers to interchange use, based on the disruption of
quality perception that this infrastructure element can have, that we reproduce in Table
4.15. In addition, the MIMIC research project, based on 11 case studies, identied the
main barriers according to the seven categories of typologies presented in Table 4.16.
Some more research has been done in Europe and the United States regarding
quality and impact of interchanges in the efciency and effectiveness of the urban
mobility systems. From the ndings of these works6 we can synthesize that
4. As an example real time information for public transport is often placed at the boarding platforms
instead of being placed at the access point, where the user effectively has to take a modal decision. From
the cities observed in this work, Copenhagen and Chicago provide good examples of integration orientated
cities, in what concerns the conception and organization of the urban mobility system.
5. Cases were selected from eight cities (London, Utrecht, Amsterdam, Manchester, Paris, Athens,
Birmingham, and Stockholm).
6. In Europe, commissioned by the European Commission the main research projects, were INTERCEPT,
PIRATE, GUIDES, and MIMIC (http://www.eltis.org/en/concepts/, 16-12-2004). In USA considerable
work is also done by the TRB Transportation Research Board and CNT Center for Neighborhood
Technology (http://www.todcommunities.org/ and http://www.cnt.org, 16-12-2004).
Table 4.15: Contrast between different quality of interchange as perceived by the

users.
The seamless journey Barriers to interchange
Comprehensive connectedness the The need to interchange is a signicant

public transport network offers an disadvantage for any public transport
anywhere-to-anywhere service journey for which there is no direct
connection
Full advantage can be taken of journey Passengers will stay on slow, direct
time savings that are offered by a services, and will not interchange
change to a faster mode onto faster modes
Planners/operators can have maximum Sub-optimal network solutions are
exibility to match demand and supply, adopted to avoid imposing additional
and maximize the overall efciency of interchange on passengers
the network through appropriate mixing
of modes and services
Source: Adapted from GUIDES, nal report, p. 24.
Table 4.16: Barriers to interchanges.
Typology of barriers Example of barriers
Logistical and operational Poor/lack of time synchronization between

services
Incomplete through-ticketing
Psychological Fear of physical attack and violence
Thefts of cars and car radios
Institutional and organizational Poorly integrated interchange management
Competition on passengers and double
services
Physical design Steps and staircases
Lack of comfortable waiting areas and
seating
Local planning and land use Pedestrian access through unsafe areas
Lack of cycle lanes
Economic and social Lack of shops and retail activities
Cost of public transport services
Information No integrated passenger information system
Signing lacking or of poor quality
Source: Adapted from GUIDE nal report, p. 54.
interchanges are equally important in intermodal situation and intramodal cases,

where different services are integrated through interchanges (stops, station or more
complex elements, as can be observed in several the United States and South America
cities), but in both cases three essential integration elements should be ensured when
designing interchanges as nodes of an urban mobility system. These are: hardware
elements, related with the physical components of the infrastructure; software
elements, related with the logical and managerial aspects of the interchange node;
knowledgeware elements, related with training and education of management and
operational staff of the interchange.
4.4.4. Pricing and Financing Policies
4.4.4.1. Pricing and nancing mechanisms In an urban mobility system there are
essentially two main areas of expenditure that require nancing. These are the ongoing
cost of operation of the mobility system and the specic investments envisaging system
construction, expansion, or renewal. In spite of a strong recourse to public budgets
even for coverage of operational costs in the 1970s and 1980s, recently there has been
increased pressure to reduce the contribution from this source of funding as other
sectors (e.g., health, education, housing, etc.) also require public support.
As we have concluded in the research project (EC, TIS, FISCUS, 1999b, p. 9), the
inherent limits of gathering user contributions (payments from the direct users related to
their amount of use, that is revenue collected from pricing), combined with the political
limits of public expenditure from general budget (or even earmarked budgets), reveal that
further sources of nancing are needed to maintain the urban mobility system within
acceptable quality levels. In addition, we must not forget that price (a major funding
source) inuences the perceived quality of services through the expectation effect, based
on the connotation that in general high quality is associated to high prices (EC, OGM,
1998a, p. 139). This relation imposes the constraint that changes in prices should always
be accompanied by changes in quality if consistency of signals is to be achieved.
The main source of funds for the urban mobility system is revenues from mobility
activities, entailing two different types of sources with very distinct treatment from
the nancial and scal point of view. The rst are operational revenues such as Public
Transport Fares; Infrastructures Charges (Public Transport); Parking pricing; Road
Pricing (e.g., cordon tolls, area licensing, distance based schemes, time, or
congestion-based schemes), Taxes related to the amount of use (e.g., kilometer tax,
fuel taxes, xed vehicle taxes, etc.), or levies (annual circulation tax). In general price
differentiation (e.g., related to the level of congestion) is an effective tool to manage
demand due to its potential to inuence change of behaviors.
The second revenue type or source is the other nonoperational revenues that
correspond to a cross nancing internal to the agent, such as, publicity, renting
infrastructure spaces or infrastructures for telecommunication, etc. The distinction
between operational and nonoperational revenues is extremely important for reasons
of transparency, and correct assessment of efciency especially when subsidies for
public service are also applied.
In urban mobility, contributions from public budgets are often provided both for
operation and capital costs. For operating costs these are usually granted either as general
subsidies, or special subsidies for specic purposes (e.g., eet renewal for less pollutant
energies, concessionary fares for less afuent user groups), cross subsidies from other
public sector (e.g., water, electricity, etc.) and tax benets. Many European countries used
this cross nancing in municipal transport. Currently the European Union enforced
separation of activities to ensure transparency and better control of State support.
In capital costs this type of nancial assistance is usually done through transfers of
ownership of capital values and general subsidies. Subsidies can also be differentiated
by other elements (EC, TIS, FISCUS, 1999b, p. 18), in particular by the level of the
territorial authority that is responsible, where we have subsidies from local
authorities and transfers from other central or regional budgets (in federal countries,
like Brazil, the distinction is between the Union, the State, the Metropolitan Region
and the Municipality). The reference quantity is also another element of distinction.
Quantity subsidies are paid for each unit of the good (or service) subsidized, value
subsidies are price related, and lump sum subsidies, are a xed amount granted to the
receiver. The type of subsidy inuences market prices for goods and services and they
can be granted directly or by way of a competitive process.
A good example on the complexity of subsidy packaging that can be found in
Europe was illustrated in research project (EC, TIS, FISCUS, 1999b, pp. 1920) with
a German practice:
In Germany, the municipal transport nance law provides earmarked

taxes (additional taxes on mineral oil) (Gemeindeverkehrsnanzier-
ungsgesetz GVFG) for nancing the improvement/expansion of urban
infrastructure. Additional nancial resources are obtained through
debts. According to the law on national roads (FernstraXengesetz), if
urban roads are not already subsidised by the GVFG, up to 50% of the
investment costs can be provided by the national road administration.
Further subsidisation is possible through the law on urban restoration
and development (Stadtebauforderungsgesetz) and structural funds.
The GVFG species the distribution of subsidies between public
transport and road investments and contains a catalogue of project
that can be recommended for investmentsythe nancial resources of
the GVFG, which has been described earlier, can also be used for
nancing rolling stock in public transport.
Another source is contributions from indirect beneciaries, also known as value

capture, that is taxes with incidence on the additional value created by mobility,
which are usually originated by one of the three following opportunities, as
concluded in (EC, TIS, FISCUS, 1999b, p. 27):
forced concentration of people at certain places at predictable points of time due

to operational necessities of transport (certain road sections, stations and their
surroundings, interchanges, vehicles);
increased opportunities for all human activities related to distance owing to the
existence (not just usage) of transport infrastructure/services; and thus
increased level of land values in the neighborhood of transport facilities.
The benets accruing from these opportunities are usually related with the
organizations utilization of the concentration of people (e.g., advertising, retail,
restaurants, etc.), or organization taking economic advantage of distance related
activities, or real estate establishment beneting from increase in properties values.
To the previous (Viegas, 2003b, pp. 45) adds private car drivers enjoying less
congested roads if public transport has good levels of occupancy. This would be
collected through charging for the use of parking or circulation spaces in congested
areas. In fact this mechanism is a cross-nancing between activities (or services)
within the urban mobility system.
An example of this contribution from indirect beneciaries is the French
versement transport, applied to employers with more than 9 employees in cities
greater than 20,000 inhabitants; in the Netherlands and in Italy there is the possibility
to levy a special tax for infrastructural needs (EC, TIS, FISCUS, 1999b, p. 33) and in
Brazil there are similar possibilities to dene added value taxes (Trajano & Silva,
2005, working document, pp. 127). Another example is the London Congestion
Charging, a charge for the access to a limited area although still a resource (i.e.,
urban space) of the urban mobility system applies part of its revenues to public
transport (Transport for London, 2003, p. 15). As we will see later, this return of
revenues to the transport system substantially improves the acceptability of the all
package.
It is worth referring also that we observed that many of the problems associated
with cost coverage of urban mobility result from the fact that most taxes charged
over individual motorized vehicles have their receipts addressed to the central
budget, with only a reduced part staying for local governments. Consequently, local
are too dependent on parking revenues.
Finally, private nancing surges as a source with growing utilization since the
1970s. The alleged reasons for this are several but the most common arguments are
the reduction of demands over public funds, a larger source of available funds, faster
implementation of projects, more propensity to innovation and increased value for
money.
Within private nancing a particularly favored option in the recent years are the
publicprivate partnerships which is pushing the transport sector to improve quality
of management, since private funds (contrary to the current practice with public
funds) demand for pay back with acceptable return on risk levels, which is why
capital cost is always more expensive in this option than in public funding schemes.
This forces the activity (i.e., service exploitation or infrastructure development)
object of the partnership to release cash ow beyond operational cost coverage, or to
mobilize public funds to bridge the gap. Often these partnerships involve also
international nancing institutions, through grants (e.g., World Bank) or loans (i.e.,
EIB, EBRD, etc.), which is a common situation in developing countries given the fact
that these options offer better conditions in lower rates and longer terms than with
commercial investment banks.
However, the engagement of private nancing is not devoid of risk of conicting with
government. This is mostly due to a different attitude toward the incentives, for
particular projects, used by both parties. Three potential areas of conict have been
identied in the research project (EC, TIS, FISCUS, 1999b, p. 21): projects with positive
nancial but negative social Net Present Value; projects that meet private sector
objectives but conict with government policy objectives; projects which involve
unbalance bidding consortia, for example if a construction company has a self-objective
of maximizing turnover and is a dominant consortium member, they have an incentive
to over specify the project which can be against the public interest, or can contribute to
rigidication of contracts, or simply may not represent best value for money.
In this framework of publicprivate partnerships, if performance levels reach an
acceptable stage of development according to investors criteria, and thus good levels
of return on investment, it seems to be possible to evolve to a sponsorship type of
relation with private nancing where a particular service or segment of linear
infrastructure is the object of a sponsorship contract, reecting the option value
benet of having that service or infrastructure that may improve accessibility and
uidity of a specic area.
In the involvement of private nancing for mobility investments two major issues
should be carefully thought: the risk identication, dimensioning, and sharing; and
capacity to prevent contractual rigidity so that the relation between partner can be
adjusted whenever change of factors inuencing the performance of the partnership
occurs (i.e., avoid contractual completeness). Risk categories are very diverse, but
some authors have reached a rather consensual systematization of the main classes of
risk in this sector (Rienstra & Nijkamp, 1997, pp. 78):
Political risks: e.g., changes in transport policy due to change of government or

any other reason.
Financial risk: e.g., uctuations in interest rates or in exchange rates.
Constructions risks: e.g., delay, unexpected higher costs.
Operational risks: e.g., damage by accidents, vandalism, demand shortfalls, etc.
Commercial risks: e.g., wrong cost estimates, wrong estimates of trafc volumes;
unexpected competition.
(to the previous we add) Planning risk: e.g., delay or cancellation of planning
actions that hinder trafc concentration of other benecial condition, such as the
cancellation or postponement of the construction of a major trafc generator.
An important interaction exists between all these different nancing mechanisms

and we have evidence from the empirical studies undertaken in the research project
PATS (EC, TIS, PATS, D3, 2000, p. 36) that people react with various degrees of
acceptability to each instrument. Consequently, the choice of nancing mechanism,
as well its packaging into sets of consistent end-goal instruments, is by itself one of
the most important management tools since it is through the specic application of
those instruments that the system management is able to inuence behavior of

both agents and nal users. Although monitoring the demonstrations of CIVITAS
cities (METEOR) and evaluating the potential for transferability of pricing
measures, among others, we have also perceived that packaging of measures was a
condition inuencing the good performance of any measure (Macario & Marques,
2004, p. 45).
This interaction is recognized worthwhile and two examples of multisource
funding for mobility systems, that largely inspired the solution for the Portuguese
and Brazilian application of our proposed quality model, is the New Zealand Land
Transport Funding (http://www.ltsa.govt.nz/funding/nltp/index.html, 01-07-2005)
(Land Transport NZ, former Transfund) illustrated in Figure 4.4, and the Japanese
model of nancing of railways networks (Kileen & Shoji, 1997, pp. 810).
The rst provides nancing for a wide diversity of activities, e.g., development of
infrastructures and services, research to study alternatives to road use, social services,
etc. The entity managing the fund is also managing the National Program for Land
Transport. In the second case, the railways operating companies evolved into
multisectorial companies, making internal cross-funding between the four following
areas of activity: rail operation, transports (road, i.e. feeder services), real estate, and
other business. The weight of nontransport activities in the overall turnover of these
companies presents a wide variation between 20% and 80% (Kileen & Shoji, 1997,
p. 15).
4.4.4.2. Public acceptability of pricing mechanisms The changes in structure and

dimension of the urban reality added to the congestion phenomenon, the scarcity of
public money and, last but not least, a growing awareness of society about
environmental problems, are among the main factors that have lead to stronger
demands of efciency in transport systems, and consequently to the use of pricing
policies as main instrument to achieve this aim.
However, and despite a quite consensual recognition (among scientists) of all the
advantages of this type of solution, given the political sensitive character of urban
mobility there is a signicant problem of public acceptance of pricing measures
whenever they include charging for road use in urban areas, built on the following
main arguments: have to pay for what was previously free; excessive privilege
accruing to the wealthier elements of society; no rm guarantees given for a fair and
efcient application of revenues; threat to citizens privacy.
When implementing pricing measures it is therefore fundamental to consider the
perceptions and attitudes of the public and stakeholders. It would appear from the
literature, particularly from social psychology, that one method might be to
incorporate in a global participatory approach marketing and publicity strategies as
well as information campaigns to be implemented together with the technical
measures. These strategies could be targeted at different groups and different
perceptions with an aim of inuencing them in favor of the pricing measure by
increasing their perception of the problem and of the benets of the possible solution.
Fuel Excise
Crown
and customs
Account
duty
Road User Motor Vehicle

Fuel
Charges Registration ST
G Misc
Excise
(RUC) (MVR)
National Land Transport Fund New Zealand Road Safety Programme
Safety education and

management
Safety enforcement
MVR and RUC administration
National Land Transport Programme State highways
Walking and cycling
Alternatives to roading
Regional development
Local governments Passenger Transport services
Local governments Local roads
Local governments Administration
Figure 4.4: New Zealand land transport fund. Source: Author.

The analysis of the opposing arguments to the transport pricing proposals

included in the Commissions Papers7 reveals a number of additional arguments that
are of high relevance in terms of public acceptability (EC, 1999c, TIS, pp. 4751).
These included:
the equal and fair treatment of all transport modes and all sectors of the economy;
the implications of transport pricing on European competitiveness and the social
and economic environment;
the qualication and quantication of transport costs;
the adequacy of the pricing mechanism to create a signicant modal shift from the
road to more environmentally friendly modes;
the availability of the technology for accurate transport charging;
the use of the revenues;
fair and equal treatment of users; and
integration of transport pricing with other policies (e.g., urban planning/land use,
regional policy, etc.).
As we have concluded in PATS research (EC, 2001b, TIS, pp. 4956) not only the
fairness of the price determination, but also the magnitude of the price has a major
inuence on its acceptability. Consequently, transport cost calculation, which is at the
basis of a fair and efcient price is also one of the aspects to consider as of utmost
importance. However, the elds of transport costing and pricing (especially for
externalities and infrastructure) seems to be a very sensitive one since harmonization of
methods and procedures are not yet fully consolidated to enable sound comparison
between cities (Macario & Carmona, 2002, pp. 712). These methodological
uncertainties and disputes have still to be mitigated and the evaluation process should
be made coherent and transparent.
Sophisticated technical systems also play an important role in the enhancement of
fairer and more efcient pricing as they are one of the system features in the front line
of users contact.
However, they may also impose constraints in terms of acceptability due to
complexity, lack of transparency, potential for violation of privacy and mistakes,
learning efforts imposed on the users, etc.
The careful choice of which authority regulates, implements and administers any
pricing measure and the legitimacy that an agency has in the eyes of the stakeholders
may also improve acceptability. Revenues from pricing schemes attributed to the
local authority may enhance the willingness of people to pay because they expect a
more direct advantage from their application.
Moreover, the trustworthiness of the local administration may be superior to that
of the central government as decision-makers are closer to the eld and also can be
7. Analysis of reaction letters from stakeholders (individuals and organizations) done by the author in the
framework of the research project PATS.
felt as more familiar to stakeholders interests. This is specially accentuated in Federal

countries like the case of Brazil, where the structures like the already mentioned
Council of Cities are used to overcome this distance between government at the top
of the political-administrative hierarchy and the population.
Different authorities have also different potentials to integrate a pricing measure
with other policies to reduce congestion, accidents, and pollution as well as to co-
ordinate between different agents, which again constitute important elements in
terms of acceptability.
From the literature review undertaken in PATS research it clearly emerged that
whenever a pricing measure is perceived to force behaviors not in compliance with
usual habits there is a strong potential to develop resisting patterns, unless
advantages of those changes are made very obvious. If resistance occurs then the
pricing measure can have unintended, and sometimes uncontrollable, collateral
consequences and even totally fail to meet its objectives.
Public acceptability is thus a complex problem as it requires the joint
consideration of a number of scientic areas, in particular economic, social,
technological, legal, and even managerial aspects. At the top of all these aspects one
can still add the fact that the analysis of acceptability implies the observation and
control of a dynamic process of change where in one hand we have the individual
interests of each citizen and its perception of advantage and disadvantage, resulting
from the implementation of any pricing policy, and on the other hand there are still
the organized interest groups who besides the weight of their public reactions should
also be considered by their lobbying power.
Acceptability results from an interaction between political effectiveness, here
understood as the capacity of accomplishing the proposed objectives, economic
efciency in production and consumption, equity and social fairness and, last but nor
least, feasibility of implementation. Efciency and equity are thus central concepts
for pricing transport systems. The rst justies the entrepreneurial attitude of
decision-makers when dening and implementing pricing measures, the second
covers an important part of what can be considered as the constraints to
implementation.
Equity is a concept that gained predominance in the industrialized society and it is
often referred to as a distributional principle to guide economic relations between the
State and the citizens. In practical terms equity is often understood as distributive
justice entailing three possible main rules for a just distribution, that are also
alternative and frequently contradictory: equality, meaning that everyone gets the
same share of what is at stake; equity (strictu sensu), meaning similar personal input
output ratios between costs or contributions and accrued benets; distribution based
on needs or requirements, meaning everyone gets according to his needs or
requirements (Schade & Schlag, 2000, p. 45).
In research (EC, TIS, PATS, D3, 2000, p. 38) we have translated these principles
along the following dimensions of equity:
Territorial equity (or aggregate equity) corresponds to the principle of

liberty, in which the society must guarantee everywhere the access to goods and
services, thus avoidance of social exclusion, implying the consideration of the right
to free mobility.
Horizontal equity (or procedural fairness) corresponds to the principle of
equal opportunity, which concerns the equal treatment between users and the
user-pays principle, that is differentiation.
Vertical equity (or end state fairness) corresponds to the principle of difference
or distribution, which explicitly takes into account the inequalities between
affected user groups and its consequences in relation to transport.
Longitudinal equity (or dynamic equity) which represents everybodys goal of
no decrease in previously available benets (entitlements) and corresponds to one
of the major trade-off difculties in modern, ageing societies.
This multidimensional characteristic of equity (and thus of acceptability) must be

considered in the static and dynamic perspective to reect and monitor the impact of
change along the lifetime of any pricing measure. That is, while introducing a pricing
policy we have to consider not only the absolute effect over each stakeholder
personality but also the marginal effects on the social and economic statute of those
personalities at the moment of implementation and along time.
From the above it can be understood that the main elements in the analysis of
stakeholders degree of acceptability of transport pricing policies are (Viegas and
Macario, 2003, p. 174):
The identication of the various actors involved or affected in all stages of the
decision-making process.
The perception of the specic interests of those actors analyzed as personalities.
The power of inuence of the different groups of actors and their reactance
potential.
And nally, the relevant aspects that can enhance the acceptance of the different
pricing instruments and policies.
Finally, it is worth referring the main practical aspects associated with the
implementation stage that should be carefully considered in the design of the
measures and packages:
Functional interaction between institutions involved in the decision process,

implementation and management of transport pricing and nancing policies, as
well as between these and other governmental areas.
Technological solutions necessary for implementation purposes, respective impact
in citizens privacy, easiness to use and control, etc.
Legal and regulatory framework to clarify the role of the different public and
private intervening institutions.
4.4.5. Concertation with Other Urban Policies
Integrated strategies are the combination of individual policies to achieve cumulative

positive effects or to mitigate negative effects of any of them. Consequently achieving
one specic goal might well be done at the sacrice of another goal. Besides,
integrated strategies can accrue synergies between policies and that is the main reason
for policy packaging, where it is important to select policies that will reinforce each
other so that positive effects of integration can be maximized.
It is rather consensual that transport is one of the most signicant threats to
unsustainability of urban areas. The precondition for a sustainable mobility policy is a
settlement structure, together with a conguration of the mobility system, that allows
keeping generation of private motorized mobility within accepted thresholds. When
the mobility system is congured with a consistent hierarchy of networks it is relatively
easy to specify where each type of land use should be located, thus avoiding excessive
concentration of trafc through integrated planning between transport and land use. A
good example of how to prevent this situation is given by the ABC law (http://
www.epe.be/workbooks/tcui/example12.html, 07-07-05) from the Netherlands.
As reported by May, Jopson, & Matthews( 2003; p. 157) in European cities alone
trafc congestion costs are already in excess of h100B per year, with local pollution
and health impacts imposing costs of similar magnitude, and accidents in urban
roads being reported at the level of 20,000 fatalities per year.
Already some years ago the (ECMT, 1995, pp. 3742) advocated the need to
adopt integrated approaches to these problems combining land use, environmental
protection, and wider social instruments. In spite of the availability of considerable
knowledge on instruments of all these areas the design and (mostly) implementation
of integrated strategies is still a challenge nowadays, mainly because of the
aggregated performance of instruments that is a scientic domain not (yet)
sufciently dominated that the on-going accompanying measure (SPECTRUM
project of the 5th EC-RTD program) is expected to enlighten.
But even when adequate policy (and instruments) packages are designed their
implementation is subject to several barriers. As ECMT (2002a, 2002b, p. 27) and
May et al. (2003, p. 157) reported, poor policy integration and coordination,
counterproductive institutional roles, unsupportive regulatory frameworks, weak-
nesses in pricing and poor data quality and quantity are the main barriers to pursue
the policies advocated a previous report (ECMT, 1995).
To achieve the objectives of sustainable development (Banister, 2004, in ECMT,
2005, p. 131) dened four basic groups of policy measures to be considered, being:
Life-style oriented policies, where policy intervention is only of subsidiary help

since the basic element is a change of attitude toward mobility and material
consumption. Information and education play a determined role since knowing the
transport consequences (e.g., environmental damage, etc.) of a given policy or the
effect of certain choices may well inuence behavioral change. This is a rather
bottom up approach.
Market-oriented policies, which assume that people are willing to change their
lifestyle or behavior toward mobility if others do the same and no material
disadvantage will result. In these cases measures like scal reforms or changed
property rights might change the incentive structure. This approach although with
some top down elements relies on the public acceptance of price as a mechanism to
allocate services and goods.
Regulation oriented policies, which relies on legal and regulatory changes,
technical standards, and norms (e.g., speed limits, maximum weight of vehicles,
etc.), on innovative planning methodologies (e.g., spatial planning and transport
impact assessment) and on government reform. In general this approach is rather
rationalist although it can be tempered with strong participatory processes to
provide the argumentative element.
Public infrastructure/public transport, the provision of infrastructure and public
transport services is often adopted as a policy approach and seen as associated
with regulation oriented polices, to which is also often confounded.
The dominant view seems to be that there is a wide diversity of policies that must
be developed in an integrated way and understood as major instruments to achieve
the strategic objectives of the urban mobility systems. That is, the objectives set for
any urban mobility system should be deployed through the land use, transport,
environmental, and scal areas, so that through this concerted action adequate signs
will be transmitted to citizens and their behavior toward mobility inuenced.
4.4.6. Information Systems Supporting System Management
Information is an indispensable tool for management, in particular when discussing

complex systems with high intensity of interaction and diversity of agents. Scientic
treatment of Information systems is a relatively young discipline where denitions
are still rather unstable.
For the purpose of our proposed management model we consider information as
corresponding to relevant sets of organized data that serve as general input for
decision-making and contribute to better management decision through improved
knowledge of the functioning of the system being served (i.e., the Urban Mobility
System).
Conceptually, the recurrent transformational process of the urban mobility system
is supported by a feedback process in which the information system is incorporated.
Basically, the information system is a human activity that can be organized in many
different ways, according to the degree of formality or the extent of automation, etc.
To support decision in the management of the urban mobility system the
information should be organized in relation with the decision-making structure. That
is, to serve the following types of decision as illustrated in Figure 4.5:
Strategic decisions: related with long term, complex and with a less clear structure
of decision, usually taken by senior management. Information used is often ill
Figure 4.5: Conceptual mapping of information by level of decision.

Source: Adapted from Edwards, Ward, and Bytheway (1995, p. 12).
dened, required in a nonrecurrent basis, in a good part originated in sources

external to the system, gathered in informal way (e.g. newspapers, radio, informal
meetings and social events, etc.) but highly summarized through the interpretation
of the recipient.
Management control decision (or tactical decisions): related with medium and
short term often concerning comparisons with standards. Information is internally
focused, short term, historical, usually predened and required on routine basis.
Operational decisions: usually supported in well dened rules. At this decision level
information is internally focused, predened and rather precise. Frequency of
decision is very high so it is common to have a good part of it supported by
automated means.
Besides the information for decision there is also the need to generate and manage
data related with the management of the urban mobility system (e.g., contracts with
operators, requests from citizens, network assessment, etc.).
To accomplish its purpose of major contributor for UMS learning, the
information system must be provided with the following knowledge processes as
recommended by Holzner & Marx (1979, p. 75):
Construction the process through which new material is added or replaced

within the system stock of knowledge.
Organization the process through which bodies of knowledge are related to
each other, classied or integrated.
Storage the process through which a new observation or experience is stored
after passing the test of relevancy. Storage is a pre-condition for the existence of
system memory. However, given the commercial sensitiveness of some information

storage function should be complemented with a classication degree for
disclosure of information.
Distribution the process of distributing knowledge where it is needed and can be
applied. Owing to the communicative function this process has an important role
in the interaction between agents.
Application the process through which knowledge is applied and contributes to
improve performance that is characteristic of our intuitive understanding of
learning. Application can be organized in many different forms;
Figure 4.6: Information system for UMS management (conceptual diagram).

Source: Author.
This framework was applied to the cases of Lisbon and Porto Metropolitan
Authorities, where we have conceived an information system with four main pillars,
as illustrated in Figure 4.6:
Supporting technology compatible with the one installed in some of the public
transport operators.
Integrated databases entailing a wide set of information, namely: land use; trafc
generators; mobility networks; operators and service production; and customer
satisfaction.
Application to support management, designated as knowledge pivots where key
performance indicators are available. At least four applications are required for
assessment of: network (all, single and aggregated evaluation) performance;
commercial performance of the UMS; industrial performance (infrastructure and
public transport only); and aggregated performance of the UMS (all modes and
services).
User (knowledge) interchanges, for the main types of users of the information
system, that are: authorities managing the urban mobility system; customers;
municipalities; operators of public transport; and infrastructure managers.
Just like the UMS also the knowledge system must be tailor made to the reality of
each specic urban area and use a technology as well a structure that enable to
undertake an incremental building process in line with the evolution of the UMS
management competences.
Chapter 5
How to Set Up a Management Model

for Urban Mobility Systems
5.1. Underlying Rationale
Human societies have always been dependent on some sort of quality specication.
In the earliest times of human societies, quality was associated to survival capacity,
and in some places, this is still a reality in our days, but in general, the quality
concept evolved into more complex and diversied formulations along with the
complexity of society.
How to deal with quality denition has always been a problem, especially in ser-
vices, once the relevant concept here is the perceived quality, as explained in Chapter 4.
Indeed, we can easily understand that high quality is a goal common to all countries,
sectors, and societies and that everyone is able to identify it. However, the operational
denition of high quality varies with a number of factors from cultural attributes of the
evaluator to functional attributes of the object (service or product) being evaluated,
passing through the variability of criteria used in the clients judgment.
Today, there is a general awareness that the existence of human life itself is
dependent on a number of quality thresholds such as temperature, air quality, food
quality, and so on, which is largely controlled by natural and non-natural solutions or
mechanisms. These have contributed to the growth of an enormous variety of goods
and services that again challenge the quality thresholds, and the evolutionary cycle
continues passing on the experience from one generation to another, simultaneously
increasing the sophistication, rigor, and complexity in the production of goods and
services and also in the associated ows of information.
Evolution of human societies is also dependent on productivity of industries and
countries that also lies very much on quality control for both product and process
design. Even economic growth depends on the reliability of systems such as energy,
communication, and transport, that is, on their sustained quality.
Given this importance, governments have always been responsible for establishing
and enforcing quality standards, through different institutions, some of them political
such as national, regional, and local governments, others of nonpolitical character
such as corporations, trade associations, and standardization organizations. Once the
decision to regulate quality in some sector is taken, then the approach tends to follow
a process with the following sequence of activities:
Dene the targets of the quality process, that is, the object and purpose of the
process that is to start.
Dene the legal support, that is, the act where the purpose of regulation is
established and also the subject to be regulated is dened,
Dene the organizational structure, that is, the regulatory body(ies) that receive
formal powers to establish standards and to see that they are enforced. With this
purpose, the structure is materialized in one or more institutions with powers and
means to give awards and apply sanctions on the matters considered relevant for
the industry. In addition, processes are established for management and control of
quality standards and also for the interactions between the institutions taking part
in this structure.
Dene the quality thresholds and principles that are not only related with the nal
outcome but should also concern the deployment of quality performance in
intermediate processes, human and material resources, and so on.
Dene the methods for assessment of compliance with the dened quality thresholds
and principles.
After this design and denition stage, evaluation and quality labeling is done by
the entities nominated for that duty. It is worth referring that regulating quality
implies the permanent balance between protection of consumer interests and
avoidance of excess burden in productive processes that in the end will also hinder
consumer interests as it may introduce inefciencies that ultimately may lead to
higher cost and lower quality.
Therefore, quality requirements evolve in our societies as a consequence of the
evolution and needs of the society with increasing demand on quality thresholds as
we move from one stage to the next. Disruption in this process is only observed either
when the process is interrupted by some external element (e.g. wars and natural
catastrophes) or when the system reaches the regeneration stage requiring then to
rethink standards and thresholds that represent a stage close to exhaustion of the
potential for improvement along the previous path.
From what we have observed in public transport (PT), there are symptoms that
we are very close to this exhaustion stage of the current model. As we have referred
along this work, the improvements made, even when they represent strong invest-
ments, only exceptionally return some positive variations on patronage, and the most
common achievement is retention of market share.
Simultaneously, a wide consensus gained prominence, as it was also noted along
this work, that urban mobility management has to start with the location of activities,
where the mobility need is generated, which is an important part of land use strategic
management as a driver for urban development. In fact, perspectives over an urban
mobility system (UMS) are usually diversied, and often conicting, as they depend
on which stakeholders view we adopt.
Objectives and policies from urban mobility authorities also reveal a high degree of
variation. Worldwide we can nd many urban areas where the urban mobility concept
is not yet perceived and many others where, despite evidence of the understanding
of the concept, this has not been assumed in an integrated organizational and
management structure covering all the decision levels. In fact, even in the latter cases,
the most common situation is a scattered distribution of responsibilities to several
How to Set Up a Management Model for Urban Mobility Systems 199
entities, sometimes backed up by an integrated policy document. This situation results

from the fact that even where the urban mobility concept is assumed, there is often a
mist between the existing institutional design and the organizational requirements for
the management of such a complex system, which results in inconsistencies that
inuence the overall performance of the mobility system.
The framework for urban mobility differs substantially from one country to
another and even between cities of the same country, but, whatever the choices made,
in practical terms, cities are major sources of output, of productivity, of growth, and
of wealth, and this characteristic is very likely strengthened by the city size. Although
reported as not completely proven, some authors (e.g., Prudhomme, 1996, p. 174;
Alonso, 1971, p. 75) advance the hypothesis that the synergetic effect comes from the
fact that the bigger the city, the larger is the effective labor market.1
Despite this recognized potential, Prudhomme also alerts for a common pitfall
(1996, p. 176), that is, if jobs and homes are poorly located or if the transportation
system breaks down, then the city will be formed only by several independent small
markets without appropriate scale to induce higher productivity. Therefore, the good
interaction between land use and transport is by itself a factor that inuences the
potential of a city as major source of productivity (i.e., output or growth), and,
consequently, its long-term sustainability will result from a good city management.
As the city develops, the growth of some urban nucleus together with a more
intense interaction between the preexisting nucleus transforms the city into a bigger
structure, like a metropolitan area, requiring then additional mechanisms to ensure
horizontal concertation between the various local governments that compose the
metropolitan area or region. This involves the creation of adequate organizational
structures metropolitan institutions either with very wide responsibilities
(e.g., metropolitan planning organizations in USA) or with very specic functions
(e.g., transportation authorities that can be found in several European countries).
When this dimension develops, some additional ad hoc groups arise and tend to
evolve into a syndicate type, for example, the Brazilian Fora of Secretaries of
Transport, which is a specialized forum, or still the Brazilian Council of Cities
and the Irish Transport Forum, which represents the civil society organized in
stakeholder groups. These multiple examples of organisms of varying composition
are only a reex of the complexity of the metropolitan systems and of the inadequacy
of the traditional democratic hierarchic system (where only citizens as individual
persons are represented, and then only at their place of residence) to provide
adequate representation of all the interwoven interests.
In general terms, we can say that the quality of the outputs of the UMS depends
not only on the quality of its inputs but also on the overall quality of management
of that city, which thus becomes an input of the system in the sense that it is part of
1. Remy Prudhomme, Dupuy, and Boret (2004, p. 7) report a study conducted in 23 French cities and 3
Korean cities dedicated to the determination of factors governing the productivity of cities, where the
effective size of the employment market is dened as the average number of workers who have access to
enterprises in less than a given period (e.g., 60 minutes).
the potential to produce outputs. Besides, the interplay between the different
policies and institutions that steer the relevant urban processes, such as land use,
socioeconomic development, and environment, is also considered as an input of the
UMS.
We dene urban mobility as the aggregated result of the multiple decisions (and
factors conditioning those decisions) taken by individuals and economic agents as
an answer to their requirements of displacement of people and goods. Conditioning
factors are the location of social and economic activities in the urban dened spaces;
the working hours of the different activities; intensity of opportunities for social
interaction; and other cultural elements that contribute to dene the pattern of social
relations in a city. The UMS in then a structured and organized system that tries to
provide uidity in those displacements and access to the relevant urban activities,
making use of the possibilities offered by the various transport modes envisaging an
adequate balance between the several modal resources, with the ultimate aim of
contributing to the preservation of a sustainable city.
The UMS is formed by infrastructure (including superstructure and intermodal
links), networks, services, and agents, each of them by itself a complex whole that
requires further decoupling, namely,
The main infrastructures of the UMSs are roads, rails, parking areas, pedestrian
areas and corridors, cycling areas and corridors, and unimodal and multimodal
stations. Some of these are only made functional by association with services
provided by professional entities, whereas others can be explored on self-service
basis, namely, road and pedestrian and cycling infrastructures, although they can
also support professional services.
The main services are motorized transport services, services related with vehicle,
infrastructure (and superstructure) availability and use, information services,2
citizens training, and education for self-service modes. Professional services can be
provided in all networks even in the ones used on self-service basis.
The main agents are authorities, service operators, users of the various transport
modes, and other citizens.
Finally, the main networks that are formed by the interlinkage of individual
elements (infrastructure or services) are the PT network, which can encompass
several modal networks such as road, rail, and inland navigation; the network of
individual private motorized transport; and the network of nonmotorized
individual transport, each of them with potential for subdivision of modes and
services.
In all sectors and dimensions, the main responsibility of management is to create

value that in rather generic terms means to give an effective and sustainable
contribution to the improvement of the appraisal of the object of management. In an
2. Information services also require a specic infrastructure.

UMS, creating value means to act in such a way to bring the system closer to the
desired conguration and performance and improve the satisfaction of a target
population in face of the new state of the system. In practical terms, this principle
means steering the system along its value chain, that is, the sequence of activities and
information ows that a set of agents with different missions and characteristics must
perform to conceive, design, produce, market, deliver, and monitor mobility services of
a predened quality.
In this simplied perspective, each activity, developed by each agent, constitutes a
step that should increment value to the end service. This value chain concept applied
to the UMS forces us to see the entire economic and functional process as a whole,
regardless of who performs each activity, conrming this way the need to consider
the three quality levels that have been referred along this work: quality at the service
level, at the agents level, and, nally, at the system level.
However, the demands falling over an UMS are very diverse and require the
system to continuously adjust to the urban changes. Besides, clients are divided into
segments that represent different preferences, sometimes in conict. This means that
the activities that add value to a specic segment of clients might well subtract value
to other segments. Consequently, part of the steering mission has to be dedicated to
the management of these conicts that are reected because the design of the
conguration of the UMS where equity concerns among these groups has to be
considered.
For an UMS to maintain its value-driven character along time, despite its
complexity, degree of internal conict, and instability inherent to the perception of
quality, the following properties are indispensable:
Robustness, meaning long-term stability and sustainability;

Adaptability, meaning the dynamic capacity to adapt services to evolutionary
demands or new technological opportunities;
Efciency, meaning high productivity, in the capacity to transform basic resources
into service outcomes, and these into consumption units, providing the best results
at the lowest possible cost; and
Diversity, capacity to respond to the aspirations of the different segments of
customers with different types of services in a continuous adjustment between
supply and demand for urban mobility.
This underlying logic of the proposed management model is illustrated in Figure 5.1,
where the relations between the different elements are translated into quality criteria,
forming a cycle for the planned intervention, which can be either a policy, a measure, or
simply an action.
The denition of objectives starts with the decision-makers interpretation of
several elements, namely,
the importance of the needs (or aspirations) of the citizens;

the importance of the problems to be solved, measured through their impacts on
social and economic live of the city; and
Figure 5.1: Implementation cycle. Adapted from EC (1999a, Means Collection,

Vol. I, p. 89). Source: Author.
the assessment of the probability of success of each of the actions and policies
envisaged as potential solutions for those needs and problems as well as to the
superior objectives of sustainable development of the urban environment.
Therefore, objectives are dened upstream of the prioritization of actions and

policies. Indeed, whatever the context, the formulation of a strategy always requires
the establishment of a hierarchy of objectives and the setting of the level of their
ambition. Cities differ substantially in their vocation and in their development strate-
gies. Besides, even if we are dealing with similar problems, in any given moment,
each city is conditioned by the choices made in the past, which congure a different
departure point for the problem under analysis, and consequently, different per-
ceptions are derived, which are the main problems and which are the best solutions to
mitigate them.
Achievability and relevance are major concerns when dening an objective. The
degree of achievement of an objective is easier to recognize when it is veriable and
associated with a measurable indicator. Relevance of the specied objective, in turn,
implies attainability with the means made available for that specic purpose and
coherence with the different levels of intervention of the encompassing policy, which
is achieved by aligning the decoupled objectives ensuring that the objectives set at the
strategic level are correctly declined in the tactical and operational goals, as we have
described in the previous chapters.
Under the already referred CIVITAS program of the European Commission, a
number of projects were assembled having a leader city and a number of follower
cities in each research project, with the aim of gaining deeper knowledge on how to
transfer measures and policies to import a successful resolution of a given problem
from city X to city Y, as referred in Figure 5.2.
These cities provided the evidence that generic strategic objectives (e.g., good
accessibility, uidity, and low environmental aggression) are easy to transfer, but
their operational translation is not directly transferable from one city to another,
because the weights allocated to each operational objective differ, as a consequence
of the representation of stakeholders interests as well as the intervention strategy
that depends not only on those weights but also on the degrees of freedom each
system has. Through these research projects,3 we could conrm that no universal
solution exists for the different urban mobility problems; instead, we can nd types
of measures and instruments that are more likely to have a better match toward the
solution of certain type of problems.
The rationale underlying our proposed model lies on the assumption that a key
input for the UMS is the interaction between policies, namely, between land use,
environment, and socioeconomic development of the urban area, because these
aspects are upstream the generation of mobility requirements (through land use) and
the choices made by the citizens (through the pricing system, regulation on
environmental protection, scal incentives, etc.). The structure supporting this
rationale is composed of four entities: inputs, outputs, results, and impacts.
Inputs are the resources that are mobilized by processes to move the system toward
the prescribed objectives. In this sense, the concept of system inputs encompasses all
usable resources, which can be human, material, nancial, organizational, regulatory,
political, and so on.
However, inputs are not all equally controllable. Some inputs are under the control
of the transport (or mobility) authorities and thus seen from the perspective of
authorities as fully controllable, others are controlled by the service suppliers within
the mobility system (e.g., productive factors for vehicle.kilometers), and consequently,
authorities can only develop indirect control, through contract and monitoring, others
are totally external to the mobility system and, as such, no control is possible.
3. The author developed a methodology for transferability of transport policies and measures between
cities, which is currently being tested in those cities. The work was done under an EC accompanying
measure (of several CIVITAS research projects) designated as METEOR, dedicated to monitor the
progress of experiences and demonstrations in CIVITAS cities (i.e., cities participating in the CIVITAS
research project) as well as providing technical support to the cities and the European Commission with
policy recommendations based on those experiences. This work started in January 2001 and it lasted until
May 2006.
204
Figure 5.2: Cities and measures included in CIVITAS/METEOR projects. Source: Author.
There is a considerable correlation between the degree of access to information

regarding these inputs, the commercial value of that information and the degree of
control exercised by the authorities managing the UMS, which is not necessarily
equivalent to the degree of importance of the input for the management of an UMS.
Therefore, we can have situations where we have no control on inputs although there
is good information available as well as situations where the agent is under control of
the authorities but frequently withholding information that she considers commer-
cially sensitive. In short, we cannot establish a stable relation between availability of
information and capacity to control the agent who holds it.
A good illustration of this type of difculty is provided by the interaction between
land use and mobility, where the rst is a major remote cause of mobility needs with
information largely disclosed, but absolutely no control is possible by the mobility
management entities. Due to the high complexity of these two subsystems of the urban
system (mobility and land use), there are good arguments to keep them administratively
separate, which then implies that joint management control can only be achieved
through concerted decision-making between mobility and land use authorities.
These resource inputs are then built into processes. System process is the logical
organization of agents, information, and resources into activities designed to produce
a specied result under integrated decision-making. The complexity of UMSs causes
the existence of a number of several nested processes over which the strategically
dened objectives are deployed. This deployment of objectives is made in two
simultaneous ways, through the activities (activity by activity) and through the
process priority (or criticality) in terms of improvement of the overall system, with the
ones with highest potential for the improvement of effectiveness of the overall system
going rst. Process quality, in turn, is measured along the following four principal
dimensions:
Clarity enable an easy understanding of what is to be done and why and how much
of a process has been accomplished in a given moment. Clarity is a major element of
stimulation to maintain willingness and intensity of effort for long periods.
Effectiveness meeting the objective for which it has been designed.
Efciency being effective at the least cost.
Adaptability maintaining effectiveness and efciency under a changing envi-
ronment or under change of requirements.
Following the rationale and principles that support quality management theories
(Riley, 2000, pp. 6.16.21, APCER, 2003, pp. 1316), process quality management
implies the follows:
Conscious orientation toward the customers and their needs;

Specic focus on managing key cross-functional processes that affect customer needs;
A clear pattern of accountability for each key process;
A cross-functional team responsible for operating the process; and
Application of quality management principles to process management (quality
control, improvement, and planning)
Outputs in turn are the realization obtained through the transformation of inputs
supported by organized productive processes. Typically, operators of services and
infrastructure are responsible for outputs that can be divided into two main
categories, namely,
Material outputs, such as the construction of a road, rehabilitation of an old urban

area as a consequence of trafc restraint, and a walking path.
Immaterial outputs, can be the displacement of a person or good from point X to
point Y, information, training and coaching, and so on.
Results are the benets (or disbenets) that the recipients of the services delivered by
the system obtain from their utilization. It is an end-state dimension, an immediate
outcome, centered in the system user, and internal to the UMS. Results should be
subject to regular monitoring, and it is through the evaluation process that they provide
the rst information feedback for any possible adjustment required in the
implementation of an action or measure. A good illustration of a result is the
improvement of accessibility with the extension of an underground line, that is, an
enlargement of the territorial area that can be reached within a certain time threshold.
Impacts are consequences that can either affect the recipients of any process,
action, measure or policy package, or any third parties. Impacts are spread along
time and can be any socioeconomic change that accrues directly or indirectly from
any implemented action or measure. Following the methodological guide for
evaluation used by the European Commission (Tavistock Institute, 2003, Glossary,
p. 10, former MEANS project), impacts can be of three kinds:
Direct impacts, that is, specic impacts observed among direct beneciaries of the
system, which can be reected either in short term or in long term. These can be
further disaggregated in the effect they produce on the relations between the
beneciaries and the systems:
First, only by changing perceptions, which can be seen as a direct effect over
potential users and so inuencing their choices;
Second, by introducing behavioral adjustments, as a consequence of the change in
perceptions, which represents a secondary effect because they will progressively
spread throughout society.
Indirect impacts, which affect indirect beneciaries.
Global impacts, which are the ones that can be observed at macro-economic and
macro-social levels.
Finally, system evolution is the structuring effect that results from all these
impacts. Therefore, sustainable changes act as drivers of system evolution. The
feedback cycles entail an evaluation process that enables to decide whether the system
needs correction of its path and where the improvement process should be focused.
Feedback cycles assess strategic objectives against impacts and operational
objectives against results, making this evaluation complementary to the one,
previously referred, that is, made to each inner process of the UMS. This evaluation
should be based on the following set of six quality perspectives, and respective
meanings, to ensure the structural coherence of the model:
Relevance appropriateness of the operational objectives of the UMS taking into

account the context and the needs, problems, and aspirations over the system.
Effectiveness capacity to achieve the expected outputs, results, and impacts.
Efciency capacity to be effective at a reasonable cost.
Applicability adequacy of means to the achievement of objectives.
Internal coherence correspondence between the different objectives within the
different levels of the system. This implies the existence of an hierarchy of objectives
within the system, with those at the lowest levels contributing to the accomplish-
ment of the ones at a higher level, as illustrated in Figures 5.1 and 5.3.
External coherence correspondence between the objectives of the UMS and the
ones of other subsystems of the urban system. That is, for every objective of the
UMS, there is a functional relation with an objective of the urban system and its
subsystems. This correspondence will in fact contribute to ensure the vertical and
horizontal consistency of the urban system.
A set of indicators, and respective measuring methods, is required to produce this

feedback evaluation. Figure 5.4 illustrates the usefulness of a system of indicators
(EC, MEANS, 1999a, p. 211), which, for the specic case of UMSs, should have the
following attributes:
reproducible, that is, the capacity to be quantiable at regular intervals;

reliable, in the sense that the same evaluation done by two different persons will
result in the same value for the indicator;
relevant, meaning that they should be developed for the type of information that
has important implications for decision-making;
independent, that is, measured on an independent basis.
timeliness, that is, availability in due time to support effective decision-making.
Basically, an indicator can be dened as the measurement of an objective to be

met, a resource mobilised, an effect obtained, a gauge of quality or a context variable.
An indicator produces quantied information with the view of helping actors
concerned with public [and private] interventions to communicate, negotiate or make
decisions (Tavistock Institute, 2003, p. 127, our addition).
There is a wide diversity of typologies of indicators in the literature. To support
the rationale of the model presented herewith, we adopt the following typology:
Resource indicators, which cover the inputs used by the system, providing
information on the nancial, human, material, organizational, and regulatory
means used and also the ones left available for future use.
Process indicators, which cover the efciency and effectiveness of the processes
organized to transform inputs into outputs.
Output indicators, which cover the outcomes obtained.
Figure 5.3: Rationale underlying the quality management model. Source: Author.
Result indicators, which measure the advantages for the beneciaries and for the
offended, that is, winners and losers of any action or policy.
Impact indicators, which represent the anticipated consequences beyond the direct
and indirect effect over the ones affected by the system. These indicators should
contemplate positive and negative impacts and, whenever possible (on ex post
basis), should also cover unanticipated impacts, that is, spin-offs of the system
evolution. Besides, special care should be taken to ensure compatibility and avoid
redundancies with previous indicators.
Notwithstanding all the difculties regarding the universalization and even the
simple use of indicators, and following what was said in Chapter 4 about the
information systems, there are some general principles that should be adopted
regarding indicators in any UMS. These are as follows:
Indicators should support decision-making capacity, in particular, enabling

proactive action to correct the performance path of a specic element or agent
whenever signs of potential underperformance are identied.
Indicators are meant to be a learning tool for managers of the UMS providing
incremental knowledge on the behavior and performance of the different agents,
Figure 5.4: Usefulness of a system of indicators. Adapted from EC (1999a, Means

Collection, Vol. II, p. 211).
enabling a better decouple of the nal objectives of the system into targets to be
achieved by those agents.
Indicators are also an indispensable instrument to assess transferability of good
practices, ensuring careful and sound comparability between cities or urban areas,
taking into account the following concerns identied in our accompanying work of
the CIVITAS cities:
In which policy setting(s) have specic measure(s) been successful and whether
there is any ex ante requirement identied?
What are the cause-effects relationships that in each situation explain the
success/unsuccess of the particular measure?
Indicators are a main tool for contractual monitoring with the different agents and
should be included in tenders and contracts for services and infrastructures to set
minimum quality standards.
Indicators are a fundamental instrument for perceived transparency and
accountability of the system as a whole and of every engaged agents in particular,
reason why the information system supporting the UMS should be composed of
the four pivotal windows, that is, for citizens, operators of PT, operators of other
services, and infrastructures and for land use and trafc generators authorities, as
presented in Chapter 4.
5.2. Need of an Integrated Approach to UMS
Following the rationale presented in Section 5.1, one of the main inputs that
inuences mobility needs is the location of activities. That is, land use congurations
are instrumental for the subsequent denition of objectives of the system and also of
the degree of interaction needed between the other components of the system. Indeed,
all this work lies on the assumption that mobility is a system-wide phenomenon,
where an interconnected set of elements is coherently organized around a specic
purpose, which is to provide citizens of an urban area with good levels of service in the
accessibility to urban functions, as much in self-service regime as in third-party supply
regime. As we have developed in Chapter 2, the structure of a system largely
corresponds to a network of causal relations, the identication of this causal structure
being the main contribute to the denition of its boundaries.
In the research project TRANSPLUS, we have had the opportunity to validate the
causal relation between land use and transport, which is what we have specied as a
key input for urban mobility management. The observation was done based on a set
of case studies (see Annex 2) covering different cities, the typology of which was
dened along the attributes presented in Table 5.1.
The two causal loop diagrams (CLDs) illustrated in Figures 5.5 and 5.6 reveal the
tight relation between these two sectors of urban life. According to Sterman (2000,
pp. 137156), these diagrams represent entities connected by arrows, illustrating the
casual links, and denoting the inuence with each other. Entities can be either very
simple (unidimensional) in which case entity and variable have the same meaning, or
described by several variables.
In a systems dynamics context, entities are elements of unspecied quantities,
which have the capacity to affect other elements and in turn be affected themselves.
Within entities, there is still a distinction between level and rate (or ow) variables.
The former are accumulations over time, an amount of something, which are
frequently called by professionals as stocks or state variables. The latter represent
changes of the values of some variables per unit of time.
Theoretically, each causal link between entities is assigned a polarity that indicates
how the attributes (variables) of the dependent entity vary when there is a change in
the independent variable. A positive link polarity means thus that the effect variable
has a response (i.e., variation beyond its underlying trend) of the same signal than the
stimulus constituted by the variation of the cause variable. Reciprocally, a variation
of opposite signal to the stimulus occurs when the link has a negative polarity.
It is worth saying that the application of the concept of dependency is not always
straightforward in a system of the complexity (intensity and diversity of interactions)
Table 5.1: Typology of cities based on urban form.
Settlement form Densitya Spatial pattern Commuting pattern
Strong core city High A central city with a High volume of trips
compact urban form within the city
around an center and of radial
identiable core trips to and from the
center
Polynucleated city Medium- A large urban area High volume of trips
high with multiple fully between the core
developed and the other
subcenters, possibly subcenter, and also
clustered around an among the
identiable center subcenter
Urban network Medium- A cluster of separate High volume of trips
high cities with strong among the cities
interactions due to
close proximity
Spread-out city/ Low Spread-out and low Diffuse trips in all
region density distribution directions
of land uses
Source: Adapted from EC (2002a, TRANSPLUS, D1.2, p. 9).
a
Density can be measured in different ways, depending on the population chosen for the numerator and
the area taken as denominator. The most common but also rather ineffective way to measure density
is with the ratio: total population/total area. However, with this measure, only compact cities tend to show
high densities, whereas both polynuclear and spread-out cities can show low densities, if they have similar
total amounts of population and surface. More precise measures of density are needed in this case to show
the high-density patterns of polynuclear and network urban forms, for example, those that measure the
concentration of population in the built-up area. This development is considered out of the scope of the
current work.
of the UMS, because the absence of a direct link may be more than compensated by
the effects of a chain of links. Therefore, we should say that links represent identied
direct relations (associations or inuences) between variables and be less categorical
about dependencies and independencies.
CLDs are a valuable tool to understand systems dynamics. However, there are
some pitfalls, such as the fact that link polarity describes only the structure of the
system, nothing being said about behavior of entities. In addition, stocks and ows are
not distinguished in the diagram, and, nally, diagrams are neither comprehensive nor
nal, they are always a provisional map that progresses as our understanding of the
phenomenon develops, that is, the diagram is by itself a learning tool.
As Sterman (2000, p. 141) explains, Correlations among variables reect the past
behaviour of a system. Correlations do not represent the structure of the system.
That is, in synthesis, they do not describe what actually happens, rather they describe
what might happen if a change occurs. For this reason, they are very useful in the
Figure 5.5: Transport trends. Adapted from EC (TRANSPLUS, D1.3 (2002a, p. 23)).
Figure 5.6: Land use trends. Adapted from EC (TRANSPLUS, D1.3 (2002a, p. 24)).
design of policy instruments and packages as they provide the structural relationship
of the systems with the identication of potential winners and losers.
Having these limitations in mind, Figures 5.5 and 5.6 provide the illustration for
causal relations between the complex systems of land use and transport, based on the
mentioned case studies, from which we highlight the most important loops (EC,
2002a; TRANSPLUS, D1.3, p. 21, pp. 1325, D3, pp. 8089, p. 115, 119):
If car use increases, suburbanization increases as well, which leads to more land
consumption, resulting again in an increase of car use. This is called a reinforcing
loop, leading to exponential growth until some resource constraint blocks growth
of one of the variables of the loop.
An increase in car use leads to more congestion, which in turn discourages people
to use their cars. As car use decreases, less congestion will accrue, and this will
again encourage people to use their cars. This is called a stabilizing loop, leading to
dynamic equilibrium. It is worth highlighting that this stabilizing loop occurs only
exceptionally if the available PT offers substantially better conditions of uidity
(with segregated infrastructure, bus way, or railway or tunnel); otherwise, the
individual choice continues to be the use of private car.
If car uses increases, residential areas can be potentially located at greater distance
to the center of urban areas, which in some cases contributes to an increase of
suburbanization and urban sprawl. Consequently, travel distances will increase
and again will lead to increase of car use. This is another reinforcing loop.
An increase in car use leads to increased suburbanization of retail centers. A
consequence of this is the reduction of the total number of retail locations, which
leads to an increase of the average catchment area and higher distances between
home and retailer. The resulting is the increase in car dependency and car use. This
is again a reinforcing loop leading to exponential growth. In fact, we have noted
that suburbanization of retail centers is inuenced by two groups of entities.
If the availability of land or parking lots increases in inner cities, suburbaniza-
tion of retail centers will decrease. Similar effect on retail centers results from a
decrease in the price of land in inner cities.
If the number of private cars or congestion in inner cities increases, then
suburbanization will increase as well.
Car use is indeed at the core of the management of UMSs, and the CLDs
presented in Figures 5.5 and 5.6 provide the supporting evidence for the need of an
integrated approach to urban mobility, given the high number of interactions,
highlighted through the above-explained reinforcing loops. Furthermore, car
dependency is a direct consequence of ease of use or perceived marginal price and
(therefore) frequent use. The result of this is another reinforcing loop that we have
also analyzed in TRANSPLUS research and expressed in terms of resistance to car
use by the consumer citizen, which can be expressed in monetary cost, time effort,
physical effort, and so on, as illustrated in Figure 5.7.
In line with this formulation, we observed in the last years a growing interest in the
assessment of urban quality of life and the identication of its drivers done by
Figure 5.7: Components of resistance to car use. Adapted from EC (2002a,

TRANSPLUS, D1.3, p. 25).
international organizations such as UN, OECD, WHO, and also some European
Union institutions (e.g., Regional Policy and Eurostat). In these initiatives, the
interaction between land use and transport is always present through assessment
indicators.
An example of initiative reecting these concerns was the urban audit launched by the
European Union in June 1997, inviting 58 European cities to participate in a pilot
exercise. The assessment indicators were regrouped in 21 domains, for three observation
points in time (1981, 1991, and 1996). Scores were calculated at the city level, and 27
cities out of the 58 original samples were assessed at the wider territorial unit or
conurbation level. Results of this exercise are available at the urban audit web site for the
58 participating cities together with information from other international experiences
(http://ec.europa.eu/).
In this exercise, the following indicators were used:
Land use: dimension of green space to which the public has access, % of the
population within 15 minutes walking distance of urban green areas, % of urban
area unused and in main land uses, % of the urban area subject to special physical
planning/conservation measures, population density.
Travel patterns: mode of journey to work (rail/metro, bus, tram, car, cycle, and
walking), total annual distance traveled by residents (by mode, purpose), car
ownership, road accidents resulting in death or serious injury per 1000 residents,
and average car occupancy.
The indicators selected in the exercise done in urban audit hardly report the
interactive aspects between land use and mobility; therefore, this exercise reinforces
the evidence that there is a strong need to have valid and continuous information
available on the integrated aspects of the urban life. That is, cities (i.e., political
decision-makers, urban planners, and managers of urban systems) should aim
at systems of indicators able to monitor the effects of policies and programs, so
that they can ensure the dynamics of the feedback cycles, referred in the previous
section, as a main driver of the quality of decision-making. In fact, nothing of this
information for decision and management can be obtained if we ignore the holistic
character of the cities and urban areas or the symbiotic phenomenon associated with
the interaction between its main components.
The urgency in having this information for management was clear in the message
for ministers of the European Conference of Ministers of Transport (ECMT) stressing
that a policy option is meant to include a range of feasible alternatives to reach a policy
objective, including packages of policy instruments as well as specic choices
concerning detailed measures to be implemented within such packages:
sound and reliable data are the empirical basis for good policy-making
and serve as the inputs to the analytical process. Urban data, particularly
as concerns urban travel and land use and their interactions, remain
sparse, inconsistent and often of overall poor quality, however. Data are
not collected in a consistent way among cities and collection methods are
often subject to modication within a given city. (ECMT, 2002, p. 37)
However, despite this call for consistency and use of feedback cycles as a way to
ensure quality of decision-making, this can only be effective if city contextual data
can be used as a baseline scenario (the usual do nothing scenario) against which
the different effects (i.e., marginal effects) accruing from the various policy options
can be assessed.
But even if this information exists and is used, the complexity of the UMSs by
itself has a number of difculties in interpreting the variations of indicators as a
resultant of the measures and actions undertaken by decision-makers, which hinder
clarity of feedback cycles. That is,
A specic policy action (e.g., social housing investment) is likely to affect a

considerable range of policy objectives, that is, in most cases, we have one to many
interactions (e.g., local employment, transport ows, energy consumption, and
environmental preservation) instead of limiting the effect to a single area of policy
concern (e.g., volume of affordable homes or level of family savings), which would
be represented by a one-to-one interaction.
Another difculty accruing from this complexity of interactions reects in the fact
that an indicator might also be affected by more than one action, bringing
additional difculties in the identication of which policies, and to what extent
underlie the resulting effects;
At the other extreme, we have the case for simultaneous change of several
indicators, which is also a common problem resulting from limited knowledge on
cause-effect relations between the actions undertaken, the results or outcomes
achieved, and the impacts, which often occur after a time lag that can last several
years between the implementation of the stimulus and the visibility of the answer
(e.g., change of attitude toward PT). The longer the time lag more probability
exists that interference from other stimulus introduces additional complexity in the
interpretation process, difculties in a clear understanding of dependencies
between variables, as well as in the selection of measurement methods.
These problems are rather evident in the CLDs designed for the city of Bristol, in
the framework of the already referred research project TRANSPLUS and presented
herewith in Figures 5.8 and 5.9, where three policies were depicted and indicators
were selected according to their functional relationship with land use and transport
phenomena. These policies were materialized as such: PT-oriented development,
promotion of nonmotorized modes, and car-restriction-oriented development and
Figure 5.8: Causal loop diagram for land use trends. Adapted from EC (2002a,
TRANSPLUS, D3, p. 170).
Figure 5.9: Causal loop diagram for transport trends. Adapted from EC (2002a,
TRANSPLUS, D3, p. 169).
the analysis of both CDL and conrming the reinforcing loops conceptually
presented in Figures 5.5 and 5.6.
This causal analysis also highlights the fact that integration between land use and
transport has to be achieved in two dimensions. The operational dimension,4 where
the interdependencies and interactions of effects occur, measures implemented, and
results achieved, and the institutional and policy integration dimension where the
various agents responsible for policies and actions have to concert their behaviors
4. In TRANSPLUS and TRANSLAND research, this dimension was called material integration.
and agree on common rules aiming to change decision processes and the long-term
achievement of impacts.
The second dimension constitutes to a large extent, a prerequisite to the successful
control of the rst, this is why we will call it an enabling factor in the structure of our
model. In fact, there are several evidences of the need for policy integration that can
be summarized in the following arguments:
No single measure or policy is able to provide by its own a solution to current

mobility problems. Very often, some measures can only achieve optimal
effectiveness if they can be complemented by other measures inuencing citizens
decision processes (e.g., bus corridors).
Most measures have positive contributions to give for the solution of the problems,
but most measures also have adverse impacts, that is, all measures produce gains
and losses toward the specied objectives. There are also particular cases where
some measures can achieve clear benets in one area but at the expense of
deterioration of mobility conditions elsewhere (e.g., trafc calming).
All the above-mentioned reasons provide evidence that a package of measures can
be more effective than a single measure, but different types of benets can be focused
in the decision to integrate measures, such as the follows:
Measures complementing each other in what concerns the impact on users (e.g.,
provision of park and ride facilities to increase PT patronage);
Measures that contribute to the nancial feasibility of other elements of the system
(e.g., parking charges or mobility charges applied to real estate providing nancial
means for the development of infrastructure); or
Measures that contribute to soften the public acceptability of other measures (e.g.,
increasing the road pricing acceptability by applying revenues in PT investment).
However, despite the recognized advantage of integration of policies and measures,

we could observe a considerable number of cases where severe barriers to policy
integration have occurred. From these, we highlight the existence of different govern-
ment levels with responsibility for different, but interrelated, policy areas, leading to
the problem that no authority has full control in all relevant transport and land use
interaction policy areas. In addition, authorities often need to make trade-offs and
balance different interest areas, often involving another government level.
Finally, land use decisions are characterized by a long-term production of effects,
very often decades. As Viegas (2003b, p. 3) refers, even when these decisions were
taken in full integration with transportation issues, it often occurs that tensions in
transport evolve and later what was considered as a good decision becomes an
undesirable one. This means that we have an insufcient clairvoyance of the decision-
making structures that can only be partially compensated with regular prospective
studies on the evolution of the several dimensions of the urban whole.
5.3. Evolving Decision-Making Processes in UMS
During the current work, we have had the opportunity of analyzing the decision-
making processes in several European and South American cities5 surveyed in the
studies done in parallel to the current work. From this observation, we conclude that
despite the wide variety of literature concerning decision-making processes, there are
four main practical approaches in UMSs:
Vision led, when the decision-maker has (or so believes) a clear vision on the
policies and measures needed to solve the current problems and improve system
performance, in which case, she focuses all attention and required resources in its
implementation.
Plan oriented, where problems are identied, objectives are set, and the measures
and policies that best meet those objectives are specied by analysis leading to the
implementation of the plans.
Consensus building, where stakeholders are consulted and impact on these groups
assessed. At the end, measures and policies implemented are the ones gathering the
greatest support through referendum or public consultation formal process.
Mixed approaches, which involve leadership, planning, and stakeholders con-
sultation as proposed in Viegas and Macario (2003a, pp. 213225) and applied in
the cases of Lisbon and Brazil. The original problem is explained to groups of
stakeholders, alternative solutions are presented, and choices are explained and
done after assessing and presenting the pros and cons of each option. After this
phase of gaining acquaintance with the problem and possible solutions, imple-
mentation plans are done and again presented to stakeholders for a better
transparency of the process. Feedback process is also implemented and made
public. A good example of the participation of the feedback process is the case of
the London urban toll where every 6 months a detailed report is done by TfL
(Transport for London, http://www.cclondon.com/, July 6, 2005) and published.
Mixed approaches can be organized with different intensities of any of the previous
three more radical approaches.
There is no best universal approach to decision-making in UMSs; each city has its
own characteristics that inuence the options taken. However, there are some useful
references from existing experiences that can serve as indications, for example, the
obvious pitfall of a vision-led approach, which is critically dependent on the
5. Cities surveyed in the framework of the research projects ISOTOPE, QUATTRO, MARETOPE,
TRANSPLUS, and CIVITAS program, as referred in Annex 2. In addition, this aspect was also surveyed
in the metropolitan areas of Lisbon and Porto, during the work leading to the organization of the
management of these mobility systems, and in Brazil during the works for adoption and implementation of
the current model for the management of UMS in that country. This work is formalized in the project of
law known as the National Directive for the implementation of the Estatuto da Mobilidade Urbana,
which was presented to the parliamentary approval by the end of July 2005 (Annex 3).
individual developing the vision, or the plan-oriented approach that can lead to an
excess dependency on technical planners expertise without sensibility for the political
component, or the consensus-building approach that can lead to strong delays and
inaction, unless consensus is rapidly obtained. All these pitfalls contribute to favor
the mixed approach.
In a similar classication, the research project PROSPECT surveyed decision-
makers from 54 European cities and concluded that more than one-third (23 cities)
used an approach to decision-making in land use and transport issues that cor-
responded to a mix between plan-oriented and consensus-building approaches. The
survey provided also evidence that there is a higher preponderance for this mixed
approach in northern and eastern European cities, which reects a strong tradition of
participatory approaches in these countries (Mathew, 2001, PROSPECTS, Report
on Task 16, unpublished, p. 25).
In Brazilian cities, metropolitan regions, and states,6 a strong preponderance of
mixed approaches was found in cases previous to the process of implementation of
the model proposed in the current work. In those cases, we could nd similar
prevalence in two types of mixes: vision-led and consensus building; or plan-oriented
and consensus building. The incidence in consensus building in the Brazilian reality is
justied by the strong formal participatory culture in this country, namely, with the
existence of the recently created Council of Cities, representing the organized society,
within which the Committee for Transport and Urban Mobility is the body regularly
consulted (at least twice per year) on mobility policies and to whom the federal
government is accountable to. Similar formal structures exist at the state and
municipal levels of government.
Also in Europe, the consensus-building approach gained importance along past
years, for example, in France, we can nd an identical forum, the GART citizens
conference. The GART is an association of over 240 French transport organizing
authorities (i.e., cities, departments, and regions). This qualitative approach, very
similar to the one in Brazil, and generally qualied as a consensus conference
(GART, 2001, p. 1), consists of making a panel of citizens aware of a societal issue
to have a new nontechnological point of view of a complex problem and the
corresponding feedback for technicians and politicians.
In fact, both in Europe and in Brazil, we have found that there are four key
elements in the decision-making process: the actors, the type of public participation,
the instruments to induce change in behavior, and the barriers to the decision process.
The interaction between these elements denes the risk border for the decision process,
which always brings together some elements of change or transition.
Good examples of transitions within urban mobility are the change from a state-
regulated regime to a market-driven regime, the move toward the hydrogen economy,
the implementation of new sectorial institutions, such as regulatory agencies or
6. Interviews made to secretaries of transport of all state capital cities and also to metropolitan regions and
the current and former National (Federal) Secretary of Urban Mobility and the minister of cities.
monopolies and mergers commissions, or even any type of formal organization

focusing the regulation of trade and prices (less common in urban mobility but very
typical from maritime and air transport).
In urban mobility management, it is important to identify, understand, and
monitor all types of transitions. These exist out of the result of the interplay of
several relevant processes such as changes in technology, infrastructure, economic
conditions, institutions, politics, behavior, culture, natural and built environment,
and, nally, beliefs that reinforce each other. As Rotmans, Kemp, and Asselt (2001,
p. 21) state, transitions are characterized by a nonlinear behavior and multiple
causality, where the time span is a result and not a dening characteristic. In the
UMS, a transition can be thus understood as the result of long-term developments in
stocks and short-term developments in ows, with the various time axes shifting over
each other and inuencing each other.
Transitions have a life of their own that has to be taken into account in the
decision processes, even more so because as we have repeatedly said along this work
the UMS has no direct command and control instruments for the whole system
dimension; therefore, the coordination between the several components of the
system, which we have called concerted decision-making, is made through induced
behavior. In addition, the duration of (life of) the transition process can be either
expanded or compressed (although within some limits) as a consequence of
constraints imposed by the political calendar.
This evidence was made clear in the process of design and implementation of the
management model presented in Figure 5.11 in Brazil, where we have dened and
evolved along the following stages of the transition process7 that occurs in different
moments for each component of the system:
The inception phase, where little change is visible but in-depth assessment of the
existing equilibrium is done for a pre-evaluation of the feasibility of the change
envisaged. Here, participative discussions should be very intense to enable a wide
and common understanding of the reasons underlying transition, as well as to
ensure that compatibility tests with the interacting elements (e.g., sectorial policies,
specic technical rules and regulations, and constitutional rights).
The take-off phase, where the change is effectively in the eld and the equilibrium
of the system starts to shift and the transition becomes visible. In July 2005, the
Law with the Directive for Regulatory and Organizational Reform of Urban
Mobility Systems in Brazil (at the time known as Estatuto da Mobilidade
Urbana) closed the ofcial public consultation process and was submitted to the
7. Another plan of a much more reduced scope was drafted for the transport authorities of the
metropolitan areas of Lisbon and Porto in the second half of 2004, although with the change of
government the transition process was suspended, and until the moment of delivering the manuscript for
this book (January 2011), no information was available on the intensions or decisions of the new
government.
parliament for approval (through Civil House of the President), which corresponds
to a main milestone, marking the second part of this take-off phase.
The organized disruption phase, where structural changes take place in a visible
and strong way given the superposition of the transition processes of several com-
ponents of the UMS, accumulating sociocultural, economic, environmental, and
institutional change. In this phase, there will be intense collective learning processes,
dissemination and embedding processes, and methods with the different options
taken by cities and metropolitan regions. Monitoring and feedback for adjustment
are key functions to support the decision processes at this stage.
The maturation phase, where stabilization and consolidation of change occurs in a
progressive and cumulative way leading to the new equilibrium of the system. Once
entering this stage, steering becomes the main function.
Indeed, in current democratic societies, with high diversity of stakeholders

personalities and a culture of intense public participation, decision processes evolved
into the management of complex sets of transition processes, where the key elements
are as follows:
Long-term thinking (2025 years) as the basic framework to shape medium-term

policies (1015 years) and short-term action (until 4 years, which is the most
common political term). This means that ensuring a long-term change in a stepwise
way giving enough time for consolidation of transition processes is absolutely
indispensable. Applying this rationale implies that all policy and management
actions should have a twofold assessment: against their contribution to short-term
goals and against their contribution to the overall transition process.
Multidimension thinking, that is, multiactor, multidomain, and multilevel (i.e.
strategic, tactical, and operational for different scales, i.e., local, regional, national,
etc.), encompassing concern on how developments of different domains interact
with each other. This means that each decision is subject to delicate trade-offs
between all the stakeholding interests.
Backcasting capacity, that is, setting goals based on long-term sustainability
scenarios and short-term political, social, economic, and technological feasibility,
looking at the means through which end goals are achieved. This is a practice
not yet common, although very useful, for which an information system is an
indispensable tool.
The complexity of the management processes evolved in parallel to the widening

and sophistication of the governance methods and also with the opening of the
markets. In addition, all cities decisions are, to some extent, inuenced by government
authorities from different sectors, the strongest inuence in what concerns the UMSs
coming from territorially adjacent authorities or territorially overlapping authorities
(e.g., metropolitan or regional authorities). Therefore, these constraints seem to have
more impact in medium-size cities embedded in relatively dense regions, as the smaller
isolated city has less interaction constraints, and the larger cities, given the stronger
economic and social dynamics, tend to develop smaller satellite surrounding cities,
which puts them in a position of more inuence when compared to their adjacent
counterparts and also toward the institutional hierarchy.
The more intense participation of private agents in the decision-making process
deserves some attention as this has been the driver for the need to develop inductive
tools to inuence their behavior in line with the system objectives. In what concerns
development of socioeconomic systems, where system steering is done without the
traditional chain of command, there are three main inductive tools:
Pricing administration, more specically pricing mechanisms that constitute the

coordinating element, through signaling, for decisions of autonomous actors by
providing them with the information about scarcity and economic value enabling
to equate supply and demand.
Concerted planning, more specically macro-planning, dening where and how the
different urban sectors should engage in concerted decision.
System structure or institutions, which provide interpretative frames and beliefs,
collective norms, allocation of responsibilities, telling people who does what, and
what is to be done under which circumstances. That is, in brief, the governance
structure of the UMS.
Another element of difculty in cities decision-making relates with the time span
of the formal planning acts. As a diversity of approaches exists in Europe and
elsewhere, we take only two examples to illustrate this variety: France uses a 10-year
period for the plan de deplacements urbains, whereas the British local transport
plans are done for periods of 5 years but framed by a 15- to 20-year strategy. The
planning capacity, that is, a main feeder of the decision-making processes, is in itself
depending on the administrative structure of the city, which in turn is inuenced by
the policy level. Consequently, the change that occurred in policy thinking during the
past decades had also a major inuence on planning methods and vice versa.
These changes were summarized in Goodwin, Hallett, Kenny, and Stokes (1991,
p. 111) in ve principles that had developed in the previous decade:
Transportation is part of a greater urban problem, which needs to be considered at

all government levels;
Consistency of treatment between modes is important;
It is acknowledged that meeting all desires for transport is an impossibility;
The notion that transportation requires more than simply technical solutions is
acknowledged and there is a call for greater understanding of human factors and
motivation to travel; and
Some classes of trafc are necessary or desirable, and there is a call to prioritize these.
It is worth referring that transport planning as a recognized discipline is

conventionally considered to have started in the 1950s with area-wide transport
studies in Chicago, Detroit, and others (Kane & Mistro, 2003, p. 115). The planning
method that emerged during this time was a technical exercise focused toward
cost-effectiveness having the transport planner (usually a civil engineer) as an expert.
In the late 1950s and early 1960s, the development of computers inuenced the
planning methods, and computers were a part of the rational scientic planning
and transport modeling that emerged. In the late 1960s, as a response to perceived
shortcomings of that scientic approach, the systems approach emerged (Checkland,
1999, p. 59) as an attempt to view the world holistically, in clear contrast with the
previous view of the scientic method, although this reected in transport planning
only later, in the 1970s.
System thinking led to systems analysis, the development of which is attributed to
the RAND corporation (Checkland, 1999, p. 134), who described systems analysis
as entailing the following steps, which many people consider to (also) match the
transport planning method:
An objective(s) to accomplish;
Alternative techniques by which the objectives may be accomplished;
Costs/resources required by each system;
Mathematical model(s) showing the interdependence of the objectives, the
alternative techniques, the environment, and the resources; and
A criterion, relating objectives and costs or resources for choosing the preferred or
optimal alternative.
In fact, as referred by Kane and Mistro 92003, p. 117), we can conclude that in a
relatively short time span, transport planners have had to work under a considerable
diversity of frames such as scientic analogies for transportation phenomena observed
(e.g., gravity model); hard systems approaches to conceptualize the urban transport
systems; rational approaches to planning; and computing tools to simulate the
implementation of all the previous ones.
Indeed, the origin of transport planning and, consequently, the decision processes
it supported, had a focus on objective, reductionist, and strictly technical dimensions
of the mobility problems, neglecting the sociopolitical and holistic spectrum. The
dominant way of thinking about transport planning was designated by Goetz and
Szyliowicz (1997, p. 265) as rational comprehensive and described by Linstone
(1984, pp. 8789; also cited in Wachs, 1985, p. 527) and more recently in Kane and
Mistro (2003, p. 118) as addressing the following aspects:
The notion that analysis and decision-making are separable activities performed by
different actors;
The denition of problems that are abstracted from a complex world, and the
implicit assumption that problems can be solved;
An orientation toward optimization or searching for the best solution;
A commitment to reductionism;
A practice of research and study of systems that are dened by a limited number of
elements or variables and by their interactions;
Reliance on data models and combination thereof as modes of representation and
inquiry;
Quantication of information;
Belief that the analyst or researcher is outside the system she is studying and that
knowledge can be found, which is independent of the observer; and
A commitment to problem-solving as a sequence of logical steps.
This rational comprehensive mental model is currently seen as limited to

properly support decision-making on mobility and requiring reconsideration. A
rst group of reasons lies on the scope of the object of study, for example, while the
earliest studies were focused on cost-efciency in infrastructure provision, their
equivalent in our days considers much wider subjects such as accessibility, macro-
economic impact, environmental impact, equity, land use, and growth management
(Chisholm, 2000, pp. 2728). Therefore, this means that we have moved from an era
of relative simplicity of thought to an era of acknowledgment that the real world
can only be duly represented through greater complexity and more holistic
approaches, although our capacity to deal with the real number of attributes of each
entity is somehow limited and most analysis tend to simplify the characterization of
interacting entities.
An immediate outcome of this recognition was the consideration of analysis
of potential conict in the appraisal and evaluation methods. This development
forced the evolution of the decision-making process in mobility systems from a
theoretically objective process, where objectivity was mostly supported by cost
optimization as noted by Kane and Mistro, (2003, p. 118), to one where the weight
of values and norms plays a central role in the construction of each person system
value, as referred in Chapter 2, leading to different preferred decisions over the same
object. The traditional rational approach does not cope with this problem in an
adequate way.
Indeed, our observation of the decision-making processes in the cities covered by
this work converges with the conclusions taken by other authors such as Szyliowicz
and Goetz (1995, p. 364) who noted that the rational comprehensive approach is best
suited for decisions taken by a single body and, while analyzing air transport,
concluded that failure in infrastructure projects has a strong relation with the fact
that the political dimension is often underestimated and even ignored, or Meyer and
Miller (2001, p. 487) who stressed that transport planning has been insufciently
orientated to decision-making and ought to be considered as a political exercise as
much as a technical one, or Khisty (1992, p. 141) who observes that a plan is really a
political statement and, consequently, all implementation is a political act, as we have
largely validated in the cities engaged in the above-mentioned CIVITAS initiative of
the European Commission.
In conclusion, the evolution of the decision-making processes in UMSs revealed
that a systemic approach encompassing a political component have to be used, and
this implies the in-depth revision of planning methods so that systems methodologies
are incorporated, reecting the need for consideration of multiple stakeholders as a
condition for good system governance and, consequently, for sustainability.
5.4. The Structure of Model for Management of UMS
Given all the evidences from the cities observed and the rationale developed in this
work, the management model presented in Figure 5.11 is structured along the
following building blocks for the purpose of its implementation (in any city):
Contextual agenda the list of issues or problems to which governments, and

other people outside government, but in their immediate sphere of inuence, are
giving attention to in the short, medium and long term.
Purpose of the system the denition of the territorial scope (boundaries) of the
UMS and its overall strategic objectives, which as we have seen in previous
chapters is dependent on the political administrative organization of the country.
Entities decision agents involved in the different components and stages of the
system (e.g., political authorities, organizational authorities, urban authorities
(political and management), sectorial entities, surveillance and enforcement
entities, operators of services and of infrastructure, third-party suppliers, workers,
clients, and citizens).
Boundaries of decision meaning the institutional design and scope of intervention
and functional allocation of responsibilities within the system. That is, the answer to
the normative question who does what between the different existing interacting
institution and agents.
Decision processes which at the macro level of the whole system are as follows:
Strategic denition of the system
Selection and design of management (steering) instruments
Steering process with the following subprocesses:
Conguration of system supply
Contracting of system supply
Monitoring and adjustment of system evolution

Information meaning information ows and channels supporting the assessment
required to provide feedback to the steering function and enable corrective or
preventive actions to keep the mobility system in its due course.
The system model is basically the organization of activities in such a way that
the overall result is in line with its purpose and expectations. This is established
at the outset by the entities (in democratic societies) who have the representation
of the populations and consequently dene the political objectives for the UMS.
The purpose of the model presented in Figure 5.11 is thus to manage the UMS
as a whole (encompassing all modes and means) and provide the right context for
a good performance, which will in fact be done by the agents (private and
public).
Just like organizations, systems also in general should have their design matching
the purpose of their mission. Therefore, organizing the mobility system means
understanding what to do and which organization is in best position to do it well.
The rst part can be considered as rather universal, whereas the second largely
depends on the institutional context existing in each country and city, and therefore,
it corresponds to the models adjustment to the different local realities.
As we have already referred, urban mobility involves a wide diversity of agents and
we must recognize that every agent has its own agenda and purposes that may
challenge or constrain the system objectives. The inevitable consequence is that trade-
offs have to be considered. Some are part of the compete collaborate dynamics
between agents, but others have also to be set at system level between objectives,
available resources, and results. Therefore, all this implies that the USM has to be
managed with strategic thinking, that is, an implicit interactive process that steers the
UMS considering that each agent will act according to her own strategy and aiming to
maximize her private benets, which sometimes will place the agent as a rival to the
system objectives and in other moments as an ally.
The underlying management method lies in some basic rules dened at political
level and accepted (at least temporarily) by all players, such as every movement
by one agent will provoke a reaction from the others, and, to understand possible
consequences, we need a prospective view of the possible moves of all the players and
of an evaluation of the impact of actions and reactions over the system performance.
This implies a forward thinking about these movements and feedback rationale to
understand and decide which course of action is most likely to cause the reactions
that will keep the UMS in its due course. In fact, the main challenge in urban
mobility management is to manage the system according to the dened objectives but
ensuring that each agent sees an advantage in taking part of a system that is expected
to create value for all the players engaged.
With the development of management science, the concept of strategy gained the
importance of something that is widely considered as indispensable, but very few
agree on what exactly is. The teleological discussion on strategy is vast, but it is not
within the purpose of this work, so to this respect, our concern is simply to clarify our
adopted denition of strategy as being the selection of development objectives and
paths that best suit the evolution of the UMS in line with the political aims and
society needs and considering available means and contextual opportunities and
constraints. Our denition draws on the capacity of execution, which is the capacity
to make it happen in the real world, that is, the line that distinguishes a strategic
vision from an illusion.
The model presented adopts a process approach, that is, transformation of objectives
into results and further into impacts, that is, the pursuit of quality is done through
management processes, which are organized sequences of activities that produce the
intended quality results. In business environments, quality management makes exten-
sive use of three of those management processes, known as Juran trilogy (Juran, 2000,
p. 2.5), which we recognize as also valid for the management of UMSs:
Quality planning processes that dene what is to be done;

Quality control processes that monitor and evaluate performances and
alignment (or deviation) with planned objectives; and
Quality improvement that focuses on improvement of results that can be
achieved through many possible ways.
In a rather universal description of these macro quality processes, we can say that
there is for each of them a relatively stable sequence of activities that is illustrated in
Table 5.2. Crossing these macro processes of quality management with the reality of
UMSs allows identication of the specic processes to manage this system and also
to identify which decision (or planning) level is in best position to manage it. This
map of processes is presented in Tables 5.3, 5.4 and Figure 5.10 and reects the
structure of the model presented in Figure 5.11.
The adoption of a process approach in a complex system such as the urban
mobility implies that each process has to be transversal to the institutional setting of
the local environment where it will be applied, just as illustrated in Figure 5.10, and
that a large variety of agents participate in each process. These characteristics
introduce some additional management difculties that reinforce the convenience of
using some process management principles, although adjusted to this systemic
reality, namely as follows:
Only one institution should be responsible for each process.

Each process has to be clearly identied with the following elements that constitute
the process identity: designation and description; material and immaterial resource
inputs used in the transformation into outputs and clear identication of the
sources for those resources; expected outcomes (objectives); set of activities (and
task within these) that constitute the process; functional institution responsible for
conducting the process; how the process is declined in several decision levels;
interaction with other institutions; dependencies from other processes; and control
points and indicators to assess quality performance of the process (i.e., efciency,
effectiveness, and adaptability as already referred).
As we have pointed out in previous sections, the starting input and main con-
straint for the management of the UMS, and consequently to all key processes, is the
political and institutional interaction between the three following sectors that
are critical in the generation of mobility requirements and also in the denition of
political priorities for the system: socioeconomic context, land use policy, and
environmental preservation policy.
From this point forward, the management function should sequentially address
the key processes in the order presented in Tables 5.5 and 5.6 where the critical
aspects of each process are highlighted. The operational and material representation
of these processes, and so of the UMS, varies from one city to another depending on
a number of factors.
The political administrative organization of the country and consequently of the
urban area served by the system can be a city, a voluntary consortia of municipalities
(as it can be the case in Brazil or in Spain), a formal metropolitan area (as can be
found both in South America and also in Europe), or the (voluntarily formed)
metropolitan planning organization (as can be found in most US states). This
organization imposes constrains on the institutional design and consequently affects
the power of institutions engaged in the system and its performance (e.g., in
Hamburg, we could nd six different types of authorities intervening only in PT).
Table 5.2: The three main processes for quality management in UMS.
Quality planning Quality control Quality improvement

Understand political, Establish performance Justify the need to
social, and economic monitoring and improve
priorities evaluation methods Identify the policies and
Translate those Establish an measures required
priorities to the mission infrastructure to ensure Identify agents to
of the UMS feedback information trigger improvement
Establish quality targets ows Establish the required
Dene sources for the Evaluate actual infrastructure (if any)
nancial means that will performance and Establish controls on
support the system compare it with quality improvement
Identify social proles targets Identify the instruments
and requirements of Nurture feedback ows that will foster agents
customers Understand causes of actions and reactions
Dene quality criteria deviations to the Dene sources for the
that respond to performance plan nancial means that will
customers needs and Act on performance support system
expectations deviations, realign the improvement
Design networks system Assess effective results
Plan infrastructures and prospective impacts
Dene the regulatory Nurture feedback ows
and organizational (Re)align the system, if
framework for the needed
provision of services
Ensure provision of
services and
infrastructures conform
to quality criteria (e.g.,
concessions,
management contracts,
and tendering)
Dene the instruments
for the steering function
Establish process
control
Transfer the plans to the
tactical and operational
agents
Source: Author.
Table 5.3: Key process in the quality management of UMSs quality planning.
Activities within the quality management Decision levels

processes
Strategic Tactical Operational
Understand impact of political, social, SD CS AA

and economic priorities in UMSs
Translate those priorities to the mission SD
of the UMSs
Establish quality targets SD
Dene sources for the nancial means SD
that will support the system
Identify social proles and requirements CS
of customers
Dene quality criteria that respond to CS
customers needs and expectations
Design networks and services ES
Plan infrastructures ES
Dene the regulatory and organizational ES
framework for the provision of services
Ensure provision of services and ES
infrastructures conform to quality
criteria (e.g., concessions, management
contracts, and tendering)
Dene the instruments for the steering SI
function (institutional design, rules and
regulations, contracts, incentives and
penalties, etc.)
Establish process control SI
Transfer the plans to the tactical and SI
operational agents
Source: Author.
AA, assessment and alignment; CS, conguration supply; ES, ensure provision of infrastructures and
services; MP, monitoring performance; SD, strategic denition; SI, steering instruments.
The political priority given to the urban mobility aspects is materialized, for
example, through the creation of nancial means to develop implementation of PT
and infrastructures or through the inclusion of public service obligations in contracts
and its operationalization. For example, in Europe, the approach typically led to the
nancing of PT operators as a way to ensure mobility; in Brazil, in the past, the very
same concept was made operational by enforcing the nancing of employees through
their employers, recently complemented with the creation of a federal fund, largely
Table 5.4: Key process in the quality management of UMSs quality improvement.
Activities within the quality management Decision levels (who)

processes
Strategic Tactical Operational
Establish performance monitoring and SI

evaluation methods and tools
Establish an infrastructure to ensure SI
feedback information ows
Evaluate actual performance and MP
compare it with quality targets
Nurture feedback ows AA
Understand causes for deviations to the AA
performance plan
Act on performance deviations, (re)align SD (re) CS AA
the system
Justify the need to improve SI
Identify the policies and measures SI
required
Identify agents to trigger improvement SI
Establish the required infrastructure SI
(if any)
Establish controls on improvement SI
Identify the instruments that will foster SD (re) SI
agents actions and reactions
Dene sources for the nancial means SD (re)
that will support the system
improvement
Assess effective results and prospective MP
impacts
Nurture feedback ows AA
(Re)align the system, if needed SD (re) CS AA
Source: Author.
AA, assessment and alignment; CS, conguration supply; ES, ensure provision of infrastructures and
services; MP, monitoring performance; SD, strategic denition; SI, steering instruments.
based on the redirection of taxes accruing from fuel consumption (i.e., the CIDE
Contribuic- ao de Intervenc- ao no Dominio Economico); in New Zealand, the option
was to create a transport fund supported by the additional land value created by the
accessibility provided by the mobility system, and so on.
The legal and regulatory framework binding PT, road trafc, and infrastructure
management can also constitute a barrier for the implementation of an integrated
model with these characteristics. In particular, it is very common to nd road trafc
Figure 5.10: Transversal process deployment. Source: Author.
and nonmotorized (pedestrian and cycling) infrastructure under a different institu-

tional setting leading to complex decision processes for the achievement of a concerted
strategic denition of the system. This has been the case experienced in Brazil.
In Brazil, despite the difculties with the institutional design and legal frameworks,
a more conform solution was obtained due to the fact that the model was fully
implemented through all the institutional levels (federal, state, and municipal (and also
metropolitan)) enabling the required adjustments of all related regulatory frameworks
(e.g., land use and transport integration Estatuto da Cidade, regulation on urban
and metropolitan master plans, regulation on energy use in transport, regulation on
environmental protection, and scal regulation).
In Lisbon and Porto, the fact that the implementation was restricted to these
metropolitan areas and the resulting legal product was applied only to PT, limited
the potential to reach wider degrees of conformity, given the overlap with other
institutions such as DGTT (General Directorate for Land Transport), who has a
jurisdiction of national scope. Therefore, in Lisbon and Porto, the management
control processes and the information system was designed for the whole mobility
How to Set Up a Management Model for Urban Mobility Systems
Figure 5.11: Conceptual illustration of the proposed management model for UMS. Source: Author.
233
Table 5.5: Critical aspects of key processes (part one).
Process code Quality requirements Process designation Critical aspects

(Table 5.3) for each process
SD and Relevancy Strategic denition Need to involve

SD(re) Applicability stakeholders
External coherence Concertation of
decisions with other
sectors
SI Internal coherence, Denition of steering Need to have a

applicability function multi-instrument
Effectiveness approach, very
often with the
instruments of the
same package being
managed by
different
institutions
Good adjustment of
institutional design
Quality of decision-
making processes
Consistent
translation of
strategic priorities
into actions plans at
tactical and
operational levels
CS Effectiveness Conguration of Ensure intermodal

system supply integration within
public transport
and also between
public transport
and individual
modes (motorized
and nonmotorized)
Source: Author.
Table 5.6: Critical aspects of key processes (part two).
Process code Quality requirements Process designation Critical aspects

(Table 5.3) for each process
ES Effectiveness Ensure provision of Dene regulatory

Efciency system supply frameworks for
public transport
market access
Dene scope of
public transport
contracts (e.g.,
service, line, area,
and network)
Design contractual
incentives for public
transport services
and infrastructures
Dene industrial
performance levels
for public transport
services
Dene network
performance levels
for the all UMS
Dene commercial
performance for the
all UMS
MP Relevancy Monitoring Information system

Internal coherence, performance Data availability
Applicability
Effectiveness
Efciency
External coherence
AA External coherence Assessment and Scenarios of

(re)alignment evolution and
impact estimation
Source: Author.
system but with constraints related to a legislation of reduced scope that hinders
the action of the metropolitan authorities. The subsequent change in government
suspended all the implementation process, despite the fact that the process reached
the point of public discussion with operators and was approved by the two
institutions with regulatory powers, that is, the DGTT and the INTF (National
Institute for Railways), who have both subscribed the submission of the new law to
the president for further ratication.8
Another relevant aspect is the social and political sensitiveness to environmental
preservation and consequent priorities given to these issues, which has a strong
inuence in the conguration of the UMS, in particular, in the role attributed to
walking and cycling. From what we have observed, this concern tends to be higher in
the more developed areas of the world, with the Nordic European cities taking the
lead on the promotion of mobility solutions of low environmental impacts.
Finally, the conguration of the UMS is largely dependent on the following
factors, which are dened at local (municipal or conurbation) or metropolitan
level:
Degree of competitive pressure, which is intended in the supply of PT services,

network design, infrastructure, and service provision.
Financial availability to support modal diversity. We observe that while in Europe
there is a wide modal diversity in the conguration of the USM (associated to high
levels of subsidization), in other continents, in particular, South America, the more
reduced practice of state subsidization led to a limitation in modal diversity and to
the concentration on bus-based mobility solutions, even promoting innovative
solutions to explore the maximum potential of that mode (e.g., bus rapid transit
concept worldwide known through the Transmillenium case in Bogota and similar
solutions in Curitiba).
Availability of authorities to invest in integration (intramodal and intermodal) of
all mobility means, because it is proven that the main benet of wider integration is
achieved at the system level and not at the operators level.
Local barriers to the implementation of all dimensions of the integrated model and
capacity to design instruments to overcome these barriers.
Skills and competencies of responsible entities for management of complex
networks.
Available information on the demand markets and supply constraints, to enable
the management of the UMS in its integrated dimension.
In short, system management should ensure that the physical conguration of

the system corresponds to the best t between the urban characteristics and the
8. At the time of publishing this book (January 2011), these two bodies DGTT and INTF had been merged
into IMTT (Instituto da Mobilidade e Transporte Terrestre), standing for Institute for Mobility and Land
Transport.
service attributes of each mode (motorized and nonmotorized). Therefore, while

designing the physical system and considering all the options for the optimization
of the mobility chains, a concern on the social, economic, and functional impact of
each conguration should be taken into consideration and be object of quality
assessment.
These concerns were addressed in the implementation of the model in Brazil by
enforcing the transparent demonstration of four requirements: the short- and long-
term impacts produced in the envisaged prole of urban development, the cost-
benet relation of the several alternatives being considered, the opportunity cost of
any use of public money and the variations produced in the environmental impacts
when compared with the ex ante situation. Given the limitation imposed by the
characteristics of the Brazilian Federation, this enforcement was done by regulating
the access to federal funds and other national and international credits to the
conformity with the model, reected in the principles and norms introduced in the
new law, which is why concertation with nancial institutions and respective ministry
was also required in this process. Parallel to this, a strong an extensive capacitating
program (from the second half of 2005 until 2010) for managing authorities was
drafted and will be implemented in all cities together with the production of a manual
for integrated management of UMSs.
5.5. Assessing Quality in UMS
Besides the inner quality of each process (i.e., efciency, effectiveness, and
adaptability of the process), there are still some critical aspects related with the
quality requirements that each particular management process is expected to fulll
so that the whole system reaches the envisaged objectives, that is, critical aspects
related with the decoupling of quality along the different processes and the different
decision levels. These aspects are presented in Table 5.6.
Assessing quality on the whole UMS has to do not only with the absolute
metric resulting from the application of sets of indicators but also with the
variation that can be measured through comparison of indicators along time and
with the performance constrains caused by the contextual urban conditions where
the UMS is embedded. This means that the performance of UMS has to be
interpreted in the city context (or urban area) prole that is served by that
system.
Despite the universal character of the six quality perspectives dened in the previous
sections (i.e., relevance, effectiveness, efciency, applicability, internal coherence, and
external coherence), the corresponding operational denition of the indicators for
monitoring is extremely dependent on the effective objectives and conguration
established for the UMS and even on the urban context. In fact, the latter implies the
utilization of contextual indicators (i.e., ex ante the application of any new model,
policy, or measure) to ensure comparability of cities and even ne tune the
measurement used in the quality assessment.
These contextual indicators apply to the general characteristics of the urban area
(e.g., geographical, orographic, demographic, architectural, and cultural), which may
inuence the performance or even confound the assessment of the UMS performance,
given the presence of exogenous factors interfering in the interpretation of the
measures taken. Contextual evaluation enables a rst screening of the type of prole
in which the city or urban area can be included and the identication of indirect
factors that can condition the performance of the UMS. This evaluation should
consider three domains of variables physical, economic, and institutional
because very often the conicts of planning and policy prioritization have a central
responsibility in the underperformance of the UMSs.
Examples of those contextual indicators are density of population (in the business
center, and in the all city), area of the city (or urban area), area of the business center,
area of the historical center, number of households (in the center and in the all
city), number of cars, average income, distribution of population per age groups,
unemployment, and so on. Another example of the use of contextual indicators to
prevent misinterpretations caused by straightforward measurements is given by the
above-mentioned research project urban audit,9 where the following indicators
among others were used to qualify urban quality of life, illustrating the inuence that
the urban prole can have on the mobility choices:
Green space to which the public has access;

Percentage of population within 15 minutes walking distance of the urban areas;
Percentage of urban area unused in the main land uses;
Percentage of urban area subject to special physical planning/conservation
measures;
Total annual distance traveled by residents (by mode, purpose); and
Average car occupancy.
The case of Bangkok, which is reported by Daniere (1995, p. 37), provides also a
good illustration of conict within the soft side of the system. In 1991, in Bangkok,
there were 11 agencies involved in transport planning implementation, all under the
authority of either the Ministry of Transport and Communication or the Minister
of Transport, and four agencies involved in road constructions, ve involved in PT,
and three in trafc management. In such a case, a clear barrier exists, which is the
inability to concert political decision and implement a consistent integrated policy.
But even if the inuence of contextual factors is properly isolated, the systemic
and synergetic characteristics, and inherent complexity, of the urban environment
itself reveals the existence of some pitfalls that hinder the reliability of interpretations
based on urban indicators. On the one hand, these constraints justify the need for an
integrated approach, but on the other hand, additional caution is required to prevent
9. Results of the project are available at http://ec.europa.eu/.

the risk of abusive and misleading interpretation. We have identied the following
problems in (EC, 2002b, TRANSPLUS, D3, p. 84) research survey:
Most specic policy action (e.g., housing investment) affects not only a single area of
policy concern (e.g., volume of affordable dwellings) but a wider range of domains
(e.g., local employment and transport ows).
Consequently, change in one indicator may be the effect of several causes, hindering
our capacity to understand which policy action is responsible for that effect.
At the opposite of the previous, the difculties in establishing clear cause-effect
relationships between the planned policies or actions, results achieved, and impacts
produced are often due to the compound nature of the urban condition itself that
causes common variation of different indicators and raises the risk of multi-
collinearity.
Impacts are sometimes difcult to measure because, contrary to results, these are
measured outside the operational context and often not perceived in an obvious
way. Moreover, and as we have already noted, some impacts occur only after
considerable time lag making more difcult the clear understanding of the cause-
effect relationship.
In addition, there is also the difculty of aggregating all indicators at the same
level. That is, for example, nancial indicators are easy to aggregate whereas
physical ones are extremely difcult.
Following the rationale presented in Figure 4.2, we present in Tables 5.7 and 5.8
a set of indicators of perceived quality for the different assessment aspects
contemplated in the results and impacts of the management model proposed in
this work. For any specic implementation, each of these aspects and corresponding
indicators should be decoupled for the different services and infrastructures provided
by the system.
In all cases, the quality gaps identied in Figure 4.2 and further explained in
Table 4.8 have to be considered and aligned with local perception of quality standards
(i.e., meaning of qualication of good and bad varies considerably from one place
to the other, given the strong inuence of cultural and socioeconomic factors).
Moreover, the thresholds referred in the perceived quality indicators vary with the
size and functional conguration of the city (i.e., monocentric, polycentric, etc.).
5.6. Steering Instruments, Control, and Enforcement
Public policy instruments are basically the set of tools (techniques, norms, procedures,
etc.) used by governments and public authorities to exert power in their attempt to
produce any social change. For policy-makers and public administration, it is of
utmost importance to have a clear view on the possible forms these instruments can
take to choose the appropriate combination (i.e., packaging) in the planning of their
actions.
Table 5.7: Indicators for perceived quality (part one).
Assessment stage Assessment aspects Perceived quality indicator
System results Clients efciency in Generalized cost function

consumption Improvement of accessibility
at all levels (time, economies
of scope, etc.)
Infrastructure performance In road trafc

infrastructure: saturation
(weekday, weekends);
accidents per week;
percentage of vertical and
horizontal signaling with
good conditions of
maintenance and visibility;
noise; air quality
In public transport
infrastructures: comfort;
customer care; available
information; available
ticketing facilities; security
Percentage of walking and
cycling areas with good
conditions of pavement,
signaling, and security
Percentage of available total
length of linear
infrastructure for walking
and cycling
Percentage of parking
spaces with good conditions
of pavement, security,
signaling, and pricing
equipment
Percentage of public
transport interchanges with
parking facilities
Percentage of loading/
unloading spaces with good
conditions of pavement,
security signaling, and
pricing equipment

Operators performance Physical accessibility within

including self-service modes acceptable walking distance
(time) from home, work
places, and main trafc
generators
Functional accessibility
given by working periods
and frequency
Financial accessibility, that
price as an acceptable % of
average income, to avoid
exclusion
Cleanliness and comfort of
vehicles
Punctuality
Security (number of
occurrences per months
below and acceptable
threshold)
For self-services, number of
day per year with observed
malfunction in any
equipment used by nal
customers kept below an
established threshold
Source: Author.
As reported in Bemelmans-Videc and Vedung (1998, pp. 124126, in Bemelmans-

Videc and Vedung, 1998, p. 4), instruments can vary according to the degree of
intervention: from reserved, cautious, minimalist, and subsidiary intervention, or more
intrusive, active, and development forms of intervention; from repressive to preventive
(i.e., ex post corrective actions regarding degenerated behaviors) to preventive action,
implying the creation of conditions that will benet the development of what is thought
to be the preferred behavior.
The choice of instruments is clearly a decision-making process that is much
inuenced by the policy style, and this is the main reason why some instruments
are favored by a specic country or sector. A good example of this phenomenon in the
transport sector is the use of subsidies as an economic instrument, which is widely use
in Europe, with exception of UK, and scarcely used in United States or South America.
Table 5.8: Indicators for perceived quality (part two).
System results Performance of off-line agents Number of day per year

(cont) with observed malfunction
in any equipment used by
nal customers
UMS performance Industrial performance (for

operators, infrastructures,
and other supply agents),
concerns productive
efciency measured in terms
of transformation of input
into outputs for all services
and infrastructures provided
Network performance,
assessed from a specic
zone, means the number of
destinations (main urban
attractors) that can be
reached within a certain
amount of time, using all
available modes and means,
and including waiting and
transfer times
Commercial performance,
meaning clients satisfaction
reects in variations of
modal share measured
against the objectives set
These indicators are adapted
from the work done by
Viegas (2001, p. 7) for public
transport
Externalities Number of days per year in

which the maximum
threshold for the externality
factor was not observed
(e.g., air quality, noise,
space consumption in
traveling and parking, and
accidents)

System impacts Perceptions Changes in citizens opinions

and expectations
Behavioral adjustment Changes in market shares

for the different mobility
means and modes
System evolution Changes in UMS

performance
Source: Author.
In this work, we follow Richardson (1982, p. 13) who dened policy style as the
characteristics of a government approach to problem-solving (active or reactive) as
well as its relation to other actors engaged in policy-making and implementation.
Policy style is of course conditioned by the cognitive orientation of the decision-
makers, which in turn is inuenced by cultural aspects. As observed by Freeman
(1985, p. 474), there is the need to distinguish between political elite preferences
and outcomes, therefore recognizing the existence of an objective and a subjective
dimension in the process of instrument choice, because the educational and
occupational background of elites may shape their approach to public problems
and consequently inuence the style of national decision-making.
Consequently, depending on the type of political arrangements, and the predomi-
nant political style, existing in a country or a sector, several actors can be involved in
the formulation and choice of policy instruments. Policy instruments are concrete
and specied forms of intervention by public authorities; consequently, consultation
of stakeholders can start at the design stage, or, in alternative, they may be involved
only later at the implementation stage.
In the past decade, we have observed a greater involvement of stakeholders in the
design stage, enabling a better understanding of the problems and consequently an
improved acceptability of the solutions (Viegas & Macario, 2003b, p. 221).
Given the relation between policy style, instruments, and quality of governance, the
selection of instruments tends to be prompted by aspects close to the ones used in the
appraisal of governments (Van der Doelen, 1998, pp. 130131), such as the follows:
Effectiveness, which represents the capacity of the instrument to realize the

specied goals, considering also the side effects that might result from the use of
the instrument.
Efciency, which relates to the inputoutput/outcome ratio of the instrumentation
process.
Legality, which refers to the degree of correspondence (or tness) of administrative

action in the designing and implementation with the relevant formal rules, as well
as with the principles of sound administrative processes, which can raise concerns
of equity and motivation.
Democracy, which refers to the degree to which administrative action in design
and implementation is in accordance with the existing norms for the relationship
between citizens and governments.
In terms of taxonomy of instruments, we can nd two main approaches in the

literature: the continuum versus the resource approach and the maximalist versus
minimalist approach as observed in Howlett (1991, p. 3). In the rst approach, the
issue is focused on whether the instrument should be classied in the perspective of
basic choices that decision-makers can take (including the do nothing scenario)
or, alternatively, whether this classication should focus on the resources needed for
a decision already taken. In the second approach, the dichotomy lies in whether a
full list of possible policy instruments is provided or, alternatively, only some
fundamental types under which all instruments should be categorized.
The continuum approach (spectrum of choices) or, as Bemelmans-Videc and
Vedung (1998, p. 23) called, the basic tools of the trade of statecraft, is to our opinion
closer to reality as it is also explained by Anderson (1977, pp. 5677, our emphasis):
When we face a public problem, there are really only four sorts of
things that we can do about it y Which ones we will decide to employ
depends largely on how much freedom and how much compulsion we
think as appropriate in the particular situations.
1. Market mechanism. We can let the outcome depend on what

individuals decide to do, without any interference or direction from
government;
2. Structural options. We can create government programs y that
individuals are free to use or not as they see t;
3. Biased options. We can devise incentives and deterrents, so that
individuals will be guided, voluntarily, toward the desired ends of
public policy;
4. Regulation. We can directly control, setting up constraints and
imperatives for individual action, backed by the coercive powers of
government.
At one extreme of this typology, we have the case where the government has no
interference and leaves all decisions to the consumer (market mechanism). The next
case is when policy-makers specify which alternatives exist, but the nal outcome is
dependent on individual choices (structural options). The third case reects the
situation where the policy-makers devise incitements and impediments to inuence
individuals so that they follow the direction of the public policy objectives (biased
options). The fourth and nal case in the continuum is when individual freedom of
choice is delimited by statutes, rules, and regulations. This classication makes

evident that the degree of compulsion to be used in the steering and control function
is a major issue in public policy design, in particular, in the design of instruments and
the planning of their sequence of implementation.
It is worth highlighting that as Elmore (1987, p. 175) explains, policies are
typically composed of a variety of instruments, and frequently the logic by which
these elements are stuck together has more to do with coalition politics than with
their operating characteristics, or their basic understanding of expected effects.
Sometimes, however, policies form a more coherent package, combining what might
seem at rst like incompatible elements in a way that carries a distinctive message
and considerable impact. This means that any given policy may include multiple
instruments, as referred in Chapter 4, and, in fact, the multiobjective character of
UMSs justies the joint (sometimes simultaneous, sometimes not) use of several
instruments.
The continuum approach presented earlier also reveals that the effectiveness of a
policy is given by the degree to which the chosen policy instruments themselves
contribute to achieve policy goals. To this respect, packaging of policy instruments
can be formed in three different ways (Bemelmans-Videc and Vedung, 1998, pp. 257
268): vertical, horizontal, and chronological packaging.
In vertical packaging, one instrument is directed at the implementation of another,
assuming that governments use several layers of actors to reach their policy
objectives, in a cascade effect where higher level actors are expected to exert inuence
on lower level actors. In horizontal package, two or more policy instruments are
directed at the same target thus reinforcing each other. In this case, the risk of having
horizontal packages with conicting instruments is common.
Chronological packaging, in turn, implies a time order in the selection of diverse
instruments, the impact of which is aimed to be sequential and progressively more
intense. According to Paisleys (1981, pp. 1540) work on how to pursue feasible and
effective policies, a policy problem is usually addressed over time in three different
ways: it starts by provision of relevant information, that is, education; after it goes to
the application of selective incentives, that is, what he calls engineering; and nally, it
enters the stage of establishment of rules and regulations, that is, the enforcement
phase. This sequence is sometimes used to suggest increasing degrees of constraints,
although we could not nd evidence on the prevalence of this pattern in the
literature. However, the underlying logic of this type of packaging is that in solving
social problems, authorities apply instruments of increasing strength in successive
stages.
Complementary to this approach, we have found in more recent literature, a strong
tendency to emphasize the importance of combining organizational arrangements
with policy instruments, arguing that organization is a prerequisite for the provision
of regulatory, economic, and informational instruments (Arentsen, in Bemelmans-
Videc & Vedung, 1998, pp. 211231), and, of course, the way policy implementation is
organized inuences both impact and legitimacy of policy instruments (EC, 2001b,
PATS, D4, p. 80; Viegas & Macario, 2003a, pp. 178179). This is also reinforced by
Norton (1990, p. 58) who states that for an institutional environment to induce
credible commitment, it is indispensable to entail not only the instrument but also the
institutional framework to improve measurement and enforcement.
In our model, we also recognize the need to combine organizational and
managerial instruments with policy instruments. We consider management (or
steering) instruments any means through which it is possible to interfere in the
evolution of the system at any of its decision or planning levels, that is, strategic,
tactical, and operational as identied in Chapter 3. These instruments enable to
monitor and correct the evolution of the UMS, and, consequently, they go beyond
the policy instruments previously referred.
As we have said earlier, there is a wide diversity of possible instruments, and it is
the role of the entity managing the system to choose the ones most adequate
considering the four aspects referred earlier (i.e., effectiveness, efciency, legality, and
conformance with democratic regime) regarding the specic problem to be handled.
Therefore, following the analytical work on barriers and instruments that we have
developed in (EC, TIS.PT, 2002b, D6, p. 32), we believe the following types of
instruments are required to support system management:
Instruments of institutional character (e.g., institutional design such as the creation

of metropolitan authorities, observatory of urban mobility, committees for
stakeholders representation, and political fora);
Instruments of political character (e.g., integrated measures aiming to complement
the policy dened at strategic level and induce behavioral adjustment in the agents),
which as we have seen are the ones with a wider diversity. In the research project
EC, TRANSPLUS (2002a, D6.2, p. 15), and also proposed by Viegas (2003a, p. 3),
we have divided these political instruments into the three following categories:
Supply-side instruments, addressing the quantitative and qualitative character-
istics of supply;
Economic instruments, also referred as demand management instruments,
inducing the behavior of the agents through the price mechanisms; or
Regulatory instruments, addressing market rules.
Instruments of contractual character (e.g., incentives and penalties related with
performance, extension or abbreviation of the duration of contract or concession,
and issues related with formal interaction between agents);
Instruments of procedural character (e.g., tendering processes, technical licensing,
monitoring processes, surveys for clients satisfaction, and quality assurance
mechanisms); or
Instruments of persuasive (or informational) character (e.g., all means to inuence
people through transfer of knowledge by way of communication of the rationale
supporting the arguments and persuasion.
Instruments always have alternative applications, either with a corrective or with a

preventive character. In the rst application, they enable correction of undesired
deviations in the evolution of the system; in the second, they serve to facilitate
(or push) the evolution of the system toward a given objective or in line with a
predetermined strategy.
Tables 5.2 and 5.3 provide examples of instruments of political character that have
been identied in the survey done in the research project TRANSPLUS and also
reconrmed by the research project SPECTRUM. Their specic selection for a city
depends on the type of problem or aim the instrument is expected to answer as well as
of the adequacy of the organizational setting institutional framework to ensure
proper control of its implementation.
In fact, some authors (McDonnell & Grubb, 1991, p. 16) made evidence that the
choice of the implementing organisation does have considerable inuence on the
performance of the instrument through the following factors:
How the problem is understood, dened, and translated into the operational
program, which reects the institutional capacity and is also largely conditioned by
the cognitive aspects referred earlier, reason why it may generate inconsistency
problems;
How the client population is served;
The efciency of the implementation; and
The institutional exibility to handle the change circumstances in the adjustment
of the problem over time.
Tying the instrument to the contextual characteristics is also a key aspect we have
observed in the transferability of successful policies and measures between cities
within the CIVITAS program (Macario & Marques, 2004, p. 8). In addition, another
contextual factor is the potential mismatch between institutional personality and the
values implicit in the policy, that is, to be implemented. In fact, each organization has
its own order and logic, values, core beliefs, and institutional memory. It is part of
the quality of the policy-making process to select implementing agencies with key
values that do not conict with the implicit values of the policy at stake. This very
often leads to the question of weather to use an existing institution or to create a
new one.
Another important element is the chain of implementation and the level in that
chain where the instruments are aimed at. To this respect, we can observe a cascade
effect, with the government using instruments to control the central services, which
in turn devises other instruments to control the local services, which in turn use
other instruments to ensure compliance with the citizens. Moreover, the instruments
directed at the various levels are very often of a different kind (Tables 5.9 and 5.10).
The decision to create or not a dedicated institution is extremely risky from the
political point of view, and very often, it is taken without due consideration for
the following risk factors, some of which have also been observed by Rist (1998,
pp. 160161).
Creating a dedicated organization implies consideration of the following:
Need for available funding for startup;

Availability of competent person to organize and manage the new organization;
The possibility to hold the problem allowing sufcient time lag to have the new
organization fully operational;
248
Table 5.9: Examples of instruments of the UMS part one.

Supply element Use Political instruments
Supply side Regulatory Economic
Land use Browneld Location policy (ABC- Land taxation

development like) Value capture
Mixed-use Protection of sites Development in-kind
development from development requirements
Public transport- (green belt) Public land banking
oriented development Road corridors Developer fees
Pedestrian and cycling development control Development funding,
friendly site Transfer of disbursements
development development rights
Decentralization of Building regulation,
nonservice building permits
employment Density standards
Decentralization of Purchase, preemption
retail rights
Infrastructure Walking Creation of pedestrian Car-free zones

(transport areas
networks)
Cycling Expansion of existing
bike lanes
Individual car use Expansion of existing Restricted access at Road pricing

road network certain times On-street parking
Road maintenance pricing
and clearing priority
Trafc calming Restricted access to Off-street parking
facilities certain types of pricing
Access control devices vehicles
Designation of on- Speed limits
Parking time

street parking supply
Regulation of off- constraints
street parking supply Enforcement of
parking measures
Parking regulations in
building codes
Car pooling Dedicated HOV lanes Company mobility

plans
Taxi Dedicated lanes
Public buses Bus lanes Bus prioritization Public transport fare

Quality regulations level
Information provision
and marketing
Trams, light rail, and rail Expansion of existing Quality regulations Public transport fare
and metro lines network Information provision structure
and marketing Concessionary fares
Subsidies to operators
Infrastructure charges
Source: Adapted from TRANSPLUS guidelines (2003, p. 16).
249
250
Table 5.10: Examples of instruments of the UMS part two.

Supply element Use Political instruments
Supply side Regulatory Economic
Infrastructure Bus stations Expansion of park and Infrastructure charges

(transport Railway stations ride lots Space renting,
terminals) multifunctional use
Metro stations Park and ride facilities Exploitation of real estate
Bike and ride facilities value capture in
catchment areas
Freight terminals Building/expansion of
terminals for city freight
distribution
Vehicle eets Private vehicles Pollutant and noise Fuel taxes

emission standards Vehicle ownership taxes
Incentives for alternative
fuels vehicles
Variable vehicle-related
fees
Public buses Fuel taxes

Incentives for alternative
fuels vehicles
Information Private transport ITS driver information

technology
Public transport ITS driver information
ITS eet management
and control
ITS real-time passenger
information system

Land use Browneld development Location policy (ABC- Land taxation
Mixed-use development like) Value capture
Public transport-oriented Protection of sites from Development of in-kind
development development (green belt) requirements
Pedestrian and cycling Road corridors Public land banking
friendly site development development control Developer fees
Decentralization of Transfer of development Development funding,
nonservice employment rights disbursements
Decentralization of retail Building regulation,
building permits
Density standards
Purchase, preemption
rights
Source: Adapted from TRANSPLUS guidelines (2003, p. 16).
251
Given the fact that organizations do not operate in a vacuum context, it is also
required to ensure a sort of institutional tolerance to allow the establishment of
interactions between the new organization and the ones already existing that either
support or compete with the previous in any way. This factor is specially relevant
when partial or total transfer of competencies from one incumbent organization to
the new one exists (e.g., the transfer of competencies from DGTT to the new
metropolitan transport authorities); and
Sound justication for the political repercussions accruing from the nonselection
of existing organizations.
Alternatively, locating the new instrument in an existing organization implies

consideration of the following:
Acceptability of the stakeholders of the existing organization;

Capacity of the organization to expand, given the risk of dysfunctionality with the
changes in structure resulting from the accommodation of the new task;
Cost of reorganization;
Winners and losers of the process;
Effects on interaction with the contextual network of related organizations; and
Sound justication to select a given organization to host the new function.
Enforcement methods to ensure the full application of instruments are also

dependent on the institutional settings and the political style of the different
countries. Their performance is also related with the best design of the organization
in charge of the implementation of the instruments. Informal mechanisms exist, and
they should not be ignored; on the contrary, they can play a fundamental role.
However, once again, it is the institutional design that will determine the structure
underlying these mechanisms of enforcement, because these will basically occur at
the level of the different agents participating in the system. Here, both formal and
informal controls are equally important, and, once again, the management style is the
factor with strongest inuence in the way management control, communication, and
interaction with other agents occurs.
An example of particular relevance is the current awareness of the need to have
organizations and instruments enable to impose car drivers the respect for circulation
and parking rules. It is often the case that a good part of this mission is usually
allocated to the security forces at least in the European countries, as we have surveyor
in a study on road trafc regulations developed for the European Commission.
However, we have also identied that when those security forces are entailed in an
organization, which is part of the central government, there are often difculties in
their alignment with the local agencies.
Following Anthony and Govindarajans (1998, pp. 9598) management princi-
ples, control and enforcement in the UMS can be classied into two main parts:
The process by which system managers inuence other agents to implement strate-
gies leading to the accomplishment of system objectives, where the instruments
discussed earlier have a key role. This process involves a variety of activities,
including planning system evolution; promotion of concerted action between
agents; communication with agents; evaluation of individual agents performance
and overall system progress; deciding what action and which instrument is to be
taken, given the feedback information from the eld, to correct any eventual
deviation in the evolutionary path of the system; and inuence agents and citizens
to change behavior toward an intelligent (information-based) organization of
mobility chains in favor of the city sustainability. In this process, it is the
information system that provides the detector role to activate the formulation of
corrective or preventive action.
The rules, that is, all types of formal instructions and control mechanisms. These
include standing instructions, recommended practices, quality standards, and so
on. Rules have a variable coercive character; they may be prohibitions or only
positive requirements to undertake a certain action. Some examples are as follows:
physical controls, for example, impediments for trafc circulation or limitations
to certain level of emissions, and so on;
guidelines, that is, agents are permitted and indeed expected to act in such a way
in specic circumstances or at her own judgment whenever suitable for the
system and agents interests. As an example, we have manuals for tenders and
contracts, where considerable judgment is needed to decide what should stay
or not in a contract, which parts should be guidance instead of xed rules, and
so on;
system safeguards, which should be built into the information processing to
prevent accuracy of information ows. These include monitoring and cross
checks; and
activity control, which control agents performance.
5.7. Allocation of Functional Responsibilities and Organizational

Requirements
The institutional allocation of responsibilities within each decision level depends on

the political and administrative organization of the city (or metropolitan area or
relevant territorial space) and consequently can never be the object of a model
meant to be of general application. Indeed, we have observed a bit all over the
world an intense change of institutional roles and responsibilities within the UMSs
in the past two decades, which has had a strong impact on the structure of these
systems.
In some places, functions that were previously performed by authorities have
moved to the hands of operators (e.g., farebox revenue collection), and in other
places, some planning skills that were retained by public operators, under monopoly
concessions, have been shifted into the hands of organizing authorities (e.g., network
planning or market studies), and so forth.
These movements have been largely accompanied by organizational and

regulatory reforms,10 and the asymmetry of information together with the dynamics
of principalagent relationships required an adjustment of institutional structures to
enable governmental agencies to shift their roles, from direct control of state-owned
service providers to indirect guidance through inuence of market agents behavior.
In fact, we observed that the cross-effects between horizontal and vertical speciali-
zation in an UMS leads to a complex network of institutions with differing degrees
and forms of interaction, which should be linked through a chain of consistent quality
performance objectives; a set of regulations framing agents behavior; and a control
mechanism based on feedback ows with information regarding performance
monitoring and assuring accountability of all institutions involved.
Given this tight net of interactions, a major concern in institutional design should
be the good t between purpose and action. The social structure of the system is
given by the regular pattern of relations between concrete actors and positions or
social roles within the network of institutions. This means that each actor, by the
simple fact of occupying a position in such a network, will have relational links to
other actors who also have structural positions in the same or related network.
System complexity grows whenever a new set of ties to preexisting positions is
created. An example that illustrates this process of growing complexity was observed
in the case of the creation of rolling stock leasing companies, where a newly born
agent created new patterns of relationship with existing agents, eventually changing
some of the previous principalagent arrangements and creating new potential to
develop barriers to change, because one of the attributes to determine institutional
network complexity is the number of interactions observed (Macario, 2003, in
Hensher, 2005, pp. 543558)
Despite the dependency on local constraints, it is possible to formulate some
guidelines that should be considered as good practices helping to avoid barriers to
system evolution and underperformance factors found in some of the cities observed.
These are as follows:
Ensure clear distinction between the three levels of planning and control (strategic,
tactical, and operational), or decision levels, with different organizational
requirements and functional roles (Anthony, 1989, p. 212; EC, TIS.PT, 1997b,
p. 26) and a clear allocation of these roles to different institutions, whenever
possible.
10. For example, France 1982, Brussels 1990, Sweden 1989, France 1993, UK 1985, Danmark 1987, Spain
1985, all reported in EC, TIS.PT (1997b, pp. 9093) and also in Belo Horizonte (Minas Gerais, Brasil) in
1997, reported by Gomide (2003, in Hensher, 2005, pp. 765776), and Australia and New Zealand, 1989
and 1990, respectively, as reported by Iallis (2003, in Hensher 2005, pp. 241253), and South Africa in
1996, reported by Walters (2003, in Hensher 2005, pp. 271289), and USA, San Diego in 1979, reported by
Cox (1995, http://www.publicpurpose.com/utx-us.htm) and USA, San Francisco in 1970 reported by
Chislom (1989, pp. 217223).
Within these different decision levels, activities and processes should have their
responsibility allocated to institutions with clear functions and effective power
in those levels. Otherwise, it should be ensured that at least the institution
accumulating roles has a design that makes different persons decide in different
areas.
Management of UMSs requires the following characteristics, which correspond to
the characteristics of good governance in a multistakeholder environment:
Centrality on citizens requirements and generators of those requirements, which
implies the existence of regular market surveys to enable their identication and
observation.
Stakeholders participation and responsiveness to their requirements, which
requires the adoption of participatory processes beyond the market surveys.
Effectiveness and efciency in all functions and processes, which implies the
existence of management control function at system level.
Transparency, that is, procedures and methods of decision should be open and
well understood by stakeholders. As expressed in Chapter 4, and following the
principles of good governance, transparency lies largely in the ability to provide
access to institutions, processes, and information, for those who are interested
in it.
Accountability of decision-makers and decision processes, implying an adequate
implementation of verication and reporting on quality performance to stake-
holders and to society in general.
Decision-making guided by impartiality toward markets, which implies a
permanent function of market and industry monitoring enabling to understand
the strategic behavior of agents and anticipate any perverse moves through
preventive action. This is one of the most important elements to provide system
management with effective intervention capacity.
With these requirements, it is recommended to have the same agency managing

the whole mobility system (all modes and means) at the tactical level and subjecting
the strategic requirements to political decision, that is, to entities that have the
political representation of citizens.
To achieve this, we consider the best option to be a complete institutional
separation between strategic and tactical missions, as it ensures a better functional
specication, and therefore, it is likely to produce a more effective and efcient
decision process, although we cannot provide relevant scientic evidence of this
advantage. An alternative is to encompass strategic and tactic functions in the same
entity but having a clear specialization of functions within its internal structure.
This single institution design was the option in the case of the Lisbon and Porto
metropolitan authorities, where the agency was designed having an executive body
with responsibility for the tactical function and a council, with the political
representation of the central government and all the municipalities present in the
metropolitan area. However, there was no full coverage of the UMS scope because
road trafc and infrastructures were left out of the agencys competencies.
5.8. Where and How to Start
From a theoretical point of view, it is always possible to conceive the implementation

of this model in a fresh start situation. However, this is certainly an articial problem
with the exception of very few new cities. Everywhere, we are confronted with
existing UMSs that are fractioned in independent and often divergent management
agencies. In such cases, the system as such is not assumed, even when its existence is
recognized. Therefore, by formulating a general sequence of concerns that should be
addressed to implement this integrated model, we ignore for a moment the transition
costs that certainly exist when moving from the existing situation to the one that is
envisaged by the model.
In almost every city, we will be faced with the double mission of solving the
current mobility problems (i.e., the corrective mission) while adjusting the existing
institutional and organizational setting (i.e., the preventive mission), as well as the
management control processes, in such a way that the system, as such, starts to exist
from the managerial and functional point of view. This means that there is no other
alternative to xing the system while running it and bringing its structure as close as
possible to the one recommended by the model. In a real-world situation, transition
costs should be considered at each step of the sequential change process, and success
of implementation has to be assessed more by the importance of the incremental
change than by the absolute results in each moment.
To assess the barriers to the implementation of the model, the following sequential
steps should be considered:
Characterize the city or urban area, allowing a rst screening of the range of
existing problems. Three major areas of concern should be addressed: physical
description, observing degree of integration between modes, economic character-
istics, and institutional characteristics.
Assess the intersectorial interaction at policy level, where the strategic denition of
the system starts and identify the barriers to the process of political co-decision, in
particular, whether reallocation of powers is needed. The main type of barriers
faced in most cities are as follows:
Information barriers/informal politics, meaning the lack of knowledge on how
the UMS can be organized, in the physical, institutional, functional, and
political aspects.
Financial barriers, meaning the lack of nancial resources to proceed with
change, which may also be materialized in the lack of power to mobilize the
necessary nancial instruments.
Institutional barriers, meaning lack of coherence between aims and aspirations
of the different government levels.
Understand the operational meaning and implication of adopting the principles
envisaged by the UMS model herewith presented and assess the feasibility of its
implementation. This assessment has to be done in three different dimensions:
At policy (aggregate economic) level addressing the economic and social
effects concerning economic efciency, in consumption and production, cost
coverage of public expenditure, distributional effects, of having the integrated

management of the UMS, which inevitably implies some risk transfer to the
entity managing the system.
At stakeholders level understand who will be the winners and losers of the
change process within the agents directly or indirectly affected by the change.
At practical feasibility level understand feasibility of implementation in the
following perspectives, and which instruments might contribute to smoothen
this process: legal, regulatory, technical, nancial, and managerial issues.
Design the adjustments needed for the implementation of the model within the
selected context, identifying which variations to the model are desirable and which
instruments are needed to achieve a good t to the expected purposes.
Start the process of gaining acceptance of stakeholders for the change process that
leads to the implementation of the model at the problem awareness stage and build
solidarity over the agreed solutions.
Validate and congure the key processes and allocate these to the entities
responsible for system management.
Dene the management entity, keeping in mind that the whole UMS management
functions are distributed along three decision levels: strategic, tactical, and
operational.
Chose the model implementers and dene consistent packages of instruments to
initiate and maintain system management.
Deploy instruments along the different implementers. Depending on the different
regulatory and organizational frameworks, these can be governmental institutions
or public and private agents.
Set the continuous monitoring and information system.
Organize the readjustment processes and activate it whenever monitoring
information indicates the system evolution is deviating from the intended track.
Keep a very close observation of the strategic behavior of the different actors.
Chapter 6
The Need for a Strategic Approach

to Urban Development
Along this work, we have looked for satisfactory justications of why quality
improvements done at the company and service levels do not ensure higher quality of
the whole urban mobility system. This objective was pursued by decoupling and
observing the interactions between the different elements of the system and also
between these and the surrounding context where urban mobility is embedded. On
the basis of this analysis, we concluded that the management of urban mobility
systems requires a holistic approach encompassing all mobility elements if we aim to
achieve an overall performance of high quality.
Four major aspects have been central to our analysis and form the main pillars of
this work:
the recognition that urban mobility is a system nested within a hierarchy of systems;
the fact that within our system, a high diversity of agents exist and interact, with
considerable autonomy of action calls for the need of a clear and sound institutional
design to clarify decision levels, roles, and functions, in particular, an indispensable
steering and function at system level;
the fact that this diversity together with functional interdependencies (e.g., between
land use and transport, and between individual and collective transport) brings
high complexity to the system, and consequently, its qualied sustainability can
only be achieved by providing the system with structural consistency (i.e., vertical,
horizontal and cross-effects accruing from the previous) as a precondition for its
success.
the recognition that urban mobility systems behave as living systems and as such
they have universal properties (i.e., robustness, efciency, and dynamic diversity)
that have a specic materialization in each urban area providing it with a satisfactory
system conguration.
6.1. The Vicious Circle of Public Transport Decline
In all cities observed, both in developed and in developing countries, we have

identied that public transport is an essential component of the urban mobility
system. Irrespective of having or not a tradition toward subsidization, in all cases, we
conrmed that public transport was either unable to fully cover its cost (e.g., Europe)
or, although covering it (e.g., Brazil), was unable to retain passengers, given its poor
quality when compared to private car, and consequently, market shares present a
steady decline along the past decades. It is worth noting that very often, the poor
quality is more a consequence of the soft elements of the service, as referred in
Chapter 4 (i.e., scheduling, inadequacy of route in relation to the mobility aspirations
of citizens, and poor integration) than of the hard elements (i.e., vehicles, etc.).
The most direct consequence of this wide spread problem is that the poor quality of
public transport contributes to increase market share of private motorized trafc,
which in turn leads to increased congestion, which substantially hinders any
improvement made to surface public transport (e.g., new eets and bus priority
corridors) due a major resource restriction the urban space. That is, there is a positive
feedback loop between the poor bad quality of public transport and road congestion.
Parallel to this, there is a raising awareness on the role of public transport to support
social cohesion, and in some countries, this is even giving place to social disturbances, as
it is the case in Brazil with the movement of the citizens excluded from motorized
mobility (started in 2004).
The political pressure resulting from this social character of urban mobility led
governments to favor general application of low fares, through price cap mechanisms
associated with nancing of production costs. In developed countries, this biased
interpretation resulted in the application of a at subsidy to operators irrespective of
citizens socioeconomic needs, because it is applied on supply side instead of being
demand oriented and consequently failed to provide a satisfactory answer to the
segments of population to whom the subsidies were originally addressed. In developing
countries, where patronage levels are much higher and there is less tradition on the
practice of subsidies to production, similar concerns led to nancial bottlenecks due to
the fact that prices are limited to prevent social exclusion and operators tend to recover
their margins by reducing production costs to the minimum, which are reected in
declining quality and incapacity to develop infrastructures and service that provide
adequate answer to citizens needs.
In general, the nancial problem has become more acute as other sectors (e.g.,
education, social security in old age, and health) gained higher political importance
by the amount of population affected, as opposed to the declining of public transport
market share, and started to compete for public funds with clear advantage and
stronger public support. In brief, this cause-effect phenomenon (i.e., declining market
share and declining political importance) has reduced the scope of urban mobility
issues by an excessive and almost exclusive focus on the dichotomy public transport
versus private transport.
This limited view pushed public transport to a value reduction closed loop
characterized by the following causal sequence, already identied in Europe in the
1970s: public transport is unable to deliver sufcient quality when compared to private
transport; loses market share and gets into insufcient cost coverage from farebox; loss
of market share produces the political downgrade of the sector; availability of public
funds to compensate lower farebox revenue is reduced; quality degenerates due to lack
of nancial capacity; and, again, public transport is further unable to deliver sufcient
quality when compared to private transport.
The Need for a Strategic Approach to Urban Development 261
6.2. Widening the Circle
In this loop, the interactions with the urban system and other mobility means and
modes are ignored, and public transport is addressed as an autonomous sector with a
self-standing purpose. Evidence from the real world points at the opposite direction,
that is, public transport is only one component of a wider system urban mobility,
which in turn is one part of the even wider urban system. Therefore, the description
of the close loop presented in the previous section, which represents a very common
rationale within the sector, reects a wrong formulation of the problem (although
indeed a common occurrence).
In fact, everywhere, there is evidence that in the large majority of cities, the
diversity of origins and destinations, associated with a wide variation of geographical
and functional densities, constraints public transports ability to give a universal
satisfactory answer based only on its own direct services between the mobility poles
of its clients. That is, we observe that urban mobility is the aggregate result of the
choices made by individuals and of the logistic organization of those choices into a
mobility chain.
Integration of services arises as the instrument to compensate such public
transport limitations, enabling the user to obtain economies of scope, through its
three main perceived dimensions: physical, logical (information-related), and tariff.
However, to ensure the effectiveness of these dimensions, another one much wider,
although invisible, is indispensable organizational integration. It is through this
background transversal dimension that the urban mobility system achieves what we
have called vertical consistency, composed of coherence, efciency, and account-
ability, as formulated in Chapter 2.
Coherence is given by a hierarchical design of networks and corresponding
infrastructure for transfer of ows resulting in the levels of accessibility and uidity
dened at the strategic level. Efciency is given by respecting modal adequacy at the
different levels of the organized network, in face of the specic demand proles.
Accountability is achieved by way of an appropriate monitoring system that will
provide feedback and wide diffusion on how well each agent is complying with the
objectives and rules that have been dened contractually (public transport operators)
or by law and regulation (private car drivers). It is through the constant assessment
of this vertical consistency that the steering function denes the preventive and
corrective actions that are required to keep the system within the desired track.
But we cannot forget that the one of the key factors for the complexity of the
mobility chain is the geographical location of activities that represent the desired
origin destination pairs and the justication for displacement. This means land use
is the departure point, and consistency between land use, transport and environ-
mental policies, and programs and plans is the critical rationale for the organization
of the urban mobility system, what we have called horizontal consistency of the
system, that is, equinality of action at policy level.
It is this three-tier aggregate driver that denes at the outset the level and diversity
of supply (infrastructure and services for motorized and nonmotorized modes and
means) on the basis of which mobility chains are dened, and it also through this rst
consistency test that we establish the rst distinction between policy-making and
execution as main guidance for a good institutional design, with a clear separation
between strategic (political and strategic denition), tactical (planning and
organization), and operational (service running) functions and responsibilities within
the system.
The cross-effects between vertical and horizontal specialization are also a concern
of the structural consistency of the system. The resulting effect is a network of
institutions (i.e., authorities, operators, and third parties) linked by varying degrees
and forms of interaction. Contractual arrangements and quality performance plans
are instrumental in the relation between agents, to ensure their behavior develops in
line with the strategic denition of the system, and also to maintain the consistency of
the whole network.
Consistency is thus an indispensable attribute to achieve an overall system
performance of high quality, but it implies institutional tness of purpose and action
to ensure that each agent performs its allocated role at her best potential. This is the
practical effect of a sound institutional design (and good management practice at
each institution) that should also be subject to regular assessment and adjustment.
6.3. The Time Dimension

We are aware that the change of paradigm in the management of urban mobility
system that we propose in this work requires a long-term process, which is why
structural consistency is so important to support such a wide diversity of agents
interacting with considerable degrees of freedom.
The period that denes long term in a given system is limited by the following
variables as a consequence of the intrinsic characteristics of urban mobility: electoral
cycle for the political decision-makers on urban mobility, which usually corresponds
to the political commitment with populations (typically 4 years); period correspond-
ing to the succession of active generations, with deep impact on social and economic
dynamics, upstream of the change in mobility patterns (typically 15 years); average
time for absorption of innovation by the society, which is related with the propensity
to adopt new forms of technology (not only in mobility) and very much dependent
on the characteristics of the local society (Utterback, 1994, pp. 189197).
Therefore, in practical terms, long-term quality planning should be dened for a
temporal horizon bounded by the duration of the electoral cycle and the period
between two subsequent active generations in society, that is, between a minimum of
4 years and a maximum of 15 years, the nal term depending on the propensity to
innovation of the society served by that system.
Qualied sustainability is achieved through the system capacity to dene and
maintain a certain level of quality, preserving its consistency in all dimensions. This
quality level is strategically dened and corresponds to the political decision that
responds to the question how much (and what kind of) quality do we want to
provide in our mobility system, that is, followed by the tactical question how do
we achieve it and through which means, and, nally, by the operational monitoring
that conrms fulllment of the objective or identies deviations requiring corrective
action to achieve stakeholders satisfaction, that is, each agent must know the answer
to two questions: how should I best adapt to the rules dened and what is my
responsibility and contribution to the overall system performance, the response to
the latter being more important for corporate entities than for individuals, as they are
subject to formal evaluation of their performance.
Finally, the universal properties of the system are reected in its capacity to
maintain qualied sustainability despite the growth of the urban area served (i.e.,
robustness), providing whenever needed additional or modied services (i.e., dynamic
diversity), either within the main trunk services or by creating additional services
through improvement in integration or simply by adding new feeders (i.e., efciency).
System properties are largely reected in this network articulation and the permanent
capacity to match demand and supply, adjusting all system components and, in
particular, its three main networks individual motorized transport, public transport,
and individual nonmotorized transport.
6.4. Institutional Design

To benet from the complementarity of all modes and provide citizens with
alternatives that match the different market proles, an integrated management (and
organization) of the whole system is required so that consistent signals can be
communicated to the population, such as the adequate use of relative prices of
private and public transport.
However, from the cities observed, we also conclude that some key requirements
exist regarding the institutional design of the urban mobility system, which reects
also in the regulatory framework of public transport, if we want to benet from the
advantages of such an integrated system. We elaborate on these requirements in the
following.
Strategic denition of the system is a function of a political nature; therefore, it
should be done by an entity with political representation of the citizens served by the
mobility system at stake, who should also have the capacity to allocate nancial
means to support the system, either by direct or by indirect sources of nance, as we
have seen in Chapter 3. This function can be done either by a single entity (e.g.,
political body of a metropolitan entity) or by a partnership between several entities
(e.g., consortia of municipalities) through concertation of interests and objectives, in
particular, ensuring political conditions for horizontal consistency between trans-
port, land use, and environment.
Mobility planning must be executed preferably by a single entity, under the
political surveillance of the strategic entity. That is, a managing (organizing)
authority, with technical competence and sufcient nancial means to undertake all
tactical functions, such as the denition and implementation of steering instruments,
the deployment of quality plans and standards, the design of the conguration of
basic supply (networks) based on those plans and standards, the contractual
arrangements for the operation of services and infrastructures of all means and
modes, the performance monitoring and consequent adjustments needed, and the
criteria to accept additional services proposed by the operators or requested by the
citizens.
This managing authority can be selected on competitive basis, because it has a
technical character, but we can also admit the existence of partnerships in the
constitution of this body, but the executive power should be clearly allocated to a
single body to ensure adequate levels of accountability.
Theoretically, we could propose also, as an optional alternative, building this
entity out of concerted decision mechanisms, to avoid concentration of the planning
function; however, there is not enough knowledge on effective instruments to ensure
good performance, system adjustment, and due accountability with such an
institutional design, which could offer a good guarantee of overcoming barriers
raised by problems of superposition of powers between the political administrative
entities involved.
Service and infrastructure operation should be contracted out, under a regime of
controlled competition, with incentives for development of additional and innovative
services as well as incentives to promote patronage of public transport. From all the
possible regulatory options for the organization of market access, up to this moment,
this is the one that proved to be more compatible with the strategic and tactical
planning requirements previously presented to ensure stability of supply and social
cohesion within the system. The alternative option of concentrating the provision of
all services into a single entity is possible, but to ensure levels of efciency similar to
the ones that can be provided by agents under direct competition, a very effective
system of indirect competitive pressure is indispensable, with tools such as
continuous benchmarking associated to the threat of contract suspension and
consequent outsource whenever performance falls below a prespecied level.
Network integration together with efciency aims are very likely the constraint
element that forces the selection of controlled competition as the best suited to
manage such a complex system as it facilitates the implementation of a network
planning approach allocating each mode and service to its best vocations. Indeed,
there is clear evidence that no signicant network economies result for the operators
of urban environment, as a consequence of integration in wider network integration
if a good part of the network lines are under the control of other operators.
Consequently, no spontaneous movements toward integration can be expected if left
on volunteer coalition basis, because in this conguration, the main benets exist at
system level and not at agent levels.
Each city (or urban area) has a suitable conguration for the development of its
urban mobility system, which largely depends on the geographical and functional
prole of the city. Besides, the urban mobility system should be subordinated to
the objectives and priorities of urban development, which is the superior system in
which mobility is nested. Consequently, no conguration for urban mobility systems
can be recommended as universal solution. Instead, we consider that this work
provides a robust structure, processes, and guiding principles for the adaptation and
implementation of the proposed management model to any city or urban area.

In fact, this is what we have done and herewith describe in the following.
6.5. Applicability
The applicability of this model was validated with the cities mentioned in Table 1.4,
as referred in Chapter 1, from which sufcient information (Annex 3) was available
to undertake a theoretical falsiability test. That is, we have simply conrmed
through the available information whether there were any insurmountable barriers to
the implementation of our proposed model.
In fact, while developing this work, many other cities contributed to provide
additional information on specic details resulting from their common experiences
and selected good and poor practices. These details enabled a deeper understanding
of the relation and constraints imposed by the systemic interaction between the
urban mobility phenomenon, the urban context, and the interests of the different
productive agents and other stakeholders that directly or indirectly contribute to the
functioning of the urban mobility systems.
The set of research projects done at European level, and referred in Chapter 1,
also allowed the analysis of transferability of experiences in several complementary
domains for the cities observed (Annex 2), such as interaction between transport and
land use; regulatory and organizational options for the urban mobility systems;
contractual design and quality performance; costs evaluation; pricing and nancing
options; and change processes in the organization of urban mobility systems.
This diversied research work offered us incremental knowledge as we went from
one project to the next, and while doing it becoming acquainted with one city after
another. This incremental process enabled the progressive completion of the model
with each round of new cites (or new aspects in the same cities) simultaneously
providing incremental knowledge to feed the next step of development of the model
and validation of the construct built on basis of the information acquired in the
previous steps.
In addition to this theoretical process, three practical applications of this model
(during the period 20042005), with very different scopes of implementation, provided
the practical supporting evidence on the applicability of the proposed model. The two
rst cases were the metropolitan areas of Lisbon and Porto, where the application was
limited to contractualization and nancing of public transport, and the third case was
Brazil where a full implementation of the proposed model in the 27 states was
envisaged. In Annex 4, we provide the project of Law that materialized the application
of the model for the Brazilian cases. The project law for the Portuguese cases was
produced and delivered. The change of government suspended this process before
formal approval of the law, and, in November 2005, the position of the subsequent
governments regarding the management model for the metropolitan authorities of
transport is not yet clearly known. This text is not presented in an annex because it is
still under client privilege.
A previous application case existed by the end of 2001, using a rst formulation of
this model (Macario, 2001a, 2001b, pp. 112) for application in the Great Area of
Mendonza in Argentina, but here, the political instability stopped the application
process at a very early stage, when only a presentation of the model had been done by
the author to the parliament. This resulted in the political approval for further
development and adjustment to the local reality; however, this has never occurred
due to the political and social crisis in that country.
In Lisbon and Porto, the geographical scope of implementation was the respective
metropolitan areas, which have been dened in the 1990s. The metropolitan area of
Lisbon includes 18 municipalities with a total population of 2,661,850 inhabitants (as
per 2001 Census) and a total geographical area of 2866 km2. The metropolitan area
of Porto currently includes 9 municipalities (additional enlargement with 5 more
municipalities is foreseen) with a population of 1,260,680 inhabitants (as per 2001
census) and a total geographical area of 8148 km2.
The application in Lisbon and Porto entailed the design of the management model
for the metropolitan transport authorities (created in 2003), the regulatory
framework for public transport, the denition of quality criteria for design of public
transport networks, the design of contracts between public authorities and operators,
the identication of nancial solutions to undertake change, the structure for
harmonization of operators cost accounting, and the design of the information
system to support the management control function for the mobility system.
The design and implementation process was conceived to have a participatory
process with discussion of the management model with the main stakeholders, that is,
several agencies within the Ministry of Transport, with transport regulators, with
public transport operators, with municipalities, with unions, and with associations of
consumers. The model and the resulting legislation was presented and discussed with
the two rst groups for Lisbon and Porto and with public transport operators only in
Lisbon. In all these groups, the model was well accepted and found adequate for the
two realities. This process started in May 2004 and nished in December 2004.
The participatory discussion was stopped due to the fall of the government and
consequent loss of legitimacy of the commissions for implementation of the transport
metropolitan authorities to pursue the reform.
Some major limitations were found in the experience with these two Portuguese
cases. These were as follows:
Lack of a clear nancial framework to support the activity of future metropolitan

authorities and to enable the transformation of old public transport concessions
into new contracts.
The institutional design of the transport authorities limited their intervention to
public transport instead of bridging the whole mobility system.
In Brazil, the reform (known as the Statute of Urban Mobility following the
previous Statute of Cities) had a much wider and deeper scope of implementation. The
objective set by the federal government, through the Ministry of Cities (Secretary of State
of Urban Mobility and Transport SeMob), was to make the harmonized application
of the complete model, and associated concepts, to all the cities, conurbations, and
metropolitan regions with more than 100,000 inhabitants in the 27 Brazilian states. This
represented at the time 379 municipalities plus 62 urban centers with a total population of
87,743,495 (as per 1998 census), representing 55.85% of the total population of Brazil.
The reform process involving the implementation of the model was planned to be
implemented until the end of 2005 and was designed with four main pillars: the
production and approval of a law, formally the regulatory and organizational
framework for urban mobility (Annex 4); identication of barriers and design of
instruments to overcome these implementation obstacles; training program for
municipalities and metropolitan regions to facilitate implementation of the model;
and development of a manual for integrated management of urban mobility systems.
For the adjustment of the model to the Brazilian reality, the author undertook
interviews with authorities and operators from 22 cities, made 21 participatory
discussion sessions with stakeholders, and 17 interviews with political key informants
and with the main operators (eets above 4000 vehicles) serving more than one city and
with some small operators (eets around 200 vehicles). In addition, a questionnaire
was launched in 95 cities, gathering the most relevant parts of the questionnaires
launched in Europe in the already mentioned research projects ISOTOPE,
QUATTRO, TRANSPLUS, and MARETOPE.
The process was accompanied by extensive consultation with stakeholders and
realization of thematic seminars on the issues considered most sensitive, such as
regulatory and organizational options, nancing urban mobility system, and con-
tracting and tendering for services and infrastructures.
In the consultation phase (AprilJune 2005), the proposed model was discussed in
the referred stakeholders sessions and in general obtained the conrmation of its
adequacy for implementation in the different Brazilian cities. The following categories
of stakeholder organizations were consulted: transport associations (authorities and
operators); regulatory agency: parliament: forum of transport secretaries of all states;
metropolitan regions and respective transport secretaries; associations of municipa-
lities; organized society through the council of cities; National Confederation of
Transport; associations of producers of the transport industry; most relevant federal
universities with transport studies (Brasilia, Recife, Natal, Rio de Janeiro, and
S. Paulo); and transport research associations.
In addition, discussions were held with national and international development
banks for the denition of nancing programs, managed by the union as a stimulus
for the implementation of the integrated (meaning all modes and means) manage-
ment of urban mobility systems. From the central budget, a fund (FUNAM) was
created using 11% of the fuel taxes collected each year. The nancial source is known
as CIDE, meaning contribution to economic development, and already exists with
the specic purpose of supporting the development of transport.
Internal to the federal government, the model was discussed in Casa Civil of the
president (who approved the submission of the project Law to the Congress); the
Ministry of Mines and Energy; Ministry of Environment; Ministry of Finance; and
the three other secretaries of the Ministry of Cities. It is worth highlighting that the
federal government does not have any authority to force the implementation of the
model by the municipalities, who are the entities responsible for urban mobility. This
means that the model is being implemented nationwide based on inducement.
The main difculties found in the adjustment of the model were due to the institutional
conguration of the Brazilian Federation, where we have found some overlap of powers,
namely, between metropolitan regions and municipalities. A particular difculty is raised
by the fact that the last revision of the constitution gives to the municipalities the statute
of federal entities, ignoring the hierarchy between the four levels of government (union,
states, regions, and municipalities). This creates a number of inconsistencies with
previous laws made in the perspective of hierarchical levels of government, but so far,
there is no consensual solution on how to solve the problem.
These difculties with institutional conguration were the main reason to design
the implementation with this strong component of training and dissemination of
information (in particular, good practices). In Brazil, there is a strong tradition
of adopting change processes through training and education processes, sometimes even
ignoring the need for legislation. A good example is the current calculation of operational
costs of urban transport, used for setting public transport fares, and also for payments to
the operators in contracted services, which is adopted as if it was a legislation, when in
fact it was implemented only through training and has never been legislated.
At the moment of publishing this book, the project law is already in its public stage
and being discussed in the Senate, after 5 years of maturation, and the capacitating
program is ongoing. We recall that the implementation process follows a three-tier
approach: regulation (including law for urban mobility systems, plus adjustment of
other laws; federal economic incentives); competence building through nationwide
training and education; and public awareness campaign (information leaets from the
Ministry of Cities and thematic seminars). Action from municipalities will occur in two
distinct moments: the rst by joining capacitation and the second by submitting their
application to the nancing of integrated projects by the union.
6.6. What We Still Need to Learn

From the theoretical and empirical work done, we conclude that the proposed model is
sufciently robust for implementation in diversied contexts. However, we have also
concluded on the need for further research development in areas relevant for the ease of
implementation of this type of systemic model in urban mobility environments.
The areas where we have detected insufcient knowledge, with a particular
relevancy for the further development and improvement of the proposed manage-
ment model, are as follows:
To support the strategic denition of the urban mobility systems:

Quality of decision-making is critical for the good conguration of the system.
So far, there is scant knowledge on how to measure quality of decision processes
applied to the transport sector.
Information systems are still very insufcient to support management of such

complex systems. Technology seems to have sufcient development to answer
these demands, but not enough knowledge exists on the functional requirements
applied to the multi-institutional and dynamic environment that characterizes
urban mobility.
The recognition of the existence of different segments of customers of the urban
mobility system was consolidated in the past decade. However, in the large
majority of urban systems, no action is yet taken to create stable and lasting
relations with those customers. Relational marketing seems to be an emerging
method with potential to improve better denition and conguration of the
mobility systems, enabling deeper knowledge on the characteristics of society
and population as clients of the urban mobility systems.
To support the steering, control, and enforcement functions:
Policy instruments must be implemented within specic organizational arrange-
ments. However, we still miss sufcient information to elaborate on the joint
decisions on instruments for the UMS and institutional designs for the best
implementation of policy instruments. This leads to the following research
questions:
How far can institutional design act as driver of system performance for the
whole system.
How do the contextual characteristics of a city or urban area impact on the
way instruments should be packaged vertically, horizontally, and chronolo-

gically.
What is the consequence that the choice of instruments has on the
effectiveness and efciency of policy implementation and how far are these
impacts related with the choice of regulatory frameworks.
What are the factors that should drive the choice of instruments in different
cities, based on their different impacts on effectiveness, efciency, legitimacy,

democratic participation, and symbolic effects.
To support network design and conguration of the urban mobility systems:
an urgent issue is related with the difculties of having integration initiatives
done by the agents, thus preventing a reduction of the current dependency on
planning by public authorities. Transferability of experiences from other modes
have to be explored, such as in air transport, where alliances are the instrument
that support network integration. These movements occur due to economies of
scope and density that are provided by the market dynamics itself and controlled
by the operator along the full network. In urban mobility, these network
economies are neither signicant nor controlled by the same operator, as we have
referred. In addition, in urban mobility systems, the intervention of authorities is
stronger and their nancial capacity to support integration costs is much weaker
than in air transport. Consequently, integration does not occur on voluntary
basis. It is then worth to dedicate some research effort toward a better
understanding of which are the factors that promote and hinder integration in
urban mobility systems and whether preconditions found in other system can be
reproduced here.
Current dependency on planning by public authorities raises a critical issue, that
is, how to provide planning agencies with capacity to accept, and even stimulate,
innovative service development by operators, without hindering the efciency
and effectiveness of the existing network conguration.
Abstract
Careful analysis of Urban Mobility reveals that consistent and effective policies
can only be well dened and implemented if the various components of the system
and their interrelations are considered. The denition of the Urban Mobility
System (UMS) goes far beyond the provision of public transport and should
entail all services, infrastructure, and trafc management that in its whole enable
citizens to satisfy their mobility requirements. The complexity and diversity of
dimensions of the conurbation and agents involved in an UMS imply focusing the
analysis of its performance on the symbiotic relationship between its main
components. Quality factors and processes should be set up in a coherent
organizational framework, providing adequate interaction mechanisms for
policies, and intervening institutions. The research work that is now presented
used the observation of several cities around the world to conrm that quality
improvements done at company and service levels are insufcient to ensure
signicant improvement of performance of the UMS. This objective was pursued
by decoupling, observing and understanding interactions among the different
elements of the system and between these and the surrounding environment. The
research concludes by stating the need for an holistic quality approach to urban
mobility management and presents a model along those lines.
Keywords: urban mobility; urban system; policy; management; integration;

quality
Annexes
Annex 1: The use of multimethodology and the combination of methods
Table A1: Most popular pair of methods (Rosenhead & Mingers, 2001, p. 305).
Method 1 Method 2 Frequency
Simulation Statistics 13
Forecasting Statistics 9
SWOT/PEST SSM 9
Simulation SSM 8
Inuence diagrams SSM 8
Strategic choice SSM 8
Critical systems heuristics SSM 7
SSM Interactive planning 7
SSM Cognitive mapping 7
Statistics SSM 7
VSM SSM 7
Maths. modeling Statistics 7
Maths. modeling Simulation 7
Structured analysis & design SSM 6
Maths. modeling Heuristics 5
Decision analysis Strategic choice 5
Decision Analysis Cognitive mapping 5
Statistics Cognitive mapping 5
Inuence diagrams VSM 5
Inuence diagrams SSM 5
Strategic choice Cognitive mapping 5
Interactive planning CSH 5
Strategic hoice Interactive planning 5
288 Annexes
Table B1: Most popular triads of methods (Rosenhead & Mingers, 2001, p. 305).
Method 1 Method 2 Method 3 Frequency
Strategic choice SSM Interactive planning 4

Maths. modeling Simulation Statistics 3
Maths. modeling Simulation Heuristics 2
Statistics Inuence diagrams Cognitive mapping 2
Statistics SWOT SSM 2
Statistics SSM Cognitive mapping 2
Statistics Project networks Forecasting 2
Statistics Forecasting Inventory 2
SSM VSM Strategic choice 2
SSM VSM TSI 2
SSM VSM CSH 2
SSM Interactive planning CSH 2
SSM Scenarios CSH 2
Cognitive mapping Delphi Scenarios 2
Hypergames Delphi Scenarios 2
Cognitive mapping Delphi System dynamics 2
Cognitive mapping Decision analysis Strategic choice 2
Cognitive mapping Inuence diagrams System dynamics 2
Legend: SSM, Soft Systems Methodology; CSH, Critical Systems Heuristic; and VSM, Viable Systems
Model.
Annex 2: Cities/urban regions observed in the various research projects
Table A2: (109) Case studies of ISOTOPE research (19951997).
Country Very Large Medium Small (F) Total

large (A) (BC) (DE)
(W2 (between 1 (between (o200,000 (Factual Q./

millions inh.) and 2 million 200,000 and inh.) Opinion Q.)
inh.) 1 million inh.)
Austria Graz* 1/1

Belgium Wallone Brussels* Charleroi*, Bruges* 7/6
(region)* Lie`ge, Ghent*
Flandres
(region)*
Denmark Copenhagen Aarhus Randers, 6/0
Bornholm,
Odense,
Vejle
List of Figures
Figure 1.1 Urban mobility dynamics (conceptual diagram). . . . . . . . . . . . . 4

Figure 1.2 Elements of the control process. . . . . . . . . . . . . . . . . . . . . . . . . 7
Figure 1.3 DPSIR framework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 1.4 Decision levels-conceptual diagram. . . . . . . . . . . . . . . . . . . . . . 22
Figure 1.5 Evolution of quality concept. . . . . . . . . . . . . . . . . . . . . . . . . . 25
Figure 1.6 Customer as borderline line sensor of company performance.. . . 27
Figure 1.7 Conceptual dynamics for a quality approach to UMS. . . . . . . . 29
Figure 1.8 Research and Advice projects relevant for this work.. . . . . . . . . 34
Figure 1.9 Methodological combination used in this research work. . . . . . . 35
Figure 2.1 Conceptual illustration of a system intervention based on
Vickers appreciation concept. . . . . . . . . . . . . . . . . . . . . . . . . . 52
Figure 2.2 Interactions of decision-making process. . . . . . . . . . . . . . . . . . . 57
Figure 2.3 Distribution of stated barriers in MARETOPE case studies (EC,
TIS, 2002c). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Figure 2.4 Boomerang effect of barriers and tools (EC, TIS, 2002b) . . . . . . 77
Figure 3.1 Conceptual identication of the typology for institutional
interdependencies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Figure 3.2 Chain of key decisions across planning levels. . . . . . . . . . . . . . . 83
Figure 3.3 Market structures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Figure 3.4 Regulatory and organizational congurations. . . . . . . . . . . . . 107
Figure 3.5 Learning process for qualication and selection of operators
(Macario, 1999, p. 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Figure 4.1 Decoupling of quality denition. . . . . . . . . . . . . . . . . . . . . . . 144
Figure 4.2 Quality gaps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Figure 4.3 Strategic nature of performance goals. . . . . . . . . . . . . . . . . . . 159
Figure 4.4 New Zealand land transport fund. . . . . . . . . . . . . . . . . . . . . . 188
Figure 4.5 Conceptual mapping of information by level of decision. . . . . . 194
Figure 4.6 Information system for UMS management (conceptual
diagram).. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Figure 5.1 Implementation cycle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
Figure 5.2 Cities and measures included in CIVITAS/METEOR projects . 204
Figure 5.3 Rationale underlying the quality management model . . . . . . . . 208
Figure 5.4 Usefulness of a system of indicators. . . . . . . . . . . . . . . . . . . . 209
Figure 5.5 Transport trends. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
Figure 5.6 Land use trends.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
Figure 5.7 Components of resistance to car use. . . . . . . . . . . . . . . . . . . . 214
xiv List of Figures
Figure 5.8 Causal loop diagram for land use trends. . . . . . . . . . . . . . . . . 216
Figure 5.9 Causal loop diagram for transport trends. . . . . . . . . . . . . . . . 217
Figure 5.10 Transversal process deployment. . . . . . . . . . . . . . . . . . . . . . . 232
Figure 5.11 Conceptual illustration of the proposed management model for
UMS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
List of Tables
Table 1.1 Government versus governance.. . . . . . . . . . . . . . . . . . . . . . . . 10

Table 1.2 Logic elements behind the rationale for transport and land use.. 11
Table 1.3 Subsample of cities observed for development of inductive
method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Table 1.4 Cities used for theoretical validation in this research work . . . . . 36
Table 2.1 Comparison of ve open-system theories of organizational
change. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Table 3.1 Nature and roles of entities interacting in an urban mobility
system (part I).. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Table 3.2 Nature and roles of entities interacting in an urban mobility
system (part II) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Table 3.3 Types of failure in the concertation processes . . . . . . . . . . . . . 102
Table 3.4 Allocation of conicts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Table 4.1 Evolution of quality related concepts. . . . . . . . . . . . . . . . . . . 140
Table 4.2 Valuation features in urban mobility systems. . . . . . . . . . . . . . 142
Table 4.3 Applicability of the EFQM excellence model to the quality
management of Urban Mobility Systems (Part 1 of 4). . . . . . . 151
Table 4.7 CEN quality aspects for public transport. . . . . . . . . . . . . . . . 161
Table 4.8 Oslo Sporveir quality guarantee. . . . . . . . . . . . . . . . . . . . . . . 162
Table 4.9 Contribution of marketing tools to close quality gaps.. . . . . . . 165
Table 4.10 Principles for market segmentation and market targeting. . . . . 166
Table 4.11 Physical integration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
Table 4.12 Logical integration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Table 4.13 Tariff integration.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Table 4.14 Responsibilities in integration initiatives. . . . . . . . . . . . . . . . . 180
Table 4.15 Contrast between different quality of interchange as perceived
by the users. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Table 4.16 Barriers to interchanges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Table 5.1 Typology of cities based on urban form. . . . . . . . . . . . . . . . . 211
Table 5.2 The three main processes for quality management in UMS. . . . 229
xvi List of Tables
Table 5.3 Key process in the quality management of UMSs quality

planning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 230
Table 5.4 Key process in the quality management of UMSs quality
improvement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Table 5.5 Critical aspects of key processes (part one). . . . . . . . . . . . . . . 234
Table 5.6 Critical aspects of key processes (part two). . . . . . . . . . . . . . . 235
Table 5.7 Indicators for perceived quality (part one). . . . . . . . . . . . . . . . 240
Table 5.8 Indicators for perceived quality (part two). . . . . . . . . . . . . . . . 242
Table 5.9 Examples of instruments of the UMS part one. . . . . . . . . . 248
Table 5.10 Examples of instruments of the UMS part two. . . . . . . . . . 250
Bibliography
Ackoff, R. L. (1957). Towards a behavioural theory of communication. In: W. Buckley (Ed.),

Modern systems research for the behavioural scientist (pp. 209218). Chicago: Aldine.
Ackoff,, R. L. (1959). Games, decisions and organization. General Systems, 4, 145150.
Acutt, M., & Dodgson, J. (1997). Policy instruments and greenhouse gas emissions from
transport in the UK. Fiscal Studies, 17(2), 6582.
Alonso, W. (1971). The economics of urban size. Regional Science Association Papers no. 26,
pp. 6883. Wiley Blackwell, London.
Anderson, C. W. (1977). Statecraft: An introduction to political choice and judgment.
New York: Wiley.
Ansoff, I., & Slevin, D. (1968a). An appreciation of industrial dynamics. Management Science,
14(7).
Ansoff, I., & Slevin, D. (1968b). Comments of professor Forresters industrial dynamics
After the rst decade. Management Science, 14(9).
Anthony, R. N. (1965). Planning and control systems: A framework for analysis, studies in
management control, division of research. Boston: Graduate School of Business Adminis-
tration, Harvard University.
Anthony, R. N. (1989). The management control function. Boston: Harvard Business School
Press.
Anthony, R. N., & Govindarajan, V. (1998). Management control systems. Boston: McGraw-
Hill Companies, Inc.
APCER, Associac- ao Portuguesa de Certicac- ao. (2003). ISO 9001:2000, pp. 1316.
Ascher, F. (2001). Les noveaux principes de lurbanisme La n des villes nest pas a` lordre du
jour. Collection Monde, Serie Intervention, Edition de lAube.
Ashby, W. R. (1968). Principles of self-organising system. In: W. Buckley (Ed.), Modern
systems research for the behavioral scientist. Chicago: Aldine.
Athens Charter. (1933). Available at http://sweb.cityu.edu.hk/sm3132/Documents/
Charters_Creeds.doc (March 12, 2005).
Bain, J. (1968). Industrial organization (2nd edn). New York: Wiley.
Banister, D. (2004). Transport and spatial policies: The role of regulatory and scal incentives,
Keynote paper in ECMT Round Table 124. In: European Conference of Ministers of
Transport, Economic Research Centre, Paris (pp. 117152).
Barabasi, A. L., & Bonabeau, E. (2003). Scale free networks. Scientic American, 288(5),
6069.
Barnard, C. I. (1971). The functions of the executive. Cambridge, MA: Harvard University Press.
Bartos, O. J. (1974). Process and outcome in negotiation. New York: Columbia University
Press.
Baumol, W., Ordover, J., & Willig, R. (1996). Parity pricing and its critics: Necessary condition
for efciency in provision of bottleneck services to competitors. Working Papers 96-33,
C.V. Starr Center for Applied Economics, New York University, NY, USA.
Baumol, W. J., Panzar, J. C., & Willig, R. D. (1988). Contestable markets and the theory of
industry structure. Orlando, FL: Harcourt Brace Jovanovich, Publishers.
Bayliss, L. E. (1966). Biological control systems. San Francisco: W. H. Freeman.
Beesley, M. (1997). Privatization, regulation and deregulation. London, UK: Routledge.
272 Bibliography
Bemelmans-Videc, M., Rist, R., & Vedung, E. (Eds) (1998). Carrots, sticks and sermons
(pp. 2158). New Brunswick NJ: Transaction Publishers, The Netherlands.
Benevolo, L. (1993). The European city. Oxford: Blackwell.
Benton, A. A., Kelley, H. H., & Liebling, B. (1972). Effects of extremity of offers and
concession rate on the outcomes of bargaining. Journal of Personality and Social
Psychology, 24, 7383.
Bernal, J. D. (1957). Science in history. London: Watts.
Bernal, J. D. (1957). The supply of information to the scientist: Some problems of the present
day. Journal of Documentation, 13(4), 195208.
Bertalanffy, L. (1968). General system theory. New York: Braziller.
Boulding, K. E. (1953). The organizational revolution. New York: Harper & Row.
Brasileiro, A., Henry, E., & Turma, (Eds). (1998). Onibus nas Cidades Brasileiras. Sao Paulo:
Cultura Editores Associados.
Brundtland Commision. (1987). Our common future. Report of the World Commission on
Environment and Development. Oxford: Oxford University Press.
Brown, A. (2005). J. D. Bernal: The sage of science. Oxford: Oxford University Press.
Buchanan, C. J. (1963). Report on trafc in towns. London: Her Majestys Stationery Ofce,
HMSO.
Buchanan, C., & Partners. (2002). Study on good practices in contracts, commissioned by the
European Commission.
Buchanan, C. (2002). Guide to contracts and contracting in Public Transport, commissioned
by the European Commission.
Buchanan, J. (2001). Game theory, mathematics, and economics. Journal of Economic
Methodology, 8(March), 2732.
Buckley, W. (Ed.) (1968). Modern systems research for the behavioral scientist. Chicago: Aldine.
Burt, R. S. (1978). Cohesion versus structural equivalence as a basis for network subgroups.
Sociological Methods and Research, 7(2), 189212.
Burt, R. S. (1982). Towards a structural theory of action: Network models of social structure,
perceptions and actions. New York: Academic Press.
Carr, J. D. (1997). Light touch regulation for the privatized bus industry. In: J. Preston
(Ed.), Proceedings of the 5th international conference on competition and ownership in
land transport, pp. 118. University of Leeds, Institute of Transport Studies,
UK. May.
Castells, M. (1996). The information age: Economy, society and culture, Vol. I, the rise of the
network society. Oxford, UK: Blackwell.
CEN. (2002). European standards for quality in public transport, TC320/WG5/13816-2002 EN.
Checkland, P. B. (1972). Towards a systems-based methodology for real world problem
solving. Journal of Systems Engineering, 3(2), 87116.
Checkland, P. B. (1974). What the systems approach means in education. Education
Management, Supplement to Education, 144(3), ivvi.
Checkland, P. (1983). OR and the systems movement. Journal of the Operational Research
Society, 34, 661675.
Checkland, P. (1985). From optimizing to learning: A development of systems thinking for the
1990s. Journal of the Operational Research Society, 36, 757767.
Checkland, P. (1994). Systems theory and management thinking. American Behavioral
Scientist, 38(1), 7591.
Checkland, P. (1999). Systems thinking, system practice. Chichester, UK: Wiley.
Checkland, P., & Scholes, J. (1999). Soft systems methodology in action. Chichester, UK: Wiley.
Bibliography 273
Chisholm, D. (1992). Coordination without hierarchy: Informal structures in multiorganizational

systems. Berkeley, CA: University of California Press.
Chisholm, M. (2000). The long march from realism to reality. Report 6 in the Series supported
by the Rees Jeffreys Road Fund. Landor Publishing, London.
Chislom, D. (1989). Coordination without hierarchy Informal structures in multiorganisa-
tional systems. Berkeley, CA: University of California Press.
Christensen, K. S. (1999). Cities and complexity: Making intergovernmental decisions, cities &
planning. Thousand Oaks, CA: Sage.
Ciufni, F. M. (1995). Transport and public spaces: The connectivity tissue of the sustainable
city, In: The sustainable city: A European tetralogy. Dublin: European Foundation for the
Improvement of Living and Working Conditions.
Clifton, K. J., & Handy S. L. (2001). Qualitative methods in travel behaviour research.
International Conference on Transport Survey Quality and Innovation. August 510,
Kruger National Park, South Africa.
Coase R. H. (1937). The nature of the rm. Economica 4(16), 386405.
Costa, B., et al. (1998). Onibus e lotac- ao, uma experiencia de conv vio regulamentado em
Porto Alegre. In: A. Brasileiro, et al. (Eds), Onibus nas cidades brasileiras. Brasil: Cultura
Editores Associados.
Cox, W., & Duthion, B. (2001). Competition in urban public transport: A world view.
Presented at 7th Conference on competition and ownership in Land passenger transport,
Molde, Norway, in June 2001. Available at http://ses.library.usyd.edu.au/bitstream/2123/
6312/1/thredbo7-workshopA-Cox-Duthion.pdf
Crampes, C., & Estache, A. (1997). Regulatory trade-offs in the design of concession contracts
(preliminary version). Washington, DC: World Bank.
Daniere, A. (1995). Transportation planning and implementation in cities of the third world:
The case of Bangkok. Environment and Planning C: Government and Policy, 13, 2545.
David, F. R. (1989). How companies dene their mission. Long Range Planning, 22(1),
9097.
Davidson, F., & Lindeld, M. (1996). To integrate or not to integrate: Developing a model for
effective integration. In: K. Singh, F. Steinberg & N. V. Einsiedel (Eds), Integrated urban
infrastructure development in Asia (pp. 3146). UK: Oxford & IBH Publishing Co.
Dawkins, R. (1989). The selsh gene. Oxford: Oxford University Press.
Dewey, J. (1927). The public and its problems (p. 126). New York: Henry Holt & Co.
Dobzhansky, Th. (1967). The biology of ultimate concern. New York: New American Library.
Drischel, H. (1968). Formal Theory of Organization. Selected Papers on Sentrics and Manfred
Clynes, Canada.
DuBos, R. J. (1964[1967]). Man adapting. New Haven, CT: Yale University Press.
Dudley, G., & Richardson, J. (2000). Why does policy change? Lessons from British transport
policy 194599. London: Routledge.
Economides, N. (1996). The economics of networks. International Journal of Industrial
Organization, 14, 673699.
Edwards, C., Ward, J., & Bytheway, A. (1995). The essence of information systems. London:
Prentice Hall.
ECMT. (2002). Sustainable urban travel implementing sustainable urban travel policies:
Applying the 2001 key messages, CEMT.
ECMT. (2005). ECMT Round Table 124, transport and spatial policies: The role of regulatory
and scal incentives. European Conference of Ministers of Transport, Economic Research
Centre, Paris, (Complete Edition: ISBN 9282123219).
274 Bibliography
EFQM. (1999). Are the renaissance balance scorecard and the EFQM excellence model
mutually exclusive or do they work together to bring added value to a company? Final draft,
pre-publication version, released exclusively for the EFQM Common Interest Day,
December 9, Gaelle Lamotte, Renaissance Worldwide & Geoff Carter.
Ellickson, R. (1987). A critique of economic and sociological theories of social control. Journal
of Legal Studies, 16, 67100.
Elmore, R. F. (1987). Instruments and strategy in public policy. Policy Studies Review, 7,
174186.
Esser, J., & Komorita, S. S. (1975). Reciprocity and concession making in bargaining. Journal
of Personality and Social Psychology, 31, 864872.
Estache, A., & Quesada, L. (2001). Concession contract renegotiations: Some efciency versus
equity dilemmas (preliminary version). Washington, DC: World Bank.
European Commission. (1996a). OGM, Survey and data collection nal report
deliverable D1. QUATTRO Quality approach in tendering urban public transport
operations. Transport Research, Fourth Framework Programme, Urban Transport,
DGVII.
European Commission. (1996b). TIS.PT, Report on political research work package 2.
ISOTOPE Improved Structure and Organisation for Urban Transport Operations of
Passengers in Europe. Transport Research, Fourth Framework Programme, Urban
Transport, DGVII.
European Commission. (1997a). OGM, Tendering, contracting in UTP service
Deliverable D3. QUATTRO Quality approach in tendering urban public transport
operations. Transport Research, Fourth Framework Programme, Urban Transport,
DGVII.
European Commission. (1997b). TIS.PT, Final report ISOTOPE Improved Structure and
Organisation for Urban Transport Operations of Passengers in Europe. Transport Research,
Fourth Framework Programme, Urban Transport, DGVII.
European Commission. (1997c). TIS.PT, Report on Economic Research-Work Package 3.
ISOTOPE Improved Structure and Organisation for Urban Transport Operations of
Passengers in Europe. Transport Research, Fourth Framework Programme, Urban
Transport, DGVII.
European Commission. (1998a). OGM, Practitioners handbook QUATTRO Quality
approach in tendering urban public transport operations, Final Report. Transport Research,
European Commission. (1998b). TIS.PT, Report on methodological framework to assess
nancial schemes Deliverable D1, FISCUS Cost Evaluation and Financing Schemes
for Urban Transport Systems, Fourth Framework Programme, Urban Transport,
DGVII.
European Commission. (1998c). NEA, Examination of community law relating to the
public services obligations and contracts in the eld of inland passenger transport.
DGVII.
European Commission, MEANS, (1999a). Evaluation design and management (Vol. 1).
Luxembourg.
European Commission. (1999b). TIS.PT, State of the art synthesis on price acceptability
Deliverable D1. PATS Price Acceptability of Transport Systems. Transport Research,
European Commission. (1999c). TIS.PT, Design of new nancing schemes for urban transport
Deliverable D4. FISCUS Cost Evaluation and Financing Schemes for Urban
Transport Systems, Fourth Framework Programme, Urban Transport, DGVII.
Bibliography 275
European Commission. (2000). TIS.PT, Empirical studies on price acceptability deliverable D3.
PATS Price Acceptability of Transport Systems. Transport Research, Fourth Frame-
work Programme, Urban Transport, DGVII.
European Commission. (2001a). TIS.PT, Synthesis of case studies Deliverable D4.
MARETOPE Managing and Assessing Regulatory Evolution in Local Public Transport
Operations in Europe, Fifth Framework Programme, Urban Transport, DG TREN.
European Commission. (2001b). TIS.PT, Design of alternative solutions for pricing packages
Deliverable D4. PATS price acceptability of transport systems. Transport Research,
European Commission. (2001c). TIS.PT, Handbook Deliverable D5. FISCUS cost
evaluation and nancing schemes for urban transport systems. Fourth Framework
Programme, Urban Transport, DGVII.
European Commission. (2001d). TIS.PT, Intermediate report on development of case studies
Deliverable D3 (complete version, condential). MARETOPE Managing and Assessing
Regulatory Evolution in Local Public Transport Operations in Europe, Fifth Framework
Programme, Urban Transport, DG TREN.
European Commission. (2001e). TIS.PT, Recommendations on transport pricing strategies
Final report. PATS price acceptability of transport systems. Transport Research, Fourth
Framework Programme, Urban Transport, DGVII.
European Commission. (2002a). ISIS, Technical report Deliverable D1. Transplus
Transport planning, land use and sustainability. Fifth Framework Programme, Urban
Transport, DG TREN.
European Commission. (2002b). TIS.PT, Maretope Handbook. MARETOPE Managing
and Assessing Regulatory Evolution in Local Public Transport Operations in Europe, Fifth
Framework Programme, Urban Transport, DG TREN.
European Commission. (2002c). TIS.PT, Report on barriers to change and tools to assist key
players in the process of change Deliverable D6. MARETOPE Managing and Assessing
Regulatory Evolution in Local Public Transport Operations in Europe, Fifth Framework
European Commission. (2003a). ISIS, Barriers, solutions and transferability Derivable D4.
TRANSPLUS Transport Planning, Land Use and Sustainability, Fifth Framework
European Commission (EC). (2003b). ISIS, Final report, TRANSPLUS Transport
planning, land use and sustainability, Fifth Framework Programme, Urban Transport, DG
TREN.
European Commission. (2003c). NEA, Integration and regulatory structures in public transport
Final report. DG TREN.
European Commission, Directorate General Transport. (1998). Transport and land-use
policies Resistance and hopes for co-ordination. Transport Research Cost 332, Cost
Transport.
European Conference of Ministers of Transport. (1995). Urban travel and sustainable
development. Available at http://www.oecd.org. Retrieved on 16 December 2004.
European Conference of Ministers of Transport. (2002a). Transport and spatial policies The role
of regulatory and scal incentives. Conclusion of the Round Table 124, Paris, 78 November.
European Conference of Ministers of Transport. (2002b). Implementing sustainable urban
travel policies. Final report.
Fainstein, S., & Campbell, S. (1996). Readings in urban theory. UK: Blackwell.
Forrester, J. W. (1958). Industrial dynamics A major breakthrough for decision makers.
Harvard Business Review (JulyAugust), 3766.
276 Bibliography
Forrester, J. W. (1964). Common foundations underlying engineering and management. IEEE

Spectrum, Vol. 1, No. 9, pp. 6677. Also appears as Chapter 4, pp. 6180, in the authors
Collected Papers, 1975, Waltham, MA: Pegasus Communications.
Forrester, J. W. (1968a). Industrial dynamics A response to Ansoff and Slevin. Management
Science, 14(9), 601618.
Forrester, J. W. (1968b). Principles of systems. (Preliminary edition, ten chapters). Cambridge,
MA: Wright-Allen Press, Inc.
Forrester, J. W. (1969). Urban dynamics. Cambridge, MA: MIT Press.
Foss, J. N. (1996). Firms, incomplete contracts, and organizational learning. Danish Research
Unit for Industrial Dynamics, DRUID Working Paper no. 962. Danmark.
Freeman, G. F. (1985). National styles and policy sectors: Explaining structured variation.
Journal of Public Policy, 5(4), 467496.
Frey, W. H., & Spear, A., Jr. (1995). Metropolitan areas as functional communities. In: D. C.
Dahamann & J. D. Fitzimmons (Eds), Metropolitan and non metropolitan areas: New
approaches to geographical denition. Working Paper no. 12. US Bureau of the Census,
Population Division, Washington, DC.
Friedman, J., & Miller, J. (1965). The urban eld. Journal of the American Institute of Planners,
31, 312319.
Gambetta, D. (Ed.) (1989). Trust: Making and breaking of cooperative relations. Oxford:
Blackwell.
GART. (2001). Citizens conference, Associations of Organizing Authorities, Ille de France,
Paris.
Goetz, A. R., & Szyliowicz, J. S. (1997). Revisiting transport planning and decision-
making theory: The case of Denver international airport. Transportation Research, 31A,
263280.
Gomide, A. (2003). Transporte urbano e inclusao social: Elementos para pol ticas publicas.
Bras lia, Ipea, 2003. (Texto para Discussao n. 960).
Goodwin, P., Hallett, S., Kenny, F., & Stokes, G. (1991). Transport: The new realism. Report
to the Rees Jeffreys Road Fund. Transport Studies Unit, University of Oxford.
Grossman, S., & Hart, O. (1986). The costs and benets of ownership: A theory of vertical and
lateral integration. Journal of Political Economy, 94, 691719.
Gubbins, E. J. (1988). Managing transport operations. UK: The Institute of Logistics and
Transport, Kogan Page.
Gulick, L. (1937). Notes on the theory of organization. In: L. Gulick & L. Urwick (Eds),
Papers on the Science of Administration. New York: Institute of Public Administration,
Columbia University.
Gwilliam, K. M. (1987). Setting the market free. Working Paper no. 259. Institute for
Transport Studies. University of Leeds, UK.
Hagerstrand, T., & Clark, E. (1998). On the political geography of transportation and land-use
policy coordination. Cost 332 Intermediate Report. Brussels.
Hall, P. (1969). Transportation. Urban Studies, 6(3), 483503.
Harbour, J. L. (1997). The basics of performance measurement. Oregon: Productivity Press.
Hardin, R. (1982). Collective action. Baltimore, MD: John Hopkins University Press.
Hardjono, T. W., ten Have, S., & ten Have, W. D. (1997). The European way to excellence.
Brussels, Belgium: DGIII Industry, European Commission and EFQM.
Hart, O. (1995). Firms contracts and nancial structure. Oxford: Clarendon Press.
Hart, O. (2003). Incomplete contracts and public ownership: Remarks, and an application to
public-private partnerships. The Economic Journal, 113(March), C69C76.
Bibliography 277
Hawley, A. H. (1971). Urban society: An ecological approach. New York: The Ronald
Press.
Hayek, F. A. (1945). The use of knowledge in society. The American Economic Review, 35(4),
519530.
Hensher, D. A. (Ed.). (2005). Competition and ownership in land passenger transport: Selected
papers from the 8th international conference on competition and ownership. Elsevier/ Emerald,
ISBN: 9780080445809.
Hensher, A., & Macario, R. (2002). Organization and ownership of public transport services.
Transport Reviews, 22(3), 349357.
Hensher, D., & Prioni, P. (2002). A service quality index for area-wide contract performance
assessment. Journal of Transport Economics and Policy, 36(Pt. I), 93113.
Hibbs, J. (2003). Transport economics and policy. The Institute of Logistics and Transport,
Kogan Page.
Hirshleifer, J. (1897). Economic behavior in adversity. Chicago: University of Chicago Press.
Holmstrom, B., & Milgrom, P. (1991). Multi-task principal agent analysis: Incentive contracts,
asset ownership and job design. Journal of Law, Economics and Organization, 7, 2452.
Holzner B., & Marx, J. (1979). Knowledge application: The knowledge system in society.
Boston: Allyn-Bacon In: Galliers, R. D. & Leidner, D. E. (2003). Strategic information
management: Challenges and strategies in managing information systems. UK: Butterworth
and Heinemann.
Howlett, M. (1991). Policy instruments: Policy styles, and policy implementation: National
approaches to theories of instrument choices. Policy Studies Journal, 19, 121.
Iallis, Y. (2003), cited in Hensher (2005).
Jaynes, E. T. (1965). Gibbs vs Boltzmann Entropies. American Journal of Physics, 33(5),
391398.
Juran, J. M. (1990). Planning for quality. Wilton, CT: Juran Institute Inc.
Juran, J. M. (1998). Quality control handbook. New York: McGraw-Hill.
Juran, J. M. (2000). Jurans quality handbook (McGraw-Hill international editions: Industrial
engineering series) (5th revised ed.). Washington, USA: McGraw-Hill.
Kane, L., & Mistro, R. (2003). Changes in transport planning policy: Changes in transport
planning methodology? Transportation 30, New York, USA: Kluwer, pp. 113131.
Kartacheva, A. V., & Quesada, L. (2000). Contract renegotiations on concessions in Latin
America and Caribbean region: An economic analysis and empirical implications. Mimeo.
Keer, N. L. (1986). Motivational choices in task groups: A paradigm for social dilemma
research. In: H. A. M. Wilke, D. M. Messick & C. G. Rutte (Eds), Experimental social
dilemmas (pp. 127). New York: Lang.
Khandwalla, P. N. (1979). The design of organizations. New York: Harcourt Brace.
Khisty, C. J. (1992). Matching planning styles and roles in transportation to conditions of
uncertainty and state of technological know-how. Transportation Research Record, 1364,
139143.
Kileen, B. J., & Shoji, K. (1997). Diversication strategy and urban transportation: The case of
Japan. 5th International Conference on Competition and Ownership, Leeds.
King, S., & Pitchford, R. (2001). Private or Public? A taxonomy of optimal ownership and
management regimes. July, Australian National University, mimeo.
Klein, B., Crawford, R., & Alchian, A. (1978). Vertical integration, appropriable rents and the
competitive contracting process. Journal of Law and Economics, XXI(2), 297326.
Knoke, D. (1990). Political networks: The structural perspective. USA: Cambridge University
Press.
278 Bibliography
Knoke, D. (2001). Changing organisations. Foundations of Social Inquiry. USA: PressBoulder

Westview Press.
Knoke, D., & Kuklinsky, J. (1982). Network analysis. Berlim, Germany: Sage Publications.
Kotler, P., & Armstrong, G. (1991). Principles of marketing. Cambridge, USA: Prentice Hall.
Kuhn, T. S. (1962). The structure of scientic revolutions. USA: Chicago Press.
Kuklinski, J. H., Paul, J. Q., Jennifer, J., & Robert, F. R. (2001). The political environment
and citizen competence. American Journal of Political Science, 45(2), 410424.
Laffont, J. J., & Martimort, D. (1947/2002). The theory of incentives the principal agent
model. USA: Princeton University Press.
Lamotte, G. (1999). Are the Renaissance balanced scorecard and the EFQM excellence model
mutually exclusive or do they work together to bring added value to a company? Exclusive
for the EFQM Common Interest Day, December 9, Renaissance Worldwide & Geoff
Carter.
Landau, M., Chisholm, D., & Webber, M. (1980). Redundancy in public transit, Vol. I: On the
idea of an integrated transit system, University of California. Berkeley, CA: Institute of
Urban and Regional Development.
Leccese, M., & McCormick, K. (1999). The charter of new urbanism. New York: McGraw-Hill.
Le gates, R. T., & Stout, F. (Eds). (2000). The city reader. USA and Cannada: Routledge.
Levi, M. (1990). A logic of institutional change. In: K. Schweers Cook & M. Levi (Eds), The
limits of rationality (pp. 402418). Chicago: University of Chicago Press.
Levitt T. (1960). Marketing Myopia, reprint in Harvard Business Review (HBR), Special
issue, Best of HBR, JulyAugust, 2004.
Lewotin, R. (2000). The triple helix, gene, organism and the environment. Cambridge, MA:
Harvard University Press.
Liebert, R. M., Smith, W. P., Hill, J. H., & Keiffer, M. (1968). The effect of information and
magnitude of initial offer on interpersonal negotiation. Journal of Experimental Social
Psychology, 4, 431441.
Lindblom, C. E. (2001). The market system. UK and USA: Yale University Press.
Linstone, (1984). Multiple perspectives for decision making: Bridging the gap between analysis
and action. New York, USA: Elsevier.
Loasby, B. (1994). Understanding markets. University of Stirling, Stirling, mimeo, in Foss N. J.
(1996).
Lopes, A. S. (1984). Desenvolvimento regional, problematica, teoria e modelos. Lisboa Portugal:
Fundac- ao Calouste Gulbenkian.
Lorenz, E. H. (1989). Neither friends nor strangers: Informal networks of subcontracting in
French Industry. In: D. Gambetta (Ed.) (pp. 194210). UK: Blackwell.
Macario, R. (1999). Quality criteria applied to the tendering life cycle. 6th International
Conference on Competition and Ownership on Land Passenger Transport, September, Cape
Town, South Africa.
Macario, R. (2000). Upgrading quality in urban mobility systems. Managing Service Quality,
II(2), 2001, MCB University Press, pp. 9399.
Macario, R. (2001). Mapping key elements of a quality management model for Urban
Mobility Systems. 6th Total Quality Management Congress, June, St. Petersburgh,
Russia.
Macario, R. (2003). A scalidade automovel: das diferenc- as nacionais para uma harmonizac-
ao inteligente. In: J. Viegas (Ed.), Mobilidade para todos. Lisboa: ACP 100 years
commemoration book.
Bibliography 279
Macario, R. (2004a). Legal, institutional and contractual issues. In: J. M. Viegas (Ed.), Inter-urban
charges for trucks in europe (Vol. 11, pp. 4374). Research in Transport Economics. Elsevier.
Macario, R. (2004b). Institutional frameworks, the regulatory agencies and the land passenger
transport industry: Recent evolutions. Reference paper, Proceedings of the 8th Conference on
Competition and Ownership in Land Passenger Transport, Elsevier.
Macario, R. (2004c). Relations between authorities and operators. Regional Training
Programme, Union International des Transport Publics, UITP and Union Africaine des
Transport Publics, UATP, Dakar, Senegal, 2526 March, Published in CD by UITP.
Macario, R. (2004d). Integration in urban mobility systems: Quality upgrading or competition
blockade. X WCTR Conference, 59 July, Istanbul.
Macario, R., & Carmona, M. (2002). As contas de transporte. in O Economista Anuario
da Economia Portuguesa, Ordem dos Economistas, Polimeios, pp. 227230.
Macario, R., & Marques, C. (2004). Transferability of transport policies and measures in
CIVITAS. Working Document, METEOR, Accompanying measure for CIVITAS
demonstrations, Deliverable 6, European Commission, EC.
Majone, G. (1989). Evidence, arguments and persuasion in the policy process. USA: Yale
University Press.
Malhotra, K. N., & Birks, D. (2003). Marketing research. USA: FT Times and Prentice Hall.
Margolis, H. (1982). Selshness, altruism and rationality: A theory of social choice. Cambridge,
MA: Cambridge University Press.
Mason, E. (1939). Price and production policies of large-scale enterprise. American Economic
Review, 29, 6174.
Mathew, B. (2001). European Commission Fifth Framework Programme, Urban Transport,
DG TREN. PROSPECTS research project, Report on Task 16, unpublished, p. 25.
May, T., Jopson, A., & Matthews, B. (2003). Research challenges in urban transport policy.
Transport Policy, 10, 157164.
McDonnell, & Grubb (1991). Education and training for work: The policy instrument and the
institution. London, UK: Rand Publication.
Meyer, M. D., & Miller, E. J. (2001). Urban transportation planning A decision-oriented
approach (Series in Transportation). USA: McGraw-Hill.
Milgrom, P. R., North, C. D., & Weingast, B. W. (1990). The role of institutions in the revival
of the trade: The law merchant, private judges, and the champagne fairs. Economics and
Politics, 2, I23.
Mills, J. S. (1972). A system of logic (8th edn.). London.
Mitchell, J. (1991). Hierarchies. In: G. Thompson, J. Frances, R. Levacic & J. Mitchell (Eds),
Markets, hierarchies & networks (pp. 105107). USA: Sage.
Mohring, H. (1972). Optimization and scale economies in Urban Bus Transportation.
American Economic Review, 62, 591604.
Monteiro, J. A. P. (2003), Poder e Obediencia uma Teoria do Consentimento,
Universidade Tecnica de Lisboa, Instituto Superior de Ciencias Sociais e Pol ticas.
Morin, E. (2001). Seven complex lessons in education for the future (Education on the move).
Genebra: UNESCO Publishing.
Morrison, A. (1996). Alternative information technologies for the provision of spatial
information to public transport passengers in France, Germany and Spain. Transport
Reviews, 16(3), 243271.
Musgrave, R. A., & Musgrave, P. B. (1984). Public nance in theory and practice (4th edn.).
USA and Canada: McGraw-Hill.
280 Bibliography
NEA. TIS, ERASMUS (1998). Examination of community law relating to the public
service obligations and contracts in the eld of inland passenger transport. EC-DGVII,
1997/1998.
Newbery, D. M. (2000). Privatization, restructuring, and regulation of network utilities.
Review of Industrial Organization, 19(4), 497500.
North, D. (1981). Structure and change in economic history. New York: Norton.
North, D. (1990). Institutions, institutional change and economic performance. USA: Cambridge
University Press.
North, D. C. (1991). Institutions. Journal of Economic Perspectives, 5, 97112.
Norton, D. (1990). Institutions, institutional change and economic performance. Cambridge:
Cambridge University Press.
Oostergard, S., & Vautier, P. (2000). Quality in a marketing perspective: How to balance
quality of communication with quality of product. Public Transport International
(February), 512. UITP publication.
Oster, C. V., Jr., & Strong, J. S. (2000). Transport restructuring and reform in an international
context. Transportation Journal, 39(3), 1832.
Ostrom, V. (1971). The political theory of a compound republic: A reconstruction of the logical
foundation of democracy as presented in the federalist. Blacksburg, VA: VPI, Center for
Study of Public Choice.
Paddison, R. (Ed.) (2001). Handbook of urban studies. UK: Sage.
Paisley, W. J. (1981). Public communication campaigns: The American experience. In:
R. E. Rice & W. J. Paisley (Eds), Public communication campaigns (pp. 1540).
Beverly Hills: Sage.
Parsons, W. (1995). Public policy: An introduction to the theory and practice of policy analysis.
Cheltenham, UK: Edward Elgar Publishing, Inc.
Parsons, W. (2000). Analytical frameworks for policy and project evaluation: From welfare
economics and public choice to management approaches. Paper presented to the rst
TRANS-TALK workshop, Brussels.
Piercy, N. F., & Morgan, N. A. (1994). Mission analysis: An operational approach. Journal of
General Management, 19(3), 16.
Platt, J. R. (1964). Strong interference. Science, 146(3642), 351.
Popper, K. (2002). The logic of scientic discovery. USA and Canada: Routledge Classics.
Potter, S., & Skinner, M. (2002). Using strategic niche management to develop and evaluate
urban transport policy instruments. WCTR Special Interest Group 10, Urban Transport
Policy Instruments, July, Leeds.
Poulenez-Donovan, C. J., & Ulberg, C. (1994). Seeing the trees and missing the forrest:
Qualitative versus quantitative research ndings in a model transportation demand
management program evaluation. Transportation Research Record, No. 1459, pp. 16.
Prigogine, I., & Kondepudi, D. (1998). Modern termodynamics. Chichester: Wiley.
Prigogine, I., & Stengers, I. (1985). Order out of chaos Mans new dialogue with nature.
Glasgow: Flamingo.
Prudhomme R. (1996). Managing Megacities. Le Courier du CNRS, No. 82, pp. 174176.
Prudhomme, R., Dupuy, G., & Boret, D. (2004). The new constraints of urban development.
Les Cahiers de lInstitut Veolia Environnement.
Pruit, D. G., & Kimmel, M. J. (1977). Prisoners dilemma. New York: Anchor Books.
Pruitt, D. G., & Carnevale, P. J. (1993). Negotiation in social conict. Buckingham, UK:
Brooks/Cole.
Bibliography 281
Pruitt, D. G., & Syna, H. (1985). Mismatching the opponents offers in negotiation. Journal of
Experimental Social Psychology, 21, 103113.
Radbone, I. (1997). The competitive tendering of public transport in Adelaide. In: J Preston
(Ed.), Proceedings of the fth international conference on competition and ownership in land
transport, May, Leeds, UK.
Raffestin, C. (1980). Pour une geographie du pouvoir. Litec, Paris. COST 332, Intermediate
report.
Rapoport, A. (1966). Two-person game theory: The essential ideas. Ann Arbor, MI: University
of Michigan Press.
Rapoport, A., & Chammah, A. (1965). Prisoners dilemma. Ann Arbor, MI: University of
Michigan Press.
Rappaport, A., & Horvath, W. J. (1959). The theoretical channel capacity of a single neuron as
determined by various coding systems. Information and Control, 3(4), 335350.
Rawls, J. (1999). Theory of justice (revised edn.). Oxford.
Reed, M. I. (1985). Redirections in organizational analysis. London: Tavistock.
Richardson, J. (Ed.) (1982). Policy styles in western Europe. London: Allen & Unwin.
Rienstra, S. A., & Nijkamp, P. (1997). Financing and operation of transport systems. Working
Paper in the EU-DGVII funded POSSUM (Policy Scenarios for Sustainable Mobility)
project. Department of Spatial Economics, Free University Amsterdam.
Riley, J. F., Jr. (2000). Process management. In: J. Juran & A. B. Godfrey (Eds), Jurans
quality handbook (pp. 6.16.21). USA: McGraw-Hill International.
Rist, R. C. (1998). Inuencing the policy process with qualitative research. In: N. K. Denzen &
Y. S. Lincoln (Eds), Collecting and interpreting qualitative materials (pp. 400424). The
Netherlands: Sage.
Roberts, E. B. (1978). Managerial applications of system dynamics. UK: The MIT Press.
Rose, G. (2001). Visual methodologies: An introduction to the interpretation of visual materials
(229p.). London: Sage.
Ross, S. A. (1973). The economic theory of agency: The principals problem. American
Economic Review, 63, 134139.
Rothstein, B. (1996). Political institutions: An overview. In: A new handbook of political science
(pp. 133166). Oxford: Oxford University Press.
Rotmans, J., Kemp, R., & Asselt, M. (2001). More evolution than revolution. Transition
management in public policy. Foresight, 3(1), 1531.
Sabatier, P. A., & Jenkins-Smith, H. C. (Eds). (1993). Policy change and learning An
advocacy coalition approach (Theoretical lenses on public policy). USA: Westview Press.
Sack, R. (1986). Human territoriality: Its theory and history. Cambridge,MA: Cambridge
University Press.
Santos, E., & Aragao, J. (2000). Transporte em tempos de reforma. Bras lia: LGE Editora.
Schade, J., & Schlag, B. (2000). Acceptability of urban transport pricing. VATT Research
Report No. 72, Helsinky.
Schelling, T. C. (1960). The strategy of conict. Cambridge, MA: Harvard University Press.
Schelling, T. C. (1980). The strategy of conict. Cambridge, MA: Harvard University.
Schmidt, K. L. (1996a). Incomplete contracts and privatization. European Economic Review,
40, 569579.
Schmidt, K. L. (1996b). The costs and benets of privatization: An incomplete contracts
approach. Journal of Law, Economics and Organization, 12referring to the discussion paper
A-287 (1990) presented at the Bonn University, pp. 124.
282 Bibliography
Schmidt, K. L. (2002). Contract Theory (Available at http://www.vwl.uni-muenchen.de.

Retrieved on 4 November 2004). Course Winter Semester. University of Munich.
Schoeld, N. (1985). Anarchy, altruism and cooperation: A review. Social Choice and Welfare,
2, 207219.
Schon, D. (1992). The theory of inquiry: Deweys legacy to education. Curriculum Inquiry,
22(2)in Sterman (2000), pp. 119139.
Schultz, T. W. (1951). A framework for land economics The long view. Journal of Farm
Economics, 33, 204215.
Schwartz, A., & Watson, J. (2000). Economic and Legal aspects of costly recontracting.
Discussion Paper 2000-05. University of California, San Diego, CA.
Scott, W. R. (1981). Developments in organization theory, 19601980. American Behavioral
Scientist, 24, 407422.
Shleifer, A. (1998). State versus private ownership. Journal of Economic Perspectives, 12(4),
135150.
Shy, O. (2001). Economics of network industries. UK: Cambridge University Press.
Silverman, D. (1970). The theory of organizations. London: Heinemann.
Simon, H. A. (1951). A formal theory of the employment relationships. Econometrica, 19, 293305.
Simon, H. A. (1957). Models of man. Oxford, UK: Wiley.
Simon, H. A. (1960). The new science of management decision. New York: Harper and Row.
Simon, H. A. (1965). Comportamento administrativo: Estudo dos processos decisorios nas
organizac- oes administrativas (2nd ed.). Rio de Janeiro: Fundac- ao Getulio Vargas.
Simon, H. A. (1969b). The sciences of the articial (1st ed.). New York, USA: MIT Press.
Simon, H. A. (1969a). A behavioral model of rational choice. The Quarterly Journal of
Economics, 69(1), 99118(February 1955).
Simon, H. A. (1978a). Rationality as process and as product of thought. The American
Economic Review, 68(2). Papers and Proceedings of the 9th Annual Meeting of the American
Economic Association, May, pp. 116.
Simon, H. A. (1978b). On how to decide what to do. Bell Journal of Economics, 9(2), 494507.
Simon, H. A. (1991). Organizations and markets. Journal of Economic Perspective, 5(2), 2544.
Simon, H. A. (1999). The many shapes of knowledge. Revue dEconomie Industrielle,
Programme National Persee, 88(1), 2339.
Simons, R. (2000). Performance measurement and control systems for implementing strategy.
USA: Prentice Hall.
Silva, T. E. (2005). Working document on Brazilian legislation related to mobility issues.
Unpublished, March.
Smith, A. (1776/1964). The wealth of nations. Dent, Everymans Library, London, in Le
Grand J. (2003).
Smith, D. L., Pruitt, D. G., & Carnevale, P. J. (1982). Matching and mismatching: The effect
of own limit, other toughness, and time pressure on concession rate in negotiation. Journal
of Personality and Social Psychology, 42, 876883.
Southern Vectis, House of Commons. (1995) Available at http://wn.com/Southern_Vectis
Spiller, P. T. (1993). Institutions and regulatory commitment in utilities privatization.
Industrial and Corporate Change, 2(3), 387450.
Stahl, M. J. (1995). Management: Total quality in a global environment. In: T. W. Hardjono et al.
(Eds), The European Way to Excellence. DGIII Industry, European Commission and EFQM.
Sterelny, K. (2001). Dawkins vs. Gould Survival of the ttest. United Kingdom: Icon Books.
Sterman, J. D. (2000). Business dynamics: Systems thinking and modeling for a complex world.
USA: Irwin, McGraw-Hill.
Bibliography 283
Stigler, G. J., & Becker, G. S. (1977). De Gustibus Non Est Disputandum. American Economic
Review, 67(2), 7690.
Stoker, G. (1995). Urban governance in Britain. Sociologie du Travail, 2.
Sugden, R. (1986). The economics of rights, co-operation, and welfare. Oxford: Blackwell.
Szyliowicz, J. C., & Goetz, A. R. (1995). Getting realistic about megaproject planning. The
case of the new Denver International Airport. Policy Sciences, 28, 347367.
Tavistock Institute GHK IRS. (2003). The evaluation of socio-economic development: The
guide. Tavistock Institute, London. Available at http://ec.europa.eu/regional_policy/sources/
docgener/evaluation/evalsed/downloads/the_guide.zip. Retrieved on 24 October 2007.
Thompson, G., Frances, J., Levacic, R., & Mitchell, J. C. (1998). Markets, hierarchies &
networks The coordination of social life. UK and Netherlands: Sage in association with
Open University.
Tirole, J. (1988/1990). Theory of industrial organization. France: MIT Press.
Tirole, J. (1999). Incomplete contracts: Where do we stand? Econometrica, 67, 741781.
Tonnies, F. (1887). Community and society. New Jersey: Transaction Publishers.
Transplus. (20002003). Transport planning, land-use and sustainability. European Commis-
sion, EC, supported project under the City of Tomorrow and Cultural Heritage key
action, within the European Commissions Energy, Environment and Sustainable
Development Research Programme.
Transport for London. (2003). Congestion charging: Six months on. Mayor of London.
Trincher, K. S. (1965). Biology and information: Elements of biological thermodynamics. New
York: Consultants Bureau.
Truman, D. B. (1995). Interest groups and the nature of the state. In: D. McCool (Ed.), Public
policy: Theories, models and concepts An anthology (pp. 3240). USA: Prentice-Hall.
Tushman, M. L., & Romanelli, E. (1985). Organizational evolution: A metamorphosis model
of convergence and reorientation. In: L. L. Cummings & B. M. Staw (Eds), Research in
organizational behavior (Vol. 7, pp. 171182). Greenwich, CT: JAI Press.
Tyson, W. (1994). Unpublished work on Regulatory and Organizational Frameworks for
Urban Public Transport, commissioned by the European Commission, DG VII, preceded
the ISOTOPE research project.
Utterback, J. M. (1994). Mastering the dynamics of innovation. USA: Harvard Business School
Press.
Varela, F. J. (1984). Two principles of self-organisation. In: H. Ulrich & G. J. Probst
(Eds), Self-organisation and management of social systems (pp. 2532). Berlin: Springer-
Verlag.
Van de Velde, D. M. (1999). Organizational forms and entrepreneurship in public transport,
Part 1: classifying organisational forms. Transport Policy (6), 147157. Lyon, France:
Pergamon Elsevier Publishing.
Van de Velde, D. M., & Sleuwagen, L. I. E. (1997). Public transport service contracts:
Searching for the optimum. International Journal of Transport Economics, 24(1), 5374.
van der Doelen, F. C. J. (1998). The give-and-take packaging of policy instruments:
Optimising legitimacy and effectiveness. In: M. Bemelmans-Videc, R. Rist & E. Vedung
(Eds), Carrots, sticks and sermons (pp. 129149). New Brunswick, NJ: Transaction
Publishers.
Vickers, G. (1965). The art of judgement (re-printed in 1995 by Sage). London: Chapman &
Hall.
Viegas, J. M. (2001). A sharper view of system performance in urban public transport. 9th
WCTR Conference, July, Seoul, Korea.
284 Bibliography
Viegas, J. M. (2002a). Estrategias urbansticas e governabilidade. Lisboa, Portugal: Fundation

Calouste Gulbenkian.
Viegas, J. M. (2002b). Mobilidade in Gestao Urbana, Passado, Presente e Futuro, Cap.
II, Gestao da Mobilidade Urbana, Parque Expo, Lisboa, pp. 216235.
Viegas, J. M. (2003a). Competition and regulation in the transport sector: A recurrent game
and some pending issues. European Conference of Ministers of Transport, ECMT
Symposium, Budapest.
Viegas, J. M. (2003b). Transport policy development in Europe 19502020. In: 50 Years of
Transport Research: 19532003 (pp. 105115). Paris: ECMT.
Viegas, J. M. (2004a). Desaos da mobilidade urbana em Lisboa: As respostas poss veis para
uma cidade competitiva e sustentavel. Sociedade & Territorio, Junho, pp. 3738.
Viegas, J. M. (2004b). From multimodality to seamless intermodality in urban transport: A
difcult transition. Symposium on Urban Mobilities organized by the Chair of the
Institut pour la Ville en Movement and Tsinghua University, October, Beijing.
Viegas, J. M. (2005). Integrated transport systems: Public/private interfaces. In:
K. Button & D. Henscher (Eds), Handbook of transport strategy, policy and institutions,
6. OECD publication, Paris: Elsevier.
Viegas, J. M., & Macario, R. (1998a). Service quality issues in mass transit. World Markets
Series, Business Brieng, Global Mass Transit Systems.
Viegas, J. M., & Macario, R. (1998b). Public transport regulation as an instrument of urban
transport policy. 5th Conference on Competition and Ownership in Land Passenger Transport,
Presented in May, Leeds, UK.
Viegas, J. M., & Macario, R. (1998c). Ainda faz sentido a intervenc- ao do Estado no
Transporte Urbano? Congresso Nacional de Sociologia, Marc- o, Lisboa, Portugal.
Viegas, J. M., & Macario, R. (1998d). Criterios de calidad para planicacion del transporte
publico urbano. 31 Congresso de Engenharia de Transportes, Marc- o, Barcelona, Espanha.
Viegas, J. M., & Macario, R. (1998e). The role of public transport regulation for the
improvement of the urban environment. Wessex Conference, August, Lisbon, Portugal.
Viegas, J. M., & Macario, R. (1998f). Pricing and nancing schemes in the different regulatory
and organizational frameworks for urban transport system in Europe. Conference TERA
2000, October, Milan, Italy.
Viegas, J. M., & Macario, R. (1998g). Legal and regulatory option to promote system
integration in urban public transport. 8th WCTR Conference, May, Antwerp.
Viegas, J. M., & Macario, R. (2003). Acceptability of price changes in urban mobility of
transport pricing strategies (Chapter 11, edited by J. Schade & B. Schlag). Oxford: Elsevier.
ISBN-13: 978-0080441993, ISBN-10: 0080441998.
Viegas, J. M., & Macario, R. (2003a). Involving stakeholders in the evaluation of transport
pricing. In: A. Pearman et al. (Eds), Transport projects, programmes and policies: Evaluation
needs and capabilities (pp. 213225). Lisboa Portugal: Ashgate Publishing Company.
Viegas, J. M., & Macario, R. (2003b). Acceptability of price changes in urban mobility. In:
J. Schade & B. Schlag (Eds), Acceptability of transport pricing strategies. Oxford: Elsevier.
Vieira, J. (2005). Aplicac- ao de Instrumentos com objectives ambientais no sector dos
transportes: Da intervenc- ao regulamentar a` multi-instrumentalidade. Dissertac- ao de
Mestrado em Transportes, Instituto Superior Tecnico, Universidade Tecnica de Lisboa.
Von Neumann, J., & Morgenstern, O. (1947). Theory of games and economic behavior (2nd
ed.). Princeton, NJ: Princeton University Press.
Wachs, M. (1985). Planning, organization and decision-making: A research agenda.
Transportation Research, 19A, 521531.
Bibliography 285
Walter, J. (2003). In: Hensher D. A. (Ed.). (2005). Competition and ownership in land passenger
transport selected papers from the 8th international conference on competition and ownership.
Weaver, W. (1948). Science and complexity. American Scientist, 36, 536544.
Weaver, W. (1949). Translation. Repr. In: W. N. Locke & A. D. Booth (Eds), Machine
translation of languages: Fourteen essays (pp. 1523). Cambridge, MA: Technology Press of
the Massachusetts Institute of Technology, 1955, in Warren Weaver Memorandum, July
1949, published in MT News International, no. 22, July 1999, pp. 56, 15.
Weber, M. (1968). Economy and society (Vol. 1). New York: Bedminster Press.
Wegener, M. (2002). State of the art of land-use and transport research. Workshop Towards
sustainable urban mobility, co-organised by the EU parliament and the European
Commission, 31 January, Brussels.
Weick, K. (1965). The social psychology of organizing. Reading, MA: Addison-Wesley.
Weinberg, G. M. (2001). An introduction to general systems thinking (Silver Anniversary
Edition). New York: Dorset House Publishing.
Weingast, B. W., & Marshall, W. (1988). The industrial organization of congress; or, why
legislatures, like rms, are not organised as markets. Journal of Political Economy, 96,
132163.
Whitehead, A. N. (1925). Science and the modern world. Cambridge: Cambridge University
Press.
Whitehead, A. N. (1925). Science and the modern world (Lowell Lectures 1925 (1925). Alfred
North Whitehead Publisher: The Macmillan Company, London, UK.
Wiener, N. (1961). Cybernetics. New York: Wiley.
Williamson, O. E. (1985). The economic institutions of capitalism. New York: Free Press.
Williamson, O. (1975). Transaction cost economics: An overview. Available at http://
organizationsandmarkets.les.wordpress.com/2009/09/williamson-o-transaction-cost-
economics-an-overview.pdf and Klein, Peter, Joseph T. Mahoney and Jackson A. Nickerson.
(2005). Williamsons transaction costs economics: Theoretical contributions and empirical
evidence. Unpublished manuscript. University of Missouri, Columbia.
Williamson, O. E. (2000). The new institutional economics: Taking stock, looking ahead.
Journal of Economic Literature, 38(September), 595613.
World Bank (1996). Sustainability transport Priorities for policy reform. Washington, DC:
The World Bank.
Yates, D. (1977). The ungovernable city: The politics of urban problems and policy-making.
Cambridge, MA: MIT Press.
Yin, R. (1994). Case study research: Design and methods (2nd ed.). Thousand Oaks, CA: Sage
Publishing.
Yukl, G. A. (1974). Effects of opponents initial offer, concession magnitude, and concession
frequency on bargaining behavior. Journal of Personality and Social Psychology, 30,
332335.
Zannetos, Z. S. (1984). Decision sciences and management expectations. In: J. P. Brans (Ed.),
Operational Research 84 (pp. 6976). Amsterdam: North Holland.
Zeithaml, V. A., & Bitner, M. J. (2003). Services marketing: Integrating customer focus across
the rm (3rd ed.). New York, NY: McGraw-Hill.

Managing Urban Mobility Systems

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Managing Urban Mobility Systems

Diunggah oleh

Hak Cipta:

Format Tersedia

MANAGING URBAN

United Kingdom North America Japan

First edition 2011

Copyright r 2011 Emerald Group Publishing Limited

Reprints and permission service

British Library Cataloguing in Publication Data

Emerald Group Publishing

List of Figures xiii

2. A Theoretical Framework for the Management of Urban Mobility

3. Simplifying Complexity in Urban Mobility Systems 79

4. Conguration of Quality Factors in Urban Mobility Systems 137

5. How to Set Up a Management Model for Urban Mobility Systems 197

6. The Need for a Strategic Approach to Urban Development 259

1.1. Evolving urban societies

The concept of urban involves a series of interrelated dimensions, among which:

extended concept of conurbation by dening its limits according to the scope

Figure 1.1: Urban mobility dynamics (conceptual diagram). Source: Author.

considering the interactions and interdependencies accruing from this three-tier

Robustness, meaning long-term stability and sustainability.

The evolution of information and communication technologies substantially changed

Les individus apparaissent ainsi comme socialement multi-appartenants,

Individualism can be considered as the self-view of the world, ltered by the

Table 1.1: Government versus governance.

Old logic Feature New logic

Expand Networks Manage and integrate

economic, social, and environmental sustainability while keeping an integrated

knowledge: strategic denition, conguration, and delivery of supply and steering

unclear and unt legal and regulatory frameworks;

unclear and/or unt application of subsidiarity and proportionality principles

1.2. Policy and management problems in Urban Mobility

high congestion problems in urban areas;

The combination of these factors clearly contributed to change environment where

it as a basic requirement in the selection of operators,21 aiming with it to provide

Pricing should be seen as a complement of regulatory and other market policies.

The operational interpretation of the previous statements should reect a price

In future, charging systems should be based on the user pays principle,

meeting individual requirements at minimum resource costs, that is market and

However, as we have seen in previous section, the denition of an Urban Transport

Figure 1.4: Decision levels-conceptual diagram. Source: Author.

The denition of objectives, and guidelines to achieve them, is a function

Transport dimension adequate balance between modes and means of transport,

Trade-offs among these four domains are conditioned by the socioeconomic

European policies aiming at an improved efciency and effectiveness of the

1.3. Quality as a driver to enhance performance of UMS

increasing domestic and global competition;

general trend toward more quality in the economy;

to set a working group (CEN TC 320 WG 5 Public Transport and Intermodality) to

implementation of strategic quality management, including market segment

Quality management implies the increased focus on individual customer satisfac-

Figure 1.6: Customer as borderline line sensor of company performance.

Figure 1.7: Conceptual dynamics for a quality approach to UMS.

Quality of the services provided.

We partially support the arguments listed by House, in particular in what concerns

ISOTOPE Improved structure and Organisation of urban transport

System management theories.

Table 1.3: Subsample of cities observed for development of inductive method.

Country Cities Country Cities

Australia Adelaide Italy Rome, Milan,

MARETOPE Managing and Assessing Regulatory Evolution in urban

CIVITAS program (http://www.civitas-initiative.org, 14-07-2005), that involves the

regime for market access, interpretation of public service concept, socioeconomic

Country Cities Country Cities

Australia Adelaide Ireland Dublin

In addition to this theoretical validation two empirical works for application of