Anda di halaman 1dari 399

IRAN'S POTENTIAL AS A LANDBRIDGE FOR FORMER USSR REPUBLICS: A SCENARIO APPROACH

by Parviz Bavarsad Ahmadi

A thesis submitted to the University of Plymouth


in partial fulfilment for the degreeof

DOCTOR OF PHILOSOPHY

of Marine Studies Taculty of Science University of Plymouth CollaboratingEstablishments: The Programmingand Budgeting Organisation, and the Central Bank (Bank Markazi) of the Islamic Republic of Iran.

June1997

Copyright

Notice

This copy of the thesis has been supplied on the condition that anyone who consults it is understood to recognise that its copyright rests with the author and that no quotations from the thesis and no information derived from it may be published without the author's prior

written consent.
This thesis may be made available for consultation within the University of Plymouth Library, and may be photocopied or lent to other libraries for the purpose of consultation.

Signed fkrv

Parviz Bavars

ii

ABSTRACT

Iran's potential as a landbridge for Former USSR republics: a scenario approach PARVIZ BAVARSAD AHMADI
The concept of a landbridge refers to different types of integrated origin-destination international movements of shipments (in various combinations of sea, land and air) under a single waybill. There are different examples of landbridges with different characteristics related to transport supply facilities, organisational structure and managerial skills. Certain limitations of existing landbridge studies are discussed. They include the failure of a comprehensive academic study to account jointly for both demand and supply of landbridge services. Most articles on landbridges discuss the Trans-Siberian Railway or east-west coast landbridges of the United States of America. No comprehensive academic studies of landbridges in general were found. The main features of the researchcan be summarised as: *A comprehensive review of literature related to landbridges * An investigation and analysis of Iranian transport supply and demand including both domestic and foreign trade. * An investigation and analysis of the demand of the Central Asian and Caucasuscountries (Former USSR republics) for transport * The development of a demand and supply model related to an Iranian Sea-landbridge (ISLB) for eight Central Asian and Caucasuscountries and Iran. * Evaluation of the impacts of demand on landbridge supply. *A comprehensive review of the scenario approach and its application to the Iranian Sea landbridge study using a regression technique. Three scenarios are developed (optimistic, most probable and pessimistic). The main result of the scenario modelling suggests that the transport system of Iran requires considerable improvement to compete effectively with other landbridges, given an increase in trade from Iran and the Central Asian and Caucasus countries.

iii

Contents:
I
1.1 1.2 1.3 14 . 1.5

Page

Statement of the problem and objectives 1 ...............00


Background ................................................................ Significance of the problem .......................................... Purposeof the study ...................................................... Previous research ......................................................... Thesis structure ............................................................ 1
2 2 3 3 5

2
2.1 2.2

Iranian trade and transport ...................................


Introduction ............................................................... Economy of Iran ........................................................ 2.2.1 Geographical features

5 5 5 6 6

2.3

Transport

............................................................... 2.2.2 Demographic and administrative characteristics ....................... 2.2.3 Economic background and composition ..............................

2.3.1 9 .............................................................................. 2.3.2 Road transport 10 ........................................................................... 2.3.2.1 Road networks and standards 10 ............................................ 2.3.2.2 Road surfacesin Iran 13 .......................................................... 2.3.2.3 Road fleets: ownership and use 13 .......................................... 2.3.2.4 Road freight rates and financing 16 ......................................... 2.3-2.5 Importance of transport modes to GDP 17 ............................ 2.3.3 Rail transport 18 ............................................................................. 2.3-3.1 Organisation and development 18 ............................................ 2.3-3.2 Network characteristics 19 ....................................................... 2.3.3.3 Locomotives and rolling stock 20 ............................................. 2.3.3.4 Operations of the Islamic Republic of Iran Railway 21 Company (HURC) .................................................................. 2.3.3.5 Rail productivity 24 .................................................................. 2.3.3.6 Rail traffic and frequency 25 ........................................... ......... 2.3.3.7 Costs and revenuesof IRIRC 26 ............................................. 2.3.4 Port transport 26 .............................................................................. 26 2.3.4.1 Administration of ports ........................................................ 2.3.4.2 The port authority directorates in coastal provinces 27 ............ 2.3.4.3 Infrastructure 27 ........................................................................ 2.3.4.4 Port storagefacilities in Iran 29 ............................................... 2.3.4.5 Port equipment 30 .................................................................. 2.3.4.6 Imbalance of international seatradein ports 31 ......................... 2.3.4.7 Performancesof ports 35 ......................................................... 2.3.4.8 Productivity and service of ports in Iran 37 ............................ 2.3.4.9 Waiting times and demur-rage 40 ............................................ iv

.................................................................. .. Organisation

2.3.4.10Maritime andland traffic in Iranianseaports 42 ..................... hinterland in


2.3.4.11 Port Iran

2.4 3
3.1

2.3.6.1 Islamic Republic of Iran Shipping Line (IRISL) 45 .................... 2.3.6.2 National Iranian Tanker Company (NITC) 47 ......................... 2.3.7 Modal split 47 .................................................................................. 2.3.7.1 Domestic trade 48 ...................................................................... 2.3.7.2 Foreign trade 51 ........................................................................ 51 Conclusions .........................................................................

2.3.5 Border crossing operations 44 ......................................................... 2.3.6 Sea transport system 45 ...................................................................

.........................................................

44

Central Asian and Caucasus(CAC) countries

53

3.2 3.3

53 Political geography of CAC countries ............................. 3.1.1Caucasus countries............................................................... 55 56 3.1.2 CentralAsian countries ...................................................... 57 Economies of CAC countries ............................................... 3.2.1 Caucasus countries.............................................................. 58 3.2.1 CentralAsian countries 59 .......................................................... 60 Foreign trade of the CAC countries ........................... 3.3.1Caucasus countries............................................................... 61
3.3.1.1 Azerbaijan ......................................................................... 3.3.1.2 Armenia ........................................................................ 3.3.1.3 Georgia ............................................................................ 3.3.2 Central Asian countries ........................................................ 3.3.2.1 Gyrkyzstan ........................................................................ 3.3.2.2 Kazakhstan ..................................................................... 3.3.2.3 Tadjikistan ..................................................................... 3.3.2.4 Turkmenistan .................................................................. 3.3.2.5 Uzbekistan ......................................................................

3.4

Transportin CAC countries ........................................... 3.4.1 Transport in Caucasus


3.4.1.1 Azerbaijan .............. ..................................................... 3.4.1.2 Armenia .......................................................................... 3.4.1.3 Georgia ..............: *,*"********'*""*"**"*"****** -* .... **--3.4.2 Transport in Central Asia ....................................................... 3.4.2.1 Gyrkyzstan .................................................................... 3.4.2.2 Kazakhstan ...................................................................... 3.4.2.3 Tadjikistan ....................................................................... 3.4.2.4 Turkmenistan ................................................................. 3.4.2.5 Uzbekistan .......................................................................

countries ............................................

3.5 3.6 3.7

Trade with Iran .......................................................... Economic Co-operation Organisation (ECO) ............ Conclusions ...............................................................
V

61 63 64 65 65 67 69 71 73 75 76 77 78 78 78 79 79 80 80 81 81 82 84

4
4.1
4.2

Landbridges: concept and practice 0 ........................


Introduction
The concept of a landbridge
4.2.1.1 Definition 4.2.1.2 Definition
4.2.1.3 Definition 4.2.1.4 Definition 4.2.1.5 Definition

85
85
85
85
88 90
90 91 91

......................................................................
..............................................
of a minibridge ..................................................... of a microbridge ..................................................
................................................ ......................................... .............................................

4.2.1 Definition of a landbridge ......................................................


of a sea-air bridge

of a air-road-airbridge of a sea-river bridge

4.2.2

The benefits of landbridges

4.3

Influences on landbridge development ............................ 4.3.1 Geographical influences


4.3.2 4.3.3
4.3.4

..................................................

92

94
94 95 95
96

Political Technical

influences influences

....................................................

............................................................ .........................................................
................................................

Organisational

influences

4.4

4.4.3 North American


4.4-3.1 4.4.3.2 4.4.3.3 4.4.3.4 Political

.............................................................. Geographical 4.4.1.1 characteristics .......................................... 4.4.1.2 Political characteristics ................................................... 4.4.1.3 Technical characteristics ................................................ 4.4.1.4 Organisational. characteristics ......................................... 4.4.2 Trans-Siberian Railway ....................................................... 4.4.2.1 Geographical characteristics ........................................... 4.4.2.2 Political characteristics ................................................... 4.4.2.3 Technical characteristics ........................................... 4.4.2.4 Organisational .................................................................

Examples of landbridges .................................................. bridges 4.4.1 Europe-Asia

96
98 98 101 102 103 105 105 107 107 108

landbridges

Geographical Technical

.................................................
........................................... ...................................................

109
110 ill. 114 116

characteristics

characteristics

characteristics ................................................. Organisational characteristics .........................................

4.5 5 5.1 5.2

Conclusions

................................................................... ................................. 00

117 118 118 118


119 122 124 129

The Iranian Sea-landbridge

5.3

characteristics ................................................ 5.2.2 Political characteristics ........................................................ 5.2.3 Technical characteristics ....................................................... 5.2.4 Organisational. characteristics ...............................................

Introduction ............................................................... Reasonsfor developing a landbridge system .............. 5.2.1 Geographical

131 Identification of the CAC countries' foreign trade routes and ISLB competitord. ........................................................ Vi I

5.4

5.3.1 European markets via Georgian and Turkish ports ................. 5.3.1.1CaspianSea-bridge .......................................................... 5.3.1.2Rail and road route to Batumi ........................................... 5.3.2 Indian Ocean and African markets via Pakistan and India ...... 5.3.3 Far East via TSR and TCR ..................................................... 5.3.4 Identification of ISLB competitor landbridges ...................... 5.3.5 Comparison of ISLB with competitor landbridges ................. 5.3.5.1Origin and destinationdistances landbridges of .................. 5.3.5.2Technologyand storagefacilities ...................................... 5.3.5.3Organisational. structureandproductivity ............................ Conclusions ................................................................

131 133 133 136 138 139 140 140 143 146 150

Scenario approach and ISLB conceptual model development ............................................................

151

6.1 6.2 6.3

6.4

151 Introduction ................................................................ Planning and analysis under uncertainty 151 .................... The scenario approach 153 ............................................... 6.3.1 Scenariocharacteristics 153 ....................................................... 6.3.2 Scenarioelements 156 ................................................................ 6.3.3 Scenarioefficiency and effectiveness 157 ..................................... 6.3.4 Scenariodimensionsand types 158 .............................................. 6.3.5 Scenarioranking, probabilities andcriteria 160 ........................... 6.3.6 Scenarioanalysistechniques 161 ................................................. 6.3.7 Scenariodevelopmentand construction 163 ................................ 6.3.8 Strategyformulation 168 .............................................................. 6.3.9 Examplesof scenarioanalysis 169 ............................................. 6.3.9.1 Examples scenario 170 analysis transport of of ........................ 172 Development of the scenario themes for the ISLB study..

6.5

Relevance of the scenarios approach to the ISLB Study ...


6.5.2 CAC and Iranian volume of demand for transport .................. 6.5.2.1 Geographical ..................................................................... 6.5.2.2 Political ............................................................................. 6.5.2.3 Organisational .................................................................... 6.5.3 Iranian supply of transport ..................................................... 6.5.3.1 Geographical ..................................................................... 6.5.3.2 Political ............................................................................. 6.5.3.3 Technical ............................................................................ 6.5.3.4 Organisational .................................................................... 6.5.4 Output ................................................................................... Vii

173
174
176 176 179 185 185 185 186 188 192 193

6.5.1 ISLB Model uncertainties .....................................................

6.6

Conclusions

......................................................................

193

7.
7.1 7.2

Operationalisation

the ISLB model ................. of

194
194 194
194
196

Introduction ............................................................... ISLB study characteristics ..........................................

7.2.1 Time period and level of aggregation ....................................

7.3 7.4

Description of ISLB methodology and data sources ...


7.3.1 Data sources ......................................................................... 7.4.1 The variables ....................................................... 7.4.1.1 General cargo and oil products of foreign trade of Iran (GOPFF) .......................................................................... 7.4.1.2 Domestic trade of Iran (DOMTI) ..................................... 7.4.1.3 Central Asian and Caucasus foreign trade (CACFF) .......... 7.4.2 Selection of independent variables ........................................ 7.4.2.1 Geographical characteristics .............................................. 7.4.2.2 Political characteristics ...................................................... 7.4.2.3 Technical characteristics .................................................... 7.4.2.4 Organisational characteristics ............................................

7.2.2 Data measurement and level of accuracy ..............................

197
197

Selection of variables ...................................................... dependent

197
197 198 202 206 208 209 212 219 227

7.5 8 8.1 8.2 8.3

Conclusions

.................................................................

233
234

Scenario analysis of the ISLB demand ................ *00 Introduction ................................................................ Objectives and regression analysis ........................... ISLB model estimation ..............................................

234 234 239


239 243 245 246 246 250 254 255 255 256
262

8.4

8.3.1 GOPFF ................................................................................ 8.3.2 DOMTI ............................................................................... 8.3.3 CACFF ............................................................................... Interpretation of the results .......................................... 8.4.1 GOPFI . ................................................................................ 8.4.2 DOMTI ................................................................................ 8.4.3 CACFr ................................................................................ descriptions ISLB scenario characteristics and ......... 7.5.1 Justification and choice of ISLB scenarios ............................ 7.5.2 ISLB Scenario descriptions ....................................................
8.5.2.1 ISLB most probable scenario .......................................... viii

8.5

8.5.2.2 ISLB optimistic scenario ................................................. 8.5.2.3 ISLB pessimistic scenario ................................................

263 263

8.6

Model forecasts

.............................................................

263
264 264 267 269

8.6.2 Optimistic scenario ...............................................................


8.6.2.1 GOPFr

8.6.1 Most probable scenario ......................................................... 8.6.1.1 GOPFF ............................................................................. 8.6.1.2 DOMTI ............................................................................ 8.6.1.3 CACFr .............................................................................

270
270 272

8.6.2.2DOMTI
8.6.2.3 CACFr

.............................................................................

271 ............................................................................. .............................................................. 273


274

8.6.3 Pessimistic scenario 8.6.3.1GOPFF


8.6.3.2 DOMTI

.............................................................................

273 .............................................................................. 275 ..............................................................................


.............................................................................

8.6.3.3CAM

8.7

Conclusions

.....................................................................

276

9
9.1
9.2

ISLB Scenarios: modal split and distribution


Introduction .....................................................................

277
277

ISLB demand scenario distribution and modal split 278 ......... 9.2.1 Individual and regional sharesof CAC countries in 280 CACFr ............................................................................... 9.2.2 Justification of ISLB interfaces and distribution for 282 combined GOPFT and CACFr ............................................. 9.2.3 Distribution of GOPFT and CACFT to ports and border 285 crossings .............................................................................. 9.2.3.1Distribution of the combinedGOPFT and CACFT to 287 ports .................................................................................. 9.2.3.2Distribution of the GOPFF and CACFT to bordercrossings 292 9.2.4 Total ISLB demands modal split 294 ........................................... 9.2.4.1Modal split of the combinedCACFT and GOPFT demand. 297 9.2.4.2Real DOMTI modal split 298 .................................................. 9.2.4.3Total demandand modal split for the Iranian transport 299 system................................................................................... 301 Future Iranian transport supply requirements ................... 9.3.1 Port traffic and capacity 302 ........................................................... 9.3.1.1Maritime traffic of GOPFT and CACFT 302 .............................. 9.3.1.2 Futureport capacityof Iran
ix

9.3

...............................................

303

307 9.3.1.3Requiredlength of berthsfor combinedGOPFF and CACFr ................................................................................. 309 9.3.1.4Requiredstorageareasfor combinedGOPFF and CACFI. 9.3.1.5Requiredinland road and rail equipment............................. 311 9.4 Conclusions 314 .................................................................

10 10.1

Conclusions Introduction

and recommendations

..................

315 315

..................................................................

10.2 Changesrequired for the Iranian transport system ......... 10.3 Landbridge demand and supply
10.4 10.5 10.6

316 317
318 320 322
322
323 324 324 325 326 326 326 327 327 327 328 329

.....................................

Transport structure in Iran .............................................. Main finding of the scenario analysis ............................ Recommendations ..........................................................

10.6.1 Recommendations to the Iranian government ......................


10.6.1.1 Recommendations to PBO 10.6.1.2 Recommendations to MRT 10.6.1.3 Recommendations to PSO ............................................... .............. ..................................

.............. .................................... 10.6.1.4 Recommendations to IRIRC ................................................ 10.6.1.5 Recommendations to IRISL .............................................. 10.6.1.6 Recommendations to Customs administration .................... 10.6.2 Recommendations to Iranian transport operators ................. 10.6.3 Recommendations to ECO ...................................................

1 10.6.4 Recommendations to for further research ............................. 10.6.4.1 Recommendations for research into the ISLB ................. 10.6.4.2 Recommendations for research into landbridges in general References .................................................................... 0

Appendices

.................................................................. 000 376 List of the Iranian Institutions .................................. 0

346

List of Figures
Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 2.1 2.2 2.3 3.1 4.1 4.2 4.3 4.4 4.5 4.5a Location and political administration of Iran in south west Asia. The state structure of transport in the Islamic Republic of Iran. The Islamic Republic of Iran railway network in 1995. Location of the Caucasus and Central Asian Countries in world. General concept of a landbridge. General concept of a minibridge. General concept of a microbridge. General concept of a sea-air-bridge. Most important landbridge countries and ports in the world Intra-European rrnibridge and microbridge, TSR landbridge sea leg between port Tunis and Vostochney (unit: km). Euro-Israel landbridge between Greece and Djibouti in Africa, and Rotterdam-Marseille-Mesin Tehran (unit: km). Trans-Chinese rail landbridge via Central Asia and CIS to Europe (unit: km) Trans-Siberian rail landbridge routes (unit: km). North American rail landbridge and minibridges between Far East, Europe, and Asia (unit: kin). Iran Sea-landbridge land and sea legs, ports and border crossing terminals in the south and north (km). Road corridors between port of Imam and Abass and northern Iran's port and border crossings. Different Iranian road landbridge and minibridge alternatives to CAC countries (km). Rail rninibridge and microbridges between southern ports of Imam and Abass, and CAC countries (km). CAC routes to European markets through Georgian and Turkish and Iranian ports, and Caspian Sea Bridge (unit: km). Pakistani and Indian alternative routes via Karachi and Bombay ports to CAC countries (km). Corporate strategy system. The broad structure of the ISLB conceptual model. The volume of general cargo and oil products (foreign trade) of Iran (000 tonnes). Scattergram,of the actual volume of the domestic production of Iran (000 tonnes). Share of different sectors in the volume of domestic production of Iran. Total volume of the CAC Asian Foreign Trade (tonnes). Growth of the population of Iran. Land under cultivation and fertiliser used in Iran. GDP per capita, population and final GDP in Iran (based on fixed prices 1982 Iranian Rials). Oil revenues in Iran Value added of all major industries and volume of GOPFT of Iran. Gross national investment of Iran: Private sector, government, and transport share at 1982 fixed prices. Goods and crude oil exports and imports of Iran at 1982 fixed prices. Employment in Iran. Sectoral contributions to employment in Iran. Locomotive numbers in Iran. Train wagon utilisation in Iran. xi

Page 7 10 19 54 86 89 90 90 97 98 99 100 106 112 120 126 129 132 135 137 169 174 202 204 205 208 211 211 213 215 216 217 218 218 219 220 221

Figure 4.6 Figure 4.7 Figure 4.8 Figure 4.9 Figure 5.1 Figure 5.2 Figure 5.3 Figure 5.4 Figure 5.5 Figure 5.6 Figure 6.1 Figure 6.2 Figure 7.1 Figure 7.2 Figure Figure Figure Figure Figure 7.3 7.4 7.5 7.6 7.7

Figure 7.8 Figure 7.9 Figure 7.10 Figure Figure Figure Figure Figure 7.11 7.12 7.13 7.14 7.15

Figure 7.16 Figure 7.17 Figure 7.18 Figure 7.19 Figure 7.20 Figure 7.21 Figure 7.22 Figure 7.23 Figure 8.1 Figure 8.2 Figure 8.3 Figure 8.4 Figure 8.5 Figure 8.6 Figure 8.7 Figure 8.8 Figure 8.9 Figure 9.1 Figure 9.2 Figure 9.3 Figure 9.4 Figure 9.5 Figure 9.6 Figure 9.7 Figure 9.8 Figure 9.9 Figure 9.10

Numberof wagonsandheavygoodsvehiclesof Iran Value addedof threemodesof transportin Iran. Value addedof the agriculturalsectorin Iran. Components annualaverage of stayingdaysfor six major portsof Iran. Components annualaverage of servicedaysat berthsfor six major ports of Iran. Specificport productivity of the six Iranianmajor seaports. Productivity of six major ports of Iran in terms of tonnesfor non-oil sea tradeper PSOemployee. in Freightton es/peremployee IRIRC. frameworkof a forecastingsystem. Conceptual Scattergrarn the GOPFTof Iran by populationduring 1979-1993. of Scattergrarn the GOPFIof Iran by final GDP during 1979-1993. of Plot of the predictedresidualsagainstactualtonnage the GeneralCargo of andOil ProductForeignTradeof Iran during 1979-1993. Scattergram the GOPFI' of Iran by valueaddedof agriculturesectorduring of 1979-1993. Scattergrarn the DOMTI of Iran by GDP during 1979-1993. of Scattergram, the DOMTI of Iran by the productivity of the agricultural of sectorduring 1979-1993.. Plot of the predictedresiduals againstactualtonnage the DOMTI of Iran of during 1979-1993. DOMTI forecasts Iran 1979-1993. for ISLB modelof distribution andassignment Distribution of the threescenarios combinedGOPFTandCACFT for forecasts over ports. Distribution of the threescenarios combinedGOPFTandCACFT for forecasts over bordercrossings. Comparisons threecombinedGOPFTandCACFT scenario forecasts for of the southern portsof Iran. Comparisons the threeGOPFTscenarioforecasts the two northern for of portsof Iran. Comparisons four southern of portsof Iran for the threecombinedGOPET forecasts 2005. in andCACFT scenario Comparisons the threeGOPFTscenarios the two northernports of Iran for of in 2005. Modal split of threescenarios GOPFTin 2005. for Total future forecastvolume(GOPFI, DOMTI, andCACFT) for Iranian transportnetworksand interfaces. Development the nominal(design)port capacityin Iran. of

222 224 225 228 228 231 232 233 235 247 248 250 252 252 253 254 269 279 287 287 288 289 289 289 296 3M 304

xii

List of Tables
Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 2.5 Table 2.6 Table 2.7 Table 2.8 Table 2.9 Table 2.10 Table 2.11 Table 2.12 Table 2.13 Table 2.14 Table 2.15 Table 2.16 Table 2.17 Table 2.18 Table 2.19 Table 2.20 Table 2.21 Table 2.22 Table 2.23 Table 2.24 Table 2.25 Table 2.26 Table 2.27 Table 2.28 Table 2.29 Table 2.30 Table 2.31 Table 2.32 Table 2.33 Table 2.34 Table 2.35 Table 2.36 Table 2.37 Table 2.38 in Shareof oil and gas,andtax in revenues Iran during 1989-1993 Sectoralcontributionsin the economyof Iran, 1982fixed prices(million Rials) National andprovincial roadsof Iran underthe supervision Ministry of of RoadsandTransportin 1992(km). in Compositionshares main roadsof Iran in 1992. Compositionandlengthof the secondary roadsof Iran in 1992(km). Shareof 24 provincesin all nationalandprovincial roadsunderthe supervision (MRT) 1992. of the Ministry of RoadsandTransportation Pavedroad surfaceconditionsin Iran after the war in 1988. Developmentof different types of motor vehiclesin Iran during 1979-1992. Growth of road vehicle carriers in Iran during 1963-1987. Number and typesof all freight trucks in Iran in 1987. The numberof goodsvehiclesgreaterthan 10 tonnesaccordingto type and carrying capacityin Iran in 1987. Compositionand numberof road vehiclesin Iran in 1991. Shareof different road truck fleet capacitiesin 1994(capacity:tonnes) Improvementand development rail tracks in Iran during the First of EconomicDevelopmentProgramme (km) Number of the locomotivesfor main routesand shuntingservicesof the IRIRC (unit). The developmentof freight railcars in Iran. Compositionof eight major typesof freight railcarsof IRIRC in 1990. Compositionand numberof the eight major types of domesticmanufactured freight railcars of IRIRC during 1984-1993. Volume and ton/km freight carried by IRIRC (Unit: Freight 000 tonnes). Volume and compositionof domesticand foreign tradeof Iran by rail (000 tonnes). Compositionand volume of the major types of IRIRC freight in 1991(000 tonnes). Volume of the freight handledin 12 rail districts of Iran (000 tonnes). Operationalproductivity of the IRIRC for passenger freight km and and traffic (units: million otherwisespecified). Daily movementsof train along five major IRIRC networksin 1992. Seasonal train traffic in Iran (million tonnes) Causes propertiesof train delaysin Iran in 1992 and Compositionof costsand revenues IRIRC, million Iranian Rials of Revenuecompositionin IRIRC during 1986-1990,million Iranian Rials. The infrastructureof six major Iranianseaports. Facilities of different seaports Iran in different berthsin 1993. of Storage of andtypesin seaports Iran in 1993(sq. metres). capacity /metres). Shareand typesof storage of of areas major seaports Iran in 1993(sq. Main mobile and fixed port authorityequipmentin Iran in 1993. Shareof different portsin main cargohandlingequipmentin 1993. Iraniannon crudeoil seabome andtotal foreign trade(000 tonnes). Volume and compositionof the Iranian imports andexports(000 tonnes). Iranian seaandtotal import/ export tradeof non crudeoil foreign trade(000 tonnes). Compositionof tonnage shareof the seatradethroughmajor Iranianports and (importsandexports)for 1993(000 tonnes).
xiii

Page
8 9 10 12 12 12 13 14 14 15 15 15 16 20 20 21 21 21 22 23 23 24 24 25 25 25 26 26 28 29 29 30 30 31 31 33 34 34

Table 2.39 Table 2.40 Table 2.41 Table 2.42 Table 2.43 Table Table Table Table Table Table Table Table Table Table Table Table Table 2.44 2.45 2.46 2.47 2.48 2.49 2.50 2.51 2.52 2.53 2.54 2.55 2.56

Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table Table Table Table Table Table 3.5 3.6 3.7 3.8 3.9 3.10

Table 3.11 Table 3.12 Table 3.13 Table 3.14 Table 3.15 Table 3.16 Table 3.17 Table 3.18 Table 3.19 Table 3.20

35 Composition of throughput for the port of imam for 1993 (000 tonnes). 35 Composition of throughput for the port of Abass for 1993 (000 tonnes). 36 Composition of imports and exports in ports of Bushehr, Chah Bahar Anzali, Nooshahr for 1993 (000 tonnes). 38 General port productivity of Iran's six Major seaports during 1988-1993 (000 tonnes except where otherwise specified). Average port turnaround time of ships visiting seven major Iranian ports in 39 1992 (days). Comparison of the ship indicators in Iran (unit: day). 40 Share of different main basic imported commodities in demurrage during 1992 41 Imports and demurrage in Iran 41 Maritime sea going traffic visiting Iranian major ports. 42 Maritime traffic in major Iranian ports (Ships/day). 43 Container ship traffic in Iranian ports (unit: ship). 43 Number of heavy goods vehicles and wagon traffic into Iranian seaports. 43 Non-oil imports of Iran (000 tonnes). 45 Growth and voyages of IRISL own and chartered ships. 46 46 Composition and dwt of IRISL fleet in 1990. Share of the IRISL and its chartered ships in the imports of Iran (000 tonnes). 47 49 Domestic productions and modal contributions in Iran (000 tonnes). Estimate of modal shares of the general cargo and oil products foreign trade 51 in Iran (000 tonnes). Caucasus countries GDP before and after independence ($ in). 58 Comparison of sectoral shares (%) in the economy of the Caucasus countries 59 in 1991. CAC GDP before and after independence ($ million) (Purchasing power 59 parities). 60 Comparison of sectoral shares (%) in the economy of the Central Asian countries in 1991 (current prices). Values of inter-republic and international trade of Azerbaijan (Rb: million). 61 62 Composition of the Azerbaijan inter-republic export/import (Rb: million). 62 Composition of the Azerbaijan international export/import trade (Rb: million). 62 The share (%)_of main foreign trade partners of Azerbaijan in 1990. 63 Value of foreign trade of Armenia (Rb: million, current prices). 63 Composition of the import/export trade in Armenia (Rb: million, current prices). 63 Share of European countries, Asia and American continent in the foreign trade of Armenia in 1990 (% of total in foreign trade Roubles). 64 Value of foreign trade of Georgia (Rb: million). 64 Composition of Georgia's foreign trade (Rb: million). 65 Share of the major European, Asian and American countries in the foreign trade of Georgia in 1990 (%_Oftotal foreign trade in Roubles). 65 The Gyrkyzstan inter-republic export/import trade (Rb: million). 66 Composition of the Gyrkyzstan inter-republic export/import trade (Rb: million). 6 Composition of the Gyrkyzstan external export/import trade (Rb: million). 67 Major sharesof the International export/import of the Gyrkyzstan in 1990 (% of the total at foreign trade Roubles). 67 Value of inter-republic and international trade of Kazakhstan (inter-republic in Million Roubles market prices and international in $ million). i 1 68 Composition of Kazakhstan's export/import trade (Rb: million; world market xiv

Table 3.21 Table 3.22 Table 3.23 Table 3.24 Table 3.25 Table 3.26 Table 3.27 Table 3.28 Table 3.29 Table 3.30 Table 3.31 Table 3.32 Table 3.33 Table 3.34 Table 3.35 Table 3.36 Table 3.37 Table 3.38 Table 3.39 Table 5.1 Table 5.2 Table 5.3 Table 5.4 Table 5.5

Table 5.6 Table 5.7 Table 5.8 Table 5.9 Table 5.10 Table 5.11 Table 5.12 Table 6.1 Table 6.2 Table 6.3

rices). Shareof the majorEuropean, Asian, andAmerica! countriesin the foreign s in tradeof Kazakhstan 1990(% of total in foreign tradeRoubles:million). I Compositionof Tadjikstanexport/importtrade(Rb: million). C Structureof theTadjikstanInter-republicexport/importtrade(Rb: million). Compositionof the TadjikstanInternationalexport/importtrade(Rb: million). Asian, andAmericancountriesin the foreign Shareof the majorEuropean, tradeof Tadjikstanin 1990(% of total in foreign tradeRb: million). Compositionof Turkmenistan export/importtrade(Rb: million). Inter-Republicexport/importtrade(Rb: million). Structureof Turkmenistan internationalexportlimporttrade(Rb: million). Structureof Turkmenistan The shareof the main inter-republicforeign tradeof Turkmenistan 1990. in in Major shares internationalforeign tradeof Turkmenistan 1990. of in Compositionof the foreign tradeof Turkmenistan 1990(Rb: million; foreign tradeprices). Compositionof Uzbekistan export/importtrade(Rb: million) Export/importstructureof Uzbekistanat the commoditygrouplevel in 1992 (Rb: million, domesticprices). Compositionof Uzbekistan export/importtrade(Rb: million). Main international in export/importtradeof Uzbekistan 1992. Transportinfrastructure superstructure the Caucasus in and countries. Transportinfrastructure superstructurein the CentralAsian countries and ). (kms. Kazakhstan railway performance. Volume of the foreign trade of Iran with CAC countries(000 tonnes). Road and rail distances (km) betweensouthernports of Iran and northern border crossingsservingCAC countries. CaspianSeaeast-west rail-bridge betweenCAC capitalsand port of Batumi in Black Sea(km). Rail-bridge betweenCAC capitals and port of Batumi in Black Sea. Road-bridgebetweenCAC capitals and port of Batumi in Black Seaand St in Petersburg Baltic (km). Rail-bridge distances betweenlandbridgeport of Mersin in East Mediterranean coastsof Turkey and CAC countriesusing CaspianSea landbridgeand Iranian rail (km). Roadand rail bridge distancesbetweenCAC capitalsand four closestport in Black Sea,PersianGulf, and Indian Ocean(krn). Rail-bridge distances betweenCAC capitalsand four main ports in Yellow and JapanSeas(km). Comparisonof ISLB rail and road minibridge systemwith the TSR and TCR landbridgesfrom Kobe, Japan(km). Comparisons the ISLB road and rail transportnetworksand fleet with of major landbridgesof the world. Comparisonof the ISLB ports infrastructureand superstructure with main landbridgeports of the world. Comparisonof the ISLB ports maritime and cargo traffic and productivity with main landbridgeports of the world. Comparisonof the ISLB containerports and terminalsownershipand managementwith other landbridgesand ports of the world. Four basictypesof scenarios. Different scenarioconstructionprocedures. Key ISLB decisionareas.
xv

68 69 70 70 71 71 72 72 73 73 73 74 74 75 75 77 79 80 82 122 13T_ 133 134 134

136 138 142 143 145 147 149 15_ 164 17

Table 6.4 Table 7.1 Table 7.2 Table 7.3 Table 7.4 Table 7.5 Table 7.6 Table 7.7 Table 7.8 Table 8.1 Table 8.2 Table 8.3 Table 8.4 Table 8.5 7a le 8.6 Table 8.7 Table 8.8 Table 8.9 Table 8.10 Table 8.11 Table 8.12 Table 8.13 Table 8.14 Table 8.15 Table 9.1 -fa-ble 9.2

Table 9.3

Table 9.4 Table 9.5

Major uncertainquestionsin the ISLB project. Data sources the ISLB study. of Loading and unloadingoperations(generalcargo, bulk tradeand oil products)in major Iranian ports (000 tonnes). The sevenbroadcategories productsof Iranian domestictrade. of Volume of sevengroupsof domesticproduction and tradeof Iran (million tonnes). Description of the convertedvolume of internationaltradeof the CAC countriesfor Asia (000 tonnes). Description of the selectedvariablesfor the scenarioanalysisof GOPFr, DOMTI and CACFr. Generalproductivity of six major Iranian seaports. in Number of employees IRIRC. Impactsof B coefficient on final GOPFr model. Impactsof B coefficient on final DOMTI model. Description of the threealternativeISLB scenarios demandand supply to of 2005 in Iran and CAC countries. for Most probable,optimistic and pessimisticscenarios ISLB development. Actual and forecastsindependent dependent and variablesof the most probableGOPFr scenarioby growth factor and time seriesprojection. ComparisonbetweenregressionGOPFr forecastand threetime series modelsfor 1994-2005(000 tonnes). Actual and forecastsof the DOMTI independent dependent and variablesby growth factoEand time seriesprojection. Comparisonbetweenregression most probableDOMTI forecastand three time seriesmodelsfor 1994-2005(000 tonnes). CACFI' independent dependent and variablesby growth factor and time seriesprojection (000 tonnes). GOPFr independent dependent and variablesby growth factor and time seriesproje tion for the optimistic scenario(000 tonnes). DOMTI independent dependent and variablesby growth factor and time seriesprojection for the optimistic scenario(000 tonnes). independent dependent CACF17 and variablesby growth factor and time seriesprojection for the optimistic scenario(000 tonnes). GOPFr independent dependent variablesby growth factor and time and seriesprojection for the pesimisticscenario(000 tonnes). DOMTI independent dependent variablesby growth factor and time and seriesprojection for the pessimisticscenario(000 tonnes). CACFI7 independent dependent variablesby growth factor and time and seriesprojection for the pessimisticscenario(000 tonnes). Comparisons the three scenarioforecastsat regional and individual levels of for eight CAC countries(000 tonnes). forecastof Volume and distribution of baseyear and multiple regression for three ISLB scenarios GOPFr and combinedGOPFr and CACFr for year 2005 at interfaces(000 tonnes). Volume distribution of baseyear and three ISLB scenarioforecastsfor for GOPFr and combinedGOPFr and CACF17 year 2005 for six major ports (000 tonnes). Distribution of GOPFr and CACFr three scenarioforecastsfor 2005 at ports and national levels for eight types of trade (000 tonnes). Volume, shareand distribution of baseyear and forecastof threeISLB scenariosfor GOPFT and combinedGOPFT and CACFr for the year 2005
xvi

175 199 200 214 205 207 210 230 232 247 251 256 257 265 266 268 269 270 271 272 273 274 275 276 281 286

290

290 293

Table 9.6

Table 9.7 Table 9.8 Table 9.9 Table 9.10 Table 9.11 Table 9.12 Table 9.13 Table 9.14

Table 9.15 Table 9.16 Table 9.17 Table 9.18 Table 9.19

at ten border crossings (000 tonnes). 298 Volume of modal split for the GOPFr and combined GOPFr and CACFr base year and regression forecasts of three ISLB scenarios for the year 2005 (000 tonnes). 299 Modal split of three real DOMTI in 2005 (000 tonnes). 300 Total demand of three scenarios for Iranian transport system in 2005 (000 tonnes). 301 Total forecast demand for Iranian road and rail transport in 2005 (000 tonnes). 302 Comparisons of three ISLB scenarios with base year situation and their !]2aritime effects on ports (Unit: ships). 304 Capacity and trade in 1993 and 2005 under three scenarios for all six major ports at different types of berths (million tonnes). 307 Capacity and trade in 1993 and 2005 under three ISLB scenarios for every of Iran (million tonnes). major port 309 Estimated length and number of berths for Iranian ports under three ISLB scenarios. 309 Direct and indirect delivery in base year 1993 and three ISLB scenarios for combined GOPFr and CACFT (excluding oil products) in 2005 through Iranian ports (000 tonnes). 310 Covered and open storage areas required under three ISLB scenarios. Number of heavy goods vehicles & rolling stock required in base and horizon 312 years under three scenarios. 313 Import/export rail freight of Iran and CAC countries in 2005 under three ISLB scenarios. 313 Annual and daily number of block train required in base and horizon years for imports nder three demand scenarios for GOPFr and CACFT. 314 Annual and daily number of block train required in horizon years for exports under three demand scenarios for GOPFr and CACFr. II

xvii

Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix

I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

of Iran in the Persian Gulf and the Caspian Sea. The dependent (Y) and independent variable values of the ISLB study for GOPFr, DOMTL and CACFr. Ports Directorates The average main ship indicators of Iranian seaports during 19791993. Matrix of the correlation between GOPFr variables of Iran during 1979-1993. Summary output and ANOVA of the preliminary GOPFr model. General feature of the estimation procedure for the best fit of the GOPFr of Iran. Summary output and ANOVA of the final GOPFT model. Matrix of the correlation between DOMTI variables of Iran during 1979-1993. General feature of the estimation procedure searching for best fit of the DOMTI of Iran. Summary Output and ANOVA Summary output and ANOVA Summary Output and ANOVA of the first estimated DOMTI model. of the second estimated DOMTI model. of the final DOMTI estimated model.

List of Appendices

Page 46 348 351 352 353 354 355 356 357 359 359 360 360 361 361 362 363 364 365 366 367 367 for three 368 370 371 372 374 375 J

Matrix of the correlation between Asian CACFr with GDP per capita and time for 1987-1993. Summary Output and ANOVA of the CACFF estimated model. Residual output of the domestic trade of Iran in the final estimated model 1979-1993. Share of regional ports of Iran in the forecast of the most probable GOPFr and CACFT scenario. Share of regional ports of Iran in the forecast of the optimistic and CACFr scenario. Share of regional ports of Iran in the forecast of the pessimistic GOPFI'and CACFT scenario. Distribution of the three GOPFr border crossings (000 tonnes). and CACFr scenario forecasts on ten GOPFr

Modal split of the most probable scenario for GOPFr. Modal split of the optimistic scenario for GOPFr. Modal split of the pessimistic scenario for GOPFT. Modal split of individual and combined GOPFr and CACFT

scenarios of ISLB study (000 tonnes). Modal split of real DOMTI for three scenarios of ISLB study (000 tonnes). Modal split of total demand for three ISLB scenarios (000 tonnes). Modal split of the total three types of demand three scenarios of ISLB study (000 tonnes). Total and modal demand for transport of Iran under three scenario 1994-2005 (000 tonnes). Maritime traffic characteristics of Iranian ports in base years 1993 and horizon 2005 for three most probable, optimistic and pessimistic GOPFr and CACFr scenarios (000 tonnes).

xviii

Acknowledgments
This study was supportedand helped by a number of people in Plymouth and Iran. I my would like to express greatand sincereappreciation: To my two supervisors who simultaneouslyguided,supervised constructivelycriticised and the research. Firstly, Dr. Richard Gray as my director of study who has provided constructiveand timely help, support and guidancethroughoutthe work. Also, Dr. Kevin Cullinane as the secondsupervisorfor his valuable guidanceand supervisionin particular for the quantitativeanalysis. Dr. My genuinethanksin Iran to my supervisors Dah-Mardehthe rector of the University of Sistan and Baluchestan,Banazadehin the Ministry of Culture and Higher Education (MCHE) of the Islamic republic of Iran and Salar in London for their great support, To Professor Nomoto in the University of Kobe and guidance and encouragement. ProfessorZade, vice rector of the World Maritime University for their early support and also to Dr. Arshi a fellow country man in Brighton who furnished me with a completeand valuablesetof the TransportJournalof Iran (Sanate-Hamlo-Naghl). To the Ministry of Culture and Higher Education,the University of Sistanand Baluchestan, and Faculty of the Marine and Navigation Sciences Chah Bahar of the Islamic Republic of of Iran for their funding. To the collaborating organisations,Programming and Budgeting Organisation, Central Bank, and Statistical Centreof the Islamic Republic of Iran, UNCTAD in Geneva,the port authoritiesof Rotterdam,and New York and New Jersey.My colleague,Mr. M. Saaudiin PSOIran for his informative supportin providing me with valuabledata. Roodabeh, Ramineh,and Sarvinehwho encouraged and to my Finally, to my daughters me, parents,brothersand sisters,and in particular Behrooz,without whosehelp and patiencein providing data,this study would not havebeencompleted.

xix

List of Abbreviations Abbreviation a ANOVA BASIC BSECOG b (B) bn b/l CACFF CAC CD ROM CP CN cis COFC CIM CIA DOMTI dwt EQ E.I.U ECO FSU GATT GOPFr GDP GNP HGVs IRU IRIRC IRISL ISO I-0 INTERAX ISLB IMF km
LASH LGVs M MRT NAFFA NITC NOVCCs

Description Intercept(unexplaineddependent variable) Analysis of Variance Battle ScenarioInputs to CorporateStrategies Black SeaEconomicCo-OperationGroup Regression coefficient (explaineddependent variables,coefficient of impact) Billion Bill of lading CentralAsian and Caucasus Foreign Trade CentralAsian and Caucasus Countries ComputerDisk Run On Memory Pacific Canadian Canadian National Commonwealth Independent SovereignCountries of ContainerOn Flat Cars Cross-Impact Matrix CrossImpact Analysis DomesticTrade of Iran DeadWeight Tonnage Equation EconomistIntelligent Unit EconomicCo-operationOrganisation Former Soviet Union Group Agreementon Tariffs and Trade GeneralCargoand Oil productsof ForeignTradeof Iran GrossDomesticProduct GrossNet Product Heavy GoodsVehicles InternationalRoadTransportUnion Islamic Republicof Iran Railway Company Islamic Republicof Iran ShippingLine Company InternationalStandardOrganisation Input-Outputanalysis InteractiveCross-Impact Iranian Sea-Landbridge InternationalMonetary Fund Kilometre
Lash Aboard Ships Light Goods Vehicles metre Ministry of the Roads and Transport North American Free Trade Agreement National Iranian Tanker Company Non-Vessel Operating Common Carriers
xx

NVO-MTO _OPEC _O/D _PSO PBO _PAS _PSD _RA e - RO/RO _Rb _SDEP SOTRA q. $ TOFC TEU TSR TCR TIR TAR TECG TEEM TDI TCDD ULCC USSR UN UNCTAD UK us USA VAA WTO

Non-Vessel -Operating Multimodal Transportation Operators Organisation of the Petroleum Exploration Countries Origin/Destination Ports and Shipping Organisation Planning and Budgeting Organisation Productivity of Agricultural Sector Port Staying Days Regressi n Analysis Coefficient of determination Roll On Roll Off Rouble Societal Discussion on Energy Policy Soyz Venesh Transit Square United Statesof America's Dollar Trailer On Flat Cars Twenty-Foot Equivalent Unit Trans-Siberian Railways Trans-Chinese Railways Traffic International des merchandisespar Route Trans-Asian Railways Trans-Europe-Container-Express Trans-Europe- Express-Merchandises Turkish Maritime Organisation Turkish Comhuriyeti Develt Demiryollari Ultra Large Crude Carrier Union of the Soviet Socialist Republics United Nations United Nations Conference on Trade and Development United Kingdam United States United Statesof America Value Added of Agricultural Sector World Trade Organisation

xxi

Author's Declaration
At no time during the registrationfor the degreeof Doctor of Philosophyhas the author beenregisteredfor any other University award. This study was financedwith the aid of a studentshipfrom the Ministry of the Culture and Higher Education and the University of Sistan and Baluchestanof the Islamic republic of Iran and carried out and in collaboration with the Programming and Budgeting Organisation,and CentralBank (Bank Markazi) of the Islamic republic of Iran. A programmeof advanced study was undertaken,two paperspreparedfor publication and externalcontactsmade. Publications:
* Transport development in Iran: a new role as a landbridge country * Long term demand and supply planning, forecasting, and evaluation for a landbridge: a scenario approach External Contacts: * Ministry of Roads and Transport of the Islamic Republic of Iran (Tehran) * Central Bank (Bank Markazi) of I. R of Iran (Tehran). * Programming and Budgeting Organisation of I. R of Iran (Tehran). * Statistical Centre of I. R of Iran (Tehran). * Embassy of the Republic of Azarbaijan (London) * Embassy of the Republic of Armenia (London) * Embassy of the Republic of Georgia (London) * Embassy of the Republic of Kazakhstan (London) * Embassy of the Republic of Russia (London)

si

.......... 0,97 Date ....... .... ...............

xxii

Chapter 1

Introduction

1. Statement of the problem and objectives


1.1 Background
investigates potential of Iran as a landbridgecountry for the eight This research the
new independent Caucasusand Central Asian countries of the Fonner Soviet Union (FSU). The concept "landbridge" refers to the different types of integrated origin-destination

internationalmovements shipments(in various combinationsof sea,land and air) under of a single waybill. The collapse of the Soviet Union in 1991 has changed greatly the geo-political environment of the world and has introduced new opportunitiesfor all the former Soviet republics and also for neighbouring countries including Iran. Prior to the collapse, the economy and trade of all former republics had been centrally controlled by the Soviet
Union, and decisions were based, not on specific requirement of the republics, but on the

Soviet Union's objectives,often of a military nature.


The Central Asian and Caucasus (CAC) countries are eight republics of which four

Tadjikestanand Gyrkyzestan)are landlocked.Three, bordering the (Armenia, Uzbekestan, CaspianSea, (Azerbaijan, Turkmenistan and Kazakhestan)have summer accessthrough the Volga-Don rivers to the Baltic and Black seas. After the collapse of the former Soviet Union, the landlocked CAC countries, which are rich in raw materials and cotton, have creatednew opportunitiesin world trade and need reliable accessto new trading outlets and routes (Tat 1995, Sajadpour 1994). For these countries, located far from international waters particularly for southward trade, the transport costs are high and there are difficulties in developing or attracting large scale tradeor manufacturingactivities (Reitsma 1980).

Chapter 1

Introduction

1.2 Significance of the problem


Iran has an ideal location as the closestmaritime country to the south and west of the CAC countries.The port of Imam in the South of Iran is about 1800km to the Azerbaijan
Armenian capitals, and the port of Abass about 1600 km from the nearest Central and

Asian capital in Turkmenistan.Iran, with direct land and sea accesscovering more than
800 km of borders with three of these countries (Azerbaijan, Armenia, and Turkmenistan),

is well placed to serve as a landbridgeusing different modes,routes and border crossings (Stone 1993,Herzig 1995).
Iran has common interests with the CAC countries in terms of culture, religion, history and national security. According to Garabaghi (1994, p. 115):
"Iran represents at once a source of cheap energy for those republics lacking energy resources, and a transportation routes to the Persian Gulf for those who seek to export oil and in order to earn much needed foreign currency". gas

Iran is morelikely to developcloselinks with severalsmallerneighbouring countrieswith such which is ideologically different (Herzig 1995).A commoninterests,than with a super-power from the tight planning issueis the future trade of thesecountries,newly released significant control of the former USSR, and in particular, the likely long-term routes for their foreign developedtransportinfrastructure trade.Iran has potential for such routeswith its reasonably the with neighbouringcountries.Until independence CAC countrieshad not made compared useof Iranian transportservices.

1.3 Purpose of the study


In the light of the precedingdiscussion, main purposeof this studyis: the
To review the landbridgeconceptandpracticesand to developa modelappropriate Iran and to the the CAC countries.This modelwill assess potentialof the transportsystemof Iran, and identify influential variablesaffectingits volume and supply.It will evaluatethe implicationsof Iran actingas a landbridgebetweenthe CAC countriesand the rest of the world.

Chapter 1

Introduction

The study focuses on the domestic and international trade of Iran as well as the predicted future

Republicsof the former Soviet Central Asian and Caucasian trade of the newly independent Union. In particular,it examines whetherIran hasa role to play asa landbridge.

1.4 Previous research


Although thereis somepublishedliterature directly relevantto this study, and included in the literature review, there is a lack of systematicstudies of landbridge the potential of Iran in this context. systems,andnoneon

1.5 Thesis structure


The Iranian Sealandbridge study consistsof ten chapters.Chapterstwo and three review in detail Iran and the eight CAC countriesin terms of information relevant to this Therefore,it looks at the economy,the transport infrastructure,the physical and research. vehicular network and operational characteristicsof modes and interfaces.For the CAC countriesthe economyand the composition, value and direction of international and interrepublic tradeis examined. Chapterfour reviews the landbridgeconcept,with examples.Different bridge conceptsare identified and the most important examplesassessed comparedunder the four broad and headingsof geographical, political, technical,and organisationalaspects. Chapter five comparesthe potential of the Iranian Sea landbridge (ISIB) model with potentially competinglandbridgesand routes,also under the four broad headingsoutlined above. Chaptersix introducesthe scenarioapproachas an analytical tool and the characteristics of
the scenario approach are studied. The key issues related to the ISLB project are identified and developed as the conceptual model of the research.

Chapter I

Introduction

In chapter seven, an operational model is developed, requiring the selection of appropriate

time periods,units of measurement levels of accuracy. and The ISLB scenarioanalysisand forecastingis carried out in chaptereight. By meansof a multiple regressionanalysisapplied to the data preparedin chapterseven,three models of demand for the ISLB are developedin a preliminary analysis.These are based on the
general cargo and foreign trade of Iran (GOPFr), domestic trade of Iran (DOMTI), and

Central Asian and Caucasusforeign trade (CACFT). Three ISLB scenarios are then developedaccordingto most probable,optimistic, and pessimisticassumptions. Chapternine assesses potential and requirementsof the ISLB transportsupply from a the baseyear of 1993up to the year 2005 for each of the three scenarios. Conclusionsrelated to the researchare drawn in chapterten and severalrecommendations madefor further are research.

Chapter 2

Iranian Trade and Transport

2. Iranian trade and transport


2.1 Introduction
The aim of this chapteris to examinethe main trendsin the economyand transport demandand supply of Iran in the context of the potential for an Iranian landbridge. It
describesrecent economic developments in both domestic and foreign trade and in transport

the Islamic Republic of Iran. It is directedparticularly to changesafter 1979 which within during andsincethe war with Iraq. occurred

2.2

The Iranian context

2.2.1 Geographical features


The Islamic Republicof Iran is a maritime country locatedin the south-westof Asia between SouthCaspianSeain the north and the PersianGulf andOmanSeain the south. the Mountainscover about 90% of its land masswith a vast plain in the south west. Iran has a
land area of 1,648,000 sqJkm Which is approximately equal to the size of Britain, France,

Belgium andDenmarkcombined.It supports populationof about55 Switzerland, Germany, a with a densityof 37 peopleper squarehectareand is the sixteenthlargestcountry in million theworld. for Gulf andIndian Ocean, The coastlineof 2700km accounts some2043 kms on the Persian by and657 km on the CaspianSea.The country is characterised threedistinctive geoclimatic The semi-aridregion hastwo-thirds of the land massandone-thirdof the population, regions. and contains most of the economic and industrial activities. The moderatemountainous region lies mainly from the north west to the central part, and the moderateCaspianregion coversa strip alongthe CaspianSea.

Chapter 2

Iranian Trade and Transport

Iran borderssevencountriesas shown in Figure 2.1. There is all-season to access ports in Russia,Kazakhstan, Turkmenistan Azerbaijanto the North throughthe CaspianSea,and and during the summerthroughthe Volga, the Don, the Baltic andthe Black Sea(Badiyee1993).

2.2.2 Demographic and administrative characteristics


The Islamic Republic of Iran is divided into 24 administrative territorial provinces

called Ostans(seeFigure 2.3) with 229 major towns,47 of which havea populationof more than 100,000. Iran had a populationof about 55 million in 1991 ( SCI 1993,p. 44) of which 57% live in urban districts and 17.9%live in greaterTehran.The other major populatedprovinces are Khorasan, EastAzerbaijanand Esfahanwith populationsof about6 million, 4.4 million, and 3.7 million respectively.

2.2.3 Economic background and composition


Iran is a mineral-richcountry, and one of the biggestoil and naturalgasproducersin The oil reserves about 88 billion barrelswith gas the world with considerable reserves. are of reserves about400 trillion cubic feet. Iran is also substantiallyrich in other undeveloped e. mineraldeposits g., copper,iron ore,copperlead,manganese, coal, sulphurandchromate.It is the world'slargestproducerof turquoiseandfifth largestproducerof zinc andcopper. The economyof Iran, like that of many of its neighbours, tied stronglyto the income from is the oil industry as shown in Table 2.1. The macroeconomy the pre-revolutionary of period because the revolutionaryprocessduring 1978-79.Oil production fell by was damaged of 85% in 1979,andagainin 1980-81at the beginningof the war with Iraq. But the economic andmonetaryreserves after the Islamic revolution andduring the eight years'war with Iraq 6

cr

tall
0a

17'

:.

to

)-A

Its

9: PCs

0
(A PA

lz

10

CD LT

Chapter 2

Iranian Trade and Transport

to resistthe costburdenof the war (Herzig 1995). were strongenough Table 2.1 Share of oil and gas,and tax in revenuesin Iran (%).
Year Oil & Gas Tax Others 1989 46.9 37.4 15.7 1990 59.8 30.1 10.1 1991 51.1 39.9 9.0 1992 52.0 38.2 9.8 1993 72.5 20.1 7.4

Source:Basedon the CentralBank (Bank Markazi) NationalAccountsReports for 1989-1993 variouspages.

Political factors,including unresolvedproblemswith the U.S.A, the eight year war, and the fall in oil prices,caused economyto havelow rateof growth,especiallyin industryandin the were the main war targets. High inflation, the falling value of the currency, oil which increased shortage hardcurrency,low productivity andthe closureof some of unemployment, industries,the lack of raw materials,spareparts,etc. wereamongthe main resultsof the war, during 1979-1988. the economy andweakened Since 1988,manylight and heavyindustrieshavebeenprivatisedto increase their efficiency the role of oil in the economy,and also to reducethe close relationshipbetween and reduce imports andthe valueandvolume of crudeoil exports(Ehteshami1995). in In termsof GrossDomesticProduct(GDP), aspresented Table 2.2, the major contributors to GDP on averagefor 1979-1993(excluding the servicesector)were agriculture (23.5%), (19.8%)andoil (16.9%). industryandconstruction The principal economicaim of the governmentsincethe war with Iraq has beento free the to caused the country'sover-stretched economyfrom strict controls,and to repair the damage by infrastructure, especiallyto its petroleumand transportfacilities. This hasbeenattempted the promotionof the privatesector.

Chapter 2

Iranian Trade and Transport

Table 2.2 Sectoral contributions in the economv of Iran. 1982fixed vrices (bn. Rials). Oil Services Final GNP Industry & Year Agricultural Mine Construction Gas, GDP Elec. & lWater 1722.0 2535.2 4964.4 10551.3 1979 10574.1 1851.2 51.7 0.5 16.4 24 47.1, 105.6 % GDP 17.6 1820.5 866.1 4855.0 9555.5 1980 1914.9 53.9 9559.6 0.7 19.1 9.1 50.8 % GDP 20 99.7 1819.5 882.6 4507.2 9320.7 1981 1952.7 55.6 9345.7 0.6 19.5 9.5 48.4. % GDP 21 99 1818.8 1947.7 4543.4 10539.8 2091.4 65.2 1982 10539.8 0.6 17.3 % GDP 19.8 18.5 43.1 99.3 71 2183.6 2006.3 5135.6 11934.7 1983 2193.0 11939.0 0.6 18.3 43 97.1 % GDP 18.4 16.8 2290.2 1625.6 5269.5 12043.8 1984 2353.7 74.1 12047.8 0.6 19 96.4 13.5 43.81 % GDP 19.5 . 2160.6 1644.4 5373.2 12072.3 1985 12057.6 2537.6 71.6 0.6 21 17.9 13.6 44.5 97.6 % GDP 1970.4 1403 4654.7 10248.9 10250.7 1986 2650.5 62.3 0.6 19.2 45.4 % GDP 25.9 13.7 104.8 2018.6 1598.7 4340.5 10368.11 1987 2715.8 65.5 10359.2 0.6 19.5 15.4 41.9 % GDP 26.2 103.6 .1988 1921.5 1754.0 4030.2 9468.01 2648.0 56.6 9451.1 0.6 28 20.3 % GDP 18.5 42.6 110 1 2050.5 1889.5 4146.5 9781.51 1989 2746.0 58.6 9797.0 0.6 28.1 21 19.3 42.4 111.4 % GDP . 63.1 2967.5 2328.7 2264.7 4499.6 10930.21 1990 10997.5 0.6 % GDP 27.2 21.3 20.7 41.2 ill 3120.2 68.4 2733.6 2516.7 4945.9 12181.21 1991 12377.9 0.6 25.6 22.4 % GDP 20.7 40.6 109.9 3351.6 72.2 2860 2553.5 5343.5 12477.81 1992 12985.6 26.9 0.6 % GDP 22.9 20.5 42.8 113.7 1 3535.7 76.8 1993 2923.2 12645.3 5743.7 13101.0 13400.8 27 0.6 % GDP 22.3 11: 20.2 43-R
Source:Basedon the CentralBank (Bank Markazi) variousNationalAccountsreportsfor 1974-1987,19881990,1991,1992and 1993. by *: Final GDP afteradjustment the CentralBank which makesthe total higheror lower than 100%.

2.3 Transport
2.3.1 Organisation
The administration of Iranian Transport is distributed among the ten different ministries of Road and Transportation,Constructive Jahad (mainly rural regions), Oil,
Agriculture, Interior, Defence, Industry and Mining, Commerce, Finance and Economy, and Justice. Figure 2.2 shows the flow chart of all the ministries and organisations involved in roads and transport in Iran, of which the Ministry of Roads and Transport (MRT) has the

Chapter 2

Iranian Trade and Transport

largestrole, with responsibilityfor all modes(exceptmerchantand tanker fleets). They are for border- crossingterminals.MRT has different internal and external also not responsible both in the national capital, and in all 24 provincial capitals;the internal body of this roles, for and ministry is mainly responsible road construction,maintenance supervisingthe traffic
management of both passengerand freight movements and vehicles on roads. The external

bodies include Rail, Road Developments,Ports and Shipping (Maritime) Administration, Airlines, Airports and their relevantindustries,and are responsibleto the governmentand operation,construction maintenance all and of parliamentfor the functioningof management, transportactivities.

2.3.2Road transport
2.3.2.1Road networks and standards The construction and maintenanceof roads within a country like Iran is not only difficult, but also costly and time-consuming. total road network of 167,156km connectsall A
provinces, cities, towns and villages, excluding roads between small villages, making a road

densityof about 10.14km per 100sqJkm. The nationalandprovincial roadswhich areunderthe supervision MRT aremainly distributed of on the basis of economic activity, maritime connections,strategiclocations, and size of the provinces.Table 2.3 showsthe detailsof national and provincial roadsunder the supervisionof the MRT. National roads are asphaltand connect24 provincial capital cities and some major industrialtowns.Provincialroadsarethosewhich link minor townswithin eachprovince. Table 23 National and provincial roads of Iran under the supervision of Ministrv of Roads and TranSDort in 1992(km).
Type of roads Secondary Main Access Total M,R'T Road length 20130 39589 12938 72T55]6 % 28 54 18 )0 Source:StatisticalCentreof Iran (SCI) (1993),Iran StatisticalYearbookfor 1992,p. 388.

10

C.

ro cm CD

F- a4

ct

OQ Ro 8 CD (.D

00

(v

cr

Rp

k4

ITj
m 0
rA

0 0
CD

CD
Co

CD

R
CD

0 M CD

CA

a, a

QX 0

Ln

CO2 CA >TI

(a

E, ,.cj

0L, 2.

Gn CD >zj

' r. 9 LE, <D gg 0 -I

C > tz0
OrQ

cI 00

0
1-41 rA

QQ

8 Av

2, .

co i

11

Chapter 2

Iranian Trade and Transport

The quality of roads ranges from ungraded dirt roads to the modem multi-lane divided

highways. Highways and expressways constitute only 1.3 % of the total road networks. Freewaysas shown in Table 2.4 form 2.36% of the total and 33% of them originate from GreaterTehran; Ordinary and secondclass roads constitute the structureof the secondary roads(seeTable 2.5).
Table 2.4 COMDOSition shares in main roads of Iran in 1992. Free ways 1 Express roads Wide roads I Ordinary Type of roads (6 lanes) (13-13.3m) (4 lanes) roads (11m) 2.361 20.91 71.71 4.971 %
Source: Same as Table 2.3, p. 388. Tnhip '21; rnmnnqitinn and lenoth nf thp 4ceenndarv rnad-. nf Tran in 1992 (kml-

Type of roads 1 Length

Wide paved (9m) I Ordinary paved (8m) I 60491 186241

Second class (7 1

Source: Same as Table 2.3, p. 388.

Khorasanas the largestprovince, and Khosestan an important industrial centrehold the as highestproportion of total roadsin the country.Table 2.6 showsthe provincial and national shareof roads(SCI 1993)amongthe 24 states Iran underthe supervision MRT. of of The main objective of the first ten years of the Islamic government(1979-1988)was to from 92,353Ian in 1979to developrural roads.Therefore,all typesof roadswere increased 167,156km in 1987.During the sameperiod only 4,831 krn of main road and 401 km of freeways wereconstructed.
Table 2.6 Share of 24 provinces in all national and provincial roads under the supervision of the Ministrv of Roads and Transnort (MRT) 1992.

Provinces
% Provinces % Provinces', % Provinces %1

Tehran

Markazi

Gilan 'Mazandaran 2.3. Fars 8.23 Hamedan 2.21 Zanian 4.161 4.7 Kerman 6 Chahar. Mahal 2.21 Senman 2.35

2.76 2.11 KermanKhosestan shaban 3.18 10.48 Sistan & Kurdestan Baluchestan 6.97 2.96 Koh-Giloyeh I Bushehr 1:3 1 2.161

East Azerbaijan 6.57 Khorasan 10.9 Lorestan

West Azerbaijan 4.42 Esfahan 5.11 Ilam

2 1.66' Yazd 1 Hormozgan 31 2.061

Source: SCI of Iran (1993), Iran Statistical Yearbook for 1992. P.388.

12

Chapter 2

Iranian Trade and Transport

2.3.2.2. Road surfaces in Iran

Roadsurfaces affect the operatingcostsof vehicles,particularlythe cost considerably and of fuel, spareparts,tyres,maintenance safety.Therewererestrictionson the maintenance during the war with Iraq, but they havebeenmaintainedto a reasonable to standard of roads match economic servicesand provide a faster transit time for trade. Table 2.7 shows an
estimate of the conditions of four classesof paved road surfacesat the end of 1988.
Table 2.7 Paved road surface conditions in Iran after the war in 1988.

Type of Roads 1 -2 3 4 Hieh-wavs National roads Provincial Others Good 40 50 30,

%Q ality of roads % Total Fair Poor 40 20 100 30 100 20 30 20 100 40, 30 1001

Source:Ministry of FinanceandEconomics(1992),unpublished report annex4, p. 5.

2.3.2.3.Road fleets: ownership and use


Transport services are provided by both the private and public sectors. Different

types of private ownership contribute widely to the movement of both passengers and freight throughoutthe Iranian road network. The proportion of trailers and lorries to vans and private carsis low andcarriersmainly pull single trailers only (with a maximum weight limit of 38 tonnes,length of 16, and height of 4.5 metres).In terms of types of commodity, road freight fleets mainly carry agricultural products, followed by minerals (construction) and industrial products. There has been a sharpincreasein the number of motor vehicles since the mid-1960swhen the first Iranian Table 2.8 showsthe growth in road vehiclesregistered car assembly plant was established. since 1979.The dramatic changesin 1987 result from the official count of existing lorries by the Islamic Republic Iranian Disciplinary Forces, and take into account out-of-service vehicles.
The supply of lorries or heavy goods vehicles (HGVs) has not coped with the demand, resulting in periodic congestion in ports. Restricted supply has resulted from restrictions on

13

Chapter 2

Iranian Trade and Transport

imports, insufficient foreign exchange for purchasing HGVs, the limitations of heavy

in goodsmanufacturingfactories,and increases the purchaseprice of all typesof vehicles, particularly goodsvehicles.


Table 2.8 Development of different types of motor vehicles in Iran.

_1979 _1980 1981 _1982 _1983 _1984 1985 1986 _1987 _1988 _1989 1990 1991

Sedan & Ambulance 1367953 1463406 1539915 1604118 1687985 1778826 1884489 1923937 1916612 1977289 1996674 2008504 2074130 2148903

Goods vehicles 260829 271510 280828 293229 306225 345010 355065 362295 260089 262068 263781 265511 280600 299554

_1992 Data for 1979-1981is basedon SCI (1982), Iran StatisticalYearbookfor 1981, 588; Source: p.

Buses & Minibuses 42324 44825 48342 50802 55550 62462 68502 72716 102455 103475 105528 106374 113720 120749

Vans 27830 36999 42355 48375 56702 75643 118554 141051 490439 496189 500409 505229 534860 575898

Motor cycles 290346 372675 435309 472138 571287 604226 633728 662919 966466 979864 993242 1000280 1024615 1062924

Total 1989282 2189737 2347125 2469092 2678259 2866167 3060338 3162918 2810929 3818885 3859634 3885898 4027925 4208028

for 1982SCI (1990), English versionNo. 6, p. 125; data for 1984-1987is basedon SCI (1989), No. 7, pp. 142-144; for 1988-1989:SCI (1991), No. 9, p. 136; for 1990:SCI (1992), No. 10, p. 154; for 1991-1993:SO (1995), No. 13, p. 125.

Table 2.9 showsthe ageof the goodsfleet in Iran during 1963-1987. The average of the age Iranian heavy goodsvehiclesfleet is over 15 years,suggestingthat 94% of vehicles are not economicallyviable.
Table 2.9 Growth of road goods vehicles in Iran during 1963-1987. 1963-67 1968-73 1973-7T Type 1978 1979 1980 Lorries 2435 6879 24071 3653 1719 2130 Road tankers 572 1322 5174 563 207 266 485 1414 7961 445 208 244 _Trailer 1452 Dump 3513 13965 1746 832 857 Reefer 14 68 430 69 25 51 Vans 1219 105721 89964 27602 101155 11128 Others 254 695 2094 457 169 185 Type" 1984 '", `-, ': 1983 11985 1986 1987 ", Lorries 3726 4014 3403 1695 392 Road tankers 2161 557 935 363 125 Trailer 2856 2144 627 442 164 Dump 2455 2015 2108 903 349 Reefer 156 102 226 54 27 48841 45661 36543 17196 5521 _Vans . 426 Others 533 683 125 1 270
Source:SCI (1991), Statisticalreflection of Iran, No. 9. p. 138.

1981 2357 404 239 883 165 13601 251

1982 2238 1017 777 1128 88 16267 293

14

Chapter 2

Iranian Trade and Transport

According to the MRT, HGVs which are engaged in long distance domestic and

international trade are categorised. the basis of higher and lower than 10 tons capacity on (i.e. heavy or light goodsvehicles).
Tahle 2.10 Numher and tVDCS all heavv twods vehicles in Iran in 1987. of Trailers Dump Reefer Vans Others Total Road Heavy goods Lorries tankers vehicles 14851 19748 34717 1672 345683 7718 487861 63472 Number _ 3 4 7.1 0.34 70.9 1.9 100 13 -% is Source: SCI (1991), Statisticalreflection of Iran, No. 9. p. 138.Inconsistency due to different sources

The most detailed availablepublished data about composition and capacities(load limits) the Iranian heavygoodsfleet was issuedby MRT in 1987and 1991(seeTable 2.10,2.11 of and 2.12). Thesegenerallyindicate that there were 97,964 operatingvehicles of which 1020 tonne vehiclesat 58% of the total number are the backboneof the fleet, followed by 2030 tonne vehiclesat 26%.
Table 2.11 The number of heavy goods vehicles of greater than 10 tonnes to tVDe and carrvine cavacitv in Iran in 1987 (unit: tonnes). 10 to 20 20 to 30 Others 30-38 38 & Type/ carrying capacity _Trailers 1614 15165 630 476 _Lorries 28865 1772 10144 _Road tankers 5944 5414 122 151 _Dump 19045 1395 483 2541 713 19 24 Reefer il 8- Others 1239 561 39 56961 25020 12221 685 Total in capacity
Source: Same as Table 2.10.

according more 1863 1084 54 106 3107 Total type 19748 _ 40781 . 12715 20923 ] 1064 2763 97994

Table 2.12 ComDosition and number of road vehicles in Iran in 1991. Van-taxi I Taxi I For-hire I Mini-bus I Bus I Heavy goods I Total road Vans vehicles vehicles ' 1 1 24374 1 1 2410911 46202 180461 59706 170370 809129 249370
(1993a),809 thousandvehiclesin servicefor road transportof Source:Sanate-Hamlo-Naghl, Iran, Farvardin, 1993, No. 116,p. 19.

The most recent statisticsof the road transport fleet of Iran (Atrchiyan 1995) as shown in Table 2.13 indicate that 50% of the total is devoted to vans with less than one tonne capacity while the HGVs of 13.5-22.5tonnes capacity moving over long distanceshave 48% of the total.

15

Chapter 2 .

Iranian Trade and Transport

Based on SCI (1994a, p. 311), applying the growth of 1992 (18,954 HGVs) and 1993

(9,651 HGVs) to the total number in 1991 (see table 2.12) the number of HGVs was for 1993 as about 198,975.The distribution of the domestictrade mainly takes estimated by trucks of lower than 13.5 tonnes carrying capacity. HGVs of 13.5-22.5tonnes place capacityhave a significant role in the import/export of foreign trade through ports carrying and border crossings(Atrchian 1995).Assuming the sharecomposition in 1993 is similar to 1994,the numberof thesetrucks was about95508 in 1993.
TAle 2-13.1ghnrenf different HGVs fleet canacities in 1994 (can2citv. - tonnes).

Capacity 1 1.751 % of total 171_

5.751 191

9.51 221

13.51 18 1

27.5 1 over 30 1 Total 181 22.51 1001 71 23 11131

Source: Atrchian, (1995), Sanate-Hamlo-Nagle, No. 140, Tir, p. 23.

The road freight industry (Sanate-Hamlo-Naghl,1993a), has about 958,623 employees to 6.42% of the total 14,934,000 employed population of the country in 1993 equal (CentralBank of Iran, 1993);about78,1457personsor 81.5%of the total employeesin the freight industry run their own vehicles. road The road censusin 1993indicatesthat 48.1% (769,849vehicles)of the total traffic, can be transport, accountedfor by freight transport,whilst 47.2% (755,754vehicles)is passenger and the remaining4.7% being of other types.The compositionof inter-urbanfreight traffic is limited to vans (47.6%), lorries with two axles (24.4%), lorries with three axles or more (13.3%),road tankers(6.2%) and trailers (8.5%) (MRT 1994). 2.3.2.4 Road freight rates and financing MRT is responsiblefor the preparationand enforcement the freight tariffs for all the of transportnetwork, while the National Economic Council is in chargeof the ratification and final approval of all transport tariffs in Iran. This Council consists of the Presidentand Vice President,the Head of Planning and Budgeting Organisationand almost all ministers of the Cabinet.

16

Chapter 2

Iranian Trade and Transport

At present tariffs are based on the combination of two systems. The list system is based on

a few tableswhich determinethe tonnes/kmrate, and explain different categoriesof goods. In this system, cargoesand goods are categorisedinto four broad classesof metallic cargoesand containers,perishableand bagged,bulk cargoes,and dangerous goods. In the in secondsystemwhich was established 1992the freight ratesare a function of the capacity
of the goods vehicle, distance travelled (extracted from a standard table), and a tonnes/km

rate of 14-18Rials. Therefore:


I Freight Rate = Vehicle Capacity x Distance x (14-18) Rials I

The Ministry of Finance and Economy is responsiblefor collecting all the state income originating from oil, gas and related products, while the Programming and Budgeting Organisation(PBO) as an affiliated body to the FinanceMinistry is in chargeof national expenditure. Transport policies are determinedand issuedby the SupremeTransportationCouncil and
the Cabinet. -Port dues and charges (i. e. tariffs) are prepared and approved by the Council of

fares through the Council of Ministers and Parliament,but inland freight and passenger Transportation.All national and provincial long and medium term plans and projects are madein collaborationbetweenthe Ministry of FinanceandEconomyand the PBO. 2.3.2.5Importance of transport modes to GDP The importance of the transport sector to Iran's economic and social life can be measuredon different basesand levels such as the contribution to GDP, employment, balance of payments and the investment created by the sector. Relevant statistics are published annually by the Central Bank of Iran (Bank Markazi) and the Statistical Centre of Iran (SCI).

17

Chapter 2

Iranian Trade and Transport

Iran, as a big country, has road goods vehicles mainly for long distances of above 1000 km

for imports and exports, but domestic traffic (particularly associatedwith farms) varies Shipmentsby manufacturersare more stable but from Tehran and the with the seasons. central provinces to all other parts of the country are mainly increasing as a result of wholesaleprovincial delivery or direct delivery to small dealersin major cities and towns.
The type of vehicle used for this kind of trade depends on volume, but usually small goods

vehicles and vans of up to 10 tonnes are used. The pattern of demand for goods and passenger movementsin Iran is basedmainly on the uneven geographicaldistribution of featureshave populationand industrial activities in the country.The country's geographical causedtransportcoststo be rather high. Long distancehaulagerequireshigh initial goods depreciation and intensive use of spare parts. This is particularly important for vehicle fleets which are continuously engagedin imports of foreign trade between the southern ports and other areas,where there are likely to be empty return journeys. Both domestic and foreign trade flows are carried out over long inland distancesas a result of the geography the country. of

2.3.3 Rail transport


2.3.3.1Organisation and development The Islamic Republic of Iran Railway Company (IRMC) consists of rail line operations and maintenance of passenger and freight services along domestic and internationalnetworksconsisting of 5,020 km standardgaugetrack in 1993,excluding the to small narrow gaugeisolatedfrom the main network and connected the Pakistanrailway. The railway is owned and run by the government.The rail network is mainly single track and non-electrified although there are extensive approved development plans in both The IRIRC was restructuredin 1987(IRIRC 1986-90)from a purely public entity aspects.

18

Chapter 2 into a public company to enable it to perform

Iranian Trade and Transport more efficiently. Operations are under

and development are directly run by the MRT. company control, while construction

2.3.3.2 Network characteristics Tehran functions as a hub for the mainly single line network as shown in Figure 2.3 and therefore the network is poor in terms of speed. In terms of connectivity it directly crosses 14 of the 24 provinces and through about 54 main stations to the end destinations in the main geographical route directions. In terms of freight it is purpose-built to serve the southern ports of Imam and Khoramshahr to Tehran and then to the east coast of the

Figure 2.3 Islamic Republic of Iran railway network, major ports border crossings in 1995. and
Turke
B4, arp 2 156

gssia? 'ia I Az. zl, ,


110 3 . 3y iin Wri: 01-

Uzl)ekistan

Astara Latfabad 421 Anzah 0 st Fort Turkmen Noo5hahr ed). . 16 rkjan 698 Kurdes r9an 5 315 Sanadej Tehran 240 114 114 Senuian 180 Qum axwar 14 a O,, Khorasan 14 Arak 62Baad K, rn,,, h, h Doroud 206 351 M v

Caspian Sea

Turkmenistan
B aj e_n

Taybad

Dlm Iraq
121

Khorw-bAr

208 Es Andmieshk 141 Ahwaz I! 121 Yasui Pori Imarn

M od W %afgh 11 3j

Athanistan

YaA

uwait

Shiraz Port 0 Bushehr

,erm%n 224Bam 560 254 Sujan Ziledw. Pakistan 398 Miriv

Persian Bahrain Saudi Arabia Gulf Q tar

Port Abass Hormuz

Chah B Oman Sea

UAE

Source: Map is based on CD ROM: World Atlas for Multimedia IBM PC and compatible versions 4,1991-1993. Distances Based on Thomas Cook Overseas Tinictab1c, March-April p.282.

1995,

Caspian Sea and main border crossing points of the country with the FSU (Q)*Lilfa)and with Turkey (Razi). The railways of Iran have are in a poor condition and unmodernised. The 19

Chapter 2

Iranian Trade and Transport

war with Iraq and over-utilisation has worsened the network and reduced the speed of

trains. From 1986 and particularly during the 1989-1993First Economic Development Programme after the Islamic revolution, a planned programme forrenewal and tracks was adoptedand executed(IRIRC 1984-1993).It can be seenin rehabilitation of Table 2.14, that during this period HURC made substantial progress (Sanat-e-Haml-oNaghl 1993b).
Table 2.14 Improvement and development of rail tracks in Iran during the. Firet Rennnmir Develonment ProLyramme (km).

Rehabilitation Renewal New lines

1989 95 20 -

1990 130 70 260

1991 120 75 82

1992 150 130 145

1993 150 150 392

Source:IRIRC (1984-1993),performance report of the Islamic RepublicIranian Railway Companyfor 1984-1993, 21. p.

2.3.3.3Locomotives and rolling stock As an oil producingcountry, 97.8% of locomotivesrun on diesel (1.5% electrified, 0.7% turbotrain). The numberof locomotives and rolling stock are given in Table 2.15 and showing the considerablevariation over the yearsin the numberof in-service locomotives in indicating the shortages driving power and also maintenance problems. and The availability of in-service locomotives for main routes in a mountainous and vast country like Iran, which mainly relays on imported equipment,has always been a critical issue and a reason for slow rail freight movement.Table 2.15 shows these fluctuations during 1979-1993 and the general decrease of rail motive power due to age and and on maintenance partly asresult of direct war damages rolling stock.
Table 2.15 Number of locomotives for main routes and shuntinLy services of the TRIRC (unit).

1,1 , 1,11 . Main routes Shunting Total in service Main routes Shunting Total in service

'

1979 128 22 150 A987 227 77 MA

1980 ''1981' '1982 120 166 102 62 105 161 164 225 237 1989 1990 1988 169 176 148 67 65 58 236 241 206

1983 169 151 320 1991 150 63 21

1984 242 76 318 1992 1 163 53 216

1985 247 73 320 1993 156 52 208

'1986 239 74 313

Sources:PBO (1989-1993), the preliminary results of the first social, cultural and economic development plan of the Islamic Republic of Iran, transportsector,unpublished,p. 1-11;IRIRC (1984-1993),p. 17.

20

Chapter 2

Iranian Trade and Transport

The capacities of all types of in-service rail freight cars in 1990 was about 623,453 tonnes.

The developmentof rail of wagonsis shown in Table 2.16. They consistof eight types, of which the proportionsare shownin Table 2.17.
Tnhip 2.16 The develoDment of freiLyht railcars in Iran.

Year ' Freight railears Year Freight railcars

1980 1979 12422 12247 1987 " 1988 13074 13074

1981 11750 1989 12439

1982 12265 1990 12224

1983 12150 1991 12629

1984 1985 12422 11750 1992 , 1993 13000 14400

1986 12205

Sources: IRIRC (1984-1993), p. 18; PBO (1989-1993), p. 1-11.

The ageof the rolling stock of IRIRC is rather high (14.9 and 18 yearsfor locomotivesand in 1993). The rate of renewal and replacementof locomotives and railcars respectively freight wagonsduring the First EconomicDevelopmentProgramme1989-1993, was rather Table 2.17 ComDosition of ei2ht maior tVDeS freiaht railcars of IRIRC in 1990. of
Railcars type Number % Covered 2900 23 Low Wided. flat sided 1995 180 16 1 High sided 3170 24 Flat 420 3 Tank 2543 20 Ballast 750 6 Ore high sided 892 7

Source: IRIRC (1986-1990), English Version. p. 46

slow (Sanat-e-Haml-o-Naghl1993b) achieving only 63 locomotives and 1771 wagons, was 65.6% and 44.3% respectivelyof the programmetargets.The composition and which the above-mentioned manufactured rolling stock is shownin Table 2.18. numberof
Table 2.18 Composition and number of the eight major types of domestic manufactured freiAt railcars of IRIRC durin2 1984-1993. Bulk Covered I Flat I Ballast I Tank I Ore I Coach I Diesel Wagon locomotives cement production 80 3801 2611 861 3 190 1 1718 1 1677 1 Number
Source:IRIRC (1984-1993),p. 22.

23.3.4 Operations of the IRIRC The comparison of the annual freight performance of IRJRC during 1979-1993 indicatesthat there hasbeena 223% increasein volume (seeTable 2.19), while in terms of commoditiesgenerally, mineral flows constitute more than 50% of the total services.The amount of freight carried by the railway during 1993 was 19.83 million tonnes, 13% of
which was trade from seaports and across borders, and intemational cargoes via Turkey.

21

Chapter 2

Iranian Trade and Transport

Along with the general task of providing relatively cheap passenger and freight services

during the war with Iraq, IRIRC has greatly supportedthe war logistics both in cargo and forces.Becausea considerable proportion of rail operationsare in the south west and very to the war zone, the transport of crude oil and its products was greatly restricted close during the war period. During the period of 1984 -1993 the averagelength of haul in rail
freight in Iran was relatively low (varying between 732 km and 912 km for 1986 and 1993

The increaseis due to the end of the war and more rail freight operationsin respectively). (seeTable 2.19). southwest provinces The composition of the sevencategoriesof commodities and products carried by HMC indicatesthat, except for 1986and 1987as peak war times, the carriageof minerals,crude its products, and imports through ports, border crossingsand also through Turkey oil and havealwaysbeenthe threemajor areasof businessfor rail freight in Iran.

Table 2.19 Volume and tonne/km freight car ied by IRIRC. 1979 1983 1984 1985 1986 '1981 -'1980 -1982 , 6138 6782 5712 7931 9973 10874 11614 12704 Freight (thousand tonnes) 3428 9700 7570 8890 7320 3124 3811 N/A Freight (nMon tonnes/ km) 1993 1991, '., 1992 1990 -1987 ,, 1988 -1989 16979 1 17649 19830 1 14788 12957 1 12334 1 14881 Freight (thousand tonnes) Freight (million tonnes/km) 1 8830 1 8050 1 79601 7630 1 7700 1 8000 1 91201
Source: IRIRC (1984-1993), p. 25; PBU (1989-1993), p. 1-11, Sul (1982), Iran statistical YearboOK tor 1981, pp. 585-586. SCI (1994) Iran Statistical Yearbook for 1993, p. 313. Sanate-Haml-o-Naghl (1982), January, p. 45; Sanate-Harnl-o-Naghl (1983), August, p. 105.

Table 2.20 showsthe volume and sevencategoriesof freight carriedby rail in Iran with the the three border crossingsfor the four years 1990-1993indicated as part of the sharesof total "others and international". The contribution of rail from the only rail-connectedport
(port Imam) is shown at the bottom of the table as the difference between the sum of the rail trade from the three border crossings and the "total others and international".

The above-mentioned seventypesof rail trade statistics for the period 1986-1991can also be describedas 22 type of goods commodities, the 1991 sharesof which are shown in Table 2.21.

22

Chapter 2

Iranian Trade and Transport

Table 2.20 Volume and composition of domestic and foreign trade of Iran by rail (000 tnnnpq)-

Oil Minerals Agricultural Foodstuffs Industrial Administration cargoes Others* & international of which: Intemational (Razi) Djulfa Mirjaveh Port Imam Total Oil Minerals Agricultural Foodstuffs Industrial Administration cargoes Others* & international of which: Intemational (Razi) Djulfa Mirjaveh Port Imam Total

1979 1899 1761 607 322 1 352 401 796 N/A N/A N/A N/A 6138 1987 2036 3487 2180 569 718 343 5423 N/A N/A N/A N/A 14756

1980 1416 1416 503 232 182 214 1519

1981 1201 1948 721 384 1 379 396 1712

1982 1983 1786 1939 2173 3281 851 884 350 445 750 1005 537 398 1561 1975

1984 2028 3978 990 387 855 569 2030

1985 2289 4383 577 464 496 484 2884

'1986 2343 3771 864 390 331 424 4548

N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 5482 6741 80081 9927 10837 1900 1989 1 1990 " -1991 2210 2661 2427 2483 6576 3990 4092 5475 1648 1535 1794 1346 530 529 540, 402 714 1450 1732 2847 448 37 25 28 3378 1832 1929 2512

N/A N/A N/A N/A N/A N/A N/A N/A 1157 1 12671 4992 1993 3089 3470 8854 8293 951 1129 452 408 2143 3534 32 297 2245 2581

N/A N/A 522.23 1 157.25 5.48 108.37 N/A N/A 993.71 1013.97 1.37 74.49 N/A N/A 14.37 32.85 4.25 28.11 N/A 398.7 1313.64 2029.30 2569.91 . 19L32J 12918 1 11902 13978 163721 17648

Source: Same as table 2.19. *: Means trade through ports and border crossings.

Table 2.21 Composition and volume of the major types of IRIRC freight in 1991 (000 tonnes). Animals Vehicles :; -Wood Passengerparcels Transit 'Army Imports 1111. -i goods goods o0d5 44 16 158 130 89 2073 28 Other Sugar Fruit/ Grain 'Cereals Other Metals Other 0 ther industrial mineral agricultural vegetable I 145 239 18 331 740 1015 5061 2107 Other oil Kerosene !--"'_-Gas Crude Salt St Coal Administration A k,. il '"6il products oil oods , 34 460, 162, 227 1813 699 1 1481 , ,
Source:Ministry of Financeand Economic (1992), unpublishedreport, p. 9.

The analysisof the availabledata during 1986-1990as shown in Table 2.22, indicatesthat among twelve rail districts in Iran, the three regions of South East Iran, South West Iran, and Tehran with 66.1% of the total rail freight are major sourcesof freight movements. South East Iranian railway as a source of iron ore to Esfahanhas averageof 26.4%; the
23

Chapter 2

Iranian Trade and Transport

South West is a main point for the carriage of crude oil and its products into the other parts

the country, and also as the most important rail outlet for foreign trade.The SouthWest of 22.9%, and Tehranwith 16.8%of the total freight have generated most traffic during with 1986-1990.
Table 2-22 Volume of freight handled in 12 rail districts of Iran (000 tonnes).

Rail districts Tehran Lorestan Arak SouthWest North North East North West Azerbaijan Khorasan SouthEast Zahedan Esfahan Total

1986 1988 347 159 3401 279 259 165 1149 158 33637 287 1376 12674

1987 2192 508 76 4040 269 253 189 1884 134 Z83 33 1126 14788

1988 2124 229 193 2371 3336 331 229 2027 123 3528 41 1428 12957

1989 2362 179 230 2411 304 334 303 1583 178 2987 61 1401 12334

1990 2671 272 268 3327 308 390 525 1574 152 3939 22 1432 14881

Source:IRIRC (1986-1990),Facts& Figures,English Version., p. 29.

2.3.3.5Rail productivity On the basis of permanent and contracted temporary employeesof the IRIRC during 1986-1993,the most important operational productivity indicators of the HURC have been computedand are shown in Table 2.23 (column 6) which generallyreveal the in decrease the number of employeesand the increasein the volume of the traffic units, rail resulting in increased productivity. Table 2.23 Operational productivity of the IRIRC for passenger and freight (units: millions otherwise SDecified).
Indicators 1
2 3 4 5 6 7 8 F9 ,0 Passenger-kin Freight-tonne-km Traffic unit (1+2) Rail length, km IRIRC employees, number Employee/kin of line (5/4), number d Traffic units/kin f l ine (3 / 4) Traffic units/ f 5 000 employee Passenger Freight- onn e

1986
4640 7320 11960 4567 41290 9 2.60 289.70 6.34 12.70

1987
3670 8630 12300 4568 40790 8.90 2.70 301.50 5.27 1 14.79 1

1988
4660 8050 12710 4568 38960 8.50 2.80 326.20 6.80 12.96

1989
4750 7960 12710 4569 38480 8.40 2.80 330.30 6.67 i2.33

1990
4570 7630 12200 4847 37170 7.70 2.50 328.20 7.81 14.88

1991
4590 7700 12290 4850 36940 7.60 2.50 332.70 8.14 16.98

1992
5300 8000 13300 5020 35780 7.10 2.70 371.70 8.22 17.65

1993
6420 912 0 ' 15540 5020 34450 6.90 3.10 451.10 9.17 1=9.83

Source: IRIRC (1986-1990) various pages; IRIRC (1984-1993) various pages.

24

Chapter 2

Iranian Trade and Transport

2.3.3.7Rail traffic and frequency While the maximum speedfor a passenger and a freight train is 80 km/h and 55 knVh respectively(Jane'sWorld Railways 1990)the average speedof a freight train in Iran in 1992 was only 4.1 km/h for a 4500 tonne train (PBO 1993).The South West railway district is the busiestroute for both freight and passenger trains as shownin Table 2.24.
Tahle 2.24 Dailv movements of trains alone rive maior IRIRC networks in 1992.
North I South West South East I North-West Train type North East 324 33 241 271 Freight 19 5.5 301 31 6.5 Passenger 19 Source:Ministry of Financeand Economics(1992), annex5, p. 6. Total 135 92

Train servicesand frequenciesvary during different seasons. analysisof the five year An (Table 2.25) shows seasonal and passenger freight operationof IRIRC during 1986-1990 that rail freight peak time is during April, October and November, while for passenger the traffic is high during August and March. The operationof HURC passenger movements and freight is subject to regular and long delays as a result of the single track layout and also becauseof technical deficiencies, poor programming, and lack of human skills (see
Table 2.26). Table 2.25 Seasonal train traffic in Iran (million tonnes). January "February March April "May 1197 1325 852 1044 1986 1090 1222 1987 1209 1441 1411 1083 878 820 1255 1988 1216 1342 998 1989 1092 855 1095 1008 1990 1365 1237 1084 1266 1251 Xugust September 'October, November' j ly -6 928 1049 1180 1054 1061 IF98 1987 1188 1300 1111 1130 1358 1988 1010 962 1097 1039 1112 1166 1225 903 1122 973 1242 1104 1142 1310 1339
Source: IRIRC (1986-1990).pp. 62-86.

'June 848 1102 1209 879 1168 December 1197 1233 1016 1066 1372

Table 2.26 Causes and nronerties of train delavs in Iran in 1992. -uman factors Causes Poorprogramming I Emergency Technical Delays % 441 36 17.5 1 2.5
Source: Sanate-Hamlo-Naghl (1993c),Issueof train delaysin Iranian Railways: No. 127,P. 35. assessment causes somesuggestions, of and

25

Chapter 2

Iranian Trade and Transport

2.3.3.8 Costs and revenues of IRIRC

Table 2.27 reflects the revenuesand costs of the IRIRC during 1984-1993and the gradualtrend toward an equilibrium as a break-evenorganisation. Table 2.27 COMDositionof costsand revenues of IRIRC. million Iranian Rials.
Costs Revenues Subsidies 1984 37660 25410 _ 14000 1985 42330 24060 11840 1986 40610 24060 10420 1987 46000 31100 8040 1988 50590 31750 6590 1989 51930 34220 5580 1990 76870 49180 4430 1991 99420 81240 2500

Source: IRIRC (1986-1990).p. 27

According to IRIRC (1984-1993,p.27), since 1984the IRIRC subsidieshave declined and it is claimed that in 1992 the company did not receive governmentaid. Revenue for passengerand luggage, freight, transport sub-revenues(warehousing and railcar hire internationaltransportand miscellaneous revenues indicatedin Table 2.28. are revenues),
Table 2.28 Revenues COMDOSition in IRIRC. million Iranian Rials.

Passengerrevenues Freight revenues Luggage Transport subsidies International transport revenues Other revenues Total

1986 5091 16749 132 1848 442 676 24938

1987 5729 20946 119 2539 906 864 31103

1988 7112 19204 135 3516 715 1069 31751

1989 1990 7512 8367 20105 31121 373 476 3357 1 4961 982 1 1238 1889 3018 34218 49181

Source: IRIRC (1986-1990), pp. 61-87. Note: The inconsistency between values in Tables 2.27 and 2.28 is due to source.

2.3.4 Port transport


2.3.4.1.Administration of ports Iran hasa coastline of about3000 km and is connected with the seafor foreign trade from both southandnorth. All seaports undera centralport authoritycalled the Ports and are Shipping Organisation(PSO) which is the owner of all seaports and is responsiblefor the maritime administrationof Iran. It is not only responsiblefor both the management and but operationof seaports, also for the constructionand maintenance all superstructure of and the infrastructureof seaports, ship registration,the examinationand issueof a certificate of for competence the establishment stevedoring of companies, shippingagencies port state and 26

Chapter 2

Iranian Trade and Transport

control, the safety of navigation and maritime pollution, communications and conventions,

of andenforcement the nationalmaritimelaw, etc. PSO is affiliated as a vice-ministerialorganisationto the Ministry of Roadsand Transport (MRT) of the Islamic Republic of Iran. It has four main divisions: Financial and
Administration, Technical and Engineering, Planning and Programming, and Ports and

Marine Operation.

23.4.2 The port authority directorates in coastalprovinces The Maritime Administration of Iran (PSO) has 13 internationallyrecognisedand into eight major and five minor commercialports in Iran. are categorised activeports which There are also two inland lake ports in the north west of the country on both sidesof Lake Uroomiyeh.Eachdirectorateadministers, protectsand suppliesoneor a seriesof other minor matters.Appendix I gives the provincial directorates the Portsand ports in all port-reIated of SWpping Organisation the Islamic Republicof Iran by coastalregion. of 23.43 Infrastructure The major present infrastructure and superstructure(including cargo handling vary amongdifferent ports, but the ports of equipmentand storageareas)of Iran's seaports Imam and Abass have superior cargo handling facilities. On the basis of analysisin this as chapterthey areselected the landbridgeports for this study.The deepberthsand anchorage are in thesetwo main major southernports. Other major ports are equippedwith facilities accordingto the natureand volume their operation.All major and minor seaports combined
have 17,256 metres of berths with a maximum depth of 12m for containers and general

cargoesand 15m for iron ores (the ports of Khoramshahr and Abadan and the lake port of Uroomiyeh are excluded as major ports since their infrastructurewas damagedduring the war). Theseports have a total numberof 100 seagoing bargingberthsandthe entireport and

27

Chapter 2

Iranian Trade and Transport

is conducted through different terminals composed of one or more berths and other operation

facilities. superstructure About 77% of the total length of berths(seeTable 2.29) is suitablefor ocean-going shipsof than 6m draughtand the rest function with a lower working depth. Among all ports, more only the two main ports of Imam and Abass can serve ships of 12 m draughts.The other only, while Caspian ports only acceptshipsof lessthan southcoastports allow to 6-8 metres; 6 metresdraft. Table 2.29 showsthe generalcompositionandcharacteristics infrastructure of of of the seaports Iran in 1993. The portsof AbassandImam, as the two most importantandlargestcommercialseaports of hold most Iran (with 68 berths)in tenns of length anddepthof berthsandchannelentrance, Table 2.29 The infrastructure of six major Iranian seaports.
1 Type of berth Ocean going: No. of berths 77 Capacities* 26.65 Depth (meters) 9-15 Length (metres) 14592

of which 61 19.45 12 11101 General 4 10 12 2051 Container 2 3.2 10-15 440 Iron ore 2 N/A 560 12 Grain 440 2 N/A 12 Liquid bulk =2 23 1-61 26 1 Barge harbours 1-15 1 17256 56 100 1 30.15 1 3 Total for Source:Basedon PSO (1993), annualperformance the Ports and ShippingOrganisation 1993,and PSO of Operational PSOestimates. Reportof the PSOfor 1992,variouspages. million tonnes, *: (1992)variouspages.

infrastructure(excludingthe war-damaged (79% of total 86 berths)of the above-mentioned and of Khoramshahr Abadan).In terms of berthsthe ports of Abassand Imarn have a ports varietyof differenttypes(seeTable 2.30). Gulf, the port of Anzali on the Caspian As well asthe portsof Imam andAbassin the Persian Sea, due to its infrastructureand facilities comparedwith Nooshahr,can also function to
serve the CAC countries for the second sea leg of the landbridge. Therefore, it is important to

studytheir presentinfrastructureand potential.A featureof port Imarnis that it hastwo huge to petro-chemical complexes alsothe Mahshahroil productterminaladjacent the port. and

28

Chapter 2

Iranian Trade and Transport

Table 2.30 Facilities of different seaports of Iran in different berths in 1993. Chah Abass Port1berths Imam I Anzali Bushehr Bahar 5 0 0 1 0 5 Terminal container 0 0 0 0 2 2 Multi urpose 0 0 0 3 Heavy cargo berths 0 4 5 4 24 4 4 17 General cargoes 01 0 5 0 0 2 Liquid cargoes .6 1 0 0 0 Iron ore 1 2 0 0 0 0 7 Silo (Grain) 8 0 0 0 0 Ro/Ro 1
9 10 Aluminium terminal 1 0 0 0 0

Nooshahr 0 0 0 2 0 0 0 0
0

Live-sheepterminal 11 1 Oil products 12 1Total

1 1

1 0 341

0 01 51

0 11 51

0 11 51

0
21

0
d 3

341

. Source:Basedon PSO (1993), annualperformanceof the Ports and Shipping Organisationfor 1993 various
Inconsistency pages. with Table 2.29 is due to sources.

23.4.4 Port storage facilities in Iran Other important factors of capacity in Iran's major seaportsare the large reserved facilities andportsequipment. After the two congestion the areas, numberandtype of storage
crises of 1974 and 1980 and the closure of the main seaport of Khoramshalu, a great attempt

in was directedtowardsthe constructionand completionof transit shedsand warehouses all link betweenimported ports, but mainly in the port of Imam to establisha more reasonable were disruptedby the and goods transferredinland, as infrastructuredevelopments goods and revolution and the war with Iraq. In 1993all six major ports had 45 coveredwarehouses transit shedsand 9 under constructionin the port of Imam. Tables2.31 and 2.32 show total facilities andthoseunderconstruction nationalandindividual port levels. at existingstorage Table 2.31 Storage capacity and types in seaportsof Iran in 1993(sq. metres).
I Open storage I Covered storage I Facilities Total I Under construction 1 4421721 143746621 Area & volume 139324901 1920001 Source:PSO(1993),annualperformance the Portsand ShippingOrganisation 1993.variouspages. for of

As shown in Table 2.32 the ports of Imam and Abasshave the highestproportion of both openand coveredstorageareasequalto 29% (Imarn)and47.5% (Abass)of the openstorage in and38.7%(Imam) and43.2% (Abass)for all storage areas Iranianports.

29

Chapter 2

Iranian Trade and Transport

Table 2.32 Share and types of storage areas of major seaports of Iran in 1993 (sqJmetres). Imam Anzali Bushehr Chah Bahar Abass Nooshahr Total Open storage 1228000 154175 443000 350000 2008000 46600 4229775 Covered 171000 22351 23000 18000 191000 15000 440351 storage Total 13990001 1765261 4660001 368000 1 2199000 1 616001 4670126
Source:PSO (1993), annualperformance the Ports and Shipping Organisation 1993,various pages;PSO for of (undated in probably 1991 An investigation the performance the IranianPortsfor 1941-1990), 64. a), of p.

2.3.4.5. Port equipment

In tenns of the superstructure ports, basically all port operationsrely on the port of authority's mobile and fixed equipment.The lack of quay cranesis obvious in Iran's major southernports; the only quay craneswhich were acquiredand have becomeoperationalin recentyearsservethe containerterminalsof Imam and Abassports. For the first eight years of the revolution and the war with Iraq the port equipmentwas usedextensivelywith little repair due to lack of spareparts,currencyproblemsand little domesticproduction.There is The fixed and mobile heavy and light equipmentand also a low level of containerisation. installationsfor all six major port authoritiesare shown in Table 2.33, indicating the great
reliance of port operations on ships' gear and cranes, and also on general cargo practices for

ship to berth andthe transferof cargointo storage. Table 2.33 Main mobile and Rxed port authority equipment in Iran in 1993.
Type of equipment Units Remarks Crane 160 Lift truck 169 Tractors 196 Transtainer 12 Only in ports of Abass (10) and Imam (2) Top lift 14 Towing head 39 Trailers 654 Gantry cranes 4 Only in ports of Abass (2) and Imarn (2) Electrical rail mounted Installation 6 Only mounted in Northern ports Pneumatic conveying installation 13 Pneumatic ship unloader 4 Only in ports of Abass (2) and Imam (2) Grain ter7ninal silos & installation 21 Iron ore installation 21 1 Source:PSO(1993),annualperformance the Portsand ShippingOrganisation 1993.variouspages. for of

30

Chapter 2

Iranian Trade and Transport

While all main gantry cranes, transtainers, grain towers, and quay cranes etc. are as specified

in Table 2.33 the shareof all six major portspresented Table 2.34 indicatesthat more than in half of the mobile cargo handling equipmentis locatedin the two major ports of Imam and Abass.
Table 2.34 Share of different ports in main cargo handling equipment in 1993. Tractor Lift truck Top lift Pneumatic unloader Crane 54 Port Imam 50 75 5 2 48 Port Abass 38 25 6 2 25 29 0 Bushehr 17 0 Chah Bahar I1 201 II 01 6 181 301 --2Anzali 261 1 1 131 17 1 Nooshahr 181 1 2
Source:PSO(1993),annualperforniance the Portsand ShippingOrganisation 1993.variouspages. for of

2.3.4.6.Imbalance of international seatradein ports Iranian maritime imports are substantiallygreaterthan non-oil product exports. In 1993 imported non-oil product commoditiesand goods through ports totalled 16,375,000 tonnesbut exportsonly 4,647,000tonnes(a ratio of 4: 1). Sincethe end of the war and up to 1993the seatrade as shownin Table 2.35 (column 6) has approximately doubledin volume
for both imports and exports. Table 235 Iranian non crude oil sea borne and total foreign trade (000 tonnes). Foreign trade 1988 1989 1990 1991 1992 Seaimports with oil products 12447 15019 17379 19378 21748 2 Seaimports non-oil 8225 12764 15034 16378 16734 3 Seanon-oil exports 1178 1204 1134 2117 3082 4 Seatradenon-oil ( 2+3) 9403 13968 16168 18495 19816 5 Seaexports with oil products 1248 1317 1338 2378 3399 6 13695 16341 18717 21756 25147 Total seatrade (1+5) 7 Total Iran non-oil imports 10742 19241 21045 21501 19091 8 Total Iran non oil exports 1224 1455 1308 2395 3351 9 Total Iran exports with oil products 1294 1568 1510 2657 3668 10 Total Iran non-oil trade (7+8) 11966 20696 22353 23896 22443 11 Total Iran imports with oil products 149 21496 23390 24501 24105 12 Total Iran non-crude oil trade (9+11) 16258 23064 24900 27158 27773 ,

1993 22651 15728 4498 20226 4924 27575 18861 5037 5433 23898 25784 312171

Source:PSO (1993), annualperfomianceof the Ports and Shipping Organisationfor 1993,p.36. SCI (1993), Iran Statistical Yearbookfor 1993,pp. 452-456.

The cargo traffic in most major Iranian ports consistsof some considerable loading and/or unloadingof oil productsby ship and shorepumping systemsthroughpipelines(columns 1

31

Chapter 2

Iranian Trade and Transport

and 5). The volume of oil product imports during the war and also during 1988-1993 through

(seeTable 2.36).The total with total tradeduring the war, was considerable ports, compared foreign trade through seaportsduring the twelve years 1967-1978was about volume of 98,007,000 tonnesand varied between2,761,000tonnesand 15,462,000 tonnesfor 1967and 0,199 1a). During the twelve yearsafter the 1978respectively; increase about460% (PS of an
Islamic revolution from 1979-1990, the total trade carried through major seaports was

announcedas 168,819,000tonnes indicating an increase of only 114.5% whereas the had doubledsincethe revolution in 1979.The lower figure for tradeof 9,623,000 population tonnesduring 1990. tonneswasin 1979andthe highestwas 20,645,000 In termsof the foreign tradein generalcargoand oil products,the four southernPersianGulf during ports contributedon averagea shareof 92.7% of the total cargo trade of the seaports (Table 2.37) while the northernports in the CaspianSeacontributed the six years 1988-1993
the rest (7.3%).

At the regional level the two ports of Imam and Abass accountedfor almost 84.5% of the
total cargo handled (119,183 tonnes) by the southern ports, equal to 78.3% of the total

(128,569 tonnes) for the ports of Iran during 1988-1993.The composition of sea trade includessixteencategories cargoes (metals,fertiliser, barley,maize,coal, sugar,rice, dairy of fish, generalcargo,chemicals,oil products,wheat, meat, soya,and vegetable products, oil) for imports and 21 for exportswhich are the basisfor annualreports.By classifyingall into thesenine broadergroupsas indicatedin Table 2.36, it is possibleto show the throughputof all six major ports of Iran during 1988-1993and for later stages(PSO 1993).Table 2.36 indicatesthe low proportion of containerised goods (1% to 3% of the total during the six years) and the high volume of generalcargo and dry bulk (grain, barley and maize), and baggedcargo (mainly sugar and fertiliser). Using the survey of 1993 as a base year (a relatively stable year after the war) it can be seenthat oil products (mainly imports) still

32

Gn

0.6 ba

0" 1" b-A =

o 00 -A

Ch tA

A. W

I.-

0
lu

O-L eD

CL

ID ol -

CD CL

tz

'0 eD j

0
*0 00 0 :Z
0-

10 -P6

I. CS 3i - 12 00
, V. -4

'D

"

L43

! F

pb

00

VI

00
'0,00 L t)

O-A

00

CN 1.0 00

Cs 0
59

VW 0"o 0-0
LA

t) w 1.0 8 ).0 ON 10

LA -14

00

ON !

00

00
%0 W tA
CN w D % CN "

Z! 'O

C) w

LA

w 00 1--

ON LA I

t-j w

w 8 0. t

I I

G
l .4. LQ
I

ll

U 52a

0"

00 w 60 -Ph 00 o 0

(7

1
v tA LA w t-j LA oo w U) W LA 0 tA .91. 00 A

LA

00 L4

00
w ol ti

1 -

0
. 1h.

0- 00 "' " tJ 0% 0

00
LA

00 (D .4 :rl.
4 LA

-j
i u 0

LA

00
33

C)

CA

'cr> F- E F-F-& E E e E 1 F-.: F-,E "0 e -oC, . Z E e A Q Irl YQ o Gn. -t -0 m Z ;g 0 0 j csju 0 M, &
1. Z

,D

EL Ici PO

tli el

(> CY\

eg

C)

(>

> IDO

LA --3

>

w
00

C: 0, - ) ,

u (D >%D

00 o
o 00 >-

m-3 w D

-4 00

Ln

t2 (A b- , 0% %0 O\ . 0 -2 00
.4

-0

CD
u3 43

ll-,
LU 10 (: tli

--a

CD

2
%0 00

O ,_ tg .
CY% Q

'21 .
00

e-

%0

00 LA

Z 82,0 Iwi 1c22sll l%


9

tA (>O CD

w 00

A 00

t2 N)

ON

Co
00

Z. 1. 00

4) (:

Z s 2

h-

k9 0 Z; SO, 'lI C) 9 1 - 75 lZq 2 oo -0 2 w 0


1 00 -

t.A %0 9oi
(CID

-4-

00 00

t-i b-

00

N. -4

bW.

I
t

-4

z -

-l']

2 00 ,0 . 9 - -

0, 0% 00 00

vb--

34

Chapter 2

Iranian Trade and Transport

the highest proportion of trade (see Table 2.38). The flow is handled directly by constitute

to other areasand also to storagetanks outsidethe port and Customsareas.In nonpipelines ) trade,traditional dry bulk cargoes(grain, etc. and then generalcargoesare the oil product main flows throughseaports. 2.3.4.7Performance of ports As mentionedearlier, most cargoesgo through the two largestports of Imarn and in Abass,followed by Bushehrand Chah Bahar. After the cease-fire 1988the port of Imam its hassteadilyincreased contributionsto such an extentthat in 1993it had 36.4% of the total shareof the port during 1988-1993 was 31% of the seaportforeign tradeof Iran. The average total seatradeand was mainly composed metals,followed by generalcargoes of and dry bulk in 1993.This situationis shownin Table 2.39.
Table 2.39 Composition of throughput for the port of Imam for 1993(000 tonnes).
Trade hnports Exports Total ,% General 1493 1660 3153 31.4 Dry bulk 2017 2017 20.1 Minerals 123 123 1.2 Bagged 959 959 9.5 Reefer 116 116 1.2 Metals 2674 1005 3679 36. 1 Total 7259 2788 10047 100

Source:Basedon PSO 1988,1989,1990,1991,1992,and 1993annualperforrnances the PortsandShippingOrganisation variouspages. of

by The port of Abassis the other large port suggested the analysisas a suitablelandbridge handledabout 63.7% of the total seaimports of the country during the war period port, and Rajayeeand Bahonar)accounted (PSO, 1991a).The port of Abass(both Shaheed 1981-1988 for between40% and 56% of the total ports' national trade during 1988-1993, equivalentto betweenabout44% and62% of the total southernports'performance.
Table 2AO Composition of throughput
Tradee rad I Imports m or Ex orts Total % General 2996 715 Dry bulk 1441 000

for the port of Abass for 1993 (000 tonnes).


Reefer 55 158 Bagged , 958 000 "r Oil products 4795 0000 i Metals 55 525

Trade Imports Exports Total %

3711 28.1 Mineral 830 86 916 6.9

1441 10.9 Vegetableoil 610 000 610 4.6

213 1.6 Total 11740 1484 13224 100

958 7.2

4795 36.3

580 4.4

Source:Basedon PSO 1988,1989,1990,1991.1992,and 1993annualperforniances the PortsandShippingOrganisation of variouspages.

35

Chapter 2

Iranian Trade and Transport

The third ranking southern port is that of Bushehr with 9.7% of the total southern port trade

Bushehr,with a total trade of 9.1% of Iran's non-oil port trade in the period 1988-1993. and in 1,586,000 tonnes(63% oil products)is mainly engaged generalcargo and baggedcargoes like rice and sugarimports,andalsois an importantport for the regionaltradeof small vessels PersianGulf. in Arab States the Southern with The fourth most important southernport, particularly for Central Asian trade, is the port of Chah Bahar on the Indian Oceanwith a 4.8% sharein the operationof the southernports It during 1988-1993. contributedonly between2.6% and 5.5% of the total annualperformance (or between3% and 5% of all national cargo throughput for ports) within southernports This port is not operativefor about three months of the year during the during 1988-1993. Trade varies but it is mainly concernedwith grain and generalcargoesas monsoonperiod. shownin Table 2.41.
Table 2.41 Composition of imports and exports in ports of Bushehr, Chah Bahar Anzali- Nooqhahr for 1993 (000 tonnes).
Ports & trade
General Dry bulk Minerals Bag Oil products Vegetable Oil Reefer Metals Total

Import . Export Bushehr % Import . Export Chah Bahar % Import _Export Anzali % Import Export Nooshahr %

207 93 352 16.81 841 21 86 11 183 42, 225 21 74 8 82 21 ,

53 00 53 3 256 000 256 31 00 00 00 00 00 00 00 00

00 150 29 000 29 150 1.6 8.4 1 00 156 00 000 0 156 0 19 0 0 0 0 0 0 0 ,0 10 00 10 1 20 0 20 5

999 000 999 67.1 320 000 320 39 577 000 577 53 ,0 232 0 232 58

3 0 3 0.17 0 0 0 0 0 0 0 0 0 0 0

00 52 52 2.91 00 00 00 00 00 00 00 00 00 1 0 0 0

000 000 000 000 000 000 0 00 266 000 266 25 62 0 62 16.

1412 174 1783 100 816 2 818 100 1036 42 1078 100 388 8 396 100

Source:PSO(1993),annualperformance the Portsand ShippingOrganisation 1372.variouspages. for of

In the north, Iran has only the two ports of Anzali and Nooshahrwhich are limited to coastal
trade with states around the Caspian Sea (including Russia), in winter, and in summer through

36

Chapter2

Iranian Trade and Transport

the Volga and the Don to the Black and the Baltic Seas.They both had about 7.3% of Iranian sea trade during 1988-1993 and 5.2% in 1993. Anzali handled 59.5% of the total northern ports' trade during 1988-1993 but nationally accounted for only 4.4% of Iran's total sea trade. Nooshahr, the second port in the north, covered 40.5% of Iran's Caspian sea trade (equal to 3% of national port cargo throughput) during the six years to 1993. The trade of Nooshahr port mainly includes general cargoes and metal products while imports of oil products from Turkmenistan and Azerbaijan during the war with Iraq and later accounted for more than half of the port operations as shown in Table 2.4 1.

2.3.4.8Productivity and serviceof ports in Iran Accordingto Honeycutt(1989,p. 24): into "In thedevelopment human three of resources, factors come play:theemployee ; thework ... three that of elements determine ; theclimate. It is theinteraction these overallproductivity, ... ... to employee satisfaction commitment theorganisation7. and The assessment Iran's seaportproductivity, where there is centralisedgovernmentcontrol of over cargooperations, equipmentand land and seatraffic, and management within ports, is a difficult task. Port productivity in Iran can be criticised for both internal inefficiency and also for the poor quality of the road/rail transportfleet androlling stock.Thereis evidenceof both in quantitative and qualitative shortages skilled labour and personnel,port equipment and procedures, which impairs the ports productivity. The generalport productivity of Iranian etc. seaports calculated is shownin Table 2.42.It is based the following equation: and on was Productivity= Annual seaports' trade/Totalberths'length/Annual working day. _j The labour productivity of ports must coversall organisations directly involved such as port authorities,stevedoring, shipping,forwarders,shippersand customsadministrationin the port Cargo handling in all three trade.But such data is only availablefor the period 1991-1993. stagesof port operations- ship, quay, and storage- is directly under the practical control of both the portsauthoritiesandthe stevedoring companies while only the stevedoring companies

37

Chapter 2

Iranian Trade and Transport

involved in the ship (hold) operationin the southern ports whereport authoritiesplay a key are in the quay and storage an and operations can be considered importantfactor in the entire role Therefore,productivity in termsof tonnes/employee, shown handlingoperations. as port cargo
in Table 2.42 (columns 8 and 10), is based and calculated on available information about the PSO personnel who receive training programmes (PSO 1993) and the maritime transport

(Atrchian 1996).The productivity relatedto the PSO indicatesa generalimprovement sector during 1988-1993, which is partly due to the coursesgiven and partly to the increasein the in trade simultaneous with a decrease the total numberof PSO employees. volume of ports The productivity measure with total maritime transportshowsa decline due to the concerned increased employmentin the sectorover the period of study.
Table 2.42 General port productivity where ntherwise snecified). of Iran's six major seaports (000 tonnes except

Year 1 Total non-oilseatrade 2 Berthslength(metres) Working days 3 4 Productivity:Per day/annum/ metres(1: 2: 3) 1tonnes/berths Total seaports 5 trade 6 Productivity:Per day/annum/ tonne/berths metres(5: 2 :3) PSOtotal personnel 7 8 Productivity:tonnes non-oilsea (in) trade per PSOemployee 9 Total marinetransport employees 10 Productivity:tonnes non-oilsea trade per total maritime (1/9) transportemployee

1988 1989 1990 1991 9403 13968 16168 18495 14592 14592 14592 14592 341 341 341 341 1.9 2.9 3.4 3.9 1 13697 17717 20849 23901 4.8 2.8 3.6 4.2 6457 1456.3 N/A N/A

1992 19816 14592 341 3.9 24830 5

1993 20226 14592 341 4.1 27575 5.5

5956 6598 6837 6757 6145 2117 2364.8 2737.2 3224.7 3395.9 N/A N/A NIA N/A

1 40817 38543
502 479 I

43225 467.9

Source: SameasTable 2.36. PSO (undated probably 1991a), An investigation into the performance of Iranian ports for 1941-1990, various pages; PSO (1993), p. 51 for PSO personnel; Atrchain (1996), for marine transportation employees. pp. 24-25. Note: assumesminimum working days per annum of 341 based on official holidays during a Iranian year.

The composition of maritime sector includes four other sub-groupsconcerning all those organisations companies and with direct influence on maritime transport(PSO,NUC, IRISL, private shipping companiesand agents),supporting services(insurance,surveying agents,

38

Chapter 2

Iranian Trade and Transport

Customsservices,etc.), industries(ship and boat repair and construction),and construction (development improvementof ports). and The quality of port performances be shownby using two classes indicators- ship and can of the cargo, and personnelindicators using different units of measurement; time, tonnage
handled, and cost for ship, quay, and storage operations. The analysis of the past and current

port efficiency is basedon data for 1992publishedby the PSO,which is the only available

source.
The five ship indicatorsmost commonly used to show both maritime and cargo traffic over ports are port or turnaroundtime, waiting time, servicetime, operationtime, and idle time. Theseindicatorsarecomputedand shownin summaryin Table 2.43.Total port, or turnaround time, is defined as the total time taken for arrival and readinessfor berthing and for the For clearingof goodsfrom or into hatches. eachship in eachof the six major ports it varies betweenI and30 + daysin 1992.The annualaverage time of a ship as shown port turnaround in Table 2.43 is the sum of the stayingdaysfor every ship divided by the numbersof ships calling to eachport during that year.Port turnaroundtime also includesthe waiting time and servicetime of shipsduring their stayin port. Table 2.43 Average port turnaround time of ships visiting sevenmajor Iranian ports in 1992(davs). ajor Iranian ports
Abassports 2 3 Bushehr 4 Chah Bahar 5 Anzali 6 Nooshahr 7 Total

I 2 3 L4 5

Ship Indicators Port days Operation davs Service days Waiting days Idle days

Imam

Rajayee

Bahonar

11.15 9.22
9.44 1.71 0.22

11.36 8.36
8.83 2.53 0.471

15.80 7.15
7.93 7.87 078 1

8.90 6.72
7.08 1.82 0.36

14.01 10.54
11.49 2.52 0.95

2.98 2.29
2.58 0.4 0.29

2.98 2.40
2.69 0.29 0.29

9.60 6.67
7.15 2.45 0.48.

bource:iiased on Fsu (1992),operationalreportof me ebu for 1992,variouspages.

39

Chapter 2

Iranian Trade and Transport

Accordingto UNCTAD (1973,p. 27) the two principal indicators- serviceandwaiting timesaredefinedasfollows: Servicetime is the total time spentat a berth, including idle time. A ship berthedalongside
for discharging or loading is liable to some interruption. Therefore, service time includes both operation and the idle time which a ship may face during its berthing. Operation time, as a

proportionof the servicetime, refersto the time which a ship is alongside berth to deliver or a receive cargo without any interruption causedby ship, quay equipment,gang failure or weatherconditions. Idle time is the difference between service and operation times, and indicatesthe total pauseand down time if the cargo operationof a ship is for any reason interrupted.Waiting time is the averagetime a ship spendsbetweenarriving at the port and arriving at the berth where cargo-handlingis to take place. It is therefore,the difference
between port and service times and the time for ships waiting in anchorageto come to a berth

for loading andunloading.In the analysisof the ships'port indicatorsin 1992as a sarnpleof port services,it must be taken into accountthat 21.3% of the total port tradesin Iranian seaportsinvolves the easy and speedytransfer of oil products. This results in the ship indicators being lower than otherwise would be the case. The comparison of the two by consecutive yearsof 1991(calculated the PSO)and 1992for threeship indicatorsat major in that times. portssuggests therewereimprovements port turnaround Table 2.44 Comparisonsof the ship indicators in Iran (davs).
Ship indicators Port days Service days Waiting days 1991 9.76 7.20 2.55 ] 1992 9.60 7.15 2.45

Source: for 1991 is derived from PSO (1992). For 1992 same as Table 2.43.

23.4.9 Waiting times and demurrage The extensivewaiting time of shipshastwice severelydamaged economyof Iran; the first early in the 1970sat the time of suddenoil price increases, then during 1981-1982 and 40

Chapter 2

Iranian Trade and Transport

because the war with Iraq and the closureof the port of Khramshahr, of resultingin the poor level of operationat the port of Imam as the other largestport and also other ports (Lane 1984).The economyhasalso sufferedfrom generalinefficiency including shortages cranes of
and road/rail facilities. According to the statistics of the Central Bank of Iran (Nooshahr, 1982), during these two greatest periods of congestion, Iran's economy suffered annually by

for (Nooshahr1982)were: aboutUS $ onebillion. The reasons demurrage in * Congestion destination ports * Delaysduring discharging times * Delaysin the openingof tradecredits. is The contributionof different importedcommoditiesto demurrage shownin Table 2.45.
Table 2.45 Share of different main basic imported commodities in demurraae durine 1992. Fertiliser Commodities Wheat Fish m=ealRice Suga; Others 40 2.7 % 27 15 9.5 5.8
Source: MRT (1994), Complete Text of the Proceedings and Approvals of the Transportation Co-ordination Group of the High Council of the Islamic Republic of Iran. Vol. 2. various pages.

Table 2.46 indicatesthat during 1988-1992 aboutUS $ million 237.9was paid in demurrage.
During a period of generally increasing imports, the extent of demurrage (US $ per tonne) has

fallen. Table 2.46Imports and demurrage in Iran.


(US $ million) _Demurrage imports (000 tonnes) _Seatrade tonne / US $ Demurrage Ships' traffic 1988 45.60 7.74 5.89 1078 1 1989 68.54 12.76 5.37 13141 1990 70.60 15.03 4.70 16441 1991 46.97 16.37 2.87 1897 1 1992 6.15 16.23 0.38 19391

Source:Basedon MRT (1994),Proceedings approvals the TransportCo-ordinationgroupof the High Council and of VariousPSOpublicationsfor 1993,1992,1990,1989,1988. of the IslamicRepublicof Iran. Vol. 2. variouspages. variouspages.

Theseimprovements bodiesand werepossibledueto the interactionbetweenthe government port industries.The PSO was able to have a more open hand in purchasingand equipping ports with new container and generalcargo technologies,and in conducting many more At training courses. the sametime, after the war normal working procedures were gradually
implemented and existing equipment was repaired and replaced (PSO 1993).

41

Chapter 2

Iranian Trade and Transport

23.4.10 Maritime and land traffic in Iranian seaports During 1979-1993,21754 seagoing vessels visited Iranianports,of which about70% Gulf andIndian Oceans visited the Persian port in the Southof Iran and30% the two Caspian Seaports (PSO 1991a,PSO 1992,and PSO 1993).The shareof eachport is shownin Table 2.47. Among the major southernports, Abass and 1[mam. accountedfor 37.4% and 18.4% respectively.After the end of the war with Iraq the volume and direction of traffic was during the previous changed considerably the port of Imam, which wasnot operational as
Table 2.47 Maritime sea going traffic visiting Bushehr Abass Chah Imam Bahar 213 1979 607 71 00 316 74 00 1980 425 486 1981 273 114 00 165 495 00 1982 124 630 1983 237 185 00 143 672 1984 114 51 . 1985 150 772 44 000 655 1986 000 135 42 164 638 1987 27 56 163 397 1988 87 43 140 460 1989 283 65 147 512 63 1990 352 609 1991 544 152 60 . 1992 138 651 72 567 _ J 160 639 1993 584 43 Total 4224 21FI 8145 539 9.7 % 18.4 74 2.5 1 Iranian m2kjor ports. Total Anzali Noo- Total South shahr North 891 219 50 269 815 304 76 380 873 221 100 321 784 310 232 542 1052 312 288 600 980 272 89 361 966 249 130 379 832 116 84 200 885 242 346 588 690 133 255 388 948 232 134 366 1074 386 184 570 1365 349 183 532 1428 345 166 511 1426 371 124 195 15009 4061 2441 6502 9 1 .9

Total Iran 1495 1395 1194 1326 1652 1291 1345 1032 1231 1078 1314 1644 1897 1939 1921 21754* 100

Source:PSO(undated report probably 1991a),An investigationinto the perforinance Iranian ports for 1941of 1990,part H. p. 9. Different PSOpublicationsfor 1993,1992,1991, and 1990,variouspages. *: the traffic of Kharamshahr Abadanduring 1979and 1980(1.1%) includedin total. and

threeyearsto 1988,graduallyre-established major position in the maritime traffic of Iran by a 1993.The daily rate of ships arrival into ports after the end of the war was increased during 1988-1993 78.2%with an average of 13. annually.The seatraffic during this period by % rate throughthe southern ports (6931 ships)is about2.4 times that of the northernregion (2862). Table2.48 showsthe average daily rateof ship arrivalsin eachport on a 365 day basis. Beforethe revolution Iranianports had an activecontainerandLASH (Lighter AboardShips)
trade. Due to the war with Iraq container traffic declined. However, with the end of the war

42

Chapter 2

Iranian Trade and Transport

andthe completionof the Abasscontainerterminalandequipment,the tradehasshownsome recovery(seeTable2.49).


Table 2.48 Maritime traffic in major Iranian ports (shipstday).
1990 1989 1991 1992 1988 1993 Ports 0.79 0.96 1.50 0.24 1.60 Iniam 1751.30 1.40 1.70 1.10 1.80 1.80 Abass 0.38 0.43 0.42 0.45 0.38 0.44 Bushehr 0.18 1 0.17 0.60 0.12 1 0.20 0.12 Chah Bahar 0.64 1.10 0.96 0.95 0.36 1.10 Anzali 0.54 0.37 0.50 0.46 0.70 0.34 Nooshahr into Source:PSO(undated a), of reportprobably 1991 An investigation the performance Iranianportsfor 1941-1990, H. p. 9. Different PSOpublicationsfor 1993,1992,1991, and 1990,variouspages. part

Table 2.49 Container ship traffic in Iranian ports (shipstyear).


Semi container Full container Container trade Semi container Full container Container trade 1978 1979 1980 4981 ", '1982 -1983 108 84 187 25 93 276 61 62 136 74 115 443 N/A N/A N/A N/A 258 N/A 1987, 1988 1 , 1989 1990 1 '1991 1992 ' 106 62 48 229 209 667 146 135 108 216 57 47 153 296 696 673 231 704 1984 91 105 238 1993 708* 797 1985 66 65 195 '1986 75 116 184

Source: PSO (undated report probably 1991a), An investigation into the perfonnance of Iranian ports for 19411990, part II. p. 9. Different PSO publications for 1993,1992,1991, and 1990, various pages. *: Data is available only for total semi and full container.

The traffic by land vehiclesinto seaports alwayshas been a critical issue as a result of the imbalancebetweenthe volume of trade and the number of heavy goodsvehicles,the long betweenports and destinationsand origins of cargo, adverseweatherconditions, distances lack of inland loading/unloadingfacilities, etc. Table 2.50 indicatesthat the ports of Imam and Abasshave dominatedthe traffic for lorries 26.5%and54.4%of the total in 1989,and29.6%and 59.1% in 1993respectively, with about while the port of Imam hashad all the rail traffic of the ports.
Table 2.50 Number of heavy goods vehicles and wagon traffic into Iranian seaports.
Port Imam Truck Rail 1989 1990 175122 209490 ther port s: heavy g Bushehr Abass ChahBahar vehicles Anzali Nooshahr HGVs Total ports

41146 22269 25041 52878 359899 10789 662103 52210 34559 27577 76250 424054 14583 811146 49657 13749 38743 71922 15446 1991 8628 1 _438373 1992 298201 23600 58320 18211 61230 481031 15998 932991 1993 264245 28900 29250 275161 28338 5277161 160041 893069 . Source:PSO (1992); PSO annualperfonnancereports for 1993,1992,1991,1990, and 1989 variouspages; Wagonfor 1992and 1993is basedon Sanate-Han-d-o-Naghl, (1995c),No. 144,p.44.

43

Chapter 2

Iranian Trade and Transport

The distribution of road vehicles for port cargoes is organised through the Vehicle

Distribution Co-ordinationCentre as part of the MRT and functions in close co-operation with port authoritiesat ports and border crossingterminalson the basis of "first come/first served". 2.3.4.11Port hinterland in Iran There are very few up-to-date,and comprehensive publishedstudiesabout the port hinterlandin Iran. Due to the concentrationof industry there, most of the port and border This is crossingimported foreign trade is destinedfor greaterTehranand its suburbanareas. followed by Esfahan. Another importantaspectof the port hinterlandin Iran is that both cargo and do not consider owners and transport authoritiesand operatorsare government-owned specificmodal costsas important.Therefore,destinationports for cargodelivery may change frequently.

2.3.5 Border crossing operations


Iran's Customs administration is responsible for the operations in different border-

crossingterminalsand inland Customszones.The supervisionof theseborder crossingsis assignedto Customs,but cargo operationsand warehousingare carried out by the Public WarehousingCompany. Since most of the foreign trade of the country takes place with Europe and Japan,the two border crossingsof Bazarganand Djulfa play significant roles. The former is mainly concerned with land transportwith Europeand Americaby road freight only, while the latter is involved with the European Far Easttrademainly by rail through and
the Trans-Siberian Railway. The trade though border crossings was influenced after the

collapseof the former Soviet Union and, as shownin Table 2.51, Djulfa's trade,due to the Azerbaijanand Armenian conflicts, declined and stopped,while trade through some other points (e.g. Astara)hasincreased significantly.The other bordercrossings with a pivotal role

44

Chapter 2

Iranian Trade and Transport

for Central Asian countriesis Sarkhswhich was connectedto the rail network to Central in 1996c). 1996(Sanate-Hamlo-Naghl Table 2.51 Non-oil imvorts of Iran (000 tonnes).
Ports & border crossings Djulfa Bazargan Astara. Razi Mijaveh Total annual border crossings Total non-oil ports Total Iran Ports & border crossings Djulfa Bazargan Astara Razi Mirjaveh Total annual border crossings Total non-oil ports Total Iran 1987 2326 979 91 579 31 4006 11161 15167 1991 2367 1821 158 153 28 4528 16973 21501 1988 2239 843 87 2745 23 5937 8239 11816 1992 109 1826 281 108 33 2357 16734 19091 1989 Total border Average % share crossings 1803 2689 11535 43.1 834 1827 8947 33.4 124 1 118 1546 5.8 445 498 4543 17 52 14 186 0.7 3258 5146 26757 100 958921 122760 78.1 100 1990

13271 15899 16529, 21045 1993 0.02, 817 687 14 5 1524 13615 17610

Source:Basedon MRT (1995),Letter: 24/2/1374.

2.3.6. Seatransport system


The sea transportsystemof Iran includes merchantships and oil tankerswhich in their size,tonnageand numberform a substantialfleet, particularly in the NEddleEast.Both fleetsare surnmarised underthe following headings: 23.6.1 Islamic Republic of Iran Shipping Line (IRISL) in Iran is nowadaysranked seventeenth the world in dead weight tonnage (dwt)
(UNCrAD Review 1993). The Islamic Republic of Iran Shipping Line (IRISL) to a

significant extent covers both the imports and exports of national trade and recently has undertaken passenger alongthe PersianGulf coast.The IRISL is at presenta state movements initially ownedenterprisethat functionsunderthe Ministry of Commerce.It was established
in 1967-68 as Arya Shipping Line with 51% public and 49 % private shares with capital of

45

Chapter 2

Iranian Trade and Transport

3175 million IranianRials (US $. 45 million). The companystartedits operationprimarily as line conference betweenthe portsof the PersianGulf andEuropewith two small coasters a of 25500 dwt, but now functions as ship owner, liner operator,ship's chartererand agent,and
container owner. Table 2.52 shows the development of the national merchant shipping fleet

since 1979.
Table 2.52 Growth and vovaLyes TRISL own and chartered shins. of
Ocean-going 42 38 40 42 59 69 73 76 69 70 72 71 72 70 N/A No. of owned s ips & dwt (000) No. of jou neys dwt (own+passenger) Passenger ships Total Chartered Own 42 525 29 91 38 496 40 159 40 678 85 146 42 799 294 140 59 1166 539 112 69 147 387 174 73 2006 259 199 76 2090 213 228 69 2093 283 N/A 6 76 2115 134 226 U-1 6 78 2132 274 7 78 1 2115 324 217 7 79 I 2119 220 49 9 79 2115 138 301 N/A N/A N/A 99 317

1979 1980 1981 1982 . 1983 . 1984 . 1985 . 1986 1987 . 1988 . 1989 1990 . 1991 1992 1993

Source: Payarn Darya (1994), No. 22; SUI (1991a), (1991b), (1992b), and (1994b), and SCI (1993).

Table 2.53 shows that bulk and general cargo carriers constitute most of the fleet both in

in termsof dwt andnumberof ships.The companyis engaged both liner and tramp shipping The liner servicesof the IRISL are represented of which the former is of greaterimportance. by conventional,multi-purposeand containerships with a carrying capacityof 1.3 million tonnes annually, in round trips betweenthe PersianGulf and north and central European South East Asia, the Far East,EastAfrica, and South countries,the U.K, the Mediterranean, America.MISL tramp shippingruns more or lesson the sameroutesas well as to Canadian, Black Sea,andWest Africa ports.
Table 2.53 Comnosition and dwt of TRISL fleet in 1990. Bulk carriers General cargo Product carriers No. of ships 22 43 3 000 dwt 343 1581 120
Source:IRISL (1990) English Report.p. 3.

Container 2 47

Reefer T, 12

46

Chapter 2

Iranian Trade and Transport

The maritime transport of Iran as mentioned earlier is mainly composed of imported goods of

which the IRISL coversa significantproportion.It hasat leasta 57% sharein the carriageof the national foreign tradedirectly or throughcharteredshipsas shownin Table 2.54. It also hasabout70% of the tradethroughports on average the period 1988-1993 for (IRISL 1990). Table 2.54 Share of the IRISL and its chartered ships in the imports of Iran (000 tonnes).
IRISL ships Chartered ships Total IRISL ships 1988 4744 2082 6826 1989 5638 7427 13065 1990 6073 7398 13471 1991 6468 4274 10742 1992 7900 3351 11256 1993 8200 2363 10563

Source:IRISL (1990) English Report. p. 5. and Payam Darya (1994), No. 22, pp. 53-54; SO (1991a), (1991b), (1992b),and (1994b),and SCI (1993) variouspages.

23.6.2 National Iranian Tanker Company (NITC) The establishment the National IranianTanker Company(NITC) datesback to the of time of the nationalisation oil in 1951. The companywas formed as a joint venturewith of British Petroleum(BP.) to carry mainly crude oil. It was reformedin 1975.The companyis affiliated to the Nfinistry of Oil of the Islamic Republic of Iran, with severalindependent for subordinate companies exploitation, production,and refining. With the beginningof the war with Iraq in 1980and the air threatsto the shippingroutesand Iranianoil terrninals,the natureof the functionsand the tonnageof the NrrC was changedto cope with the national requirementto sell Iran'soil in the world marketor to transportit into the safewaterswithin the PersianGulf. Consequently dwt of the tankerswas increased the sharply.In 1995there tankersincluding ultra large were42 vessels different sizesof which 29 were ocean-going of crude carriers (ULCC) in 1995.NITC was the biggest tanker companyin the Middle East with a fleet of 5.7 million dwt, mainly crudecarriers.

2.4 Modal split


The pattern of freight modal split in Iran is strongly under the influence of the governmentas it owns most of the trade - in particular foreign trade - and also heavily

47

Chapter 2

Iranian Trade and Transport

the modesand interfaces.The modal split has two distinctive featuresin terms of controls Rail transportis only available domesticand foreign trade flow and also modal assignment. for 14 provinceswith direct lines, contributing in 1993to the distribution of 7% of foreign trade and 8.4% of domestic trade. Therefore, the road system is the most intensively
operated land mode in the country, while most of the oil production of the country is

transferredby pipelines from points of production to the populatedareasof the country. Road carries, on average,more than 93% of the total freight of the country (estimateis basedon the performanceof rail for both domesticand foreign tradeduring 1979-1992).

2.4.1 Domestic trade


Iran, inhabited by about 55 million people in 1991, has greater Tehran as a significant pole for freight generationand consumption,as a hub of the single line railway, and one end of most motorways in the country. As shown in Table 2.55 the road mode 85% of domestictrade during 1979-1993, whereasrail, which contributedon averageabout mainly carries mineral and foreign trade, accountedfor 4.6%. However,its shareshowsan intends increasingtrend.The government, accordingto the seconddevelopment programme, to decrease role of road in the distribution of oil productswith an increasein pipeline the efficiency and new lines (PBO 1993).Iberefore, pipeline mode contribution is basedon a 59.8% sharein carryingoil productsin 1993and it is assumed hasincreased this level this to by 0.36% every year from 1979.Domestic trade usesthe full capacityof freight transport Thereforeit vehicles(LGVs, HGVs andvans)due to its varied natureand limited packaging. is not restricted,unlike foreigntrade,to HGVs of greaterthan 13tonnes.

48

)_i k2 l

1 1 1 hm N-L W Q 00 00 !n 00

%0 %0 e 00 22 00

%0 \O fs 00

c 00

\o -1

W w
w

t, ) c

t-J LA

t, ) 00

00

\o

hj 0

W W

t-J

W 00 00 -4

(A

LA

91 p !-li 911 POp - !-li ?0 't 14 U, C7N -4 >- bo . 00

LA w t,) %Z LA c, ti -a 00 w M % %Z w -4 ( w

w -0 o

110
0
t g: 0

e0

Od 0

-cs
tli w -4 w ti i,) iD \o w 0 0', w-4 c> -4 t C2Z . -i W V s- w %D 0 C) 0 N- 0 -4 ( N) 00 ).00 -4 C% L4 wW. -i sN. 1,0 NUJ %0 o O\ bw 0 W -j 00 .4 t -J tJ til 00 P. tA .p 000 IB - N"

OCJ

C) %0 tA

b-

CD

(D

29

10
c2.

EL 2
C)

t-J

t, )

t, )

C\ -0 LA N)

CD N(A tli

U) w

0 C)

NO CYNo\
c) 4. -2 C)-

>-

N-

N-

mc>

>-

-2 Lst b--

00
o C)

%M

h-

w00 (>

%c (>

N-i Lft

>w -4

No w

-a w

0 UX 0 o %M w

b (:

g tli

00 w

00

-4

C> 0

C> (>

C,

C2N(D W
(A

C7N LA

(A
tj

0% ta

ba
! j

O\ t-i

tA

4) 00

CD

CD

%cu

b5
v w

N-3 LA

e 21
CD

:g

(>O .4 ,

00

-2

b.

ta

:-a oo

w oo

al

00 C,

10 0

tp i z4 LA
t-i --1 bi -j
b--

%0

gl
wi P-

cm

CD
A (A u P, (A un f. ZO 8 00 00 ! LALALA u LA (A LAu c> 9, tA !A u ZA 2 -4 74
'43

o
CD

C%

bo

b oo , t_J,, ch

00

$OL

p po
91d

D (A 00 0% (A %0 b-

t.A

(Z) >-

%D C> C> LA

-0

(A -4

00

C> W j -3

C>

(A

,l . 0

49

Ici

Ln

W-L %0

\O

%0 \C

%0 %0 No

%0 s

E
0 K

!
rfp

W
%00

00
Gn
c> 00

00

t, )

00

00

-a

00

00 t

00

rA

w0 cr -2 < .
9c2

e15

(D

9 10

>-

(A

h-4 C)

b.til o0 (>

me

%I0 LA

,0 u) ,M CD. (D

01 w1

o 101

C)

-4

t2,W 31 l W, l

v Lli w

00 -

00 CY\

00 00

-4 %0 -

(A

(A

nd

e. n

ti

-1 1

(Am, r--'2, . '. 4,Gr= p0 w1 wa 1

2
Od

R.

cl

-, Z

EL-ci Gn r,0 (Z
o

.pe -

% Lti ii

o %, o o U Z4 Lst

U, :A. 1

,--, EI
13 "C#2 %C 0 lb . , ci ;., ta

%c 1,0

(A
bu 3

3
U)

8.-

-1

h- . to 0
w

b . w -' 0%
s-

>00

(D 0

w- . %10 _]

0-00

b-

N-

N-- 6. U) w

00 00

CD
w 2 ulg

a c 4 U

2 W :4 - b 112", -" :p. Zz 4 i


1
00 00 00 00 00 23 00

00 - :.1 O\
PC 9eA

cm
(D
N--

00

00

00

C
'C C

,>O 22u cC% 00 w Cz% WtJ N--

23

00

23

od 1 &2
Zm

Eg 0 bi olleb

c7,

00

%0

bi b-. -4
00

00

oo

00 tA w F

%0 9 00

00
00

(A

tA

5.15 ci

M , %

23

' t

, t ) 00 22

%A

b- 0

Cr.

so

Chapter 2

Iranian Trade and Transport

2.4.2 Foreign trade


The modal split of foreign trade moving within Iran is basedon the three modesof The major feature is the shareof pipelines in the transport of oil rail, road and pipelines.
from ports. The only data about the modal share of pipelines transport for oil products

is for 1993 when 59.8% of total foreign trade oil products went from ports by products (PBO 1993). Assuming that it was the samefor the entire period of the study, pipelines Table 2.56 estimatesmodal shareand clearly shows the significanceof the road mode in the foreign trade for general cargo and oil products. Transhipmentmoves are excluded from this analysisbut havebeendealt with in Table 2.37 asexports. Table 2.56 Estimate of modal shares of the general cargo and oil products foreign
tradp in Trnn (AM) tannp. q)2 1 Pipeline Noncrude oil foreign trade Pipeline %of total oil products 58.00 58.36 58.72 59.08 59.44 59.80 3 Oil product trade through ports 4(1-2) Share of rail and road Volume % of total Volume % of total 64.5 84.0 83.6 81.9 80.6 77.1 78.6 1993 5 Rail mode 6 Road mode (4-5)

3378 20.5 10632.08 4274 14010.08 1988 16489 2478.92 1832 8.0 19197.75 3119 21029.75 1989 22850 1820.25 1929 7.7 20864.90 3682 22793.90 1990 2162.10 24956 2512 8.8 23498.10 4553 26010.10 1991 28700 2689.91 2245 8.1 22389.57 5280. 24634.57 1992 3138.43 27773 2581 8.5 23579.30 7349 26160.30 1993 4394.70 30555 14469 10.3 119026.90 Total 151323 16275.90 Source: Different PSO, IRIRC, PBO and Iran Customs Administration reports for 1988-1993. *: -59.8in the assumption of the decreasing rate of 0.36 backward x column 3. with

-77-!

2.5 Conclusions
This chapter has studied the Iranian economy from the macroeconomicpoint of view and Iran's trade and transportsystemin depth. The aim was to provide sufficient data to support further stagesof the study. The main revenue-makingsectors of Iran were identified and their fluctuations were traced from 1979-1993.It has becomeevident that Iran has passedthrough a turbulent period due to the Islamic revolution in 1979, the war between 1981 and 1988 and the later reconstructionprocess.Each mode and interface in 51

Chapter 2

Iranian Trade and Transport

layout andperfonnancewas studiedin tenns of organisation,infrastructure,superstructure, and of sufficient detail to provide an insight into the advantages weaknesses the transport system.The internal modal split for both domestic and foreign trade was investigated,and it was seen that Iran is heavily dependenton the road mode for both trades. The two Abassand Imam are particularly important for foreign trade. southernports of However it is evident that Iran may have a suitable transportinfrastructurefor use by the This topic will be pursuedin subsequent CentralAsian and Caucasus countries. chapters.

52

Chapter 3

Central Asian and CaucasusCountries

3. Central Asian and Caucasus countries


3.1 Political geography
The aim of this chapteris to examinethe main trendsin the economy,foreign trade transportof the Central Asian and Caucasian countries(CAC) in the context of their and foreign trade through Iran. Since the dissolution of the former Soviet Union (FSU) in 1991 the newly independent stateshave been faced with many changes, problems,conflicts and Apart from ethnic and territorial considerations, economicsystemsof all the the shortages. FSU republics have had to replace a centralisedeconomic system with an international in but marketsystem, with little experience internationalaffairs. The eight new CAC countriesas shown in Figure 3.1 can be characterised oil-producing as countries(exceptfor Armenia and Georgia).like all other former Soviet republicsafter the collapseof the USSR, the CAC countrieshave had a hard transitionalperiod of political conflicts, decline of demand and supply and consequentlytrade, currency exchangeand paymentproblems.However, their economyand trade is under slow improvementwith the establishment new institutionsto supportand converttheir former centralplanning into an of openmarketsystem(Michalopoulos1993andFischer,SahayandVegh 1996). The CAC region, from the early yearsof the establishment the USSR, was for about 70 of yearsintegratedclosely into a very centralisedplanning policy dictatedand controlled from Moscow. The development the EconomicCo-operationOrganisation (ECO) by the treaty of of Tehran 1985,originally a new format of the treaty of Izmir of 1964but without military objectives,should bring into being a region with new internationalrelations (Gharabaghi 1994,Raeece-Dana 1995).It may help to overcomeany suspicionabout the policy of Iran towardsits neighbours.

53

00i

Ch

eD

A.

p-

lc

PO

54

Chapter3

CentralAsianand Caucasus Countries

Each of the two main geographical regions of the CAC countries (Caucasusand Central Asia) has its own political and economic features, which are discussedin the following sections.

3.1.1 Caucasuscountries
The Caucasusregion, much smaller in area comparedwith Central Asia, is also It in resources. functions as a bridge betweenthe oil-rich Middle East countriesand poorer Europethrough the Caspianand Black Seas.All three Caucasus countriesare unstableand living without peacedue to serious internal and external conflicts and do not have the common policy, therefore, to improve their fragile economic prosperity, security and independentstatus. The region, however, is a sensitive and unstable area situation and for Russia,and with other neighbouringcountries it can play "a special role in primarily (Barylski 1994,Roberts 1996). world affairs" Azerbaijanis a semi-maritimecountry along the westernshoresof the CaspianSea and is very rich in off-shore oil reserves.It has a population of about 7.2 million of which most (83%) are Muslims. Azerbaijan also contains the Nagomo Karabakh republic with an Armenian majority, resulting in disputes with Armenia. This has hindered its economic development weakened political systemsince 1991(E.I.U 1992,Rutland 1994). the and landlockedcountry in the south-west the Caucasus Armenia is a mountainous of region with tiny borderof about40 km with Iran asshownin Figure3.1. It hasa populationof about3.5 a 68% are urbanised.Most (93.3%) of the people are Christiansbut million, of which about different ethnic minorities such as Russians and Kurds and therefore,unlike Azerbaijan with andGeorgiais free of internalethnic problems(E.I.U 1992,Rutland 1994). Georgia,at the interfacebetweenAsia and Europe,is the only maritime country amongthe CAC republics,as it is locatedon the west coastof the Black Sea as shown in Figure 3.1. According to the Europa World Year Book (1994a) based on a U.N census in the Demographic Yearbook,it hasa populationof about5.5 n-dllion,of whom most areOrthodox 55

Chapter 3 Christians but with Muslim and Jewish minorities,

Central Asian and CaucasusCountries and ethnic minorities such as Russian,

Armenian, Azerbaijani, Ossetain, Greek, etc. Its ports (e.g. Batumi) have an ideal location for the European trade of all the CAC republics and Iran (E. I. U 1993) but their proper utilisation depend on the settlement of conflicts in western Georgia (Barylski, 1994). will

3.1.2 Central Asian countries


The five Central Asian countries to the north east of Iran have a much larger area than the three small Caucasus republics, with mountainous and desert regions and a dispersed low

population(Ogutcu 1995).They are strongly under the social,economic,cultural, religious, influencesof neighbouringcountriessuch as China, India, Pakistan,Iran and and political Afghanistan. They also havebeeninvolved in someethnicunrestinternally and in the border China and in the Afghanistan internal conflicts, and to some extent foreign provinces of investorsmay have doubtsabout their long term interests(Rashid 1994).China, on the one hand, may, following the collapseof the former Soviet Union, lose a potential threat to its from having growing free borders. On the other hand, it may be disadvantaged northern in of marketeconomies potentially ethnic and religious states its vicinity (Harris 1993).Iran, the most populouscountryin the southof CentralAsia, hascultural andhistorical links in the in the areas.where language is derived from Persian, such as in region, particularly Tadjikistanandmanypartsof Uzbekistan(Akbarzadeh1996). /km in Gyrkyzstanis a small and mainly agricultural mountainous'countryof 766,400 sq. the north-eastem part of Central Asia. The population (4.4 million) is composed of Muslims of mainly Gyrkyz origin, but with Tadjik and Uzbek minorities. It was the first country to leave the Rouble zone of the Commonwealth of Independent States and launchedthe Som asthe national currencyin May 1993(E.I.U 1992). is Kazakhstan the secondlargestrepublic of the former Soviet Union, and the largestwithin the CAC republics(about2.7 million sq. km). It is much bigger than Iran with a long coast

56

Chapter 3

Central Asian and CaucasusCountries

line of about 2320 km on the Caspian sea as shown in Figure 3.1. It has a population of 16.7 density of six persons/sq. km. The population is comprised 39% Kazakhs, 37% million with a Russians and 5% Germans most of whom follow predominantly the Muslim religion (E. I. U

1992). Tadjikistan is larger than Gyrkyzstanand located in the South East of Central Asia on the Pamir Mountains. The population (5 million) is composedmainly of Muslims of Tadjik but there are also Uzbek, Russianand Tatar minorities (E.I.U 1992). origin, Turkmenistanhas a relatively small population,mainly living in the southernstrip close to Iran in the capitalcity of Ashkhabad othercities including Mary, Chardzhov, the port and and Krasnovodskon the easternCaspianSea coast. 90% of Turkmenistanis desert.Its inland is along the Amu Darya river (E.I.U 1992). Through the unique location of waterway to Turkmenistan CentralAsian countries,arelinked in the north-east Iran (Barylski 1994). the Uzbekistan is the third largest country of the CAC republics, after Kazakhstan and Turkmenistan in area, and after the former in population. It is a completely landlocked Its only waterway accessis by river (1100 km length) through Turkmenistanto the state. Aral sea at the border with Kazakhstan.The population is mainly Muslim. Other than Uzbeks, it consists of 8.4% Russians,4.7% Tadjiks, and 4.1% Kazakhs, with smaller Tatarsand Crimeans(E.I.U 1992). minorities of

3.2 Economies of the CAC countries


The FSU republics and Eastern bloc after independencein 1991 have left towards a new system.Each country is faced with some serious communism and moved economicproblems (Murrell 1996).The economiesof the CAC countrieshave developed larger integrated and centralisedFSU and Eastern bloc, a region within which within a there are still dependentties, but where major steps have been taken towards a market

57

Chapter 3

Central Asian and Caucasus Countries

(UNCTAD[UN 1994, Murrell 1996). The collection of economic data for this economy. region is difficult.
"Ibe official Russianstatisticsdo not provide indices of export and import volume of changes... Statisticson Russianforeign trade only start with the emergence Russiaas an independent state,as Soviet Union foreign trade statisticswere compiled only for the USSR as a whole, without breakdownby republics. Hence,all dataon Russiantrade for the yearsbefore 1992are estimatesderived from USSR statistics,the accuracyof diminishesas one movesbackwardsfrom 1992" (UNCTAD/UN 1994,p.69). which

3.2.1 Caucasus countries


The general tendency of the Caucasuscountries is to move towards a free market from the traditionally strong economic and industrial ties with Russia economyand away and other FSU republics. had Table 3.1 showsthat the three republicsof the Caucasus reasonable growth up to 1989 the disintegrationof the USSR started.The diversity of their economicstructure,as Nyhen
well as the existence of internal and inter-republic conflicts, have had great effects on the

these countries. For instance,the industrial production of small countries like output of
Armenia and Georgia is largely dependent on the importation of raw materials and

intermediary goods from other republics, and in particular Russia,and energy (oil) from Azerbaijanand Russia. The GDP of 1993was half that of 1985(Table 3.1) due to the generalconsequences the of dissolution of the USSR, transport disruption and limitations, hostilities with Azerbaijan over Nagorno Karabakh,and conflicts betweenRussia and Georgia. Both Armenia's and due to theseincidents and this worsenedthe Georgia's rail connectionswere disconnected economic situation. At the sametime, Azerbaijan, which is rich in gas and oil resources, hasbeenaffectedto a lesserextent from the disagreement with Armenia. Tahle 3.1 Caucasuscountries GDP hefore and after indeDendence(S m).
Caucasus Armenia Azerbaijan Georgia 1985 21395 30413 41276 1986 22412 31828 37558 1987 23248 34296 38532 1988 23835 36871 42305 1989 26912 36119 45378 1990 26060 33279 42095. 1991 24718 34368 34761 1992 12102 22860 21515 1993 11198 21152 19908

Source: E.I. U (1994a) based on I.M. F: World Bank, Statistical Handbook of the FSU, United Nations EconomicCommissionfor Europe,Bulletin for Europe,Vol. 44 1992;EIU calculation.variouspages.

58

Chapter 3

Central Asian and Caucasus Countries

The structure of the econorniesin the three Caucasuscountries to some extent differ. Azerbaijanis substantiallyrich in oil and gasand highly attractivefor foreign investors. The economy of Georgia is based on non-oil minerals, agricultural products, and heavy industries. Armenia does not have significant mineral deposits, but has well-developed industries suffering from energy shortageand heavily dependenton Russian imports of intermediarygoods.The shareof different sectorsis shownin Table 3.2.
Table 3.2 Comparison of sectoral shares (%) in the economy of the Caucasus countries in 1991. Armenia Azerbaijan Georgia Sectors 54.2 48.3 35 Industry 25.7 26.1 Agricultural 37 14.6 10.6 11 Construction 1 1 2.91 ransport & communication 1__2.2_ 5.6 1 2.21 6 Trade & catering 1 3.61 4.01 7 Others
Source:E.I. U (1992),variouspagesfor Georgia,Armenia, Azerbaijan,CentralAsian Republics.

3.2.2 Central Asian countries


The five Central Asian countriesare vast in geographicalscaleand rich in oil, gas, minerals, and agriculture that is still unexploited while suffering from a water shortage (Rashid 1994). Table 3.3 presents a comparison of the Central Asian countries' GDP indicating a continuousgrowth before 1990 and a decline during and after the collapseof
the FSU.

Table 3.3 Central Asian countries GDP before and after independence ($ million) (PurchasinLy Dower varities*).
Kazakhstan Gyrkyzstan Tadjikistan Turkmenistan Uzbekistan 1985 1986 1987 1988 1989 1990 56333 54954 62874 68698 71821 74269 9223 9794 10159 11924 12939 13938 12946 13761 14004 16291 16538 16981 1 8177 1 8775 1 94201 10811 1 10512 1 1116_ 1 30465 1 31725 1 32417 1 36803 1 39668 1 46070. 1991 65731 13974 16124 11067 47769 1992 57821 10757 10757 10757 1 374 1 1993 53502 9622 9554 11060 41059

Source: E. I. U (1994a), various pages for Central Asia, based on I. M. F: World Bank, Statistical Handbook of the FSU, United Nations Economic Commission for Europe, Bulletin for Europe, Vol. 44 1992; E. I. U calculation. *: Showing the GDP and or GDP per head for all 15 former Soviet republics. It gives an indication of living standards and domestic purchasing power, free from the distortions caused by the exchange rates (E. I. U 1994a).

59

Chapter 3

Central Asian and Caucasus Countries

Kazakhstan,with potential economic output in the three major categoriesof industrial, production, has a strong economic structureand within the CAC agricultural and mineral the highestGDP as shown in Tables 3.1 and 3.3. The structureof the economies countries in the five Central Asian countriesis largely similar with oil and gasresources, agricultural
and industrial manufacturing. They are also very rich in gold and have potential for cotton growing. Table 3.4 shows the importance of the industrial and agricultural sectors in

generalfor the CentralAsian countries. Table 3.4 Comparison of sectoral shares (%) in the economy of the Central Asian countries in 1991 (current Drices). 1
Kazakhstan 41.4 35.8 9.6 & 3.9 5.5 3.8 100 Gyrkyzstan 45.3 36.4 7.7 2.8 0.01 1001 I Tadjikestan 30.6 43.9 12.8 2.6 0.01 U. 1 -! 1001 Turkmenistan 19.6 46.4 22.7 4.2 0.0 7.1 100 I Uzbekistan 33.2 43.2 10.7 4.2 0.0 8.7 100 Industry Agricultural Construction Transport communication Trade & catering Others Total

7.7 1

Source:E.l. U(1992), various pagesfor Central Asia, basedon World Bank, Statistical Handbook: Statesof FSU.

Turkmenistanand Uzbekistanwith rich hydrocarbonresources, without any externalor and domestic conflicts or problems, and Gyrkyzstan as an agricultural basedeconomy, have experienced smallest economic downturn. Uzbekistan is also very rich in gold with potential for cotton growing. Tadjikstan economyis basedmainly on agricultural products, silk and in particular cotton (Akbarzadeh1996).

3.3 Foreign trade of the CAC countries


The main feature of trade, when the CAC countries formed an integral part of the USSR, was that the inter-republic trade and trade with the Eastern bloc dominated the "foreign" trade of thesecountries.Generally this type of trade declined after independence as a result of the imposition of import and export controls and the availability of hard trade with FSU republics createdpayment currency in new markets.After independence, 60

Chapter 3

Central Asian and Caucasus Countries

from the new currenciesand differencesin the monetarysystems.For problemsresulting it is has been economic to trade with these new republics while their many countries toward internationalprices, sincethey offer cheapergoods.But as most exportsare moving the CAC countries are rich in raw minerals (hydrocarbonsand others) it seemsthat in of future there will be a considerableshift in the structure of trade of many materials and
Moving toward international prices to acquire foreign currency for imports is products.

for all CIS countriesbut rather difficult for those such as Annenia and Georgia, critical do not have raw materials and have low quality manufacturescompared with which westerncountries.

3.3.1 Caucasuscountries
3.3.1.1Azerbaijan In recentyearsAzerbaijan's tradewith countriesother thanthe CIS hasbeenin deficit (Table 3.5). It has a sizeabletrade surpluswith the other statesof the FSU, mainly on the basisof gas and oil (Table 3.6). The close geographical proximity and common featuresof Azerbaijan and Iran have led to a considerableexpansion of bilateral trade and joint independence I.U 1992). (E. infrastructure after utilisation of Food and light industrial productsconstitutethe bulk of the Azerbaijanexportsas shown in Table 3.6. Thereis a significantcontributionfrom exportsin oil andgas,ferrousmetals,light 1. industries, food productsto tradesurpluses with otherrepublicsduring 1990-199 and
Vnhipc nf inter-renuhlic and international trade of Azerbaiian (Rh.- million). T.qhh-1--; 19 1 1990 1992 1989 Exports Imports Exports Imports Imports Imports Exports Exports Trade 6429 5752 12204 11022 207756 131580 7122.9 5189.8 Total 11456 8837 6104 4248 102175 85026 6674.9 3794.3. Total inter-republic Balance +1856 +2619 +17149 +2880.6 325 1 1504 748 1 2174 1055811 46 4 448 1 1395 Total foreign 1426 59027 Balance -1179 . -948
Source: Europa World Yearbooks (1994a) p. 444 and (1995a), p. 1547.. Inconsistency is due to sources.

61

Chapter 3

Central Asian and Caucasus Countries

Table 3.6 Composition of the AzerbQan

inter-republic import/export (Rb: million).


1992* Exports

Imports Composition of trade 1450 6702 Petroleum & gas 270 24166 Ferrous metallurgy 384 2334 Non-ferrous metals 1161 15964 Chemicals 2115 29884 Machines and metalworking 0 3656 Wood and paper products 2270 7461 Products of light industry 3892 3090 Food 280 7379 Agricultural products 0 0 Other commodities _7 12_ -511 - 84 1182T Total inter-republic trade incl. other commodities I I 1 1 -1 World Yearbooks(1994a)and 1995a,variouspagesrelatedto Azerbaijanbasedon IMF, Source:71be Europa EconomicReview:Azerbaijan.InternationalFinancialStatistics:Supplement Countriesof FSU, World Bank, on AzerbaJan: Transitionto a Market Economyand StatisticalHandbook:States the FSU. *: Includesboth The of is Inconsistency dueto sources. inter-republic internationaltradedata. 1: Includesothercommodities. and

1988 1990 Exports Imports 'Exports Imports 292.4 747.3 428 886.1 206 71.2 218.5 88.5 80 100.8 91.3 102.2 518 488 601 407.1 1104.5 936.3 1119.4 1012.3 0 117.1 0 99.2 6201 1365.8 707.9 1479.1 642.3 1 1748.6 500.9 1828.5 51.8 134.1 145.8 312.6 13.9 124.8 117.1 140.5 5726.1 3943.5 6450.8 3528.5

According to E.I.U (1994a) the inter-republic trade of Azerbaijan decreased sharply after by independence 64.6% (imports) and49.2% (exports)in 1992.For exportsandimports with than FSU republics,light industrial productsand machinerywere important, countriesother while imports of food productswere important as shown in Table 3.7. Europeanand Asian countrieshadthe highesttradelevelswith Azerbaijan(seeTable 3.8).
Table 3.7 Comnosition of the Azerbaiian international import/export trade (Rb: million).

Composition of trade Petroleum & gas Ferrous metals Chemicalsand chemical products Products of machine-building industry Products of light industry (textiles) Products of food industry Agricultural processedproducts Total international trade incl. other commodities

1989 Exports Imports 102.7 1.5 0.0 82.2 0.0 94.4 105.9 167 172.5 362.9 0.0 409.8 228.4 0.01 448 1 1395.5

1990 Expo Imports 1.9 101.6 0.0 49.1 87.5 0.0 112A 348.8 61.2 403.2 431 0.0 131 0.0 ,I ' 325.1 1504 .9

is Source:Sameastable 3.6. Inconsistency due to sources. .

Table 3.8 The share (%) of main foreign trade partners of Azerbauan in 1990.
Germany Exports Imports Total 10.7 13.4 24.1 Bulgaria 9 7.2 16.2 Czechs& Slovakia 7.2 8.5 15.7 Hungary 5.9 4.7 10.6 Poland 5.7 10.9 16.6 Cuba 5.7 5.2 10.9 Europe 62.7 65.3 128 Asia 15.9 20.5 36.4 America 5.7 5.2 10.91

. . Source: E.I. U. 1992, Georgia, Armenia, Azerbaijan, Central Asian Republics, Country Profile- Annual Survey of Political and Economic Background: Based on: World Bank, Statistical Handbook: Statesof the FSU.

62

Chapter 3

Central Asian and Caucasus Countries

3.3.1.2 Armenia Armenia was an industrial republic of the FSU and was very active in foreign trade,

(see having sometimes surplusof tradebeforeindependence Table3.9). a


Table 3.9 Value of foreien trade of Armenia (Rb: million. current Drices).
19871 1986] 19881 1989 1 1985 1 5554 4115 3767 3691 4997 Exports 4249 5181 4877 4898 4979 Imports Source: I.U (1992),basedon World Bank, StatisticalHandbook:States the FSU. E. of 1990 3523 4662

According to IMF sources (E.I. U 1992) for the second half of the 1980s, imports/exports

than 50% of the total GDP. Industrial goods formed the main as were assessed more components both exportsandimports for Armenia(seeTable 3.10). of
TahIp

1986 Import Export Import Export 1054 1276 981 1301 Light industry 1568 2022 1731 2295 Machine building & metal 524 502 650 505 Food industry 481 Energy 510 388 394 443-1 445 Chemical products 42 169 49 141 Agricultural gOods 1988 198 9 Import Export Export Import 1501 1107 Light industry 1117 1427 845 1002 1005 861 Machine building & metal 946 641 584 843 Food industry 435 Energy 456 ----376 --385 338 362 Chemical products 200 212 15 10 Agricultural goods L Source:E.I.U (1992),basedon World Bank, StatisticalHandbook:States the FSU. of Trade

rnmnn. ritinn of the iMDort/exDort -3-10

trade in Armenia

(Rh. - million.

current

nrices)-

1985

1987 Import Export 1066 1550 929 1027 546 572 374 363 4 146 15 1990 Import Export 1166 15.01 975 828_ 412 _754 304 FT 370 2 290 12

According to World Bank statistics(E.I.U 1992)the principal tradingpartnersof Armeniain countriesof which Bulgaria(exports)and Germany(imports) externaltradeare the European in have had the highestshares 1990.Asian and Americancountriesare also prominent(see Table 3.11).

Table 3.11 Share of European countries, Asia and American continent in the foreign trade of Armenia in 1990M of total in foreism trade Roubles).
Czech& Slovakia Poland Germany Romania Bulgaria _11ungary Exports 8.7 6.8 11.7 10.7 5.8 3.9 9.1 Imports 15.7 9.6 7.6 5.6 0 17.8 16.4 Total 26.4 11.4 19.3 3.9 Source: E.I. U (1992), p.4 1, based on World Bank, Statistical Handbook: Statesof the FSU. Asia 10.5 16.9 27.4 America 7.7 16.2 23.9

63

Chapter 3

Central Asian and Caucasus Countries

3.3.1.3 Georgia

Georgia,as an industrial republic in the FSU, usedto be very active in trade having the USSR, but an imbalancewith other countriesleading to a sometimes surplus within a deficit asshownin Table 3.12. considerable
Table 3.12 Value of forehm trade of GeorLyia(Rb: million). 1988 1989 1990 5901 6084 Exports 5983 6469 6493 6839 Imports 385 Deficit 592 857
Source: EIU (1992),p.27, based IMF EconomicReview,Georgia. on

] 1991 6112 7266 112

Heavy and light industrial goods, minerals, constructionmaterials,chemicals,agricultural


goods, forestry products and food are the main components of both imports and exports as shown in Table 3.13. Table 3.13 ComDosition of Georzia's foreien trade (Rb: million).
Trade Oil and gas Ferrous metals Non-ferrous metals Chemical fuel Machine building & metal Timber, wood, paper Building materials Light industry Food industry Agricultural goods Others Total 16 1988 Imports Exports 413 100 489 375 102 541 316 1533 848 248 155 1275 1221 2438 1204 348 280 27 11 6493 5901 1989 Imports Exports 360 68 443 376 106 544 343 1522 869 244 148 1287 1285 1142 2573 358 190 103 105 6469 6084 1990 Imports 285 430 97 576 1580 279 117 137 ,2 1174 498 140 6839 Exports 68 318 1 339 804

1260 238 404 93 5983..,

Source: The Europa World Yearbooks (1994a) and (1995b), various pages for Georgia, based on IMF, Economic Review: Georgia. International Financial Statistics: Supplement on Countries of the FSU, World Bank, Anncnia : Tbe Transition to a Market Economy and Statistical Handbook: States of the FSU.

The principal tradingpartnersof Georgia,accordingto World Bank statistics(E.I.U 1992),in externaltradeare the European countrieswith about71% of the total, of which Germanyand Bulgaria (exports) and Germany and Poland (imports) had the highest sharesin 1990, followed by Asian andAmericancountriesasshownin Table 3.14.

64

Chapter 3

Central Asian and CaucasusCountries

Table 3.14 Share of the major European, Asian and American countries In the foreign trade of Cf. r-, -.nra; n in I ()()nI nf tntal fnreifrn trade in Rnuhles).
Exports Imports Total V Ex c E!! Im ( rts orts Total Tota Exports L Bulgaria 10.7 8.5 19.2 Romania 3.4 Germany 10.7 17.7 28.4 Italy 4.4 Czech & Slovalda 7.7 9.6 17.3 Netherlands 2.7 Poland 8.4 10.8 19.2 U. K 3 Hungary 6.0 6.1 12.1 Finland 0.0

Imports Total

0.0 3.4 Japan 4 4.9 8.9

0.0 4.4 Syria 0.0 3.2 3.2

0.0 2.7 Other Asia 11.2 7.8 19

0 3 Cuba 5.4 4.8 10.2

3.2 3.2 'Other America 1. 3.3 5.2

btates ol tne rbU. Source: E.l. U (1992), p.28, based on World BanK, btatiStiCal HanObOOK:

3.3.2 Central Asian countries


3.3.2.1Gyrkyzstan The economyof Gyrkyzstanis mainly agriculturalbased(E.I.U 1992).The country is in mineralsbut hasinsignificanthydrocarbon and resources is heavily dependent on very rich imports from other republics (EuropaWorld Yearbook 1994a).It was heavily dependent (about98% of trade) on foreign transactions with other republics,an imbalancewhich still in the 1990's (see Table 3.15) but there has been attemptsto restructurethe continued 1995). toward a marketeconomy(DelehantyandRasmussen, agriculturalsector
Tahip IIq Thsy,r-vrkv7-ctnn inter-renublic imnort/exnart Import I 24241 +476 8061 trade (Rh: million).

77
Total inter-republic Balance Total foreign $ million Balance

1987
Export Import 2691 2150 1 +541 54 774 1 _ -720 1990 Export Import 2446 31791 1 -733 1063 1 -1010 53

11'6-1988
Export 2900 74

19891'',
Imp rt I 2433 1 +750 883 1 Export 3183 65

Total inter-republic Balance Total foreign $ million Balance

-732 1991''. Import Export 5409 65061 +1097 13741 +1333 41 1

-818 1992 Import ,Export, 67407 1 46301 -21106 31651 +3296 6461

basedon IMF, Economic Source:E.I.U (1992),p. 81 and EuropaWorld Year Book, (1994a),variouspagesfor Liyrkyzstan, Supplement Countriesof the FSU; World Bank, Gyrkyzsstan: Review:Gyrkyzsstan, internationalFinancialStatistics: on and The Transitionto a Market EconomyandStatisticalHandbook:States the FSU. of

As indicatedin Table 3.16, exportsand imports at inter-republiclevels are dominatedby the light andmachinebuilding industriesand food products,althoughafter independence exports 65

Chapter 3

Central Asian and Caucasus Countries

of electric power coal, ferrous and non-ferrousmetals with countries outside the former (Table3.17). USSRrepublicswere significantly increased
Table 3.16 Comnosition of the Gvrkvzstan inter-reDublic iMDort/ eXDort trade (Rb: million). 1990 1991 1992 Exports Imports Exports Imports Exports Imports Trade 0 269 0 578 0 19654 Petroleum gas 67 0 188 0 3047 0 Electric power 22 38 14 85 Coal industry 1013 2137 1 167 0 292 340 4280 Ferrous metals 96 164 500 277 Non-ferrous metals 5662 2506 324 23 218 566 Chemicalsand chemical products 667 7489 industry 893 985 2018 1272 21506 i7O-23 Products of machine-building 0 114 7 183 185 2221 Wood and paper products 13 69 67 81 1445 Construction materials 736 649 1017 1901 1882 12609 5171 Products of light industry 516 663 1317 1080 3721 4247 Products of food industry 87 247 182 323 792 4185 Agricultural unprocessed products 2446 3179 6506 5409 46301 67407 Total inter-republic indude others 4242 2499 6547 6783 52762 70572 Total incl. external trade Source:EuropaWorld Year Book (1994a),various pagesfor Gyrkyzstan,basedon IMF, EconomicReview: Gyrkyzstan, InternationalFinancialStatistics. and

Table 3.17 Comvosition of the Gvrkvzstan external imnort/exiDort trade (Rb: million).
Trade Ferrous metals Non-ferrous metals Chen-dcals chen-dcal and products Machine-building and metal worldng Wood and paper products Products of light industry Products of food industry Agricultural unprocessedproducts Total (including others) Source:Sameastable3.16. 1990 Exports Imports 1 0 0 19 1 51 I1 106 0 0 9 364 8 415 0 80 53 1063 1991 1992 Exports Imports Exports Imports 0 0 340 F, 8 3454 0 0 44 273 531 8 88 559 212 7 0 185 0 3 275 1414 670 644 0 146 1200 242 0 0 AZ7 6461 3165 41 1374

The main trading partnersof Gykyzstanin 1990 at the continentallevel cover Europe and then Asia. In Europe,as shown in Table 3.18, Italy, Germanyand Bulgaria have dominated (now the CzechRepublic and Slovakia)imports. exports,and Germanyand Czechoslovakia In Asia, North Korea and Japanwere important for exports and imports respectively,while Cubawasimportantfor both tradesin the Americas. Gyrkyzstanwas amongthree former USSR republicswith an overall trade surpluswith the CIS, while in deficit with Russiaduring 1994(UNCTAD/UN 1994).In Europein the same year the UK (15.1%) and France (8.9%) and, in Asia, China (36.6%) have achievedthe 66

Chapter 3

Central Asian and CaucasusCountries

highest ranks for exports. The USA (32.8%) and China (23.1%) were the most important

importing countriesin 1992.


Table 3.18 Major shares of the international importlexport of Gyrkyzstan in 1990 (% of the
tntni in farman trade

Roubles).

Germany Exports Imports 9.0 18 Others in Europe 13.6 12.5 1

Italy 11.90 2.30 North Korea , 11.90 1 1.001

ulgaria 9.0 7.6 Japan I 1.5 1 4.8 1

Chzech& Slovakia 6.0 9.3 Others in Asia 22.01 10.91

Poland 4.5 6.2 Cuba 4.5 3.01

Hungary

Yugoslavia

Exports Imports

4.5 3.0 6.2 2.9 Others in America 11, 1 0.9 2.7

1. Source: EIU (1993c), Gyrkyzstan, p. 13

3.3.2.2Kazakhstan is The foreign tradeof Kazakhstan dominatedby other former republicsas a result of
the high level of industrialisation. Since January 1992 the liberalisation process has led to

individuals and enterprises becoming engaged in foreign trade without government permission being required, except for goods of national interest (fuel, minerals, mineral fertilisers, grain, cotton, wool, caviar and pharmaceutical products).Thesemay be exported Kazakhstanafter 1988 has had a considerabledeficit in interonly by state organisations. republic and internationalforeign trade(seeTable 3.19) . It plays a significant role within the CIS and at the beginning of 1995 had implementedtwo significant improvementsby free trade zone with Uzbekistanand Kyrghyzstan,and a Customsagreement establishinga with RussiaandUkraine (Griffin 1996).
Table 3.19 Value of inter-republic and International trade of Kazakhstan inter-republic In Rnuhles-market nrices and international In S million). million
1988 1,
Exports Imports Ba Balance Inter-republic 8337 12455 -4118 1990 Ex orts Imports Im Balance Ba Inter-republic 8443 13026 1 -4583 Internationai 1800 2015 215 . Inter-republic 13208 17363 -4155 International ($ m) 1600 2664 -1064 Inter-republic 8201 13259 -5058 1991 International 1254 1356 -]oil International 1700 2550 -850

Source: E.I.U (1992),p. 68. basedon IMF, EconomicReview:Kazakhstan.

67

Chapter 3

Central Asian and CaucasusCountries

For "ports to former USSR republics (see Table 3.20), oil and gas, and machineryare important, wWle ferrous and non-ferrousmetals are important in internationaltrade. In the caseof imports, macWneryand oil and gas are important in the inter-republic trade, and machineryand food in internationalmarkets. is The main continentaltrade partnerof Kazakhstan Europewith 64.7% of exportsand 72% imports, followed by Asia and America. At the country level (as shown in Table 3.21) Germany, Japan,and Cubahad the highestsharein the foreigntradeof Kazakhstan.
Table 3.20 COMDOSition of Kazakhstan's iMDort/ CXDort trade (Rb: million. world market nrices)-

Petroleum gas Electric energy Coal Iron and steel Non-ferrous metals Chemicalsand chemical products Products of miachine-buildingindustry Wood and paper products Construction materials Products of light industry Products of food industry Agricultural unprocessedproducts Total trade (includes others products) Total inter-republic trade Total foreign trade

1989 Exports Imports 875 1447 224 371 312 000 1077 1040 791 272 1121 1703 836 5359 000 988 000 313 1625 3113 617 1871 1146 457 9094 17569 8201-1 14571 893 1 2998

1990 Exports Imports 795 1181 233 420 306 000 1036 986 777 275 1082 1727 786 5510' 000 832 000 331 1536 3374 612 18801764 392 9350 17830 8443 14314 906 3516

Source:EuropaWorld Year Book, 1994b,variouspagesfor Kazakhstan, basedon IMF, EconomicReview:Kazakhstan, and InternationalFinancialStatistics:Supplement Countriesof the FUS; World Bank, Kazakhstan: Transitionto a Market Ile on EconomyandStatisticalHandbook.States the FSU. of

Table 3.21 Share of major European, Asian, and American countries in the foreign trade of Kazakhstan in 1990M of total in foreign trade Rnubles).
Germany Export Import Export Import .1 -''' Export Import Export Import 11.0 17.9 Romania 3.6 2.4 ,- Other Europe 12.8 8.10 North Korea 0.0 2.6 Bulgaria 9.0 8.3 Italy 3.4 3.3 Japan 3.4 2.8 Other Asia 10.4 9.70 Czech & Slovalda 7.5 9.0 U. K 3.1 0.0 India 2.2 0.0 Cuba 5.1 4.8 Poland 6.30 10.7 Yugoslavia 3.0 2.7 Vietnam 1.8 0.0 USA 0.9LL2.6 1 1 Hungary 5.0 6.2 Finland 0.0 3.4 i,, China 0.0 2.7 Others America 0.4. 1.0

bource: h. I. U (1992), p. 7U, based on World Bank, Statistical Handbook: Statesof the FSU.

68

Chapter 3

Central Asian and CaucasusCountries

3.3.2.3Tadjikstan Tadjikstan's foreign trade is dominatedby the former republicswith which (exceptfor 1991)it had a deficit of about five times during 1992 and 1993 (Sinclair 1995).The total foreign tradehad a slow growth during 1987-1991, while the increasein 1992(as shownin Table 3.22), as in other former republics,is due to independence the devaluationof the and in Roubleratherthan to an increase real values.
Table 3.22 COMDOSition of Tadiikstan eXDortsrimDorts trade (Rb! million). 1987 1988 ' 1989 Imports Exports Exports Imports ExLo Imports Total 2264 3261 2359 3300 85 1 3460 2908 1970 2025 Total inter-republic 2856 2040 2891 Balance -938 -831 -851 294 1 353 Total foreign 334 1 444 3451-569 Balance -59 -110 -224 1990 1991 1992 Exports Imports Export sI -Imports Exports imports 3816 Total 1980 37011-3668 36646 30886 Total inter-republic 1672 2212 3201 1 3067 16433 23839 _ _ Balance +134 -540 -7406 Total foreign 308 1 725 500 1 601 20213 1 704 Balance +13166 -417 -101
(1992), pp. 91-92, based on 1mr, monomic Keview: iiiajiKstan; -ne turopa woria Year 130oK. (1994b), based on IMF. Economic Review: Tadjikstan, Financial Statistics: various pages for Tadjikstan, and international Supplement on Countries of the FSU; World Bank. Tadjikstan: 7be Transition to a Market Economy and Statistical Handbook: States of the FSU. Sources: LIU

Tadjikstan'strade is mainly basedon agriculture,one of the most important being cotton, for U which accounted I I% of the total former USSR production(E.1. 1992).Inter-republic for transactions accounted about 81.8% of exports and 90% of imports in 1990.To a great extent the commodity structurein Tadjikstan, as shown in Table 3.23, is similar to other CentralAsian countries.The other principal exportsare aluminium,raw cotton, textiles, fruit and vegetables, and marble, while its major imports are fuel, manufactured silk goods,and food products (Europe Yearbook 1994b). Primary and manufacturedgoods dominate its exports while intermediategoods, including capital goods and energy, are important and in mainly suppliedby CIS republics.After independence 1992,petroleumand gasconstituted 25.6% of the total inter-republicimport value, followed by chemicalproducts,and iron and
69

Chapter 3

Central Asian and Caucasus Countries

steel. At the same time exports to former republics mainly consist of non-ferrous metals

(29.1%),machinery(28.4%),andraw materials(25.8%). Exports from Tadjikistan to outside the CIS after independence showedgreaterexpansion than imports, which to someextent is due to the devaluationof the Rouble (UNCTAD/UN Non1994).However,the samereport stresses difficulty of obtainingaccurate the statistics.
Table 3.23 Structure of Tadjikstan inter-republic export/import trade (Rb: million). 1990 1991 1992 Exports Imports Exports Imports Exports Imports 272 000 000 364 000 6093 & Petroleum gas 63 000 71 000 000 000 Electricpower 000 110 000 191 Iron andsteel 000 2952 298 193 955 79 4782 524 Non-ferrous metals 119 322 72.0 258 Chemicals chemical 787 4761 products and , 228 796 486 475 4661 Machineryandmetalworking 2926 000 124 000 197 000 820 Wood andpaperproducts 29.0 181 000 225 Construction 000 materials 1061 558 954 574 4233 2777 Products light industry of 405 424 491 707 Foodandbeverages 1091 1455 92.0 182 000 AgriculturaIunprocessed 107 000 697 products ' 000 59.0. 142 000 000 Othercommodities 000 TIMTt - 23839 1 3359 2378 3201 3067 Total inter-republic trade
Source:Europa World Year Book (1995b), p. 2952. basedon IMF, Economic Review: Tadjikistan, and international Financial Statistics:Supplement Countriesof the FSU; World Bank, Tadjikistan:Ile Transition to a Market Economy on andStatisticalHandbook: States the FSU. of

ferrous metalsmadeup almost 99.5% of exports,while agriculturalproductsformed about 96.4%of the total imports in 1992(seetable3.24).
Table 3.24 Composition of Tadjikstan international import/ex ort trade (Rb: million). 199 2 1990 1991 Imports Exports Imports Exports Exports Imports Trade 39 168 488 000 20107 00 Non-ferrous metals Chemicals chemical 000 57 000 1.00 00 10 pWd-ucts and 000 113 000 2.00, 00 13 Machineryandmetalworldng 298 2.00 198 126 76 124 Products light industry of 000 000 Foodand beverages 151 272 95 000 4.00 81.0 000 126 6790 Agricultural unprocessed 00 products Other conunodities 4.00 29.0. 000 2.00 00 110 1 7M Total 308 768 1 500_1 601 20213
Source:Sameastable2.23.

Among the former USSR republics,Russia,Turkmenistanand Kazakhstan greatlyexpanded their imports in 1992 by 46.7%, 14.6%,and 12.2%respectively.At the sametime Russia,

70

Chapter 3

Central Asian and Caucasus Countries

Kazakhstan, and Ukraine took 47.5%, 14.7%, and 11.7% respectively of the exports of

Tadjikistan(EuropaWorld Yearbook 1995b). Tadjikstan foreign tradewith At the internationallevel, Europeancountrieshave dorninated for 61.1% and 67.6% of exports and imports respectively, while Asia accounted 18.5%and At 17%respectively. the country level, Germanyand Bulgariahad most exportsand imports Tadjikstan (see table 3.25). Cuba and Japan,were prominent among American and with Asian countries. internationaltrade took a much higher shareand accountedfor 55.2% After independence (exports)and32.8%(imports)of the total tradeof Tadjikstan(EIU 1994a). Table 3.25 Share of major European, Asian, and American countries in the foreign trade of Tadiikstan in 1990(% of total in foreign trade Roubles).
Germany Bulgaria

Poland Chzech & Hungary Slovalda 5.1 9.0 7.7 3.1 1.10 Export 7.3 00 8.1 5.2 15.7 Import U.K Austria Italy, Finland France 0.0 3.6 3.1 2.7 2.7 Export 0.0 4.1 4.7 3.7 0.0 Import Cuba USA Other America China Other Asia Japan Other Europe 56, 0.0, 15.1 0.01 14 3.4 12.8 Export 5. 3.7 2.7 1Itt 2.5 18.8 Import Source:E.I. U (1992),p. 93, basedon World Bank, StatisticalHandbook:States the FSU. of

3.3.2.4 Turkmenistan
Before 1990Turkmenistanhad a trade deficit with the former USSR republicsand to 13.2%of GDP asshownin Table 3.26 (E.I.U 1992). alsowith othercountriesequivalent Table 3.26 Composition of Turkmenistan export/import trade (Rb: million).
1989
Total Total inter reZbl-Lc Balance Total Foreign Balance Exports 2660 2418 -326 242 1 -348 Imports 3334 2744 5901 1

1990
Ex orts 0 1 2641 2 69 -454 172 1 -513 P Im orts 3606 2 3 685

1991 Imports Exports


7906 6785 +2179 1121 1 +232 5497 4608 . 889

Source: ElU (1992), p. 115, basedon IMF, Economic Review: Turkmenistan;Europe World Year Book, (1994b), p. 3072. based on IMF, Econon-dcReview: Turkmenistan,and International Financial Statistics: The Transition to a Market Economyand Supplement Countriesof the FSU; World Bank, Turkmenistan: on StatisticalHandbook: States the FSU. of

71

Chapter 3

Central Asian and CaucasusCountries

But from 1991, as a rich hydrocarbon producing country, the situation was changed mainly

due to the export of oil and gas to former Soviet states,with which there has been a The value of both inter-republic and international trades increased considerablesurplus. 1. by abouttwo times (exports)and four times (imports)during 1989-199 significantly The composition of exports in terms of value with both ex-USSR republics and in The most important item internationallyis presented Tables3.27 and 3.28 for 1989-1991. is energy(oil andgas)followed by industrial products.
Table 3.27 Structure of Turkmenistan inter-reDublic eXDOrMMDOrt trade (Rb: million).

1990
Exports Petroleum gas Electric power Iron and steel Chemicals Machinery and metal worldng Wood and paper products Products of light industry Food and beverages Agricultural unprocessed products Other commodities Total inter-republic trade 743 70 152 1076 85 128 93 2418 Imports 100 84 209 949 125 453 478 34 99 2744

1991
Exports 696 67 147 1 1083, 206 124 74 2469 Imports 79 106 203 959 97 551 445 132 98 2923

1992
Exports 2563 152 400 2767 300 370 80 6785 Imports 125 167 320 1509 152 867 700 208 154 4608

_ Source: Europe World Year Book (1995b), p. JUIZ basea on IMtf. Lconoffuc Keview* 'lurkmenistan ana international

Financial Statistics: Supplement on Countries of the FSU; World Bank, Turkmenistan: The Transition to a Market Economy and Statistical Handbook: States of the FSU.

TnhIp 3-7R. Rtmehirp nf Tnrkmeni., dan international

einort/imnart
Exports 9 7 93 12 243 144 51 590

trade (Rh.- million).

1990
Trade composition Petroleum & gas Iron and steel Chemicals and petroleum products Machinery and metal worldne Wood and paper products Products of light industry Food and beverages Agricultural unprocessed products Other commodities Total international trade Exports 2 Imports 19

1991
Imports 7 120 9 255 185 60 685

1992
Exports Imports

445
9 35 576 17 42 1121 155 12 330 240 78 889

223 2 5 8 242

139. 7 4

Source: Europa World Year Book (1995b), p. 3U72. based on IMF, Economic Review: Turkmenistan and international Financial Statistics: Supplement on Countries of the FSU; World Bank, Turkmenistan: Tlie Transition to a Market Economy and Statistical Handbook: States of the FSU.

In terms of individual inter-republic trade, the shareof countriesis shown in Table 3.29 wherewithin the CIS exportsto Russiaand Ukraine arethe mostprominent.

72

Chapter 3

Central Asian and Caucasus Countries

Table 3.29 The share of the main inter-renublic foreign trade of Tiirkmpnktnn Kazakhstan Russia I Ukraine I Uzb-ek-ista-n-l Belarus Exports 2.59 49.811 7.411 2.06 27.78 1 3.99 41.741 15.601 Imports 5.67 1 3.59
Source: Europa World Year Book, (1994b), p. 2974.

In 104)n 1 Others10.78 31.40

Europe dominatesfor other trade with 71% of the total imports and 63.6% of exports, followed by Asia with 20% of exportsand22.1% of imports (seetable3.30).
Table 3.30 Maior sharesof international foreien trade of Turkmenstan in 1990.
Export Import Export Import Export Import Germany 11.4 25.4 Romania 4.3 0.0, Cuba 5.7 2.8 Bulgaria 9.3 7.2 Finland O. o 3.2 USA 0.0 3.0 Chzechs & Slovalda 7.1 8.1 Former Yugoslavia 0.0 3.2 Europe 63.6 71.0 Hungary 5.7 0.0 India 2.9 5.7 Asia 22.1 20.0 Poland 5 9 Japan ol 3 America 6.7 7.2 China 0.0 2.7

Source: F-I. U, (1992), various pages for Ilurkmenistan, based on World Bank, Statistical Handbook: States of the FSU.

Imports from foreign countriesother than the former Soviet Union were only about 11% of
the total. Trading partners within the former USSR were mainly Central Asian countries. As

indicated in Table 3.31 the highest value for inter-republic imports is for capital goods followed by intermediateproducts,while foodstuffs are the most important imports from othercountries. Table 3.31 Composition of the foreign trade of Turkmenistan in 1990(Rb: million: forehm trade vrices).
Major commodity group
Capital goods
Energy

Inter-republic 37
1760

Others 2 23 3
2 0 4 114

Food
Intermediate Intermediate Tranufactures primary goods goods

116
711 50 99 2773

Total 39 1783 119


793 50 103 2887

Other consumer Total

Source:E.I. U (1992),p. 116based World Bank, StatisticalHandbook:States FSU. on of

3.3.2.5Uzbekistan The economyof Uzbekistanduring the Soviet era was mainly basedon cotton and Karakul skins both for domesticindustriesand foreign trade, although its land has a high potential for other agriculture using artificial irrigation (E.I.U 1994aand Lerman, Garcia-

73

Chapter 3

Central Asian and Caucasus Countries

GarciaandWichelns, 1996).The foreign tradeof Uzbekistan,like all other fonner republics, heavily dependson Russia, accountingfor 65% of imports in 1993 (Griffin 1996). This in considerable deficits. Table 3.32 shows the inter-republicand foreign trade and resulted
trade balance of Uzbekistan. Trade outside the CIS is not significant (see table 3.32). Table 3.32 ComDosition of Uzbekistan exDort/imDort trade (Rb: million). 1990 1991 1989
Export Total Total inter-republic Balance Total foreign trade Balance 10169 8542 1 Import 14158 12046 Import 14158 12046 Export 9351 8169 Export 19535 17339 Import 21475 17766 Import 191885 162246 -39110 29639 1 -2257

1992
Exports 150518 123136 27382

-3504.4 1628 1 2112 -486

1 -3694.7 2798 1 1182 -1613

1 -427 2196 1 3709 -1513

Source: EIU (1992), p. 103, based on IMF, Economic Review: Uzbekistan; huropa World Year Book (1994b), p. 3267. based on IMF, Economic Review: Uzbekistan, and International Financial Statistics: Supplement on Countries of the FSU; World Bank, Uzbekistan: The Transition to a Market Economy and Statistical Handbook: States of the FSU.

As shown in Tables3.33 and 3.34 in terms of the compositionof foreign trade,Uzbekistan producesa surplusin oil and gas, non-ferrousmetals,and light industriesincluding cotton
and its products while it has considerable deficits in food and ferrous metals.

The main trading partnersof Uzbekistanduring 1992 for both exports and imports were Russiaand Ukrainewith 53% and 13%of exports,and 14%and53% of importsrespectively (EuropaWorld Yearbook 1995b). At the intra-CIS level, during 1994it had a surplusof about US $246 n-0ion, particularly with Russia.At the non-CIS level Uzbekistanis very 94% of tradewith the active in the cotton marketsof the world. For example,it constitutes UK (UNCTAD/UN 1994).
Table 3.33 Exporblimport structure of Uzbekistanat the commodity group level in 1992(Rh-million. domesticnrices).
Type and composition of trade Oil & Gas Machinery & metals Light industries (incl. cotton) Non-ferrous metals Food industry Ferrous metals Total Source: l. U (1993c),Uzbekistan, 19. E. p. Exports 16956 22614 72143 15355 000 000 127068 Imports 886 3292 1938 000 26762 26255 59134

74

Chapter 3

Central Asian and CaucasusCountries

Table 3.34 Composition of Uzbekistan export/import trade (Rb: million).


Composition of trade Petroleum & gas Electric energy Iron and steel Non-ferrous metals roducts Chen-dcals chemical and Products of machine-building industry Wood and paper products Construction materials Products of light industry Products of food industry Agricultural unprocessed products Total trade including other products Total inter-republic trade Total foreign trade 1989 Exports Imports 646 1032 214 000 000 676 468 424 894 1111 1190 3553 000 725 000 228 4659 2761 795 2156 757 661 10169 14158 8542 12046 1628 2112 1990 Exports 598 207 000 447 853 1231 000 000 4242 824 447 9351 8169 1 1182 1 Imports 888 000 661 409 1147 3625 560 205 2963 1983 1309 14662 _ 11864 2798

Source: Europa World Year Book (1994b), p. 3267. basedon IMF, Economic Review: Uzbekistan and International Financial Statistics:Supplement Countriesof the FSU; World Bank, Uzbekistan:The Transition to a Market Economy on and StatisticalHandbook:States the FSU. of

datafor its non-CIStradepartners shownin table 3.35, is The latestcomprehensive aggregate indicating the important roles in 1992 of the former CMEA and UK for the exports of
Uzbekistan, and Switzerland for its imports.

Chart 3.35 Main international import/export trade of Uzbekistan in 1992.


Belgium Imports 00.0 Exports 12.9 Tormer CMEA Imports 4.8 3.8 Exports 14.2 8.91 Source: EIU (I 994a),Uzbekistan, 19. p. UK 00.0 13.5 Turkey, Germany 3.00 10.8 Others 46.7 32.7 Switzerland 21.2 00.0 China 7.3 0.0 Other EU 6.6 9.2 .

3.4 Transport of the CAC countries


All CAC countriesare connected eachother and other neighbouringcountrieswith to to a non-standard network. Armenia, Gyrkyzstan,and Tadjikstanhaveno access the sea, rail while Georgiais the only full maritime country of the bloc. Azerbaijanwith Kazakhstan and Turkmenstanhave direct accessto Russianand Iranian ports through the CaspianSea and
through the Volga Don Canal to the Black and Baltic Seas.

75

Chapter 3

Central Asian and CaucasusCountries

According to Mc Donell (1995,p. 14):


Sincethe independence, traffic on the Black Seahasbeenalmosttotally disruptedand the Ile formerly thriving rollthe previousfleet is in disrepairor hasbeendispersed. most of Turkmenbashi (port of Krosnovosk)andBaku hasbeen on/roll-off shuttleservicebetween Even so, somenew tradesare developing.SeveralIranianports, virtually discontinued. cargoes with otherCaspianlittoral notably Nooshahrand BandarTurkmen,areexchanging countriesand, ... ."

Kazakhstan the Far Eastgatewayof the CentralAsian railway connectingChina to Iran and is
other Middle East and Western countries while Georgia is the gateway for Europe.

Demand for transportin the CAC countrieswas severelyaffectedby the consequences of independence both internally and externally. This resulted from existing significant interand republic economic dependence new border crossingbarriers, although a fundamental movementtowardsco-operationhas taken place since the end of 1991.While the previous the central planning designed road systemonly to act as a feederserviceto rail in the CAC countries,the new situationinevitably forced thesecountriestoward the expansionof roads and a road freight systemcapableof taking vehicleswith highercapacities about20 tonnes of (Jenkins,1994). by As stated,road transportin the CAC countriesis characterised feederservices rail. The for eight republics have 42.9% and 17% of the total former USSR 903,000 km of roads and 147,360 km of rail network. Road freight traffic is mainly composedof light and short
distance traffic on narrow roads. (Jenkins 1994). Since the break up of the fornier USSR

thereis no reliable dataaboutthe inventory of rolling stock and road freight vehiclesof each of the CAC countries.

3.4.1 Transport in Caucasus countries


The Caucasus countries,although smaller, have more integratedtransportnetworks than the Central Asian republics.They have 7.5% of the rail and 3% of the total roadsof

76

Chapter 3

Central Asian and Caucasus Countries

former USSR.The hard surfaceroad and rail networksof the Caucasus region (Table 3.36) areaboutone-fifth (115)that of the CentralAsian republics. Table 3.36 Transport infrastructure and superstructure in the Caucasuscountries.
Road vehicles Rail networks, km Road networks, km Armenia (1992) 825 7700 Azerbaijan (1992) 2040 28600 Georgia (1989) 1570 31200 Total Caucasus 4435 67500

for EuropaWorld Year Book (1994a),variouspages Caucasus countries.

3.4.1.1 Azerbaijan

Azerbaijan, a semi-maritimecountry on the CaspianSea,has partial accessto the Baltic and Black Seasin Sumner through the Volga-Don waterwayin Russia.The regular betweenthe port of Baku and all other ports of the coastalcargo/passenger operate crossings CaspianSeaand are designed takepassengers, wagons,trailers,andrefrigerated to rail goods (Agazadeh, 1994).
In terms of rail connections, Azerbaijan is linked through the broad gauge easternsystem with 'Ibilisi (Georgia), to the Russian Autonomous Republic of Daghestan in Makhchkala, and

Yerevan(Annenia) through Nakhichvan.The total length of the link is 2040 km of which 806 km is electrified and about 1270km. is double track. According to an agreement with Iran in 1992,Azerbaijanwill be connected with the AutonomousRepublic of Nakhichevan throughIran, bypassing Armenianterritory. The rail network in Azerbaijanis claimedto the handle40 million tonnesof freight and20 million passengers annually.Azerbaijanhasa road in network of 30,400km of which 28,600krn (seeTable 3.36) were hard-surfaced December 1989. More thansixty per cent of the truck capacityof the former USSRwas devotedto vehiclesof trucking systemandthe importance the eight tonnesor lower, emphasising poor long distance of the roadmodeasa feederserviceto rail (Holt 1993).

77

Chapter 3

Central Asian and Caucasus Countries

3.4.1.2Armenia Armenia as a landlocked country is connectedwith road, rail and air modes to neighbouringcountries.In 1990it hada total rail track of 820 km and 100diesel-electric and 80 diesel locomotives.In terms of internationalconnectionsthere are rail links to Turkey, Iran, Georgia and the AutonomousRepublic of Nakhjavan.The motor roads in Armenia mainly have hard surface.As a part of the former USSR it suffers from the lack of an independentrail gateway with Iran (all trade was carried out through the Autonomous Republic of Nakhjavanof Azerbaijan)and it was disruptedwhen the conflicts concerning NagornoGarabakh A for started. major positivedevelopment Armenianforeign tradewas the completion of a bridge over the Araks river along the Iranian border in 1996 in the Noor Dooz (Megri) providing normal flows of trade to the PersianGulf and Europerespectively
for Annenia and Iran (Sanate-Hamlo-Naghl 1996a).

3.4.1.3Georgia In terms of internationalconnections there are rail links to Turkey, Azerbaijan,and the RussianFederation. Georgia,as the only maritime country of the republicsbeing considered, is connected with road andrail modesto neighbouringcountriesandothersthroughthe Black Seaports of Batumi and Sukhumi.Georgianrailways in 1990had a total lengthof 1570km. The road network in 1989comprised 35,100krn of which 31,200km werehardsurface.

3.4.2 Transport in Central Asia


The CentralAsian transportsystemcoversvast areas. five countriesin this region All different levels asshownin Table 3.37. They have35.4% arc connected with rail and road at (903000km) of the total public roadsof the former USSR and 14%of the rail network (Holt 1993).Rail forms the most important mode for the transportof freight and passengers for long distances.In Central Asia, two countries have accessto the CaspianSea, although,

78

Chapter 3

Central Asian and Caucasus Countries

unlike Georgia,none have accessto the open seas.Another featureof Central Asia is the the Aral Seaand the two navigablerivers of Amu Draya and Syr Darya,which existence of constitute important intra-Central Asian freight communications for different shipping companies.
Table 3.37 Transnort infrastructure and SUDer structure in the Central Asian countries (km)-

Transport infrastructure Rail networks Road networks

Kazakhstan (1991) 14148

Turlanenistan (1989) 2120 22600

Gyrkyzstan Uzbekistan (1990) (1990) 370 3460 28400 73100

Tadjikstan (1990) 480 28500

Total 20578 319464,

_166864 Sources: Europa World Year Book (1994a) and (1995b) various pages.

3.4.2.1Gyrkyzstan Gyrkyzstanis a mountainous country and hasthe smallestrail and road network among the Central Asian countries.The limited rail network as shown in Table 3.37 connectsthe network only, but it is linked by road to Tadjikstan capital of the country to the Kazakhstan
Kazakhstan. It has only 600 km of inland waterway. and

3.4.2.2Kazakhstan It has a poor quality road systemexcept in the north-westempart of the Uralsk (Hanter 1995).The availableperformance the broad gaugerailway as given in Table 3.39 of indicatesa declinein freight and passengers. Most of the freight of the countryis transported by rail (E.LU 1992). The north-eastward rail line of Kazakhstanis joined to the TransSiberian Railway (TSR) while its eastwardline is connectedto the Trans-ChinaRailway (TCR). Kazakhstanis an important rail and road link betweenRussia and the other four countriesof CentralAsia. The road network in December1991was shownin Table 3.37, of with Russia(46 border which 114,636km have a hard surface.It is mainly road-connected crossings)and with Uzbekistanand Gyrkyzstaneach with 7 links, and China through 6 border crossings.Inland waterwaysare operatedby different companiesand in 19893,857 million tonnes-krnof freight were carried by this mode (EuropaWorld Yearbook 1995a). This includesnavigationthroughthe Syr Daryariver to the Aral Seaand also domestictrade

79

Chapter 3

Central Asian and Caucasus Countries

along Lake Balkhash. Kazakhstanhas two main ports along the northern coastlineof the CaspianSea,both of which are active in coastalshipping trade with other statesalong the Caspian andalsowith European countriesduring Summer. sea
T.qhlp IAR Kazakhstan
Passenger-km (nifflion) Freight net ton-km (rrdllion)
Source: The Europa World Yearbook

railwav

nerformance.
1989 18,921 409,573 1990 19,734 406,963
Yearbook. p. um.

1991 19365 374,230

(199-3b), Based on UN btatisticai

3.4.2.3Tadjikstan

Tadjikstan has a shipping waterway of about 200 km on the Amu-Darya between Term, Sarava,and Jilikulan and also through Syr Darya both of which, after passing Turkmenistanand Uzbekstanto the west and Uzbekstanand Kazakhstanto the north, end in the Aral Sea.It has 28,500 km of roads with a total railway of about 480 km in 1990 connectingthe national capital to the other rail networksof the Central Asian countries. 3.4.2.4Turkmenistan
Turkmenistan has fewer transport problems than other landlocked CAC countries as a

result of having a long border with Iran (Akiner 1995).The CaspianSea,with a controlled to the Black and Baltic Seasin Summer,is the only sea outlet of Turkmenistan. access Another waterwayis the Amu Draya river of about 1300km, which providesdomesticand ending at the Aral and Kazakhstan regional trade routes from Turkmenistanto Uzbekestan Sea.Turkmenistanhas a broad gaugeEasternsystemrailway network of 2,120 Ian. The rail fleet consistsof 1280freight cars with speeds up to 80 to 90 km/hours and 400 coaches of having a speedrangefrom 95 to 100 km/hour. The road networks of about 23,200km (of in which 18,400werehard-surfaced November 1991)are mainly narrow band.Turkmenistan will play an essentialrole for other Central Asian foreign trade through Iran, in particular is when the bilateral rail project of Sarakhs-TaJan completedat the end of 1996.It should

80

Chapter 3

Central Asian and CaucasusCountries

accommodate annualfreight load of eight million tonneswhen provided with appropriate an to technology(Tehrani, 1996).It will give access the PersianGulf and the Indian Ocean,to Pakistanand the Indian sub-continent, Ahamad, and also to Europevia Turkey (Shernshad 1995).

3.4.2.5Uzbekestan
Uzbekestan is connected by road, rail and river to all its neighbours. It has the second

largest rail and road network of the CAC countries after Kazakhstan(see table 3.38) and acts as a junction for the rail connectionsof the easternand westernparts of Central Asia. It has extensivenavigablewaterwaysthrough the Amu-Darya and Syr-Daryarivers and the Aral Seacoast.

3.5 Trade with Iran


imports Iran, as a potential and close market to most of thesecountries,established
and exports with the CAC republics soon after the collapse of the former USSR. Iran also

has extensivetrade with the Europeanand American continentsvia Turkish and Georgian ports. Therefore, another advantagefor Iran and the Central Asian countries is that the transit route to the Black Seaport of Trobzan in Turkey is about 800 km longer than routes 1995b). within Armenia to Batumi in Georgia(Sanate-Hamlo-Naghl Iran's trade with the CAC countries expandedrapidly after the break-up of the former USSR in 1992.They had 38.8% and 62.7% of Iran's total exports to and imports from the CIS with a value of about US $ 401 million (Ministry of Economics and Finance, 1993). Table 3.39 showsthe foreign tradeof Iran with all the CAC countriesfor the first two years of independence thosecountries.In only two yearstherewas a growth of about 107%in of total trade, with a net imbalancein the volume of trade with the CAC countriesinvolved. Exports to the CAC countriesin 1993 accountedfor about 12% by weight (15% by value) 81

Chapter 3

Central Asian and Caucasus Countries

for of the total Iranian generalcargo exports.The equivalentpercentages imports were 7% 1994b). and4% respectively(Sanate-Hamlo-Naghl Table 3.39 Volume of the foreien trade of Iran with CAC countries (000 tonnes).
CAC countries I Armenia 2 Azerbaijan & NaWiavan 3 Georeia 4 Gyrryzstan 5 Kazakhstan 6 Ta ikstan 7 1 ur enistan 8 Uzbekistan 9 Total Imports 1.73 733.76 0.0009 0.035 9.44 0.13 2.96 000 748.05 1992 Exports 0.80 26.79 0.07 0.00 0.15 0.45 10.19 0.09 38.52 Total 2.52 760.54 0.07 0.035 9.59 0.58 13.15 0.09 786.57 Imports 44.61 1278.94 0.92 0.00 44.17 13.53 52.98 0.09 1,436.02 1993 Exports 12.88 117.91 0.54 0.3 8.07 0.34 4.85 0.59 189.08 Total 57.49 1396.85 1.46 0.30 52.22 13.87 101.43 1.49 1,625.10

Source: For 1992 Customs' Administration of Iran (1992) annual reports for imports and exports. For year 1993 based on Sanate-Han-do-Naghl, (1994b), No. 135, pp. 15-16 based Iran's Customs imports/exports detailed reports for 1992; Ministry of Economic and Finance (1993), Assessment of the Economic Co-operation of the of Iran with CIS in 1992, various pages.

During 1991, the three Caucasus countries, according to the Ministry of Economics and Finance(1993), had 29% of the total export value of Iran to the CIS and 59% of imports to Iran. The Central Asian countrieshad 9.8% and 2.2% of total CIS exports and imports with Iran. The performanceof freight transit for the CAC countries during the first eight months of the Iranian solar year in 1995 (starting from 22nd March) accounted for about 872 thousand tonnes. This trade was composed of rice, sugar, tea, oil products, chemical products, metals, machinery, household furniture, fertiliser, fibres and synthetic fibres (Sanate-Hamlo-Naghl 1995d).

3.6 Economic Co-operation Organisation (ECO)


It will take a long time for the CAC republics, neighbouringcountries,and the world to the understand assess future interestsand objectivesof the CIS and in particular Russia and after the USSR.The CAC republics are looking for support for economicdevelopmentand even defenceagainstRussia(Baba 1994).All thesecountrieswill inevitably be dependent on the Russian economic and industrial system for a long time, but they also seek and 82

Chapter 3

Central Asian and Caucasus Countries

welcome other alternatives both internationally and regionally to secure their present and

future situation. ECO is a new but non-military version of the Regional Co-operation for Development (RCD) and the Izmir treaty. ECO was originally set up in 1985 betweenIran, Turkey and Pakistanwith a permanentcentre in Tehran to facilitate econon-Lic, technical, and cultural developments co-operation.Its formal structureis composedof the Council of Foreign and Ministers as the main decision makers,and meetingstake place by rotation at least once a Ahmad (1995) ECO is now, basedon population, after the According to Shemshad year. European Union (EU), the second largest regional organisation in the world. It has branchesof co-operationassigned each follow three major specialised to committeeswhich of the following founder members: * The ECO banking systemis basedin Turkey. * Insurance issuesare discharged Pakistan,and to 1995d). * Transportissuesto Iran (Sanate-Hamlo-Naghl ECO gainedsevennew membersin 1993(six CAC countriesand Afghanistan)covering a region with 7 million squarekm. and about300 million population (Haghi 1995).Armenia in is an observermember of the treaty and frequently attendsECO sessions Tehran while Romania is also seeking admission. ECO has therefore strengthenedits position as a regional co-operation centre, developing its relations with other regional, Islamic and international agenciesgreatly (Gharabaghi1994, Yearbook of International Organisations 1995). ECO aims to eliminate Customs tariffs among member states (initially with a 10% reductions of tariffs), to establish an investment and developmental bank, to expand tourism, and to facilitate transportand communications(Baba 1994).

83

Chapter 3

Central Asian and Caucasus Countries

The ECO shipping organisation in 1995 had five members; Iran, Pakistan, Afghanistan, Turkmenstan,and Kazakhstan.Its operationnow not only covers the CaspianSeabut also other internationalwaters.The first ship of the ECO shipping organisationwas provided by Iran and is now operatingin the Indian sub-continent, Far East,and the South of Europe the
(Tarjoman 1995).

3.7 Conclusions
This studyhasdealt with the CAC countriesfrom the macroeconomic point of view and has a twofold researchaim: first to investigatepotential areas affecting the demand for internationaltradeto provide an insight andsupportfor otherpartsof the research;second, to indicatethe existing transportsituationof a region which requiresthe use of a new outlet for
foreign trade. Therefore, it has attempted to investigate these countries in terms of political,

foreign trade,and transportlevelsand alsothe compositionandrole of the ECO as econornic, a regionalco-operationorganisationfor the CAC republicsand other members.The insight gained from the review, despitethe lack of fully comprehensive statistics,can be used as supportfor further investigations study(ISLB). It becomes relating to the Iran Sea-landbridge evident that the CAC countriesare passingthrough a transitionalperiod and, due to sharp economic, political and trade changes, will also need developmentsin infrastructure provision. They have many common problems,but eachcountry has its own potential and for weaknesses it is impossibleto represent completeand single future perspective all and a
countries combined.

However it would appear feasible to develop a general model based on aggregate internationalforeign tradevaluesas a basefor the long tenn evaluationof the potentialISLB flows from the CAC countries.

84

Chapter 4

Landbridges:

Concept and Practice

4. Landbridges: concept and practice


4.1 Introduction
The purposeof this chapteris to examinethe conceptof a landbridge,and to identify a suitable form to investigate the specific landbridge applicable to Iran and the CAC countrieswhich will be describedin the next chapter. This chapteradoptsan approachbasedon threemain stages: * The conceptof a landbridge * * Influences(controls) on landbridgedevelopment Examplesof landbridges

Different authors spell the term as "land bridge", "land-bridge" and "landbridge". This work will use the last mentioned spelling except when giving a direct quotation where anotherform of spelling may be used.

4.2 The concept of a landbridge


The landbridge phenomenonis not a new concept (Hayuth, 1987) and dates back to travelled betweenEast and West for centuriesand ancienttimes. e.g. the silk way caravans before the invention of modem transportsystemswas a well organisedsystememploying mainly land, river and short sea modes. The landbridge concept suffers from a lack of detailed academicstudy and, therefore, the limited number of cited writings have been tracedas far back as 1969.

4.2.1 Definition of a Iandbridge


A landbridgeaccordingto Hayuth (1987,p. 86) is:
"The utilisation of the land transportfor a part of what would normally be entirely an oceanvoyage".

85

Chapter 4

Landbridges:

Concept and Practice

Miller (1977, p. 64) more specifically defines it as: "A serviceprovided betweentwo or more modesto transportgoodsoriginating from a foreign point, transiting domesticpoints, and finding its destinationat anotherpoint".

The sequence activities for the generalconceptof a landbridgeis shown in Figure 4.1. It of is as an integratedinternational transportservice betweenorigin and destinationcountries, the infrastructure of any number of intermediary countries to connect the two sea using passagesat each end under a contract of carriage. In this sequenceintermediary and destination country might be the sameor differ. This is sometimesevidencedby a single bill of lading (B/L) between shipper and carrier, although the use of another single documentsuch as a waybill is more relevant to the origin-destinationnature of landbridge The B/L has a preciselegal function which is not directly relevant to this work. practices. Therefore,the more generaland less specific term "waybill" will be used,following a brief discussion.
Fipurp 4-1 General concent of a landbridim.
Consignment in country'A (origin) "Ist-"., Sea mode Consignment in transit countries 1st
port

Consignment by rail / road

Consignment in transit countries 2nd port

2nd Consignment Sea 'q 'In'country B mo4e (destination)

Singlewaybill
The BAL, as a traditional and commonly used documentof carriage of goods by sea,has functions (Wilson 1993,pp. 126-162);as evidenceof the contract three characteristics and of carriage,as a receipt for goodsshipped,and finally as a documentof title. According to Wilson (1993, p. 163):
"rhe waybill differs from the bill of lading in that, while it actsas a receipt and provides evidenceof the contractof carriage,it lacks the third characteristic(as mentionedabove)in that it doesnot constitutea negotiabledocumentof title".

Somecharacteristics the bill of lading are that anyoneholding the documentcan receive of the shipment, since it is a negotiable document. The shipper must despatchthe bill of lading directly to the consignee post or deliver it to a bank. A bill of lading needsto be by

86

Chapter 4

Landbridges:

Concept and Practice

sent to the consignee much faster than the shipment otherwise problems arise. A bill of

lading is called a 'shipped bill' if a shipmentis receivedon ships, otherwiseit is 'received for shipment' and issuedat any inland point where shipmentis received(Day 1980,Todd

1990).
According to Todd (1990, p. 252):
"Tbe InternationalMaritime Committeehastaken the view that in the great majority of cases for at cargo a bill of lading is not necessary all. Negotiability is not of the essence of general goods,which are not re-sold during most liner shipments,and shipmentsof manufactured transit".

A waybill is a non-negotiablecontractualdocumentwhich acts as evidenceof the carriage of goods, a receipt for the goods by the carrier and the carrier is obliged to deliver the shipmentsto a namedreceiveror cargoowner. It was developedfirst for use in land and air transportmodeswhere there was no or limited time for negotiations;but later was adopted in seamodesin the 1960swhen there was a considerable reduction in the journey time of shipmentsby sea becauseof faster ships and cargo handling operationsin ports through containerisation,etc. This resultedin shipmentsarriving aheadof the legal cargo delivery documentsand the waybill was introduced to avoid the wait. It is designedto cover a freight and to overcomethe transit problems under traditional bill of ladings (for simple which may delay documentsin transit). Unlike bills of example communicationproblems lading, the consigneeat the destination cannot present a waybill to take delivery of the shipmentsunlessthere is proof of identity of the namedconsigneeor his authorisedagent at the port of discharge(Hopkins 1982,Todd 1990,Wilson 1993). According to Wilson (1993,p. 163):
"The modem waybill follows the patternof the short form bill of lading in that it is a short form documentwith a blank back but with a specific clauseincorporatingthe carrier's standardterms and conditions.As with the bill, the waybill is available either in a proprietary or commonforne'.

However, the important concept in different landbridge operations is that a shipment should move under a single document along the entire routes from origin to destination 87

Chapter 4

Landbridges:

Concept and Practice

Negotiation over shipments is not required during the transit period but the carrier is

Therefore, a waybill is an appropriateterm obliged to deliver goodsto a namedconsignee. to be usedin this chapter;althoughit shouldbe understoodthat a BALcould be relevant. Despite the fact that all bridge operations(even short ones) use some kind of road or rail landbridgesmay be classifiedunder a numberof headings, which are: mode, * Minibridge * ICcrobridge * Sea-airbridge * Air-road-air bridge, and
* Sea-river bridge

Thesewill be consideredin sequence. 4.2.1.1Derinition of a minibridge According to Hayuth (1987, p. 88) a minibridge is:
"Ile movementof containersunder a single bill of lading from one country via a vesselto a port in anothercountry, thenceby rail minibridge to a secondport city, terminating at the rail carrieesterminal."

According to Norris (1972, p. 41) a minibridge is:


"rwo or more modesworking in tandem- fully co-ordinatedand producinga plannedthrough serviceusually on a through pricing basis".

The minibridge is an integratedtransportoperation. It involves transportbetweentwo countries(or more) with only one seatransit during the entirejourney. The journey endsat the second port in the destination country after an inland transit (Hayuth 1987) but nowadaysis continued from the first port to other countries by road or rail modes. The stagesof a minibridge are shown in figure 4.2. According to the traditional definitions minibridge services are less likely to cross more than two continents compared with landbridges. They appear to have been first offered in 1971, when transatlantic cargo destined for a US Gulf port was off-loaded at an East Coast port (Hayuth 1987, p. 88).

88

Chapter 4

Landbridges:

Concept and Practice

Miller (1977) and Hayuth (1987) restrict their discussion of the minibridge mainly to

journeys startingor ending in the United States.


FiLyure 4.2 General concet)t of a minibridee.
Consignment In country'. A, (origin) at an inland point (IP) Consignment by domestic mode (road rail) Consignment in the port in country of origin (A) Sea Consignment j , ,, in 1st mode,. port in country B Consignment by domestic mode (road rail) Consignmentin, 7 2nd port In country B (final stageof minibridge)

Single waybill, The developmentof minibridge traffic according to Hayuth (1987) has been mainly for threereasons: * The standardisation containerunits, of * The growth of intermodal transport,and * The greatimprovementsin rail transportservices. The differencebetweena Iandbridgeand a minibridge is that landbridgesare composedof sea-land-sea movementswhile the latter is restrictedto sea-landmovements.According to Hayuth (1987) the advantages a minibridge can be observedfrom both users' (shippers') of (shipping lines') viewpoints. Shippersare in favour of minibridges because and carriers' they provide them with better servicesand alternative transport modes,ports of call and Norris ship schedules, while carriershavethe choiceof thoseports most suitedto demands. (1972, p. 41) discusses minibridge concept from the carriers' and shippers' points of the view. For carriers:
"It may expandthe market of either, both, or all of the carrier or other transportorientedparticipantswhich, of course,reflects the stimulus of carrier competition" (page4 1).

On the other hand,for shippers:


"The minibridge, like its kin, intermodalism,represents new - an additional- transport a that option which, in the relatively few minibridge arrangements exist to date hasproved, accordingto claims, a very effective techniquein meeting someof the demandin transportation"(page41).

89

Chapter 4

Landbridges:

Concept and Practice

According to Miller

(1977) the microbridge (see Figure 4.3), is more limited in

geographicaldimensionsthan the minibridge. The main difference betweenthe mini- and the microbridge is that the latter is not involved in the double port containeroperationsin the destinationcountry. According to Hayuth (1987, p. 94) a microbridgeis definedas a:
"Through container serviceto and from an interior point in the United Statesand involving at least two modesof transportation, generally oceanand rail."
Fioure 4-3 General concent of a microbridee.
Consignment in country A (origin) at an inland point Consignment by domestic mode (road / rail) Consignment in the portin country of origin (A) Sea mode ConsignmentIn the port In destination country (B)

Single waybill 4.2.1.3 Definition of a sea-air bridge A sea-airbridge (Figure 4.4) is a pattern of transportbetweentwo countrieswhere the consignmentuses the sea mode to leave the country of origin and to reach a particular
seaport which is also near an international airport in an intermediary country. Then, the consignment is air freighted to the destination country. FiPurp 4-4 rpneral
Consignment In country A (Origin) at a port

concent of a sea-air bridLye.


Consignment In the port of a transit country (C) Shuttle modcs between port &airport In country (C) Consignment In the airport of transit country (C) Air -' ' mode Consignment In the airport In country B (destination)

Sea mode

Single waybffl`. ` The sca-airbridge conceptstartedas a combinedmode in the 1960sbut did not particularly interest shippersuntil the mid 1980s(Eller 1987). According to Eller (1987) Air Canada and KLM were the first air carriers to develop the sea air bridge concept correctly. According to Raguraman and Chan(1994,pp. 380-381):
"The sea-airconceptinvolves the movementof goodsusually by seaon the first consolidatedsector,and air on the latter, with the transferbetweenmodestaking place at a hub".

They point out that the great advantage sea-airtransport over other combination modes of is that it is less influenced by physical or geographicalbarriers. Another advantage seaof air, is that it is cheaperthan air freight aloneand much quicker than seafreight on its own. 90

Chapter 4

Landbridges:

Concept and Practice

Co-ordination at points of transhipment is essential for the efficiency of sea-air bridges, and

there is a need for official agreementsbetween sea, air and customs authorities (Raguramannand Chan 1994). These agreementsconcern preferencesin loading and
unloading ships, facilitated warehouse cargo clearance, and minimal paper work.

4.2.1.4Definition of air-road-air bridge


The combined air-road-air pattern was introduced to bridge a through flow between

Asia, and South and Central America via the USA road networks (UNCTAD, 1993). Miami Airport played a central role in receivingroad traffic from suchWest Coastports as San Francisco and Los Angeles. There are four main reasonsfor the viability of such a system: Industrial developmentof Latin America
Reduction in tariffs * Development of North American Free Trade Agreement (NAFTA), and Elimination of trade barriers

4.2.1.5 Definition of sea-river bridge This is an old establishedform of transportespeciallyin Europeusing navigablerivers.
For example, the Rhine-Danube project (TNT Press 1992) will provide another sea-river

bridge with a long river transit of about3500 km betweenthe Black Seaand the North Sea bulk sea traffic. Another sea-riverbridge connectsthe north east Baltic Coast with new from St. Petersburg with the CasPianSeavia the Volga-Don and also to Black Seaports through Russianterritories.

91

Chapter 4

Landbridges:

Concept and Practice

4.2.2 The benefits of landbridges


When a bridge systemis employedits type depends the geographicallocation of the on origin country and origin port, the volume, size and shapeof shipments,and the difference betweenthe cost of transportby land, seaand air modesto the destinationcountry. The main objectives for landbridge systems,accordingto Hayuth (1987), are to saveboth the transit time of unit loads and also total transport cost with fewer ports of call. This however that shippers' options over choice of ports will be reduced.Therefore, means, to have confidencein the landbridge systemused.Miller (1977, p. 65) shipperswill need indicatesthat:
"The objective of a land bridge is to provide the foreign shippers,and the foreign consumers (in the long term), the most efficient and economicalmeansof carrying goods".

Rijn (1981) indicatesthat transportby water is cheaperthan by rail, but the advantage can be compensated an overland route introducesa significant reduction in distanceand in if the freight rate. The capability and profitability of different modes, either singly or in combination,are obviously influencedby the distancetravelled. According to Bensonand Whitehead (1985) in using landbridge servicesthe three following points must be taken into account: freight rates,time andcargoneed. According to Talley (1988,p. 150):
"With the adventof deregulation,price and servicecompetition amongoceancarriershave intensified. In order to competein this environment,oceancarriers have entered into joint (in particular, railroad) in order to gain central control on arrangements with other modes intermodal systemconsistingof oceanand inland carriers. cargoesmoving within an Specifically, the competition of land-bridgewith all water serviceshave intensified".

According to Mahoney (1985) shippersprefer landbridgesbecause the ratesand service of they have obtainedin the wake of deregulation.Mahoney also discusses advantages the of the bridge conceptbeing under a single bill of lading comparedwith the traditional method of transportwith the issueof severalbill of lading or transit documentsfor eachstagefrom

92

Chapter 4

Landbridges:

Concept and Practice

of origin to destination.In short, the development a landbridge results from the following reasons: for * response the world trading requirements fasteroperationdelivery to * the generalrise in seatransportfreight ratesasa result of oil prices and wage levels * reductionin damageof goodscomparedwith all-waterjourneys of containerisationresulting in faster loading and unloading of products at growth and origins and destinations, intermediatepoints. * increase the supply of and demandfor high value and sensitivegoods. in * developments improvementsin overlandnetworksand vehicles and * improvementof regional and internationalrelations. * integrationof ownershipand/ormanagement intermodaland sea/landcarriers. of
* improvement and unification of the legal aspectsof the long distance carriage of goods such as the single bill of lading and waybill.

for Miller (1977) identifies three advantages both shippersand operatorsof landbridges pilferage" for the former, and "easeof with "efficient through service", and "decreased for handling" as the main advantage the latter. Miller also suggests two following the cargo for advantages the US as a landbridgehostcountry: both domesticand international. * Moneys accrueto US companies, * Diplomatic and national prestige for US carriersas those who can provide efficient and economicchannelsof trade. The development of the landbridge concept internationally has encouraged other developments. According to Mayer (1973, p. 154) in Canadaand the United States(two important landbridgecountries):
"The combinationof transatlantic,transpacific,and through landbridgetraffic has stimulatedthe long-haul rail-roads of the high-speedservices,including unit trains and run through trains specialisingin rail-oceancontainerand piggyback services. In many instances through bills-of-lading are availablebetweeninland and coastal points and overseas origins and destinations".

93

Chapter 4

Landbridges:

Concept and Practice

4.3 Influences on landbridge developments


The developmentof the landbridge concept has taken advantageof two important technological developmentsof the second half of the twentieth century; international in standardcontainerisationand swapbodies Europe,which have allowed the development of intermodalism.However, successfulapplicationof the landbridgeconceptneedsto take into account a wider rangeof important influences which can be placed under four main headings: Geographical Political Technical Organisational Although these classes of influence will be considered separately, they are often interconnected.

4.3.1 Geographical influences


The geographical natureof the overlandpart of a landbridgeto a greatextent determines the efficiency of the servicesprovided. e.g. The fact that Holland as a flat country is an important factor in the success of the port of Rotterdam as a landbridge port for neighbouringcountries.Suchphysical barriersas mountainshave implications both for rail and road in terms of lower speedsand frequencyof services,the need for more powerful trucks and locomotives, and an increase in freight and other costs. Harsh climatic conditions may restrict a service such as the Trans-SiberianRailway (TSR) particularly during the frozen Winter and melting Spring periods. The successful location of a landbridgerequiresreasonable market conditions betweentwo origin and destinationpoints. For example, the former Soviet Union lies betweenthe Far

94

Chapter 4

Landbridges: Concept and Practice

East and Western Europe with extensive interactive trade flows. Although the TransSiberian Railway was opened in 1907 (Hayuth 1987) it only become a successful betweenformer Soviet agencies landbridgeroute in 1971after bilateral formal agreements (e.g. Souyuzvneshtrans) Japanese Europeanfreight forwarders(Miller 1978). and and

4.3.2 Political influences


The political acceptabilityof the host country is anotherimportant factor in the growth of landbridges.Poor political and ideological relations with surroundingcountries cause instability and lack of credibility. The flexibility of the geopolitical conditions of the landbridgehost country is anotherfactor, which in practicecauses growth and success the of a landbridge. Liberalisation.in the use and operation of transport infrastructure and the ownership of ports and terminals (e.g. privatisation of services)have had a significant impact on the quality and quantity of servicesprovided, when comparingthe North American and TransSiberianlandbridges(Hicks 1994). The TSR landbridgeof 13200km can deliver a servicewithin 24 days from Yokohamato Hamburg.This should be comparedwith the North American landbridgethrough the New York-San Francisco rail link with a transit time of 25 days for 21,222 km. between Yokohamaand Hamburg (Damas 1992a,Bonney 1991).The harshweatherand the special geopolitical conditions of the TSR create many shortages in rolling stock, and the are management information systems lessreliable comparedwith the USA or Canadian and
landbridges for similar origins or destinations.

4.3.3 Technical influences


Inadequate infrastructurealso affects greatly the existenceor operationof landbridges. As mentioned in section 4.3.2 the lack of credibility of the TSR results from a lack of

95

Chapter 4

Landbridges:

Concept and Practice

railcars and rolling stock, signalling andinformation technology,etc. (Davies 1991,Damas 1992a).

4.3.4 Organisational influences


A landbridgeis likely to be a more complexorganisationalstructurethan other types of transport system. An efficient landbridge system will largely depend on flexible coordination linking a road fleet, port ownership, rail operation and ownership of rolling stock, and on the number and operationalcapability of the operatorsinvolved. Therefore, national and international organisationscan play significant roles in their viability and operations.In the Canadianlandbridgeboth CanadianPacific (CP) and CanadianNational (CN) had a successfulapproachto the intermodal concept (Hayuth 1987). Both of these Railway ownershipand operationbut companiesare not only involved in Trans-Canadian trucking, and airlines which make decision also in other modesand interfaceslike seaports, making and options over shipmentsmuch easier(Hicks 1994).In the samecontext, TSR in a different organisationalenvironment and as a government controlled organisation has sufferedfrom insufficienciesasdescribed earlier.

4.4 Examples of major landbridges


Different landbridges exist at inter-and intra-continental levels with geographical, political, technical, and organisationaldifferences. Important landbridgesoperate within the three following areas: * Europe/Asia * Trans-Siberia
* North America Figure 4.5 shows the most important ports and inland points used for landbridge purposes

in this study.

96

rr .4m

.r
ft

to

9'

It 1?
C,
co

od

Fl,

t: rl

v 110

z, tz:
97

, III
Chapter 4 Landbridges: Concept and Practice

4.4.1 Europe-Asia landbridges


4.4.4.1 Geographical characteristics The specific featureof Europeis the variety of countriesand economies, maritime its to geographyand its closeness two other continents,which hasmadeit suitablefor all types by of land, sea,and air integratedbridges(TNT Express1992).Europeis surrounded eight seas, causingthe continentto be a sourcefor all the current major landbridgesin the world, landbridges(seeFigure4.5a for an example). and many intra-European
Figure 4.5a Intra-European landbridge, minibridge and microbridges, TSR landbridge with sealeg between port Tunis, Odessaand Vostochney (unit: krn). &Z., ptterdam "_. *, "-Marseilles Calais 306 rail 4P Intra-European microbridge Inter-European minibridge 0 1667rail atunu (Ge: gi3a)6 Seame 3641 41 37 Seamode 693 rail \2974 Seamode 2076 rail an

rail 814 Mersin (Turkey) Tehran

Birmingham

Dover

40 Inter Europe-Asialandbridge(6086 from Birmingham,5864 from Rotterdamto Tehran via Mersin), (6344 from Birmingham, 6122 from Rotterdamto Tehranvia Batumi), all-water through SuesCanal from Birmingham-Doverand Rotterdamto port Abassand by rail to Tehran 13029and 12928.

Port Tunis

Port Odessa Moscow

Vostochney
(Far EastRussia)

Kobe
(Japan)

(D
2128 Seamode

ED4
1572TSR rail

09

9347 TSR rail

OEW

tT-I-

1487Seamode

Inter European-TSR landbridgebetweenFar Eastand Africa 14534,via SuezCanal 16603

Legend: Rail Sea


Source: Rail distances are based on Thomas Cook European and Overseas (1995). Timetables, vol. I& 11, various pages. Sea distance are based on Caney and Reynolds (198 1), Reed's Marine Distance Tables, various pages. Gavan (1987), The Ships Atlas 1987, (ed.). p. 4.

Minibridge services,restrictedto one sealeg, exist betweenNorth and South Europeand also to and from Asia and Africa with short sealegs from and to the Mediterraneanand the North Seaports. Longer distanceminibridges are also operativefrom theseareasto the

98

Chapter 4

Landbridges: Concept and Practice

Asian coast of the Black Sea and Mediterranean Sea via Turkey, Israel and Georgia (Eller 1987). These services compete with short sea trades operating between Scandinavian countries as well as the U. K and mainland Europe using road/rail networks to connect with second ports (Containerisation. International 1995).

The Caucasus area and Eastern Turkish coastlines have long and short distance rail/road landbridges with the Baltic and North Sea or between the Black Sea, Mediterranean Sea, and Azof Sea as shown in Figures 4.5a and 4.6. The Caucasus, in its location along the west coast of the Caspian Sea provides a landbridge sealeg to Central Asia for Europe and America via Turkey or Georgia (Figures 3.1 and 4.5a). Figure 4.6 The Euro-IsraeI! landbridge (unit: kin). Thesoloniki Ashdod Eflat ED 1472Meditcn-anean leg 300by road Sea 2308 Sea Red leg Djibouti ai

Ile Euro-Isracli landbridge, from7besoloniki Djibouti, Suez 4080 Canal 3978. to via
bource: Sea distances are based on Caney and Reynolds (1981). Reed's Marine Distance Tables, p. 71. Cavan (1987). The Ships Atlas 1987, (ed.). p. 13. Turkey maritime as the important country interface between Europe and the Asian Middle Marmara, to Eller East is a large Aegean, (1987), and this

surrounded

on three sides by four seas (Black, krn of coastline. to function According

Mediterranean locational

Seas) and has 8,300 has helped Turkey

advantage

as a key landbridge

country.

Different origin-

Euro-Turkish destinations between Mersin

landbridges compared

are able to save a distance the Suez Canal route

of about 6943 krn (similar to the port of Abass)

with

for journeys

Birmingham-and

Tehran

using two sea modes for Dover-Calais and Mersin-Tehran.

and MarseillesTurkey also is an

and two rail modes for Calais-Marseilles landbridge

Operational

via Rotterdam-Marseilles-Mersin-Tehran and Iraq (Figure 4.9).

(Figure 4.5a) and between

North American

ports-Mersin-Iran

99

Ch2pter 4

Landbridges: Concept and Practice

The Euro-Israeli landbridge (see Figure 4.6) takes advantage of Israel's location on the Red Sea (port of Eilat), and the ports of Ashdod and Haifa on the Mediterranean Coasts with

300 km road.The Israeli or Kedemlandbridge(Hayuth 1987)doesnot offer distance about advantageover the all-water Suez Canal transport between Europe and Asia, but technically it operateseffectively as a landbridgewith a highly co-ordinatedport-railshippingsystemoffering economicservic es.
Over longer distances, the main destination of the Trans-Chinese Railway landbridge (TCR) is Europort of Rotterdam (UNCTAD 1992) (see Figure 4.7).
Fligure 4.7 Trans-China rail landbridge via Central Asia, Caspian sea and CIS to Europe (unit: km). Rotterdam Brest(Poland) Ber in 78)
1083

1100 Moscow 1517 970 via Gomel- insk


Orenburg Kiev

3343 [Novosibirsk Ulan Ude 657 1121 2304 Ulan Bator


1561 ktogay (Kazamtan) Beijin ii

752 ney to Chita 1406


Harbin

Rostov 1319vi Gudemed


aku

1643via Gudemesd DPort 4111 Guriev


1849 Alma-Ata

079 via Urumqi 57


(KazaUmm)

iink tV 1160 569 'i

ang ellow Sealeg 1570

(China)U) Kobe "'% engzhou u 148

Caspian leg Charduz Samarkand426 9 (D Lugovov __ __ 4 Krasnovosk1141 60 926


Crurkmenistan) 369 502 Dushanbeh
n)

1782
Bishak (Gyrkyzstm) V'kv.

Shanghai
Hong Kong

225

cradja,

ermez(Afghanistan) TCR landbridge from Kobe-Lianyugang Rotterdam through:1- Aketehay(Kazakhstan)-Novosibirskto 13606.3-AketehayMoscow-Brest 12442.2-Aktehay-Orenburg-Moscow-Brest-Berlin-Hannover-Koln Alma-Ata-Charduz-Kmsnovosk-Baku 13919.4-Alma-Ata-Charduz Sea (Caspian bridge)-Rostov-Kiev (Turkinenistan)-Guriev-Rosrov-Kiev-Misk-Brest 14701.
Source: Rail distances are based on Thomas Cook European and overseas (1995), -1-imetables, vol. I& 11, various pages. Sea distance are based on Caney and Reynolds (1981), Reed's Marine Distance Tables, various pages.Gavan (1987), 71c Ships Atlas 1987, various pages.

Ile TCR which connects the port of Ijanyugang on the Yellow Sea to Europe and South West Asia via the CIS and Central Asian railways tries to compete with the Trans-Siberian

100

Chapter 4

Landbridges: Conceptand Practice

Railway (TSR) in Russia by offering 3000 krn shorter journeys, and is more suitable for trade to/from Hong Kong.

4.4.1.2 Political characteristics Europe, with a population of about 500 million (excluding Turkey and the former Soviet Union) is composed of twenty-seven mainly industrialised countries and is a major focus of international trade flows, both in terms of volume and value, making use of various single and integrated modes of transport. The existence of the EU (European Union) has facilitated greatly the use of multi-country networks, the movement of vehicles and the easing of Customs restrictions through different conventions and co-operative arrangements(Eller 1992). The existence of international centres, particularly ports such as Rotterdam, Hamburg, etc., are key elements in origin/destination landbridges. The present unsettled political situation in the Caucasus region has made different landbridge and minibridge operations less reliable after the break-up of the former Soviet Union. For example, there are disputes between Georgia and Russia over the Sukhumi region, and between Azerbaijan and Armenia over Nagomo Karabakh, which led to the stoppageof rail services from Georgian and Russian ports to Iran via Armenia after 1992. Turkey has a population of 56 million of which 7% are Kurds who have a long standing conflict with the central government. They live in the eastern regions close to Armenia, Iran, Iraq and Syria, and can influence almost all Turkish landbridge rail/road routes. The other major regional issue for Turkey as a landbridge country for the Mddle East is the dispute over the Euphrates and Tigris waters with Jordan, Syria, and Iraq (Europe World Year Book 1994b). Turkey, with Iran and Pakistan, is an important member of the ECO (Economic CoOrdination Organisation) which, after the break-up of the USSR, gained the membership of the success of the European

101

I
Chapter 4 Landbridges: Concept and Practice

qj

six new CAC countries as well as Afghanistan. Turkey also is a member of the BSECOG (Black Sea Economic Co-operation Group) composed of eleven countries. Both of these organisations are likely to provide strong support for the operation of different Turkish landbridges. As a result of Israeli-Arab conflicts the Kedem landbridge (through Israel) does not serve neighbouring countries and in order to compete with movements through the Suez Canal, is heavily supported by government subsidiesand technical services.

4.4.1.3Technical characteristics Europeancountrieshave an extensivenetwork of high performance multibandroads


and double track rail lines as potential landbridge levers (average 69 km/hour road freight speed and 37 km/hour rail). According to Hayuth (1992) one reason for the success of international and inter-European container traffic is the utilisation of swap body containers for inter-European traffic and ISO marine containers for international traffic. European countries have 43.8% of total world shipping tonnage (dwt) and 39.4% of the ships, and are the second largest ISO container producers in the world (UNCTAD 1993). The Caucasus landbridge, which owns only 3% of the total former Soviet Union rail network (Holt, 1993), not only suffers from regional conflicts but also from operational and maintenance problems. There are also delays in the return of inter-republic rolling stock. This results in poor transit time performance in the order of, for example, about 20-35 days from St. Petersburgto Iran (Hicks 1994). The Trans-Chinese Railway (TCR) is 7865 km in length between the ports of Lianyungang and Rotterdam (Strauss 1993). According to Speeceand Kawahara (1994, pp. 57-58): "Much rail capacityis antiquated, manyof the rail lines areold. Despiteconsiderable and less Progress upgradingsomelines in the pastdecade, thanone-quarter routeswere of at doubletracksby 1990.Only about 13per cent wereelectrifiedrailways.Only aboutonethird Over40 per of rail routeswereservicedby locomotives engines. with internalcombustion centof locomotives by the CentralRailway Department steamengines, while twoare run " thirds of thosein useat the local level aresteamengines. 102

Chapter 4

Landbridges: Concept and Practice

One of the main problems of the TCR, like the TSR, is that it requires two changes of railcars in each journey, both at the Central Asia-Kazakhstan borders (from standard to broad gauge) and at the CIS or East European borders (reverse action) which causesdelays and costs. Turkey has 10413 kin of rail track and 60,840 kin of all-season roads. It offers a landbridge operation through the main ports of Istanbul, Mersin, Izmir, and Samsun (Turkish State Railway 1994). Turkey is able to offer some strength in the landbridge sealeg services as it held I A% of the total world ships and 1.2% of the total dwt in 1993 (UNCTAD 1993). The road system of the Kedern landbridge gains from sophisticated and efficient port equipment and managerial skills, but the route suffers from a lack of rail lines (Containerisation International Year Book 1991).

4A. lA Organisational characteristics European transport modes both in operation and ownership have for a long time been encouragedthrough competition to develop integrated modes, with seatransport companies undertaking land transport and vice versa (Hayuth 1987). According to Eller (1992) four co-ordinating bodies are involved in rail/road container transport and terminal operation. across Europe at both domestic and international levels. The most important one, Intercontainer was established in 1967 and functions as a traffic facilitating and coordination ccntre (Whittaker 1975) while each member state is responsible for its own traffic. The other is the International Union of Rail-Road (IRU) which works as a combined company and was established in 1990 with 10 member states to function as an association for operators/carriersof piggy-back trailers and swap bodies (Containerisation International 1995). The organisation of rail terminals as major ground hubs for landbridge and other transit purposes is through two co-ordinating centres (Whittaker 1975). The Trans-Europe-

103

Chapter 4 Express-Merchandises

Landbridges: Concept and Practice (TEEM) is responsible for local domestic European markets, while (TECE) acts for the international flow of containers

Trans-Europe-Container-Express from/to Europe. Different scale and there are

bridge services are offered by shipping companies on a global about 105 Europe-based of which Non-Vessel-Operating Multimodal

Transportation

Operators (NVO-MTOs)

about 70% are located in the UX

(Containerisation International Yearbook 1991).

China, with its large population, and its closed socio-economic system, tends to deploy human rather than mechanical power or more advancedtechnology which causestransport problems (Speece and Kawahara 1994). Therefore, investment in an international landbridge may not bring the same returns as more labour-intensive domestic investment. Poor communication infrastructure and lack of expertise in English and other major trading languages leads to inefficiency, particularly in ports, which are key elements in a landbridge operation (Containerisation International, 1994). According to the Korean Maritime Institute (1992, p. 86): "TCR has lack of know how in managementof facilities, cargo tracking, booking and other parts of transportation services". There are about five thousand Chinese corporations involved in foreign trade of which sixty function as freight forwarders. The two official and largest state-owned forwarders are Sinotrans and (UNCTAD 1989) then China Shipping Company (COSCO) undertaking integrated multimodal joint ventures with Japanesecompanies. Services in different major Ports of China have low quality and long delays with bribing. However, foreign expertise is now being allowed to work in different areasof the transport sector. e.g. container trucking, freight forwarding, etc. (Speeceand Kawahara 1994). The transport services in Turkey are mainly provided and supervised by the ministry of transport and communication, and also through a board within the council of ministers. The seven major ports which mainly handle landbridge services are operated by the Turkish

104

Chapter 4

Landbridges: Concept and Practice

State Railway or TCDD (Turkeiye Comhuriyeti Devlet Demiryollari). They include the three ports of Izmir, Istanbul, and Mersin, which are equipped with gantry cranes and other container facilities (ffiggs 1995). Other minor ports, which are less active in landbridge movements, include Trabzon and Giresun which are run by the Turkish Maritime Organisation (TDI). Turkish landbridges have a strong potential for offering services to surrounding countries, but have some problems with equipment, organisation, and port

in (Higgs 1995).Sometimes, the caseof traffic to Iran, therehasbeena sudden congestion increase prices,preventingdeliveryof shipments Iran (Sanat-e-Haml-o-Naghl 1989). in to
In Israel, the Kedcrn landbridge is affiliated to the Zim Navigation Company while ports are governed by the Ports and Railway Authority, which works in close co-operation with the Kedem landbridge. This tightly co-ordinated system has led to reduced tariffs and the effective return of empty containers by Zini Shipping, adding to the success of this landbridge (Hayuth 1987).

4.4.2

Trans-Siberian

Railway

4.4.2.1Geographicalcharacteristics
The Trans-Siberian Railway (TSR), which originally was constructed during 1892 and 1916 (Strauss 1993). is the oldest and longest transit route in the world in terms of through services (Bonney 1991, Noble and King 1991). In April 1907 it started operating as a landbridge between Japanand Europe (Hayuth 1987) (seeFigure 4-8). It runs through a vast area with a harsh climate far from international waters. According to Damas (1992a) it saves 7000 km distance when compared with the all-sea Suez Canal route to Europe and provides an ideal route to Japanand other Far East countries. The TSR main and sub-branchesprovide services which start at the ports of Vladivostok, Nakhoda and Vostochney on the Far East Coast of Russia, and end at the Azof, Black, Mediterranean, Baltic and CaspianSeaswith more than seventeen port connections. 105

Chapter 4

Landbridges: Concept and Practice

Fligure 4.8 Trans-Siberian rail landbridge routes (unit: krn). Stockholm Baltic Sealeg 715 St Petersbur g)Rotterdam 1871 Bres tr
)4-

2176 1167
r* 3752

1100 1 00

650 Mosco Kurg 517 51 1133

Novosibirsk

Ulan Ude I 2304

Vostochney

Berlin 789 .)

11,70 70 via 872 Gome lv. 1 IA

228

11487 Bator / Y4'6 1076


/

I Chop (Hungary)6F //1


654 /108%

3218 --'UkBamaul

WFXoljzograde \ 883 714 822 Port Guriev


(Kazakhstan)

Odessa Rostov 10 Port Batumi 406 549 492 Yerevan 200 Djulfa to Europe Zurkey via 14 736 blafz* z TT ehran 180 Jq?,Oum 7 7 Port Imam Ic 1086 1163

Gudemesd 557 498 1849 Alma-Ata 311 Baku 0 Caspian Sea leg r Krasnovosk 555 Ashghabad 117 360 Sarakhs Charduz Termez 225 502 Tedzan 343 Chu 115 LIgovov (Pv460 572

TCR to the port of Lianyungang

Black Sea Thillisi in Turke 74 to

fWBishak

Tashkend 354 Samarkand 369 Dushanbeh

Port Abass (Persian Gulf)

I TSR landbridge to: Kobe-Vostochney-Moscow-St. Petersburg-Stockholm 12251.2. Kobe-VostochneyMoscow-Brest-Berlin-Rotterdam 13646.3. Kobe-Vostochney-Moscow-Kiev-Odessa 12412.4. KobeVostochney-Moscow-Minsk-Brest (Poland) 11986.5. Kobe-Vostochney-Moscow-Kiev-Chop 12412. 6. Kobe-Vostochncy-Moscow-VolgogradeRostov-port Batumi 13526.7. Kobe-Vostochney-MoscowAstarakhan-Baku-Djulfa 13853.8. Kobe-Vostochney-Novosbirsk-Ala-Ata-Tashkand-Tedzan-Sarakhs (Iran) 1155 . '5ource: Rail distances are basedon Thomas Cook European ana L)verseas(I Yva), i imeEaDies, oi. JL JLI, v &various pages. Seadistances are basedon: Canet and Renolds (198 1), Reed's Marine Distance Tables, pages 36 & 55.

106

Chapter 4 4.4.2.2 Political characteristics

Landbridges: Concept and Practice

The TSR and the railways of the former USSR worked under a highly integrated political system with quasi-military control (Holt, 1993). The break-up of the USSR in 1991 led to the emergence of a number of new republics and a loss of the tight control of Russia over the operation of TSR branch lines. The number of TSR major ports was reduced from eighteen to seven with disruption caused by new borders. There was also an increase in the total transit times and in tariffs and greater inefficiency in railcar

management (Holt 1993).Tradein different partsof Europeandin CAC countrieswas etc. lost. The reliability of TSR services as was questioned different republicswantedto focus After the break-upof the USSR,the moreon their own freight andpassenger requirements. resultingpolitical conflicts affectedTSR customers such as Iran, due to the closureof its importantrail outlet in Djulfa following the disputebetweenAzerbaijanandArmenia.The TSR is usedextensivelyasa vital partof domesticandstrategic routes(carryingabout80% traffic in 1990)with somelimited outlet for foreigncountries(Holt 1993).Any of domestic new routesor increasein capacitywould be a very long term project, particularly in the in light of thecurrentfinancial andpolitical problems the CIS, andespecially Russia. in 4.4-2.3Technical characteristics The TSR stretchfrom Vostochney Moscowis 9297km of broadgaugerequiringthe to
time-consuming practice of liffing railcars at borders which transfer to another gauge. The TSR benefited greatly from subsidies from the former Soviet Union to enable it to earn hard currency and compete with all-water transport, by providing large size feeder shipping services in the Black Sea and the Baltic Sea (Containerisation International Yearbook 1991) and particularly for large market such as the Far East, South eastAsia and Australia (Nfiller 1978 and Rijn 1981). After the collapse of the USSR in 1991, the fleet was divided between different independent countries, but still the CIS countries hold 12.1% of the total

107

Chapter 4

Landbridges: Concept and Practice

ships in the world and 3.2% of dwt (UNCTAD 1993). This provides considerable support for any landbridge service. According to Fraser (1994), Russia has more than seventeen ports in the Far East region but the TSR is linked only with the three ports of Vladivostok, Nakhoda, and Vostochney. The latter two are the more commercial and containerised ports and carry about 73% of the total trade, and usually there is congestion. Most of the foreign trade along the TSR is containerised, and the imbalance of trade between the Far East and West has resulted in a stock of containers in the West and the return of empty trains to the East (UNCTAD 1991,1992 and 1993). To improve TSR services and overcome old and new problems, the Ministry of the Railways of the CIS, with the assistanceof the other TSR operators, has made the following arrangements: * The introduction of block trains, new flat cars, local and total transit time tables. * The introduction of improved container control and recording systems particularly when trains approachborders. * The improvement of the port of Vostochney in the Far East as an intermodal port (Damas 1992b, Holt 1993).

4.4.2.4 Organisational characteristics The co-ordination of the operation of the TSR landbridge services before the break-up of the USSR was controlled by the Ministry of Foreign Affairs. But from 1991, the TSR has been run by the Ministry of Railways of the CIS, mainly through the two operators SOTRA (Souztransit) and SOYZ VENESH TRANSrr-SVT, (Damas 1992b, Hicks 1994, Containerisation Yearbook 1995). According to Transue and Little (1990), SOTRA in itself is composedof other different independentcomponentswhich co-ordinate the services. In the service organisation of all types of TSR bridges, foreign operators, mainly Japanese, German and American-based companies, are active (Bonney 1991). The organisational characteristics of the TSR are strongly under the influence of the former Soviet Union's

108

Chapter 4

Landbridges: Concept and Practice

centralised system of government. During the USSR period the service quality and reliability of the TSR did not have a good image and according to Bonney (1991) it suffered from delays. The post-Soviet structure in fact worsened the service problems by allowing more freedom of the railway regions in their control of the locomotives, and the effect, along TSR routes, was to causedelays and higher transit costs (Holt 1993). The new Ministry of Railways has very little control over the operation of trains in the given region in the CIS. According to Fraser (1993, p. 102) Russian officials stated that:
"We note that the rail economy is becoming unfit for use, the locomotive pool is slipping owing to a shortage of repair capabilities and that there is a shortage of rolling stock."

According to Holt (1993, p. 76) in the CIS confederation the


"Current organisational structure is not suited to providing the flexible, efficient, and reliable service required in a market economy".

According to Mc Donell (1995, p. 36):


"Institutionally, the rail sector (in the FSU) has been fragmented. The Central Asian railway has been broken up into four national systems in which, defacto, many of the pre-existing operational agreementsand conventions between republics remain in place; but the delay, costs and uncertainties resulting from the new arrangementsare pervasive".

UNCTAD (1993) claims that the TSR output declined during 1993 and pointed out the four following reasons: An increasein the number of TSR operators A significant increasein tariffs Port congestion, and The intervention of state authorities.

4.4.3 North American landbridges


Ile distinctive industrial and geographical conditions together with a liberal trading

enviromnent, as well as the high quality road and rail networks and rolling stock, has led to well-developed landbridge operations, particularly the application of the minibridge

109

Chapter 4

Landbridges: Conceptand Practice of the double stack train

concept, in the USA and Canada (Hayuth 1987). The introduction

since the early 1980s has been particularly successful (Hayuth 1992). Further development by raising tunnel heights and reducing the tare weight of railcars, has created an extensive range of minibridge services with the Far East, Europe, Africa and South Asia.

The continent of America makes a long north-south barrier between Europe and Asia, located between the Pacific and Atlantic Oceans,and is in an ideal situation with extensive co-ordinated rail and road networks to function as a landbridge between two major trading centres (Containerisation International 1991). The USA, Canada and Mexico are three American countries each with one active landbridge (Hayuth 1987) (see Figure 4.9 for two examples through New York-Houston). They compete with the all-water intercontinental passage through the Panama Canal and are about 2000 km shorter in distance than the Canal (Padelford and Gibbs 1975). They also function as competitors with the TSR landbridge, while at the same time competing with each other. 4.3.3.1 Geographical characteristics Canada is about 5000 km in width, with a few rail bridges between the east and west coasts mainly through the three ports of Halifax, Montreal and Saint John (McCalla 1994), but also through Toronto in the east and Vancouver in the west. The deep penetration of the five lakes of Ontario, Erie, Huron, Michigan and Superior into the Canadian heartland (as well as the USA) has created an ideal location for more sea leg landbridge services from Europe through Canada by using the ports of Montreal, Quebec, etc. and shortening the overland rail bridge distanceconsiderably. The Canadian landbridge has a distance advantage over USA landbridges as the port of Halifax is the closest point with a distance of 2705 nautical miles to Europe (Rotterdam) (Figure 4-9). Also some major cities of the USA (e.g. Chicago and Detroit) are part of the hinterland of the ports of Halifax Saint John or Montreal (Whittaker 1975). ,

110

eb

-3 cr
oil
rD

:i

:p

2!
CD

kO m

10
wo

Ein.
G9 z, th

Irl,

RI
Ato

I /' -T
a

\\

III

Chapter 4

Landbridges:

Concept and Practice

The USA, with its location between Canada and Mexico, has significantly developed the

landbridge concept in all aspects.The high industrialisation, population, production and and trade with Europeand Far East regions,as well as the vast areasof consumptionrates for domesticand maritime geography, amongthe important reasons the developmentof are
types of landbridges functioning at national, regional and international many complex levels. The shortest raH distance for a USA landbridge is 4471 km between Atlantic City

on the east coast and San Francisco on the west (Figure 4.9 shows United States intercontinental landbridges).USA shippersand carriers use the Mexican landbridge for trade betweenthe Gulf coast region and Mexico as it can offer shorterjourney distances than the USA landbridge to locations such as Miami (Ashar 1995). The geographical feature of Mexico, with only 333 krn between the Pacific and Atlantic Oceans,forms a and unique rail and road bridge betweenthe ports of Coatzacoalos Salina Cruze (Mahoney
1985 and Hayuth 1987). It has about a 1405 krn distance advantage over the Panama Canal

which is 80 km in length (seeFigure 4.9). 4.4.3.2Political characteristics Canadahas a small population (about26.5 million in 1990) for such a vast area,and is one of the sevenmost important industrial countries, able to provide a strong capital and technicalbasefor the provision of effective landbridgefacilities. Landbridges,and particularly mini-and micro-bridge services,have increasedin the region as a result of transport economiesand also becauseof extensivecompetition with all-sea Canal (Padelfordand Gibbs 1975). transportthrough the Panama The transatlantic and transpacific minibridges extensively cover the trade between the USA, Canada,Europe,and Far East as about 63% of the total USA population is living in the 19 maritime states.Sevenwest coaststateswith high industrialisationhave 26% of the total population (65.3 million), as well as high consumptionand production. For Far East

112

Chapter 4

Landbridges:

Concept and Practice

countries there are fourteen statesin the east and Gulf coastswith a high population of about 93.3 million (37% of total in 1990), with even more industrial and economic activities and foreign tradethan the west coast. Of all the intermodal overlandservicesin the world, perhapsnoneare more extensivethan
those offered in the USA (Hayuth, 1987). This outcome could not have been achieved without a major change in the USA to deregulation in the early 1980s. Previously, railways

had beenslow to adoptcontainerisation intermodality.The radical changein regulation and led to new rolling stock, dedicatedcross country and double-stackcontainer trains and, most importantly, increased co-operation with seaports shippinglines. and Mexico as a developingcountry with a high population of 88.3 million in 1990,has social, economic and infrastructure differences compared with the other two major industrial North American countries, and is in heavy debt to the USA. As a result, the Mexican landbridge has less investmentopportunitiesand developments. has a relative distance It Canalbut, accordingto Mahony (1985), in spite of advantage 1405km over the Panama of the large amount of investmentin equipmentand constructionof intermodal ports and rail networks,it hasnot beenable to attracta constantand significant traffic from either the Far East or Europe. The North American Free Trade Agreement(NAFTA) among the USA, Canada,and Mexico is intended for interregional developmentand co-operation of the memberstates,including minibridge operations.According to VIA International (1993, p. 9):
"rhe treaty calls for public telecommunications transportnetworks and and and servicesto be availableon reasonable non-discriminatoryterms and for firms or individuals who will use thosenetworksfor the conductionof their business ... ... .

NAFrA, in fact, aims towards the creation of a single hinterland among North American countries(Hayuth, 1992)by using all potential ports in the region commonly.The Mexican

113

Chapter 4

Landbridges: Concept and Practice

landbridge, through more investment given under NAFIA

for transport technology and

in management, shouldmakeit a strongeralternativeto the USA and Canada future. 4.3.3.3.Technical characteristics The first Northern American landbridgecame into existencein 1972through Canadian rail piggyback services (Hayuth 1987). It is a very effective landbridge and extensively servesUSA inland marketsaccordingto McCalla (1994). Three factorscontributesto this: the deep water at Halifax, the opening of new facilities, and the efficient handling of containersboth at docksideandby CN (CanadianNational) Rail. To make the port of Halifax more competitive and to strengthen domesticand international services,CN has developeddouble stack servicesand Electronic Data Interchange(EDI) (UNCTAD 1991). Another Canadianmain rail and landbridgeoperator,CanadianPacific (CP), has startedto raise rail tunnel heights to permit the passage double stack trains of across Canada(Hicks 1994). According to Slack (1994) the dominanceof the northern ports of Canadaand particularly Montreal in serving USA midpoints is based upon the inability of US railroads to provide a comparableCOFC (ContainerOn Flat Cars) service, although the TOFC (Trailer On Flat Cars) concern has developedsignificantly since the 1950s(Slack 1994). The American minibridges have received extensiveinvestmentin infrastructure,terminal Massive investmenthas. beenput into hub centressuch as systemsand double stack-trains. Tacoma and Oakland as a means to compete effectively with all-water transport (ContainerisationInternational 1994). The existence of the extensive rail tracks mainly with double-stackcapability has also played an important role in the developmentof USA landbridges.Someports suchas Oaklandare servedby threerailroad companiesallowing a numberof options for eastbound inland trade. or

114

Chapter 4

Landbridges:

Concept and Practice

Rail and road networks have been developed within Mexico and particularly in the

maritime provinces of the Pacific and Gulf coasts in south-north directions with seven crossings at the northern border with the USA. Among the six main ports of Mexico, Coatzacoalcos Salina Cruz are landbridgeports and have the most important container and infrastructure and facilities (ContainerisationInternational Yearbook 1991).The NAFTA agreement has emphasisedthe provision of technical assistancefor the operation of
Mexican port facilities to develop poor existing services including passenger services

Timetables 1995). According to VIA International (1993, p. 9) (Thomas Cook-Overseas Mexico will: under this agreement
"Immediately allow 100per cent Canadianand U.S.A investmentin, ports, and operationof, facilities suchas cranes,piers, terminals and stevedoringcompaniesfor enterprises that port handletheir own cargo.For enterpriseshandling other companies'cargo, 100percent Canadianand U.S. ownershipwill be allowed after screeningby the Mexican Foreign InvestmentCommission.T'heU.S. and Canadawill continueto permit full Mexican participation in theseactivities." As a result, the financial and technical assistance of NAFrA has had a positive influence

the Mexican minibridge and the operationof internationalcontainerrail movementshas on shown 16% growth in 1992 and 18% in the first half of 1993 (UNCTAD 1993).Another important factor in the North American minibridge developmentis that major shipping carriers own and operate their own terminals. As a result of the development of minibridges in many cases,they have suspended reducedall sea carriage through the or PanamaCanal,and have deployedtheir fleets in shortersealegs offering greaterfrequency to both west and east coast trade with the Far East, West Africa and Europe (Containerisation Intemational 1994).

115

Chapter 4

Landbridges:

Concept and Practice

4.4.3.4 Organisational characteristics Canadian National Railways (CN) and Canadian Pacific (CP) are the authorities in

chargeof landbridgeoperationsin Canada.There are different types of port ownershipin Canadabut major ports on the east,west and deepinland ports are controlled by different national and municipal boards (Hicks 1994, Ircha 1997). According to Slack (1994), the intennodal serviceoffered by CN and CP asrailway operators, resultedin the attraction has of more minibridge trade from European markets to Halifax and Montreal for inland terminalsin CanadaandUSA comparedwith USA eastcoastPorts. The organisationof the North American landbridgesis governedby independententities utilising ports and railways and also active in seatransport.In terms of port operationsthe landbridge organisationin the USA is the same as in Canadaand developedEuropean countries.However, rail networks are operatedthrough nineteendifferent companieswith their own rail track, intermodal equipment, trailers, electronic data exchangesystems, terminals and storage,rolling stock, etc. (Hicks 1994 and UNCTAD 1989). Therefore,
different shipping companies and Non-vessel operating common carriers (NVOCCs) have

arrangements agreements and with different rail companiesand many even have their own trucking or rolling stock along their routes in the USA. (ContainerisationInternational Yearbook 1991). The microbridge concept involves fewer modes than minibridge or landbridge services,and shipping companiesand IVOCCs are able to offer regular and direct services.Minibridge and landbridge servicesare mainly provided on a round trip basisalong certainports of call (ContainerisationInternationalYearbook 1991). The organisation in charge of the Mexican landbridge is Servicio Multimodal Transistimico (Semultra) while rail services are operated by the Mexican National Railways (FerrocarrilesNacionalesde Mexico-FNM). In contrastto the USA the Mexican
service is not offered by competing organisations.

116

Chapter 4

Landbridges: Concept and Practice

4.5 Conclusions
This chapter has introduced the general concept of a landbridge and describedthe specific types of landbridge(minibridge, microbridge, and airbridge). It has examinedthe key influences on the operation of landbridges, which are geographical, political, It technological,and organisational. has appliedtheseinfluencesto major landbridgezones in the world. This approachwill now be applied to the Iranian Sea-Landbridge(ISLB) which is the main topic of the research.

117

Chapter 5

Iranian Sea-Landbridge

5. The Iranian Sea-landbridge


5.1 Introduction
The purposeof this chapter is to review the potential for landbridge development through Iran to and from the Central Asian and Caucasus countries(CAC). To identify to such potential it is necessary take into accountboth supply and demandfactors, the
foreign trade outlets of CAC countries and to compare the Iran Sea-landbridge with

of other competitors.The physical and economic characteristics Iran were discussedin detail in chaptertwo, but it is necessary identify certain more detailed factors which to have particular relevance to the landbridge concept both for Iran and the CAC republics.

5.2 Reasons for developing a landbridge system


Iran and the CAC countrieshave a commoninterestin developinga landbridgesystem. The collapse of the fonner Soviet Union as a supeipower and its disaggregationinto different nations, is likely to re-establishold links betweenIran and the CAC republics (Herzig 1995,The Economist 1997).Iran had a long cultural, territorial, religious, and social relationship with the CAC countrieswhich was broken during conflicts between Iran and Tsarist Russiaduring the 19th century. Therefore,greaterco-operationis seen as one meansto eliminate any future threat of the re-establishment the former Soviet of
Union. Another reason is the economic locational advantage of Iran which was

restrictedduring the period of the former Soviet Union. The landlockedCAC countries may, in particular, see accessthrough Iran to open waters as a strengtheningof their independence from Russia. The unique maritime location of Iran, following the collapse of the former Soviet
Union, has created a new role for Iran as a transit country, and potential for the

118

Chapter 5

Iranian Sea-Landbridge

development of multi-legged integrated transport bridges between the Persian Gulf and the Caspian, Black, Baltic, Mediterranean, and Red Seas for three different continents,

and also through the northern land border crossings(Ogutcu 1995).For the three CAC countries on the west, east and north-east of the Caspian Sea (Azerbaijan, Turkmenistan,and Kasakhstan respectively)thesetransportbridgeswould serveforeign origins linked by seatransport routes to the two southernports of Imam and Abass in
the Persian Gulf and by using the northward Iranian road/rail routes to the Iranian port

of Anzali on the south of the CaspianSeafollowed by a secondseatransport(Tarjoman 1995,Stone 1993).

5.2.1 Geographical characteristics


The geopolitical characteristics Iran are a key elementin the connectionof southof west Asia and the Middle East to other parts of Asia and Europe (Haglund 1986).Iran the Central Asian and Caucasian republics form nine countriesoccupying 13.24% and of the total land areaof Asia (Upshalet al. 1994). The landbridgesprovided by Iran betweenthe CaspianSeaand the other international waters of the PersianGulf and the Indian Oceancould be economicfor trade with the Southand Far East,African countriesand also the PersianGulf region. The selectionof the three Iranian sea-landbridge transit ports of Abassand Imam in the south and Anzali in the north is basedon the maritime and cargo operationalpotentials of theseports as the largest regional ports in operation and capacity. Details of these ports and a justification for their selection were given in chapter two. Each port has for special advantages eachregion of the CAC countriesin terms of proximity (Figure 5.1 shows the land and sea leg distancesbetweenIranian ports and border crossings, and CAC countries).The port of Abassin the mouth of the Straitsof Hormuz is closer

119

,x

>o

2.

i %

b ' . w

kl

jA 10

CM

. a

12 .g .
cr.

y
S
OCI 0
cr N

(Z

Z.

0'

I.
a

120

Chapter 5

Iranian Sea-Landbridge

to central Asian countries while Imam in the north-east of the Persian Gulf

is much

closer to the Caucasus region. Port Anzali in the north is the biggest and most used port

Iran in the south-westof the CaspianSeaand in fact is alreadyplannedto becomea of landbridge and minibridge port for the states bordering the Caspian Sea (Ports DevelopmentInternational 1993). Iranian minibridges are those which originate from inland locations outside of the Iranian and the CAC region using the ports of Imam or Abass,and then go directly by Iranian road or rail ending in Anzali, or at the Iranian border crossingsor directly into
CAC demand centres.

Iran has nine active border crossing terminals directly linked to three of the CAC countries as shown in Figure 5.1 Four of these terminals, are connected with Turkmenistanin the north-eastof Iran in the provinces of Khorasanand Mazandaran.
Three of these (Sarakhs, Lotfabad, and Bajgiran) are suitable located to serve countries

in different parts of the four other Central Asian countries.The closest,and one of the
busiest, is Bajgiran which is 44 km from Ashghabad, the capital of Turkmenistan.

Given its links to road and rail networks,Sarakhsin the extremenorth-eastis the focal crossing ten-ninal for shipments to the other four Central Asian republics (SanateHamlo-Naghl 1993a). Lotfabad is linked to the Turkmenistan railway, but road is the only mode of access from Iran (Payarn Darya 1994). The fourth and last crossing in this region is IncheBoroon which serves the south-west areas of Turkmenistan and is close to the Caspian

Sea.Among the other five land border crossingsin the north-westservingthe Caucasus republics, Astara and Pileh Savar are directly linked with the main part of Azerbaijan, while two others (Djulfa and Pol Dasht) servethe Azerbaijan autonomousrepublic of Nakhjavan and one (Noor Dooz) is the border crossing for Armenia. Djulfa, which is

121

Chapter 5

Iranian Sea-Landbridge

III I

to the Trans-Iranian Azerbaijan railway, together with Astara are the most connected

important border crossingsin this region. Of these five border crossingstwo (Pol Dasht and Dejulfa) are respectivelyclosestto Georgia(Tbilisi) and Armenia (Yerevan). The distances the crossingsfrom the southernIranian ports are shownin Table 5.1. of
Table 5.1 Road and rail distances (km) between southern ports of Iran and northern hnrflpr ernednae eprvinty VAr rmintrip. qBajgiran Inche Borun Port Anzali Lotfabad Sarakhs Border crossings , 2014 1947 1513 1265 Port Imam 2110(2014) 1607 1811 1749 1674 Port Abass 1579(2656) Djulfa PolDasht Noor Dooz Pileh Savar Astara Border crossing 1475 1376(1810). 1483 1689, Port Imam 1414 , 2197 1 2098(2452)1 2205J Port Abass 18981 21731
Source: MRT Road Distances Guide book, 1983. Rail distances are shown in bracket based on 71bomas Cook (1995), March-April, various pages.

The major obstaclesto the extension of the presentrail networks are the Zagros and Alborz mountain ranges running north-west to north-east, and to the south-west respectively.A 10,000km rail project of which about 2500 km is under construction, will increasethe potential of Trans-IranianRailways and will by-passTehran and make the port of Abass about 700 km closer (at present 2656 km) to the Central Asian region also hasplanneddirect links countries.The port Imam rail route to the Caucasus from Qum to Ghazvin and Zanjan (seeFigure 5.1).

5.2.2 Political characteristics


Iran has left the critical war time complications and accordingto Roberts(1996, p. 10) as a strong alternativeamongthree countries for the Caspianpipeline "is arguably the most politically stable main export route". Since the independence the CAC of to countries,Iran has establishedan open policy focusing on assistance theserepublics to strengthen and support their independence through different direct bilateral negotiations and even economic, trade and transport agreements.The minibridge operationwith the CAC countriesis influenced by the political relationshipswith these

122

Chapter 5

Iranian Sea-Landbridge

countries, particularly with Turkmenistan, Azerbaijan, and Armenia which are directly

to connected the Iranian network (Herzig 1995). Apart from the conflict betweenArmenia and Azerbaijan, and the internal problemsof Tadjikestanthere is not a major issue which restricts trade. Six of the CAC countries agreedin 1992to becomemembersof the ECO (EconomicCo-operationOrganisation) Ahmad 1994). The ECO treaty was treaty with Iran, Turkey and Pakistan(Shernshad establishedin 1984 (The Middle East and North Africa 1994) and now functions as a significant geopolitical link between old and new parties, and in areasof economic activity such as transport, communication,and financing, banking, and energy.At its annualmeetingin 1992in Tehran,membersagreedto a ten per cent cut in trade tariffs and eventually to remove all tariff and non-tariff barriers. In January 1993 the High Council of ECO Ministers approved a ten year development programme in
communications and transport for the member countries (Ghaem Maghami 1994). They

placed a lot of emphasis on the Mashad-Sarakhsrailway for the Central Asian countries,the Kerman-Zahedan railway to link the Trans-AsianRailway (TAR) in Iran, maximisation of the present inftastructure and developmentof multimodal transport within memberstates,particularly CAC countries.Iran agreedto the Azerbaijanrailway being connectedto the autonomousRepublic of Nakhjavan through its land, while a bridge is underconstructionfor proper access Armenia over the river Araks. to The Transport International de Marchandises Route (TIR Camets)convention (the par CustomsConventionon the InternationalTransportof Goodsby Road),of which Iran's Customsis a member, is an appropriatetool for sealedtransit shipmentsof the CAC countries(vehiclesor containers)through Iran, but most of thesecountriesstill havenot adopted the convention (Sanat-e- Haml-o-Naghle 1994a). Consequentlythe present practiceis solely under the administrationof the Iranian Customsauthorities.Iran has a

123

( )p

Chapter 5

Iranian Sea-Landbridge

road freight agreement with more than thirteen European and Asian countries, while for

another eleven European countries, including the CIS and the CAC republics after leadingto improved relations. 1993,therehavebeensomemeasures The other implication for CAC countriesis the movementof their standardtrucks along Iranian roads in accordance with the International Road Transport Union (IRU) convention.This conventioncoversphysicaland technical specificationsof road freight vehicles and is in fact a pre-requisitefor the TIR convention.The countrieswhich were formerly part of the Soviet Union needto apply and adoptthis conventionindividually.

5.2.3 Technical characteristics


A total of 16,445rail cars pulled by 436 locomotives are in operationover TransIranian Railways (Sanat-e-Haml-o-Naghl,1993b).This mode has carried 17%, 20%,
and 12% of total foreign trade during 1990,199 1, and 1992 respectively. The rail

network has poor connectivity basedon a spoke format with the hub in Tehran and a total length of about 5800 km.in 1994after the completion of the railway to the port of Abass.Rail freight previously supplied about 70% of the total freight traffic, and road (Shishe-chiba1994).After the completion of transportwas not used for long distances. the 560 km rail line in the south eastof Iran, the presentIranian rail landbridgeto the CAC countrieswill be a sub-branch the Trans-AsianRailway (TAR) (World Atlas of of Railways 1978). This will improve the connection of the Indian sub-continentwith Europe, the CIS and the CAC countries through Iran as a potential alternative rail landbridgefor TSR. Both southern landbridge ports are connected to the Trans-Iranian Railway and to consequently all CAC countriesthrough two border crossingsat Sarakhsand Djulfa. As statedbefore,the hub of the presentrailway network of Iran is Tehranand both links from the south-west(Port Imam) and the south (Port Abass) first passthrough the hub 124

Chapter 5

Iranian Sea-Landbridge

and then continue to the CAC countries. In spite of the recent restructuring in 1987

there are still operational and technical implications. As mentionedin chapter two, a single line system,and also the hub and spokeformat of the rail network havemadethe present system slow for freight trains and also much longer than road transport, in particular those services required by the five Central Asian countries. As a pure
government entity, the railway cannot respond to the short term requirements of a competitive transit fteight system while there are large technical insufficiencies in terms

1995c). (Sanate-Han-d-o-Naghl of integratedtransportsystems Iran has an extensiveroad network of 167,000km (SCI 1993)linking all capital cities, towns, and small villages. Of these,105,000km are main and secondary roadsof which 80,000 km are asphalt(Torkan 1995).The quality of main and secondary roads differs slightly where the purposeand traffic volumes change.The following four categories
are identified by MRT (1983). Corridors related to the Iranian Sea-landbridge study fall

into the first and third category(seeFigure 5.2). mainly


Foreign trade corridors

Thesearethe most important fifteen roadsconnectingthe sevenmajor ports and the key bordercrossingterminalswith GreaterTehranor with eachother.Their composition differs but includesmain roadsand free- and expressways with four or more lanesor wide ordinary roadswith 11-13.3metreswidth all of which are asphalt.
* Tehran-capital city roads

Theseare main roadswhich connectTehranto all 24 capital cities and other main cities with economicimportance,and accountfor eighteenroute origins from Tehran.

125

C-1
94 1.0

Its
N cr, cn

00
110
.I

0-1

Gn.

woorr

i
01

00
a
9.

J. *
CA

ff

cr

Iw
W rs

126

Chapter 5

Iranian Sea-Landbridge

Inter-provincial

roads

Those are roads which link provinces and main cities together and account for 37 nodes.
* Intra-provincial roads

The 24 provincesof the country accountfor 118intra-provincial roads.Iran is renewing its major ports and in particular the ports of Imarn and Abass to serve CAC countries (Ogutcu 1995). The corridor from the port of Imam to Armenia and the Autonomous Republic of Nakhjavan (Azerbaijan Republic) is a main national and inter-provincial trunk road, passing through six Western provinces of Khosestan, Lorestan, Eam, Kermanshahan, Kurdestan,and East Azerbaijan. It has a main length of 1226 km to Tabriz and three sub-branchesto Djulfa, Poldasht, and Noor-Dooz. This corridor,
particularly after the first 433 Ian, has lighter traffic but is rather mountainous. The

secondcorridor from the port of Imam endsat Anzali, Astara, and Pileh Savar,and has a main route of 898 km with heavy traffic up to Boein, then dividing into two routes. One sub-routeruns north and servesthe port of Anzali in the CaspianSeaand then as a secondroute along the coastto Astara.The third sub-routegoesnorth-westto the Pileh Savarborder crossingto the Azerbaijanrepublic via the two capital cities of Zanjan and Ardebil. For the corridors from the port of Imam to the Central Asian border crossings,Tehranis the main common route for all four Customsborder terminals as shown in Figure 5.2. One sub-branchto the north along the Caspiancoast ends at the Inche-Borunterminal in the province of Mazandaranin the south west of the Turkmenistan. It has a good quality structure,althoughabout200 km from Tehranit becomes mountainous.Another sub-route from Tehran along the northern desert borders goes to Sabzvar in the Khorasanprovince and then into two other routes. A secondaryroad connectsthe two

127

Chapter 5

Iranian Sea-Landbridge

border crossings of Bajgiran and Lotfabad to the main road in Sabzvar. The more direct

and better route from Sabzvar,after passingMashadthe capital city, endsat Sarakhsas the gatewayto the Central Asian countries.Thesetwo corridors from the port of Imam heavy traffic due to Gorganin the north and to Mashadin the north-easthave seasonal to foreign and domestictrade,leisure and holy places,etc.. The port of Abass has road advantages over the port of Imarn for the Central Asian countries.Its road runs first to the city of Sirjan as the biggestinland distribution centre in the country and the junction for Central Asian and Caucasus countries. About 533 km. of the route links this port to north-eastterminals for the Central Asian republics, and crossesthe desert which is a main inter-provincial road. Becausethe copper and other mineral industries are along it, it is one of the most crowded roads over the first 400 km (MRT 1993)while from Kerman the traffic is considerablylighter. Its main leg
ends at Mashad after 1211 km with three prongs towards the three major border

Lotfabad, and Bajgiran. crossingsof Sarakhs, The route from port Abassto the Caucasus region after leaving Sidan is joined with the port of Imam traffic at Qum (1174 km) as a distribution point for the Inch Borun border crossing in the north-east along the Caspian Sea and, for the Caucasusand the port Anzali region as shownin Figures 5.2 and 5.3. A road fleet of about 147,000 heavy vehicles (13.5-22 tonnes) supports the present flows betweenports and border crossing terminals with a modal share of 77.7% for 1993(Atrchian 1995). As discussedin chapter two, Iran has a reasonablyhigh sea-goingcapacity (1.22% of world total dwt and 0.55% of total vessels in 1993) (UNCTAD 1993). After the collapse of the Soviet Union in 1992, the Islamic Republic of Iran Shipping Line

128

Chapter 5

Iranian Sea-Landbridge

(IRISL) extended its operations into the Caspian Sea. The growing Caspian Sea fleet

consistsof threecargo shipsand one passenger ship of 10885dwt. Figure 5.3 Different Iranian road landbridge and minibridge alternatives to/from
UAU countries (unit: km). 428 (Black Sea) Port Batumi Tbilisi 46 Yerevan Djul 273 48 No 188 137 Tab 452 155 Sanandej 38 315 Pol Dokhtar Sirjan 42J 560+ Port Imam (Persian Guy) 112145T Legend: o Road leg Kobe (E) 897 Busan 3 PIO ortAbass (Persian Gu1j) 90 E) Sydnay Arak Dooz Anzali 236 265 Boein Bajgiran 40 aveh Tehran K, Z ,"Quchan 85 56 19 133 Qum 88 Sabzvar 22 1084 Mashad Sarakhs 172 185 553 Bishak Baku 340 30 Astara (CaspianSea) 149 430 Krasnovosk574 (D 4 61otfa 08 I Mary 156 Ashghabad '7 6 Tashkand I 224 Dushanbeh Z113 101 Samarkand 639 Alma-Ata

210 Tedzan

Sources:Distancesare basedon: 1. MRT(1983), Handbookof the Iranian Road. 2. Different Persianand English road Atlases and mapsof Iran: map of Iran, Gita Shenasi,No. 165, Published in 1989,Tehran. Auto Atlas of Iran, Gita Shenasi, 213, Tehran. For CAC countries Map of the Former USSR, Verlagsellchaftmbh, Frankfurt/main.Germany. World Map: Iran. Publishedin Germanyin 1994,GeoCentrej,RV ReiseundVerkehrsverlag,Berlin 1993/94.

5.2.4 Organisational characteristics


The Council of Transportation Co-ordination is the highest organisationalbody affecting the transport system of Iran, while for all prices and tariffs the Ministerial Council of Economicsis also involved. Sea ports are governedby the Ports and Shipping Organisation(PSO) as a subsidiary
ministry to the MRT, based in Tehran and operated by its representative port authorities

129

Chapter 5

Iranian Sea-Landbridge

in the different southern and northern ports. The private sector is involved in

stevedoringbut is influenced by the PSO. The national shipping company (IRISL) is also an independent public entity underthe Ministry of Commerce. in The rail industriesin Iran which were restructured 1988are government-owned. The operationand maintenance rail (both infrastructureand rolling stock) are run by the of I.R. Railway companybut constructionof new rail projectsis underdirect MRT control and rail manufacturingis private, althoughmainly sharedwith the railway company.For rail to function effectively there is a need for more direct rail lines to both regions of the CAC countries, an increasein the number and composition of rail cars and operating locomotives,and a fasterservice.
Freight forwarders and road carriers in Iran are based mainly in the private sector although there are some government companies acting as domestic or international

carriers, also carrying a part of their own foreign trade. Of 2340 passenger freight and companies in 1992,1355 are involved in freight of which only 220 function as internationalcarriers under the "Syndicateof InternationalTransportationCompanies" (Sanat-e-Haml-o- Naghl, 1989). Shipping servicesand agenciesare privately owned. The IRISL, as the national shipping line, operatesits own and charteredships which constitute most of the Iranian maritime traffic, but there are about 52 shipping agents engagedin ship owning and forwarding servicesof a maritime trade nature (Sanat-eHaml-o- Naghl 1993c). Since rail networks at presentcannot directly influence Iranian landbridge flows to the main ports in the CaspianSea (seeFigure 5.4), the road freight industry must play an importantrole as an overlandmode.The road freight industry,mainly privately owned,

130

Chapter 5

Iranian Sea-Landbridge

works independently, or for carriers issuing their own bill of lading. The last major link

in the chain of the Iranian landbridgeconceptis the Customsadministration.All border crossingsare supervisedand operatedby Iran's Customsas an affiliate to the Ministry of Financeand Economics. The CustomsAuthority of Iran are signatoriesto the TIR convention.Therefore, within Iranian territories and gatewaysthe transit shipmentsof foreign countries are under an international regime with MRT following IRU regulations.

5.3 Identification of the CAC countries' foreign trade routes ISLB competitors and
For the Caucasus to countriesthe access free internationalwatersis much easierthan for the Central Asian republics as they are closer to the Black and MediterraneanSeas and to the port of Imam in the PersianGulf. Apart from Iran, the corridors with road /rail modes and also of a landbridge nature are road/rail Georgia, the Russian Federation,the Caspian Sea-bridge,Turkish-Iran bridge, Pakistan, India, China, and finally TSR. The bridges most likely to have an impact on the ISLB project have the following geographical characteristics: * Euro-Georgian and Turkish landbridges, Caspian Sea bridge, India and Pakistani landbridges,and Far Eastlandbridges.

5.3.1 European markets via Georgian and Turkish ports


The Caucasus region is much closer to Europeanmarketsas it is near the Black and MediterraneanSeasdirectly via Georgia and Turkey or through the Iran-Turkey roadlandbridgeport of Mersin. rail bridge for the southernTurkish Mediterranean The Georgianports of Batumi and Sukhomi on the Black seacoast (if current disputes are settled)will provide all the other CAC countrieswith servicesfrom Europe,Africa 131

Cb cr

, lzk

t2
a

l'

f e. Z9 GO2 bi

rA

jg

0%

<

Ici

: 1-,
00

tA

Eg*

cm

94

03

CD

%0

5. t02 W..

a'.

i ..

tA

132

Chapter 5

Iranian Sea-Landbridge

and America. Armenia (Yerevan) and Azerbaijan (Baku) via Northern Tbilisi routes

also through Yerevan can be servedby the port of Batumi directly for European and traffic as shown in Figures5.5 and 5.6 and Table 5.2 . The five Central Asian countries have accessto Batumi as the nearestport in the Black Sea through the two may and alternativesof the CaspianSea-bridge the rail androad route to Batumi.
Table 5.2 Caspian Seaeast-westrail-bridge between CAC capitals and port of Batumi in Black Sea (km). CAC countri s capitals Th5ii7r Yerevan Baku Dush- Ashgh- Tash- Bishak Alma ta abad kand anbeh 955 3216 1760 2843 3575 3841 780 4061 Distance from port Batumi
Source:Sameas table 5.1

5.3.1.1 Caspian Sea-bridge There is a ferry system operating between the port of Baku in the west and Krasnovosk in Turkmenistan in the east with a total sea leg of about 250 km. The Central Asian countriesto Batumi via the CaspianSea-bridgeis closestcapital city of Ashghabad and the furthestAlma Ata (seeTable 5.2). 5.3.1.2Rail and road route to Batumi The second route is an overland rail (also road) route which connects Batumi through Tbilisi, and along the north west Caspiancoast via Baku, Astarakhan,Guriev (Atycrou), Makat Urgench and Chardzhou,to all five Central Asian countries. The longest distanceas shown in Table 5.3 is to Alma Ata and the shortestis Dushanbeh. The sametable shows that betweenthe sameorigin and destination the Trans-Iranian Railway via Yerevan,Djulfa andTehranis considerablyshorter.
Table 5.3 Rail-bridge between CAC capitals and port of Batumi In Black Distance from port Batumi through: CAC countries capitals Baku Dush- AshghTbilisi Yerevan anbeh abad CAC rail networks 780 955 5220 4827 406 Not Not 4521 Not 3408 Trans-Iranian Railwayl 1 applicableapp 1 licable applicable , Sea (km).

Tash- Bishak Alma kand Ata 5579 5579 5845 4148 4880 5146

Source:Rail distancesbasedon ThomasCook (1995), March-April, various pages.1. Rail bridge through Trans-IranianRailway via Djulfa- Tehran-Mashad, Sarakhsto CentralAsia.

133

Chapter 5

Iranian Sea-Landbridge

Road distances from Central Asia to the port of Batumi are considerably shorter than

there are two alternatives.The first route first runs north and then along the rail and Caspian Sea to Tbilisi and Batumi. The secondroute is through Iran via Batumi to Yerevan, Djulfa, Tehran, Sarakhs.This route is also considerablycloser than all the and northernroadsvia the north Caucasus Caspiancoastsasshown in Table 5.4.
The Turkish corridor can also serve as a direct minibridge for the Caucasus region,

for Central Asia there are two alternativesusing the Caspian Sea bridge and while Iranian rail/road modes.The route using rail/road modes to Baku creating anothersea leg over the CaspianSeato the port of Krasnovoskto reach Central Asia is shown in Figure 5.5 andTable 5.5. The secondalternativeis to use the Iranian road/ rail links via the Bazarganand Razi land border crossingsin the west and Sarakhsin the north-east ofIran.
Table 5.4 Road-bridge between CAC capitals and port of Batumi in Black Sea and St Peters burg in Baltic (km) CAC countries capitals CAC countries Baku Dush- Ashgh- Tash- Bishak Alma TbUisi Yerevan capitals kand Ata anbeh abad 362i 4222 4456 4080 3433 Batumi 981 428 467 3523 4099 4329 3605 2380 Batumi and Iran _iI59 4738 5001 3869 4006 3807 3793 St Petersburg via 3533 Moscow- harduz II

Source: Same as table 5.1.1. Road bridge through Trans-Iranian Roads via Yerevan, L)julta-; I'abnz, Tehran-Mashad (Sabzvar, Quchan, Ashghabad) Sarakhs to other Central Asian countries.

Table 5.5 Rail-bridge distances between landbridge port of Mersin in East Mediterranean coastsof Turkey, and CAC countries using Caspian Sea landbridge and Iranian rail (km). CAC countries capitals Baku Dush- Ashgh- Tash. Bishak Alms Tbilisi Yerevan ta 1 anbeh abad kand 1484 1976 4580 2781 4207 4939 5205 Caspian Sealandbridge 1550 3634 4374 5106 5372 Trans-Iranian railway 4747
Source:Sameas table 5.1

134

>

Z:

ri. A

0%

I co

Gn

er ..

0,

Iti

CA

"0

2.

C>

%D

;3
be

eti

er

CD

bi

c02

I
E D

22

135

Chapter 5

Iranian Sea-Landbridge

5.3.2 Indian Ocean and African markets via Pakistan and India

Two routes, either consisting of all-road or combined rail-road, exist betweenthe port of Karachi in Pakistanand CentralAsia throughAfghanistanas shown in Table 5.6 and Figure 5.6. The combined rail-road route has a common interface in Peshawarin Pakistan. From Peshawarthere are two alternatives: one road link is connected to Dushanbehthrough Kabul to Termez where again there are rail links and warehousing facilities availableto CentralAsian Rail networks.The secondchoice on the sameroute is to use the road at Termez for east Central Asian areas.Road conditions in Pakistan are relatively good but connectionsin Afghanistan are poor, unsafe and mountainous and require frequenttenminalhandling.
Table 5.6 Road and rail bridge distances between CAC capitals and four closest ports in Black Sea, Persian Gulf, and Indian Ocean () To/From
Port Port Port Pod Port Port Port Port Port Port Karachi (road) Karachi' Karachi3 Bombay (road) Bombay Abass(road) Abasss Imam (road) Imam 6 Batumi (road) Ashghabad 2297 3873 2 4129 3188 4898 1647 3116 1987 2474 3433 Tash. Dush. kand anbeh 2345 2802 2801 3258 3411 2780 3693 3236 3449 4180 2685 2603 4229 3856 3216 3134 3214 3587 4080 3623 Bishak 3378 3857 4143 4269 4912 3179 4588 37 10 394 6 4222 Alma Baku Ata 3608 4486 4091 4419 4499 5377 5178 3413 22T8r 4854 2944 3944 17547 H12 2302 42 981 4456 56 Yerevan 4451 Tbilisi 4724

5342 22618 2652 183831 2010 467

5615 25349 3026 21119 2384 428

Source:Sameas table 5.1. 1: Via Rail to Peshawar all roadthrough 2: Afghanistan Central Asiancountries. Rail to Peshawar Via and and and Kabul,Qandhar, Harat,Kushka. Rail to Peshawar Via Kabul.Termez roadandthenby rail. 4: Rail from 3: by and Bombay Peshawar Lahore, Roadfrom Peshawar Termez into 5: to to changes rail at Termez. Rail from port via Abass Sarakhs Tehran lines.6: Rail fromport Imamto Sarakhs in to via via with changes borderinto broadgauge Tehranwith changes borderinto broadgauge lines.7. throughAstara8: throughNoor Dooz.9: throughNoor in Dooz-Yerevan.

For the most direct road routes from Karachi to CAC countriesas shown in Figure 5.6 use is made of the junction where one route deviatesto Qandaharand then to Central Asia, and the other to Zahedanin Iran for the Caucasus region. The most important Indian port, which can serve CAC countries and particularly Central Asia, is Bombay. India, which has an extensivecountry-wide rail network with a connectionin Armistar as shown in Figure 5.6, is able to serveCentral Asia via Lahor 136

I
-3 cr
eb

C-4

bi

0 cl

:i

rA

to 1
eb

-4 Vt kA

c:r

0"cl-

ON I-ON : Do

a
. .

U' U'

tj

tv

? -0001

137

Chapter 5

Iranian Sea-Landbridge

the same routes for Karachi. The. most direct road route from Bombay is through over

Karachi in Pakistanwith an extra leg of about 891 km. Generallythe Indian corridor is longer than those of Pakistanand Iran as shown in Table 5.6 and it has more border during thejourney. handling or modal changes crossings, 5.3.3 Far East via TSR and TCR The Far East outlets of the CAC countriesare limited to the TSR and the TCR. The former is more compatible and longer but the latter has a different gauge.For the TCR Khazakhstanwill be the major distribution point. Apart from the landbridge port Lianyungangin China and Hong Kong, the two ports of Shanghaiand Dalian have the to serveCAC countriesas shown in Figure 4.7. All legs are connectedby rail potential in the major city of Zhengzhouand the rail lines intersect the TSR branch to Central Asia at Aktehay, about557 km north of Alma-Ata as the gatewayof the TCR in Central Asia (seeTable 5.7). The other corridor used by CAC countriesfor trade with the Far East, Russia and Japanis the TSR. It is assumedthat currently a long distance road
infrastructure of a high quality is uneconomic. Table 5.7 Rail-bridge distances between CAC capitals and four main ports in Yellow and Japan Seas(km). TSR T CR Port Hong Kong Shanghai Lianyungang Dalian Vostochnex 11108' 10857 10223 9644 11014 1125F 10857 10366 9787 11157 10559' 10308 9674 9095 10465 9785 7842 7208 6629 7900 9813 6863 7497 6284 7654 8732 6416 5782 5203 6573 6004 8320 5370 47911 61611 7L3j 4784, 5418 42051 55751 1

CAC countries capitals Tbilisi Yerevan Baku Ashghabad Dushanbeh Tashkand Bishak Alma Ata

Sources:Basedon lbornas Cook (1995), March-April 1995,CIS, China, and Hong Kong sections. Measurements basedon Junior Atlas (1979), WHSMITH, John Bartholomew & Son Ltd, Edinburgh. and EsselteMap Service,Stockholm,. 1. Shortestroute betweenAstarakhan, Guriev, Oktyabrsk,Orenburg,Kurgan,Omsk and Vostochney.

The Central Asian countriesreceive four main rail lines of which two leave the TSR at Petropavand Novosibirsk, both of which end at Chu in Kazakhestan and are 1514 km 138

Chapter5

Iranian Sea-Landbridge

1989 km long respectively. Another rail line originates from Moscow and at and Oktyabr'sk diverts into two links, passing the north-east and north west sides of the Aral Sea. It is connected to the Central Asian rail lines at Chu (3746 km) and Charduz (3283 km) in Kazakhestan and Uzbekestan respectively as shown in Figure 4.8.

5.3.4 Identification

ISLB competitor landbridges of

landbridges (TSR,TCR, the five mostcompetitive ISLB will be compared The with distances, in Caucasus) terms of origin-destination North American,Turkish and
structure. technologyand storagefacilities, and organisational The ISLB is seento serveprimarily eight CAC countrieswhich were previously usedto trading via the TSR. Therefore, the TSR is the first real and potential competitor for ISLB becauseof its strong background, and compatibility with the CAC countries' The ISLB hasgreatpotential for future operationswith transportsystemand geography. Russia and Ukraine, and other CIS countries, and with Turkey and even European the PersianGulf, and the Black and MediterraneanSeas.European/Asian countriesvia landbridges already have a long operational history at the international level and bridge; but they are not likely sometimes the overlandmodesof many countriesas a use to presentany competitionwith the ISLB for CAC countries. The North American landbridges have many similarities with European/Asian landbridges for technical, political and organisational issues. While the distance is longer, they could divert the Far East trade of the Caucasuscountries including Iran they through North America via the Mediterraneanand Black Seaports. Consequently, countries can be consideredas competitive with the ISLB to serve the three Caucasus through the Georgian port of Batumi or even the Turkish Mediterraneanlandbridge for Mersin or Eskandarun tradewith the Far East. ports of

139

Chapter 5

Iranian Sea-Landbridge

China is a vast, highly populatedcountry andcomposed many local economies.It has of a poor transportinfrastructureand sometimesusesother markets(e.g. Hong Kong) for the distribution of the domestic trade betweennorth and south (Speeceand Kawahara, 1994). Also, the first ISIB transit shipment (1906 tonnes cotton) in 1994 was consignedfrom Turkmenistanto China (PayamDarya 1994).However,the TCR, which has been operatingfrom 1992,can be consideredas the secondcompetitor for Central Asia as it is linked to the Kazakhstanrail networks; while the TSR is the main competitor due to the compatibility of its networks,past experiences, and the common economicbackgroundof Russiawith the CAC countries. The landbridgesthrough India and Pakistanhave the major problem of transit passage through Afghanistan (which is interface betweenCentral Asia, and Pakistanand India) to ensure accessto Central Asian countries. Due to the lack of rail networks in Afghanistan, shipments originating in Karachi or Bombay must have a double unloading stage and move by truck to or from the Central Asian Republic borders, which in terms of time and cost of servicesprovided can be regardedas uncompetitive with ISLB.

5.3.5 Comparison of ISLB with competitor landbridges


In this sectionthe ISLB is comparedwith the four main competitor landbridgesin termsof:
* Origin and destination distances

* Technologyand storagefacilities * Organisationalstructure 5.3.5.1 Origin and destination distances of landbridges A comparisonis madefirst betweenthe two ISLB ports of Abassand Imam and then for eight capital cities of the CAC countriesbetweenthe ISLB systemand the TSR and
140

Chapter 5

Iranian Sea-Landbridge

TCR as shown in Table 5.8 (Macomber 1969, Wronski, 1975, Padelford and Gibbs,

1975,and Hayuth, 1982). Table 5.8 indicatesthat, of the two ISLB ports, the port of Imam with a total 24129 km rail distancesfrom all CAC capitals is much closer than the port of Abass with 29265 km but due to its maritime distance(1037 km) offers longerjourneys in total to all eight CAC capitals. Port Abass can offer much shorterjourneys when the direct rail BafghMashad project is completed.At regional CAC levels the port of Abass is closer as a road minibridge to Central Asia, while Imam is closer for the Caucasus area.For road minibridges the port of Abass in both total overland distances and minibridge operations(as shown in Table 5.8) is closerthan the port of Imam. Comparing the ISLB minibridge systemwith the TSR and the TCR, the results show
that the ISLB offers longer journeys as routes have a much longer sea leg and quite

short rail or road journeys. Road transport is faster and more expensivethan rail, but both are more expensivethan water transport.Therefore,the ISLB can act as a feeder system offering mass sea transport trade in cheaper and generally faster services comparedwith TCR and TSR. This is due to much shorterrail and road routes which provide more frequent journeys and particularly when the block train method is employedby the ISLB. Assuming that block trains operateon all the landbridge routes of Table 5.8, another comparisonwith ISLB can be madewhen there are containerships with 16 knots or 30 km speedmoving along sealegs (e.g. for TSR a sealeg of 1487krn can be travelled in 2.1 day: 1487/30/24.seecolumn I of Table 5.8 for all landbridges). Two transferpoints for each landbridge with two days operations; the block trains transit time for all landbridgesis basedon Damas(1992a) for 14 days betweenVostochney-Chopas the main rail leg of the TSR with about 11,173krn distance.Therefore,the block trains

141

Chapter 5

Iranian Sea-Landbridge

Table 5.8 Comparison of ISLB rail and road minibridge system with the TSR and TCR landbridees from Kobe. Javan MO. Overland distance, time & final destination First sealeg distance Distance and transit Final leg distance and time (days) transit time and time From Origin to 2nd port Rail distance Road Final Rail distance Road and destination and transit days transit days (column days 3+2 for each border crossingport) or 3 4 5 6 1 2 7 10559 13.2) (days Baku 12046 (days 19.3) Yerevan 12738 20.2) 11251 (days 14.1) (days 11108 13.9) Tbilisi (days 12595 20) (days 12.3) Ashghabad 11272 9785 (days TSRvia 18.4) (days Tashkant 10219 Vostochney 8732 (days 10.9) (days 17) 9813 (days Dushanbeh 11300(days (twodays 12.3 18.4) 8320 Bishak 9807(days (days 10.4) terminal 16.5) 7734 9.7) Alma Ala 9221(days (days 15.8) operations) 77302 89198 TotalCAC (Average distance 18.2) 9095 Baku 11.4) 10665 (days 17.6) (days Yerevan 11357 148 9787 (days 12.3) 18.5) (days (2. days) Tbilisi 9644 12.1 11214 (days ) 18.3) (days TCRvia 6629 Ashghabad 8199 (days 8.3) (days 14.5) 5203 6.5) Tashkand 6773 (days Lianyungang (days 12.7) 6284 7.9) Dushanbeh7854 (days (twodays (days 14.1) 157 5791 7.3) Bishak 7361 (days (days terminal 13.5) (2. ays) operations) Alma Ala 5775 (days 4205 5.3) (days 11.5) 55638 TotalCAC 69198(Average distance 15.1) (days 1754 Baku 2302 14584 2.9) 24.0) 14036 (days Kobe (days 2010 1838 Yerevan 14292 12282 2.5) 23.6) 14120 (days (17.1 (days 3) 2111 Tbilisi 14666 2384 (days 24.1) 14393 ays) 3.1) (days 1987 Ashghabad 14756 ISLBvia 2474 24.2) 14269 (days 3134 3214 4) Tashkand 15496 (days Imam (days 25.1) 15416 3587 4.5) 3216 (twodays Dusbanbeh 15869 (days 25.6) 15494 (days 3946 4.9) (days 3710 Bishak 16228 terminal 26.0) 15992 (days 3944 Alma Ata 16494 4212(days 5.3) 26.4) 16226 (days operations) 24129 21694 TotalCAC 122385(Average 119946 distance 24.9) 2238 Baku 2944 (days 14189 3.7) 23.3) 13483 (days 1124 2652(days ) 2261 Yerevan 13897 3.3 22.9) 13506 (days (15.6 days 3026(days 2534 Tbilisi 14271 3.8) 23.4) 1377T (days ISLBvia 3116(days 1647 Ashghabad 14361 3.9) 23.5) 12892 (days 3856(days 2603 Tashkand 15101(days 4.8) Abass 24.4) 13848 4229(days 2685 (twodays Dushanbeh 15474 5.3) (days 24.9) 13930 3179 4588(days Bishak 15833 terminal 5.8) (days 25.4) 14424 34131 Alma 4854(days Ata 1 16099 6.1) (days 25.7) 14658 operations) 1 119225(Average 110520 29265 20560 TotalCAC distance 24.2)

142

Chapter 5

Iranian Sea-Landbridge

798.1 km daily (11155114 798.1 or about 33.3 km per hour) along all move roughly = landbridges. The results indicates that there are on average for eight CAC sample countries 15.1,18.2,24.2 and 24.9 days transit times for a block train for TCR, TSR, ISLB Ab.. and ISLB ,, mrespectively. Due to longer sealegs for both the ports of Imam and Abass, transit times for these two ports are much longer than those for TSR and
TCR, while the 14 day transit times for the latter two landbridges seems to be very

The importance of faster container ships, with a speedof higher than 16 optimistic. knots, can effectively reduce ISLB sea leg transit times which in terms of current shippingtechnologyis easily obtainablein Iran and the CAC countries. 5.3.5.2Technology and storage facilities The technical comparisonswere made between ISLB and six other landbridg countries and areasin terms of four major available measuresas shown in Table 5.9 (Miller 1978). Table 5.9 Comparisons of the ISLB road and rail transport networks and fleet with maiar landhridge.c.of the world.
Transport indicators and rankings Road networks, km IAngth of railways Rail freight tonnes-km per year (million) Road freight vehicles Iran 157,794 4,847 8,002 (1992) 875,452 1 USA 6,237,290 199,900 1,480,205 1 1 47.095,000 Landbridges 2 3 Canada European 4 Turkey 5 China 6 FSU 1,737,000 258600 3,827,700

countries 367,409 1,056.707 392,000 4,287.888 (25countries) 10,386 53200 194,000 227,065 (24major countries) 342,534 16,000,000 1,037.295 242,439 (22major tonnes(7571 I countries) tonnes km in 1989) 3,680,000 25,052,339 861,005 N/A (22major I countries)

3,250,000

Source: The Economist-Pocket World In Figure (1993), edition. p 56 and 59. Forllurkey. Holt (1993), p. 91 for former USSR. Europa World Yearbook (1995), Volumes. I& II various pages. Jane's World Railways (1990), for rail freight tonnes-krn of Turkey.

The ISLB was compared with other major landbridges in terms of relevant port infrastructureand superstructure, road and rail networks, rail freight performance,and road freight vehicles. Ports play a key role in the operation of any landbridge,as they are a major determinantof two important elementsof the landbridge operations(cost 143

Chapter5

Iranian Sea-Landbridge

and time) in all countries. The number of landbridge ports depends on the maritime geography of the relevant countries. Iran, like the TSR, the TCR and the Caucasus, is dominated by one or two ports suitable for landbridges, while Turkey is in a better situation and north American landbridges are served by a series of capable ports on the east, west, and Gulf coasts. The output of landbridge ports largely depends upon container facilities. Comparisons were made of the seven elements of container berth length, number of gantry cranes, depth of water at container berths, covered area for container stripping and stuffing, total area for containers, and container traffic in 1993. All these measures are significant indicators of a potential landbridge port and the results are shown in Table 5.10. The port of Anzali was excluded from the comparisons as it uses general cargo berths and facilities for containers.

The ISLB ports with 8.3% of the container berths rank higher than the TSR and the TCR with 5.4% and 3.9% respectively,and close to Turkey (9.8%). However, they are not comparablewith four USA ports which have about 53.3% and Canadawith 19.3% of the total number of container berths of the 18 landbridge ports considered for comparison(Table 5.10). The ISLB ports have 3% of the total 131 gantry cranesfor the 18 landbridgeports, and ranked lowest close to the TCR and Turkish ports, while the USA and Canadianports with 84% and 17% have the highest number. In terms of depth of water at container quays, both ISLB ports can be categoriscdamong the high ranking ports with deep water quaysof 12-14metres. In terms of container(TEU) traffic the ISLB ports handledabout 2% of the total TEUs. The port of Abassrankedmuch higher than many other of the eighteenlandbridgeports in terms of both TEU and tonnage,and closestto the port of Mersin in Turkey (Miller 1978UNCTAD 1991).

144

Chapter 5

Iranian Sea-Landbridge

Landbridge ports in Iran are comparable to Turkish landbridge ports in terms of

infrastructureand superstructure, communications, ownershipand management


Table 5.10 Comparison of the ISLB ports infrastructure main landbridge ports of the world. Container facilities Land bridges Ports Berth length (metre) 2 1051 1000 2051 980 195 525 1335
1664 1017 2072 4753 710 300 1414 2424 540 425 965 7056 457 748 4898 131591 Total 1 246871 12 9 12-13 9-13 11 10.5 10.5-11 10.7-12.2 11.6-12.19 12.2 12-15

and superstructure with Container throughput p.a. 993 TEU Tonnage

Depth (metre) 3 12 14 0 12 11.5 9.75 9.5 9.5-11.5


13.8-15.2 8.2 10.7-14

Gantry crane 4 2 2 0 4 4 2 4 12
7 2 (one mobile) 8 17 5 1 2 8 2 4 6 40 11 4 29 84 131

1 ISLB; Imam Abass I Anzali Total Vostochney (1990) Nakhodka St. Peters burg Total
Halifax 3 Canada Toronto Vancouver Total Mersin 4 Turkey Trabzon Izmir Total Lianyungang 5 China Shanghai Total New York/New Jersey 6 USA Houston San Francisco Seattle 1 Total 1 Grand

5 8268 82920 4083 185000 000 73620 (1992)


300933 12922 434004 84456 N/A 212,949 25000 900256 925256 1972692 541497 119324 1151000

6 50237 758278 56078 2,000, OOC 000 825890

TSR

2518603 158941 3458242 1,211,069 R/-A 1438163 52638 65243100 6576938 NIA 4617231 2904000 8445000

Sources:1. ContainerisationYearbook (1995), London: Emap BusinessCommunicationsLtd. 2. Hicks (1994), Jane's ContainerisationDirectory for 1994-1995,(ed), Twenty-sixth Edition, Jane's Information group Inc, 1340 Braddock Place,Suite 300, Alexandria, VA 22314-1651, USA. 3. Russianports except St Petersburg basedon Fraser(1993), p. 103.TEU for Vostochneyin KoreanMaritime Institute study are project p. 56. For port of Izmir in Turkish State Railways (1994), Handbook of the Turkish ports operatedby TCDD, Istanbul. Difference in the number of ports is due to the variable availability of data for any particular comparison.

145

Chapter5

Iranian Sea-Landbridge

systems, but differ from the TSR and the TCR and the North American landbridge ports. In particular the latter employ different systems of port operational management and there is comprehensive involvement of the private sector. The alternative routes and the quality of roads for the ISLB and Turkish landbridges are limited with mainly twolane asphalt roads.

landbridgesare not suitablefor commerciallong distance The TSR, TCR and Caucasus road transportwith limited road networks. Basically, road servicesare not offered and light trucks suitable for short distancesconstitute the bulk of the road freight fleets. In contrast, the composition of the road networks in North America are complex and extensiveproviding many optionsof a high quality and standard. In termsof traffic ISLB has denselyloadedtruck and rail movements mainly in a southnorth direction for both domestic and foreign trade, leaving a considerable spare capacityfor flows from the CAC countries. 5.3.5.3Organisational structure and productivity The organisationalstructureof the landbridgesconcernsmanagerialand ownership types of port ownership and operational systems,including infrastructure aspects. g. e. and productivity (UNCTAD 1989and Damas 1992a). There is a variety of ownershipand operationalsystemsin landbridgepractice and, for the purposesof this work, the two ISLB container ports are comparedwith seventeen Iandbridgeports for total trade, the averagedaily productivity of each port, and also container traffic and gantry crane output as shown in Table 5.11. The ISLB transport systemis similar to the Turkish landbridges,and to someextent similar to the TSR and the TCR, in terms of ownership,but differ from the North American landbridgeswhich have private ownership basedon competitive long term management and operational schemes. 146

Chapter 5

Iranian Sea-Landbridge

The comparison of the general port productivity for eighteen ports indicates that the

port of Shanghaiwith an averagethroughput of about 3.8 million tonnes is the most productive followed by the ports of Houstonand Los Angeles.The two ISLB ports are (Abass) and fourteen (Irnam) as indicated in Table 5.11. These ranked at eleven comparisonswere made when both ISLB ports, despite their great storagecapacity, were dependent ships' gearto load andunloadgeneralcargo trade. on Table 5.11 Comparison of the ISLB ports maritime and cargo traffic and nroductivitv with main landbridee Dorts of the world. Ports productivity 3 4 2 5 1
Landbridges
Ports Port total cargo traffic (000 MI) in 1993unless otherwise specified (both foreign & domestic) Daily /tonnes (365 days work) and ranking per container gantry crane/day (tonnesunlessotherwise specified)

Tonnes day/p. a. & ranking (5.1) 68.8 (24) 1038(11) 1369.9

TEU/day/ p. a. & ranking (5.2 11.3 (25: 113.6 (110 126.7 (11:

ISLB

Imam Abass Vostochney Nakhodka St Petersburg

11047 13224 11600(1990)

30265.8 (14) 36230.1 (11) 31780.8 (13)

32876.7 (12) 000 (27' 12000(1990) 000(25) 29863(15) 565.7(18) 57.2 ( 20' 10900 (1990) N/A 135.1 (9: 41352 (only 113293(7) New York / New Jersey foreign trade) 1150(10) 134.9(10, 71842 (only 196827.4 (2) Houston foreign trade) 3 196 (3: USA 68664 N/A Los Angeles 188668.5 (3) __ 81.7 (16: N/A N/A 1989(2) San Francisco N/A 797.8 108.7 (14: N/A Seattle 80.2(17, 356.4(19) 157046.6 (5) Philadelphia 57322 (only foreian trade) 117.8 (12' 38663(9) 985.8(13) 14112 Halifax N/A N/A 17.7 (24' Toronto 217.7 21 4 1018.4(12) Canada 16500 45205.5(8) 102.4 (15' Montreal Vancouver 60762 166471.2 (4) 1184.3(9) 148.6 (6, 5 Turkey 13966 38263(10) 663.6(17) Mersin 46.3 (21 Izmir 5268 14432.9 (15 ) 1970.1(4) 291.7 (2: Lianyungang 123947.9 (6) 72.1(23) 45241 (container 34.3 (22: 6 China tonnage 1992) 382438.4 (1)! Shanghai 4468.7(1)1 6i6.6 (14 139590(1990)1 Source: 1. Containerisation Yearbook (1995), London: Emap Business Communications Ltd. 2. Hicks (1994), Jane'sContainerisationDirectory for 1994-1995,(ed), Twenty-sixth Edition Jane's Information Group Inc, 1340 , Braddock Place, Suite 300, Alexandria, VA 22314-1651,USA. 3. for San Petersburg,Vostochneyand Nakhodka are basedon Peters(1993, p. 290). Differencein the numberof ports is due to the variableavailability of data for any particularcomparison. TSR

147

Chapter 5

Iranian Sea-Landbridge

Landbridgeports differ in the number of gantry cranes (column 5) and therefore,in the volume of containertraffic. The productivity of ISLB ports was compared(Table 5.11) with other ports on the basisof averagecontainerhandlingper gantry crane. Each of the ISLB ports hastwo gantry cranesas haveIzmir and Liagyungang.However, although the container output of Imarn and Liangyungan (Table 5.11) are similar
(column 5.1), Izmir has nearly twice the throughput rate of Abass.

Productivity of the two ISLB ports in terms of TEUs per gantry crane per day on an annual basis was made among 19 ports (Table 5.11, column 5.2). It is very low for Imam (11.3 TEU/day/gantry crane) showing that containerisation has not been developed adequatelyin Iran. The container is still considereda type of product by Iran's Customs administration (Sanate-Hamlo-Naghl1995c) and not as a unit load
device. Abass ranks about the middle with 113.6 TEU/day/gantry crane.

A comparisonwas made(Table 5.12) betweenthe ISLB and other landbridgeports. The


management and operation of the ISLB modes and interfaces are government run and

supervisedby the PSO through a deputy minister of the MRT. Therefore, to a large extent, Iran has a governmentfinancially supportedsystembasedon central planning. Iran's landbridge systemfunctions in a much simpler way than the TSR, but there are strong similarities with Turkish transport systems.In North America, landbridge ports are governedmainly at port and municipality levels. Both ISLB ports and three other Iranian major ports have one administration and management to run and supervise all operations, its own infrastructure and superstructure.

148

Chapter 5

Iranian Sea-Landbridge

Table 5.12 Comparison of the ISLB container ports and terminals ownership and management with other landbridges and ports of the world. Ty e of port and m agement Landbridge operators Operation Port ownership LandPorts bridges management PortAuthority IslamicRepublic Iranian PortAuthority Imam Shipping Line & other and national international ISLB operators Same Imam PortAuthority PortAuthority Abass as Vostochney PortVostochney PortVostochney SOTRA& differentother and national international companies Same Vostochney 2 TSR Vladivostok Vladivostok Nakhodka as Commercial Port Port Commercial Corporation ion Corpora Same Vostochney SanPetersburg PortAuthority as Various Eightdifferent NewYork/New Portauthority shipping and of NewYork & Jersey terminal operators NVOCCoperators NewJersey 2 private Various Portof Houston 3 USA Houston shipping and Authority terminal operators NVOCCoperators Onestevedoring Various SanFrancisco Portof San shipping and Francisco terminal operator NVOCCo rators Six terminal Various Seattle Portof Seattle shipping and NVOCCoperators operators Two terminal Various Philadelphia Philadelphia shipping and NVOCCoperators Regional Port operators Authority Canadian Pacific(CP) HalifaxPort HalifaxPort Halifax and Canadian National Corp& two (CN) Corporation & terminal operators Railways Various shipping NVOCC and I operators Toront; TorontoHarbour TorontoHarbour Same Halifax as Commission Commission Same HZfax Fiveterminal 4 Canada Portof Montreal Montreal as operators Same Halifax Port Three Vancouver Vancouver terminal as Corporation operators Various TurkishState TurkishState Mersin shipping and NVOCCoperators Railways Railway 5 Turkey Trabzon Turkish Maritime Turkish Maritime Various shipping and Organisation NVOCCoperators Organisation Various TurkishState TurkishState Izniir shipping and Railways NVOCCoperators Railways Railways, Lianyungang port Lianyungang port Trans-Chinese Lianyungang Chinese National Co. Container Shipping Container Co. Line & Various shipping and NVOCCoperators I Same port 6 China Shanghai Shanghai Harbour Shanghai as Container Lianyungang Bureau Ltd Terminals I I
Source:1. ContainerisationYearbook(1995), London: Emap BusinessCommunications Ad. 2. Hicks (1994), Jane's I. Containerisation Directory for 1994-1995, (Ed), Twenty-sixth Edition, Jane's Information group Inc., 1340 Braddock Place, Suite 300, Alexandria, VA 22314-1651,USA. 3. for San Petersburg,Vostochncy and Nakhodka arebasedon Peters (1993, p. 290).

149

Chapter 5

Iranian Sea-Landbridge

5.4 Conclusions
This chapter has presenteda general description of the reasonsand the national motivation of Iran in establishing landbridge services for CAC countries. It has describedthe generaland specific characteristics the Iranian SeaLandbridgeand its of different modes and interfacesin geographical,political, technical, and organisational Finally, a comparison terms.Then, alternativesroutesfor CAC countrieswere assessed. was madebetweenthe ISLB and five other main landbridgesof the world on the basis Thesewill provide the foundationfor the conceptualmodel presented of certaincriteria. in the next chapter.

150

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

6. Scenario approach and the Iranian development conceptual model


6.1 Introduction

Sea landbridge

The preceding five chaptershave examined the transport and trade of Iran and the Central Asian and Caucasian(CAC) countries, and have undertaken a review of the (ISLB). landbridgeconceptas well asthe main characteristics the Iranian Sea-landbridge of The purposeof this chapter is to discussthe scenariobuilding approachand examine its relevanceto the ISLB study.

6.2 Planning and analysis under uncertainty


Johnsonand Scholes(1988) note that uncertaintyis a result of complexity in the system within a context of. * The sheerdiversity of environmentalinfluences * The natureof management skills to cope with uncertainty * The interactionof environmentalinfluences. Uncertainty arises from several distinctive sources.According to Frenkiel and Goodal (1978) thesecan be discussed under two major areas: * Inherent uncertainties,relating to the systemcomponentswhich cannot be treatedin a deteaninisticway. e.g. pollution. * Superimposed which uncertainties,arising from modelling and observationalprocesses may lead to someerror in the structureor estimationof the parameters the model, thus of affecting the model output. Businessprojects are affectedby a seriesof internal and external elementsresulting from changesin decision making or from uncertain environments. Researchers and planners require techniquesto understandthe main attributes and variables of an uncertain future 151

Chapter6

Model Development ScenarioApproachand the ISLB Conceptual

their potential for proper decision making. According to Wigg (1982) the objectives of and for uncertainty can be categorised into two main areas: planning

* To decidewhich strategyshouldbe pursuedto the organisation'sgreatestadvantage. * To developcontingencyplans. Before the 1970sin a relatively stableera andprior to the first oil crisis in 1973,forecasters in econometric modelling employed mainly the rate of economic growth as the only uncertainty (Mandel 1983). The oil price crisis in the early 1970shad a great impact on led life, and consequently to vast technological and social normal economic and political changes in all countries. In developed nations it also increased the unreliability of projections and forecastingmethodsbasedon past data for their outcomes.This led to the developmentof alternativetechniques(Mandel 1983). In order to develop the ISLB into a reliable and modem Iandbridge, many types of problems must be identified and overcome.The CAC countries as customersand Iran as the supplier of the landbridge capacity and servicesexist under severeuncertainty. The CAC region is composedof eight countries with about 60 millions population and over but three million square/kmland which have obtaineda nominal independence politically, economically,territorially, and socially are and will be for a long time under the influence of the future changesand events in Russia,the leading country of the former USSR. Iran has unstable oil revenues and problems in its relations with the USA. Landbridge operations,with a complex international hinterland, like any other economic or service entities, have to face up to the issue of uncertaintyduring the long range planning stages. There are some techniqueslike sensitivity analysis, simulation techniques (e.g. Monte Carlo Simulation), spreadsheet modelling, etc. that deal with uncertainty. However, the most suitable technique to deal with uncertainty is the application of scenario analysis (Foster 1993).

152

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

6.3 The scenario approach


A landbridgehas many internal and external environmentalfactors confronting the its political and economic operation.The CAC countriesare passingthrough reliability of the preliminary stages of a free market restructuring while Iran is still in a war Both the CAC countriesand Iran are in hasunstableoil revenues. reconstructionperiod and their different ways, passingthrough political and economic instability. Therefore,a valid approach to the study is to consider some potential scenarios under conditions of Consequently the ISLB environment will be investigated under a set of uncertainty. (Mandel 1983,Wigg 1982,King 1986)to divulge the probable multiple plausible scenarios future situations.The plausibility of the scenariodepends the content and consistency of on the scenariocomponentsas best predictors of the future situation. A scenarioexplains a plausible future while two or more scenarios consider probable alternatives for the uncertainenvironmentunder study. Therefore,decision makerscan seea probableversion of the future and are able to designdesiredstrategies.

6.3.1 Scenario characteristics


is that havebeendevelopedin The scenarioapproach amonga variety of techniques fields where the environmental or future services are uncertain for different long-term periods. Scenarios must include sufficient data to confine the range of the different uncertainties.The reliability of the scenarioresults dependupon the nature and strengthof the data involved. Each scenariois composedof both quantitative and qualitative data and descriptionsof different natures,e.g. social, political, economic, geographical,etc. under certain assumptionsand conditions aimed at the future by the use of past and present information. 153

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

Allport et al. (1986, p. 232) mention the following definition accordingly due to Ayres

(1969):
"A logical and plausible set of events,both serial and simultaneous, with careful " to timing and correlation whereverthe latter are salient. attention

Jungermannand Thuring (1987, p. 247) quote Mitchell (1979) and (Kahn, 1965), the
pioneer of the concept, for two well known definitions: "An outline of one conceivablestateof affairs, given certain assumptions about the presentand the courseof eventsin the intervening period". (Mitchell 1979)

"Hypothetical sequences eventsconstructedfor the purposeof focusing attention of and on causalprocesses decisionpoints". (Kahn 1965)

Wack (1982,p. 18) noted that a scenario:


"Is an archetypaldescription of a possiblefuture basedon a mutually consistent grouping of determinants ....... . is intendedto be regardedas a tool to assist . understanding as a backdropto the decision making process,ratherthan as an integral part of the decision itself".

Nijkamp and Blass (1994, p. 82) have indicated the role of the scenarioand its differences from other techniques transportplanning: in
" It is one of the methodsand techniquesof perspectivepolicy research that havebecome very popular sincethe late sixties. Especially in the caseof unstructureddecision problemswith uncertainoutcomes,scenarioanalysismay be an appropriateinstrument.It doesnot only contain a descriptionof one or more future situations,but also a description of a consistentseriesof eventsthat may connectthe presentsituation with the described future situation (s)."

Shearer(1994 p. 112), in defining a scenario,compared the Delphi technique with the scenariomethod and pointed out that the Delphi technique contributes to some "idea of severalpossible futures" with a long-term time horizon but the scenarioapproachserves
for a lesser period and: "Concernsthe transition to a new long-term position rather than the final position itself and exploresthe impact of thesepossiblefutures on the company. "

154

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

Millett (1988, p. 61) in addressing the role of scenario analysis points out: "Scenarioscan be an excellent approachto someforecastingand strategymaking problems,but they will not answerall questions.No analytical tool can. There are and proper times to use statisticalmodels.Scenarios proper times to use scenarios, are best suited for long-rangetime spans(5-20 years),highly complex situationsof many factors of which someare unquantifiable,highly uncertainsituationswhere virtually every factor is a variable and none is a constant,and wherethere are few " or no reliable data available for quantitativemodels. According to Wack (1985b, p. 140): "Scenariosstructurethe future into predetermined and uncertainelements.The lies foundation of decision scenarios in exploration and expansionof the elements". predetermined

Amara and Lipinski (1984, p. 5) havedefined a scenarioas:


describingan "Representation an interconnected sequence trends/events of of internally consistentalternativefuture."

The differences between the Mitchell (1979) and Kahn (1965) definitions according to
Jungennann and Thuring (1987, p.247) lie: "In their emphasison the courseof eventsbetween the initial stateand the final state".

Scenariosas practicable and logical alternativesare clear straightforward devicesused in studies where there are uncertainties.According to Mandel (1983, p. 10-6): "Scenarios don't reduce uncertainty, they clarify it". When using scenariosthe forecaster must be certain about the existenceof the indications of uncertaintiesin each scenario,which can be found through analysisof the pastinternal and externalvariables. According to Parenteand Anderson(1987, pp. 149-150):
'Ile content validity of the scenariosis as important as the representativeness the sample. of Perhaps best way to ensurecontentvalidity is to adequatelysurveythe literature (and the Scenariostatements should expert) prior to developinga tentativelist of relevant scenarios. be no longer than 20 words and shouldbe in sucha way that they can be easily verified."

Kleiner (1995, p. 275) addsthat:


"Contrary to what many peoplebelieve about scenarioexercises, their purposeis not prediction. You don't predict what will happen:you posit severalpotential futures,noneof which will probably come to pass,but all of which make you more keenly awareof forces acting on you in the present... A scenarioplanning exerciseis a bit like a storytelling that the conventionalwisdom workshop, setup to bring forth distinction and phenomena ignores. "

155

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

Ginter and Duncan(1990, p. 97) basedon Wack (1985) concludedthat:


"Scenariosat leastas they are usedat Royal Dutch Shell are fundamentalaids to changing to the mental modelsand alerting the corporatemicrocosmin ways that allow managers levels of risk." for the future while assumingacceptable generateoptions

The scenariosand their individual sub-setsin one study must be sufficiently different from highly related to the area of interest to cover most probable each other, consistent, and
outcomes (King, 1986).

6.3.2 Scenario elements


A scenariocomposition has three elementswhich are called scenariodescription, sub-titles. The scenariodescriptionconsistsof the general scenariovariables and scenario the scenariopurposeand its future. It is a description of each scenein explanationsabout detail and may cover a few pages.Mitchell et al. (1979) mentionedthree types of variables in any scenarioas: * "Event-trend" (e.g. rates,percentages, ) etc. * "Events" which occur unexpectedlyand onceonly influencing the future output. * "Trends" which are time seriesdataor any other measuringunits. Scenario sub-titles cover different events and trends within a similar area of activity or According to Verroen and Jansen(1993) and Mitchell et al. (1979) sub-titles are sector. called "themes" and cover different internal and external environmentsof the scenario (Schnaars 1987,Shoemaker1995).Vlek and Otten (1987) developedsocietaldiscussionon (nuclear)energypolicy (SDEP) scenarios under three areasor "themes" as socio-economic policy, energypolicy, and environmentpolicy. Veroen and Jansen(1993) in an attempt to transportdevelopedtheir scenariosunder five headings:economic,socialstudy passenger cultural, technological,infrastructuresupply, and pricing and regulation.Naylor (1983) has made anotherdistinction for the scenarioelementsas: key detenninants,which are events and variables and conceptuallycan make a future situation and principal impactfactors, 156

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

According to Amara which are the result of the scenarioeventsand the adoptedstrategies. and Lipinski (1984) the external environmentaffecting scenariosof a systemhave the six following of influence with major areas different geographical characteristics

(international,national, regional,local, etc.). indicators, including macroeconomic * Economic/resources * Consumer/demographic changes, * Labour force/ work place, * Regulatory/politicalincluding government/private sectorinterfaces, * Competitive/institutional,and * Technological/capital. Methods of developing the scenarioelementsdiffer, and as pointed out by Mitchell et al. (1979) in Jungermann Thuring (1987), fall within the two following approaches: and * Bottom-up: it is inductive, the elementsare first generatedand then, using matrix and their relationshipsare gradually developed; graphtheoreticalprocedures, * Top-down: it is deductive,clustersof interrelatedelementsare usedas input for scenario developments. The uses of these methods depends "upon the agency and forecasting purposes" Jungermann and Thuring (1987, p. 248). For example, the top-down method is suggested by Kahn (1989) for long range planning studies. According to Saunderset al. (1987, p. 324), the Trend Impact Analysis approachcan be used to identify the scenariosub-titles and:
"Once the most probablescenariohasbeenchosenother boundaryscenarios be can by generated examining deviation from the core. Alternatively severalindividual theme scenarios be chosenautomaticallyor by groups". can

6.3.3 Scenario efficiency and effectiveness


Vleg and Otten (1987) havesuggested a good scenario consistsofthat 157

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

*A basic analysisof the statusquo and its developmental trends, *A projectedor a desiredimageof the future at an adoptedtime horizon, and *A descriptionof developmental (whether controllable by deliberatepolicies or processes not) by which the statusquo is going to be connectedto the imaginedfuture. Mandel (1983) suggeststhat the number of scenariosis important in the validity and effectiveness the scenariotechniqueand advisedthe applicationof two to four scenarios; of in while Howard (1990) points out that the numberof scenarios any situation can be two to the power of the number of options; but all cannot be executed and many might be infeasible. Stokke et al. (1990, p. 3) in the scenarioplanning for Norwegian Oil and Gas havecommentedthat:
"For scenariosto be more useful they must be 'decision focused'.That is, their analysis of alternativefutures must focus on the specific trends and issuesthat are important to the decisionbeing made.Accordingly, the team first selectedthe 'key decisionfactors': that is, what insights must the research managerhave about the ftiture business environmentto Thesefactors, in turn, must form the basisfor the scenarios' make the decisionneeded? storiesof the future."

6.3.4 Scenario dimensions and types


Different scenario approachesmay have different dimensions in terms of the purposeand objectivesand causeand effect. Jungermannand Thuring (1987, p. 248) note that:
"Unfortunately, the authorsof scenariosrarely state which type of scenariothey . ..... . for want to develop and why this particular type is adequate this purpose.In fact, most scenariosare not of any distinct type at all but representmixtures of the various types."

Nijkamp and Blass (1994, pp. 83-84) state that Van Doom and Van Vight (1981) have indicatedthe four following types: * Descriptive scenarios are basedon the digging of the past and present knowledge as causes into probable future situations as effects, irrespective of their desirability or undesirability. 158

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

* Nonnative scenariosare basedon the writers' or users' ideas;therefore,the future paths and pictures are selectedin advanceby giving meansor goals and obtaining others (e.g. Ozbekhan,1969 and Van Doom and Van Vight, 1981). According to Becker (1983) the future scenariois determinedand then the eventsand trendsare developedto achieveit. * Projective scenarios with both descriptive and normative characteristicsare those in into future paths as which their future forecastsare basedon the presentsituation as causes effects. * Prospectivescenarios are those with normative characteristicsand the values of their forecastsand paths leading to that future are predetermined. They work from the future to the presentsituationin reverse. Both thesetypes of approachto scenarios projective while trend scenariosforecastthe are future by extrapolationof the presentsituationand pathsfrom a feasiblepoint of view. If a scenario is constructedby the choice of the "majority of people" it is "common
opinion" otherwise it is called happyfew scenarios.

Ducot and Lubben (1980) madethe distinction betweenscenariotypes as shown in Table 6.1.
Table 6.1 Four basic tvnes of scenarios.

Scenario types Descriptive Normative II

Anticipatory Exploratory Given the causes, what are the Given the effects,what are the causes? effects? Given the goals,what means Given the means,what goals I be reached? can be used? can

Source:Basedon Ducot and Lubbcn (1980,pp. 51-54)

Howard (1990) suggeststhat the scenario technique can be developed under the five
following types in sequence:

* Status scenario: which considersthe future aswas seenin the past. quo
* Thepresent scenario: considers the changes in the status quo scenario with present time

intentionsaccordingto the interactionsof the scenariocomposers.

159

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

* Position scenario: reflects the "diplomatic negotiations" and results from the interaction

amongactorsin a given situation. * Compromise considersthe "compromisebetweentwo actors' position, ... which scenario: is preferredby eachto the other's position". * Conflictsscenario:In this scenario one actor may force others to accept his position through strikes,boycotts,deadlocks,etc. Foster(1993) hascategorised scenarios two levels: at The global scenario concerns global economic issues and variables while the industry scenariocoversthe competition and technologicaltrends. Allport (1986) hasclassifiedscenarios into two types: accordingto their parameters
* Time trend scenarioswhich are complex and composed of "an explicit dynamic

path for changes"with high costs. * Discrete which are more popular in transportplanning and appliedin events scenarios
a single "design year". Their parameters have static end-states and concentrate on

dynamicelements.

6.3.5 Scenario ranking, probabilities and criteria


Mandel (1983) suggests three typesof ranking for scenarios: * Optimistic (high growth) * Most probable(moderategrowth) * Pessimistic(recession low growth). or Foster (1993) notes that ranking is dependent upon the number of scenariosand should be relative to the baseor central scenario.Amara and Lipinski (1984) point to three tYPesof
scenario rankings as "uppee, "central", and "lower" band, all of which should be

developedaround the central scenario.Mitchell et al. (1979) note that for a scenario to This is doneby "act of faith", explain the future it must be rankedwith its best assessment. 160

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

"heuristic rules", "simulation" or "methodologicalprogrammingtechniques".According to Schnaars (1987, pp. 108-109),labelling the scenarioeventsor ranking the likelihood of the may be conductedthrough quantitative methods such as cross occurrenceof a scenario impact analysis which "attach probabilities to the scenarios they drive" or by "some judgmental schemes". In any environmentthe number of scenariosmust be limited logically to be managedand studied. Wilson (1978) quotes Zenter (1975) on scenario selection as consisting of credibility, utility, and intelligibility.

6.3.6 Scenario analysis techniques


Application of the scenariomethodmay involve the use of other techniques, and in particular a quantitative approachlinked to a qualitative scenariomethod is very common practice. The application of mathematicalmodels to the scenario gives "greater internal quantified consistency"(Cole and Chichilnisky 1978,p. 22) and any forecastingtechnique can be employed (Zenter 1982). The range of these supplementary techniques for forecastingthe future values of the independentand dependentvariables vary depending The six upon the time, availability of data, purpose and knowledge of the researcher. following techniques by were suggested Taylor (1992): * trend analysis * computcrsimulation
* decision analysis

* sensitivity analysis * Delphi study and * impact analysis. Shearer (1994) developed the scenarios as a planning model relating economic and company assumptionsto be linked and forecast through the application of a computer 161

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

spreadsheet. Tongzon (1991) applied multiple

regression to three scenarios of an

Australian ports study. Picard and Nguyen (1987) and Dagenais (1987), applied inputin techniquesto two scenarios measure impactsof changes the to the output and regression transportand economicattributes,and long-termcontainertraffic forecastingof the port of Montreal. According to Huss and Honton (1987) and Stover and Gorden(1978) the Centre
for Futures Research Group applied trend impact analysis and interactive cross-impact

analysis(MERAX)

and Mont Carlo Simulation when using scenarioanalysis,while the

Battle ScenarioInputs to Corporate Strategies(BASIC) used trend impact analysis with sensivity analysis.According to Martino (1978, p.389) there are two differencesbetween the scenarioapproachand the cross-impact matrix (CIM):
"First it is lessrigorous, sinceit is primarily a verbal description of somefuture situation. Second,it is more detailedand providesa rich descriptionrather than simply a list of events that are to take place.The scenariocan be considereda written descriptionof a single future history, which might have come from a single play of a cross-impactmatrix.

Zenter (1982) proposedtwo approaches scenariodevelopment,covering qualitative and to mathematicalmodels, distinguishablewith soft and hard methods;in the soft method the scenario is based on human capability and environment using Delphi or cross-impact analysis (CIA) to identify events and trends and their relationships and importance.The hard method is basedon computer application and is a dynamic behaviour model called system dynamics,and its componentshave cause-and-effect relationships with feedback loops. Schnaars (1987) points out that the main difference betweenthe scenarioapproach and other analyticalmethodsis: * It usually provides a more qualitative and contextualdescription rather thanjust seeking numericalpredictions. * It identifies a set of possibleplausibleeventsfor the future.

162

Chapter 6

Scenario Approach and the ISLB Conceptual Model Development

6.3.7 Scenario development and construction


Scenario construction covers the range of activities and investigations to systematicallyidentify and combine the componentsof each scenario ending with the forecastingof the future scenariovariablesandevents(Twiss 1992). A scenariomust be seenas a single framework containing components with logical socioeconomicallinks with eachother. According to Howard (1990,p. 242) the preliminary task in the scenariodevelopmentis to preparethe:
for "List of actorsand options [which] onceachieved,is a mechanism generatingscenarios. The object is to preparefor interactionwith othersby looking at what may lie ahead,what they may and you should aim at, and how to exert or resistpressure".

An aspect of the scenario is to how satisfactorily it is designed.Scenariosof different studies may use common data but they are developed to answer different questions. Leemhuis (1990) mentions that, in the development of scenario descriptions, the relationships and interactions of four components (social, political, economic, and
technological) in an internally consistent manner is of significant importance.

Different authors explain scenario construction differently and in the fonn of a few systematicstages(seeFigures in Table 6.2) leading to the application and interpretationof the scenarioresults.Scenarioconstructioninvolves the use of a seriesof systematicworks

about past, presentand future situationswhich may differ slightly in format, but are
generally common in concept. Construction and developmentof scenariostake place in different stages.Table 6.2 shows the most important cited approaches scenariowriting to by fourteenauthors. To construct scenarios, the scenario builder needs to have some knowledge -of macroeconomicsto identify the key factors and also data processing to find out the interrelationships betweenvariables.

163

ZA
M

fe

fD

ep 0

rp

5fD c

&

9,

fb .

.2 5 CD
F
0. o E

2 5.

zi cr.

:5.

1 10

tv
m rD
12
me

CD cg
l'

s. "0 >% 0
m
n fb

c,

g
c

CZ

cx

C)
1 Ct

=
n 0

ELro*
9
01

EL
c2. fD

(V
12 0. 12

H
ep
st c3 0 >ob Pm rD

<

(A n eD

121.

-1 1 m
cl Gr

m
0 2

fo

& -

c
00

99 C Olb n. 0

0 0-%

-4 Ch LA t)

- ch

4.

C4

-4

00

th

LA

-4

Ch

se. i'

Go,'

-wCb

I
I
Pd

-2

C>

N)

bi

-a

CY%

00

(A

P.

0-

ui

e 2

164

1 0%c 00 -4 N-0 00 -4 Ch "I

,u i rD c"

4 Ili 4 10
to
ct

..*. ;;, &M.1n ;, -, a0


,
1D 5

I
:3

En &0,cp CAr, 0 (A cp P, Pil -1 8 Q Q .n

mP

Ln

'A

(n

C: 110
to,

.1w

5: .
cl rt

5.
W

m N. to a. 11
FA fv I

rA

r 0
is

I 9: 61 9 1 El

4,

m cr
2
ts fb w

fo

r, ;1 fD
Ig 2
W M C:

Ed n 0 4 E d p

Ir.
M rD ld

91

113
.1

Q.
od CD 10 to Ild ts to

N
cA

rA *v
0% .

LI. 1

0
-.

r. cr. ts

5 , LT

a -1

m E;
Qd

In.
fD

C:

EP

ft

'&

fo

. o

oo -4 a, w

t4

qft

.01

00

1 92.

13

165

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

Wigg (1982, p. 79), stated regarding scenario construction that: "Planning for uncertainty typically is an iterative process,which requiresa repeated of refinementof the characterisation the strategicoptions that are available, the project specifics,and the uncertaintiesto allow better analysisof their relative preference".

in his Norse (1979) describes approachto scenariodevelopment five stepsas: * Definition of the conceptual models, in his case, concerning the socio-political and
economic situations of the OECD and third world changes after World War If;

* Selection and specification of the key assumptions,covering the qualitative and pointed out in the conceptualmodels; quantitativeassumptions * Productionof the quantitativeframework; * Enrichmentwith materialsfrom complementarystudies;and * Analysis of the dynamicpath leadingto the end state. Mercer (1995b) emphasises"simpler scenarios" both in creation and application while Ginter and Duncan (1990) point out the importanceof macroenvironmental analysisas it assesses socio-political and technologicalforecasting.According to Schofer and Stopher all (1979, p. 205):
"The ultimate objective of any scenario-generation schemeis to producea framework directly a new long-rangetransportationplanning process,including which supports generationof sensiblealternatives,forecastingtravel demandand systemperfon-nance, and impact and cost evaluation".

Schofer and Stopher also argue that as the scenario-generation technique has a significant subjective input, the identification of the key quantitative variables is of significant importance.Therefore, it must be flexible enoughto cover "the freedom of choice in the selectionof themevariables" (Schoferand Stopher 1979,p. 206). Verroen and Jansen(1993), develop scenariobuilding into a systematicprocessin two phases consistingof a thematicprocedureand a translationmodel, relating to four components:
* Themes, * Steering variables,

166

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

* Scenario variables, and

* Model variables Each of the above-mentioned componentsdescribedby Verroen and Jansenexplain from broad to specific values. For example "themes" include five major variables influential areasderived from the external environmentof transport(economics,social
and cultural, technological, infrastructure supply, and pricing and regulation).

Allport (1986) suggests threestepsin the constructionof scenarios: * Identification of scenarioparameters * Combinationof scenarioparameters * Developmentof plausibleparametervalues. According to Faheyand Narayanan(1986,p. 216):
"A scenario displays in a dramatic and persuasive fashion a combination of the future. -typically [it] includes some trends, patterns, and possibilities about to these, conditions in the current environment, and events, assumptions pertaining the dynamics that lead from the present state of the environment to some future state."

According to Saunders al.(1987, p. 324), the trend impact analysisapproachcan be et usedto identify the scenariosub-titlesand:
"Once the most probablescenariohasbeenchosenother boundaryscenarios be can by examining deviation from the core. Alternatively severalindividual theme generated be chosenautomatically or by groups". scenarios can The plausibility and consistency of trends and events are characteristics of any scenario

they determinethe future situations.Vlek and Otten (1987, p. 273), in the evaluation as arguedthat: of the four SDEPscenarios,
"Potentially the most controversialpart of scenarioconstruction,however, would be the comprising both uncertainfactors and conditional modelling of developmentalprocesses " decisionswhosecombinedprobableeffects have to be assessed.

Abeelen et al. (1984) in a study of scenariodevelopmentfor traffic and transport have suggestedtwo scenarios.One is where the context scenariois related to the factors explaining the future, and consists of the description of the problem, selection of

167

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

variables,and preparinga basic analysisand creatingthe prospectivepicture. The other is where the policy scenario covers all actions and measuresby which government seeksto influence traffic and transport,and affects both the scenariocontext and other environmentalvariables. Methods of developing the scenarioelementsdiffer as pointed out by Mitchell et al. (1979) in Jungermann and Thuring (1987). Mitchell et al. (1979) also have suggested that the "bottom-up" and "top-down" approaches to scenario construction are "emergent"and "imposed".

6.3.8 Strategy formulation


Scenariosare developedto investigate and disclose, under different political, technological,social, and other conditions, the various future situationsof an uncertain environment. After the completion of the scenario structuring, a group of related strategies should be formulated where each of the scenarios comes into practical analysis(Nutt and Backoff, 1987).The designand fonnulation of thesestrategiesare of significant importance as they must reflect support for the scenarios.Therefore, they must be interrelatedand consistentwith other decisionsand relate to the scenariofuture events and trends.The decisionsinvolved have technical, geographical,organisational and political environments. Allport (1986) hassuggested steps: two * Isolation of non- important or indirect decisions or strategies: The number of decisionsis relatedto the level of detail of the study (aggregate disaggregate), in but or either casesomedecisionsmight be isolatedrelative to their importance. * Integration and formulation of strategiesin accordance with their priorities from top to bottom.

168

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

Millett (1988, p. 62) presenteda systematic model (see Figure 6.1) for "a typical corporatestrategysystenf' which is used in formulating internal and external business strategies.
Figure 6.1 Corporate strategy system
Goalsand Measures:
Priorities 3) Management's

Options: 1)Wntification
2) Evaluation

I )Qualitative Quantitative and 2) Measures Performance of

Internal Environment

Stratcgy:
1) Plans 2) Actions

....................

External Environment: 1)Customcr/Clients

and

2) Market-place 3) Conditions 4) Allies

Trigger StrategicThinking, p.62. Source:Millet, M. S. (1988), How Scenarios

6.3.9 Examples of scenario analysis


Scenariomodelling nowadayshas emergedin the context of different long term regional, national and international studies,or at industry level. The aim of this section is to discussresearch cited in areas,particularly transport,which have relevanceto the ISLB study. However, it must be stated that most scenario studies cover general applicationsof a non-transportnaturc. Stokke et al. (1990) in a casestudy for the Norwegian oil and gas industry developed four scenarios assessed dynamic externalenvironmentof the industry which led the and to the identification and interpretation of strategy alternatives. Esterhuyse (1992) investigatedthe implications to South Africa for the transitional period and discussed the strategicproblems,priorities and future of South Africa from the political, social, and economicpoints of view for a ten year time horizon. Unsink (1995) developeda simulation model using four scenariosas a baselineand three others as alternative scenarios for the foreign exchange constraints and the economic performances developing countries; two of the alternative scenarioswere of
169

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

and capital flow nature, and another featured technology with an of an economic increasein total factor productivity growth. Goldfarb and Huss (1988) used the seven stepsof the BASIC group for GeorgiaPower's planning processin USA and developed three scenarioswith high, moderate, and low economic growth labels. These were directed to see the effects of a few future assumptionsinfluencing the purposely
company's load and energy growth under the five social, political, economic,

technologicaland physical resourceareas.Norse (1979), with six scenariosof different socio-economic aspects of the OECD and third-world countries, studied the developmentof the world economyin the 20 yearshorizon. Dewhurst and West (1990) with the constructionof two scenariosstudied the Queenslandeconomy for the year 1985/86 and then projected the base run sixteen times for eight years. Another
application was made using the Georgia econometric model by Legler and Robertson

(1976) in the USA. They first developed a macroeconomicmodel on the basis of statevariablesand forecastthe estimatedrelationships national variablesand exogenous through the application of the regression technique. Then three scenarios were developed and applied to identify the impacts of the alternative national economic policies. Galer and Kasper (1982) were assignedby Shell Australia to develop a two scenario survey study to see the effects of changeson long-term economic, social, politicalstrategicand technologicalsectorsfor Australian developmentto the year 2000 to give a "greater depth to Shell Australia 's planning scenariosfor the next 20 years" (Galer and Kasper 1982,p 55). 6.3.9.1Examples of scenario analysis in transport Mitchell et al. (1979) applied the scenariotechnique to the state government railway organisationof Australia. Wolf (1981) in a freight demandstudy used a three

170

Chapter 6 scenario the period energy with first two analysis of the truck

Scenario Approach and ISLB Conceptual Model Development for five Canadian and rail modes the effects investment. of each mode maritime provinces for

of 1980-2000

to estimate and capital

on equipment,

labour,

use, infrastructure,

It was based on nine and technological freight

assumptions The the

the three influential scenario other covered

areas of economic, situation

energy

innovation.

the current raised

of the Canadian

industries

while

scenarios

the profile

of trucking

and rail respectively

and quantified

differently Several traffic develop

against the assumptions. attempts have been made to apply A medium tool ports. for It the scenario developed the number three services. scenarios approach to ports and shipping (1991) aimed of to

and operations. an analytical Australian

term model forecasting

by Tongzon

and characteristics to identify the

ships

visiting

provided

economic shipping

determinants

of the supply

of shipping

These were based on world

trends of ship size, length, beam, and draught, and the effects of these trends on Australian ports. It forecast, with the use of a regression technique,a rangeof shipping for Australian ports. Its findings suggested internationaltrendsdo not have that services immediate impacts on shipping servicesto Australian ports, and thesetrends are most likely under the influence of demandrather than overseas technologicaldevelopments. In another study using scenariosin the ports and shipping industries, Darzentasand Spyrou (1996), by the use of simulation, applied three scenarios(with sevenvariables of ports characteristic).These scenarios focused on the design, implementation and testing of a model for the ferry traffic in the Aegean Islands in Greece.In the first scenario the current ferry fleet was compared with a hypothetical situation where all ships were replacedwith new ones.The secondscenarioconsideredand comparedthe current fleet with having new technologies,and in the third scenariotraditional routes were comparedwith new routes.

171

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

Picard and Nguyen (1987) studied the interregional freight flows for 64 commodities among 67 zonesof Canadaby the use of two scenariosaided by the input-output (1/0 ) technique.Scenarioswere applied to measurethe impacts of changesin transport and economicattributes.In the first scenarioa decrease 20% was applied to the transport of cost for every good originating from the province of Quebec;then the impacts of the In effects of the 110techniqueon the regressionmodel calculationswere assessed. the secondscenariothe effects on personalincome of a 10%increasein the demandfor one of 64 commodities within the province of Quebec were investigated. Dagenaisand Martin (1987) made a long-term container traffic forecasting study for the port of Montreal in Canadain two scenariosand phases.The study included 78 commodity groups for 7 world regionsand 11 North American regions.Verroen and Jansen(1993)
in Holland conducted a study as a "scenario explorer for passengers" with three

sequentialphases.The first was the scenariobuilding module, followed by the travel demandmodel, designedas the core phasewith four main outputs (trips and passenger kilometres, travel times, and car ownership). In the third phasethe evaluation module was designedto identify the impacts of travel demand by using such indicators as travel demand,accessibilityand traffic accidents. aggregate

6.4 Development of the scenario themes for the ISLB study


The environmentof a landbridge contains many factors influencing its viability and operation within different fields, which broadly can be divided into external and internal forces. The external forces of a landbridge include those factors which are outside the transport system,but which support the output of the landbridge system. (e.g. GDP, investment, national revenues, etc.). Internal forces are those which determinethe transportdemandand supply such as numberof berths,capacityof fleet, etc. The variables to be identified and included in any scenario approach must
172

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

adequately reflect and cover different aspects of landbridge demand and supply and also

be applicable to other quantitative techniques.Therefore, to identify uncertaintiesand and Thuring 1987) and then to carry out variables in a top down method (Jungermann scenario analysis, the landbridge environment is divided into four themes (areas of influence): geographical, political, technical, and organisational domain areas. The selectionof thesethemescoincideswith much of the literaturereview.

6.5 Relevance of the scenario approach to the ISLB study


The ISLB model mainly aims to explain how demand for transport and its changesover time in terms of internal and external factors can affect directly the supply infrastructuresystem. in requirements Iranian transportand associated
The ISLB as a long term multi-landbridge project in a transitional environment can be

consideredas a suitableproblem to be studiedusing the scenarioapproach.However, it is first necessary develop an ISM conceptualmodel as a preliminary and reliable to tool to representthe main areasof the study (Ackoff and Saieni 1968).To accomplish this task, and basedon the literature reviews undertakenin previouschapters,a number of key decision areassuitable for scenario analysis must be identified (stokket et al. 1990).These are, what insights must the researchmanagerhave about future business Thesekey issueswere identified and shown environmentto make the decision needed? in Table 6.3 and later expanded into different questionsin Table 6.4. Basedon the key issuesidentified and shown in Table 6.3, the broad ISLB conceptual model was developedunder the headingof the four adoptedthemesdescribedin section 6.4 and the uncertainties(questions)derived from the fifteen key decision areas(see Table 6.4) and attributesas shown in Figure 6.2. The purposeof the conceptualmodel is to illustrate and explain key attributesand their relationshipsin a dynamic way. It can be seenthat the ISLB conceptualmodel is influenced by CAC and Iranian demandfor 173

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

transport, and Iranian supply of transport. Each of these areas will be subjected to At the operationalisationstage (see chapter 7) these decision areas scenarioanalysis. be convertedinto measurable variablesto enabletesting of the scenariohypotheses. will Table 6.3 Key ISLB decision areas. I
2 3 4 5 6 7 8 - 9 10 11 12 13 14 15 CAC and Iran volume of demandfor transport Populationgrowth developments Land enrichmentand significant agricultural-based Iranian domesticpolitics and foreign relations CIS, CAC countriesand Russiafuture political environment Iran and CAC countries' economicgrowth Exchangerate stability Employment in Significant changes Iran's oil and the CAC countriesrevenues in Sectoralchanges human,modesand ports productivity CAC origin-destinatiOn travel distances Iran's transportinvestment Transportcapacities,servicesand competitions T -Transporttechnology 1 Institutional elementsand developments ECO of

Figure 6.2 The broad structure of the ISLB conceptual model.


CAC and Iranian Volume of DemandScenarios Uncertainities Themes: * Geographical * Political * Technical * Organisational. Iranian Sea LandbridgeOutput

Iranian Supply Scenarios

6.5.1. ISLB Model uncertainties


According There are two distinctive kinds of uncertainty, independentand dependent. to Porter(1985, p. 455):
"Independentuncertaintiesare the scenariovariableson which scenarios basedand are basisfor constructingscenariosbecause they are true sourcesof are an appropriate about the uncertainties.Dependentuncertaintiesare resolvedonce assumptions " havebeenmade,and thus becomepart of eachscenario. independent uncertainties 174

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

The occurrence uncertainevents(externalvariables)is independentfrom the internal of on the system(e.g. world oil prices are not likely to be dependent the ISLB variablesof or Iran's actions)while the growth of the economyof Iran as an uncertainfuture event is dependenton oil prices. The list of uncertaintiesin Table 6.4 are derived from the key areas and have been identified as the basis for the development of the trend Theseuncertaintiesare categorised and developed statements all the ISLB scenarios. of under the four headings of the themes (geographical, political, technical and to identify clearly the decision and impact areas ( Shoemaker1992, organisational) Shoemaker1995).They arejustified in more detail in the following section. *or uncertain questions in the ISLB project. Table 6.4 Demand uncertainties Themes Geog Phical, ' in CAC demand ISLB? be Will

1.1 1.2 1.3 1.4 2' 2.1 2.2 2.3 2.4 2.5 2.6 2.7 4 7 -1

Political-3-

through there a significant change Will there be a significant change in Iran's trade ? Will there be a significant change in CAC and Iranian population? Will there be land enrichment and significant agricultural-based developments? , 11 ", . 111 11, 11II1 I11. I11. ,11,11' I1 11 , III1 11 -1 11 ,, -"',I Will there be a fundamental positive/negative change in Iran-USA political elations? Will there be a fundamental positive/ negative change in the CAC countries foreign political relations? Will there Will there Will there Will there Will there be a fundamental positive/negative change in Iran's economic conditions? be a fundamental positive/negative change in the CAC economic conditions? be significant changes in the domestic politics of Iran and CAC ountries? be significant changes in Iranian and CAC employment levels? be significant changes in Iran's oil and the CAC countries revenues? changes in sectoral human productivity?

Organisational Will there be significant

Themes
Geographical' 1.5

" ', t, "i

Supply Uncertainties. 'i,

Will there be a significant reduction in the CAC ori gi n-desti nation distances?

2.8 2.9 3 3.1 3.2 3.3 3.4 3.5 3.6 4 4.2 4.3

investment? in Will therebe significant network priorities ISLB transport in for discounts theCAC countries usingISLB services? Will therebe significant
Technical-,

infrastructure superstructure increase thetransport in Will therebea significant and capacities? Will all rail networks double be tracked? in increase therail rolling stock? Will therebea significant in increase thenumber roadfreightvehiclesof Iran? Will therebea significant of increase thedeployment newtransport in Will therebea significant in technologies ports, of border system Iran? of of and crossings otherparts thetransport ? Will otherlandbridges competition theISLB system with significant provide Organisational 0 .":4 increase theoperational in Will therebea significant performance thetransport of system of Iran? for Will therebesignificant to success ECOtransport economic and programmes integrate states thetreaty? of with member I theCAC countries existing

175

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

6.5.2 CAC and Iranian volume of demand for transport


In Iran and the CAC countries freight transport demand analysis has not been developedas extensively as for passenger transportdue to data difficulties. According
to Ogden (1992, p. 70): "freight demandis a derived demand,the movementof goodsbeing an input in the in productionand consumptionprocess... and is expressed terms of two variablescommodities,and land use".

The CAC countries' demand for freight and their possible choice in using the ISLB is the main reasonfor investigating the Iranian landbridge concept.Therefore, service this demand,together with Iranian demandshould be varied according to changesin indicatorsand environments. different macroeconomic
According to Kanafani (1983, p. 3): "Transportationdemandanalysisis the processof relating the demandfor transportation it. to the socio-economicactivities that generate In this process,the type, level, and location of humanactivities are relatedto the demandfor movementof peopleand goods betweendifferent points in spacewheretheseactivities take place.The resultsof this of analysisare relationships,often in the form of models,betweenmeasures activity and of measures transportdemand".

Kanafani (1983, p. 285) also identified the six following points as important factors for choices in transport demand.They are total quantity, supplies, mode, shipment size, frequencyof shipment,and reorderpoint. 6.5.2.1Geographical
1. Changesin the CAC countries' demand through ISLB As with many former Communist bloc countries, the CAC republics aim to diversify their trade with free world market countries to obtain a higher standard of living and technology for better quality products (Michalak 1994). Therefore, they are recovering from the sharp decline in trade in 1992 and 1993 by expanding their exports and imports in 1994, and particularly with neighbouring non-CIS countries. The foreign

trade commodity structure of thesecountries as they pass through a transitional stage 176

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

appears to be declining to a few classes of exports: raw materials, base metals,

(UNCTAD/UN 1994). chernicalsandenergy For a long time, Iran and the former Soviet Union as neighbouringcountrieshavehad a bilateral transport agreementfor the transit of foreign trade goods using the transport modes and ports of each other. Iran rather than the FSU mainly was one of the major users(forboth Trans-SiberianRailway and also Volga-Don Canal). The recent demand of the CAC countriesfor transport through Iran is essentiallydue to the collapseof the former Soviet Union and the new political and economicregion. Factors affecting the changesof CAC demandthrough ISLB corridors (question 1.1) and are both enclogenous exogenous.The latter are those transport supply influential factors, which generally the CAC countries have options to use or not and also their
quality and reliability may stimulate the future CAC demand.

indicatorse.g. The endogenous factors are mainly national and regional macroeconomic size and economic dynamism of the CAC countriesincluding political issues;success of the CAC countries will expand the economic structure and consequentlyincrease either imports or exports or both which the ISLB can accommodatepartially to the flows of these demand; while failure of these countries in the current political and can lead to the decline of the production and consumption or economic approaches the political system.They are factors related to the ability of eachof entire collapseof the CAC countries to perform moderate, stable and growing economic activities, production and consumption functions. Trade from CAC countries through Iran may changeover time due to the differencesin commoditiesand goodswhich, accordingto Ogden (1992), may vary according to the state (bulk, general, etc.), mass, volume, number of pieces,portability, fragility, hazard,urgency and irregularity of shipments. Each of theseattributes can affect ISLB perfonnancesin tenns of annual tonnageand

177

Chapter 6 the utilisation of transport

Scenario Approach and ISLB Conceptual Model Development supplies. The impact of the CAC countries' trade on a

limited transport infrastructure

and services will also affect the quality of ISLB services.

Another major area which may have an influence on the demand for ISLB services is the extent of production costs and prices in different industry sectors. Government

in a centralised developing country like Iran may also affect the volume of expenditure trade and costs of production.

Iran's foreign and domestic trade (question 1.2) despite sharp variations during the war has followed steady growth since 1988 when war was ended. While domestic

by production capacities of the firm, intermediate import production was restricted level of rainfall, natural climates e.g. flood and earthquake, etc. it does not seem goods,

to decline andit doesgrow slowly undernormal conditions. The foreign trade of Iran (question 1.2) can be boosteddue to the increasein foreign foreign investmentor the significant improvementof the political issuesand revenues, relationshipsof the country with the USA as it can give more and longer term foreign tradecredits to Iran.
2. Population growth Demand is largely determined by the level of population (question 1.3). The these countries (CAC countries and Iran) is below the economic activities of each of future population growth. Both Iran and the CAC countries, for religious reasons and birth following the liberalisation of tight control policies after the collapse of the former, USSR in the CAC countries, are among the moderate or high fertility rate countries. Therefore, population may be an important issue in changing the volume of future demand for both Iran and the CAC countries (Sampson and Farris 1979, Lerman, Garcia-Garcia and Wichelns 1996). High population rates, may affect negatively the success of the developmental programmes as domestic production and revenues are

178

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

limited and usually cannotmatchthe demand.This is particularly important as Iran is in the processof reconstructionafter the war and the CAC countries are in transition to marketeconomies.
3. Land enrichment and significant agricultural-based developments

Both Iran and the CAC countries have vast areasfor agriculture on which they are largely dependent. Therefore,any developmentprogrammesuch as building dams will result in a higher demandfor land and better irrigation through the application resources such as fertiliser or through the use of tractors and agricultural machinery and may causehigher domestic production, imports and in particular agricultural-based exports (question1.4).

6.5.2.2Political
1. Iranian domestic politics and foreign relations

Iran is passingthrough a difficult transitional period after a long war with Iraq and continuing problemsin its foreign relations with the USA. According to Tarock (1996, p. 150)the USA is accusingIran in five areasas: * its allegedefforts to acquirenuclearand other weaponsof massdestruction; * the threat that, once these offensive capabilities are acquired, could (in fact it is assumed will) be posedto its neighbours,i. e. the PersianGulf states; * its alleged engagement and sponsoringof international terrorism with a view to in the clestabilising PersianGulf statesfriendly to the USA; * its oppositionto the peaceagreement betweenthe PLO and Israel; * its allegedpoor record on uman rights. Disagreements with Arab statesin the southernPersianGulf region about three Iranian islands appearunlikely to have seriouspolitical repercussions. The more economically strongerIran the more advantageous stable will be the PersianGulf Arab statesas and

179

Chapter 6

Scenario Approach and ISLB Conceptual Model Devclopment

they are usually third and fourth trade partners for Iran's foreign trade (EIU 1994b,

Amirahmadi 1996). The future of foreign relations with the USA (question 2.1) will dependlargely on the long tenn interestsof both countries.For the USA, Iran has a unique political location in the Middle East after the collapse of the USSR and the creation of and role independentCAC countries to its north (Askari 1994,Tarock 1996).Since the end of the war with Iraq in 1988, Iran has followed some aspects of liberalisation and democratisationin domestic politics, developmentof the private sector, assurance of foreign investors and stability in the oil and gas industry. All of these developments the future environmentof the ISLB (Agazadeh1996,An-drahmadi1996). would affect A changein the macro-policies of Iran with an impact on foreign relations may be tackled in the electionsin 1997where there is the likelihood of non-clerical candidates campaigningto becomethe future presidentof Iran (EIU 1996).The nature of Iranian domesticand foreign politics can significantly affect both demandand supply for ISLB services.
2. The CAC countries foreign political relations

in Changes power betweenpolitical partiesin the CAC countries(question2.2) can the volume and direction of demandfor the ISLB. On a larger scale, restrict or stimulate in the Russian political environment may lead to the restructuring of a new changes "USSW' with strong border crossingcontrols and restrictions on trade with or through Six neighbours. of the CAC countrieshave an Islamic heritage,while two (Georgiaand Armenia) areChristian. All of them havedifferent interestswith neighbouringcountries (and other adjacent Russian autonomousrepublics) including ethnic, religious and language issues which may, as external factors, destabilise the future of the geo-

180

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

of the CAC countries as the "Islamic-Christian crescentbetween economic conditions Asia and Europe" (Barylski 1994). 3. Changesin economic growth of Iran The developmentand improvementof the economicsectorin Iran (question2.3) as a (in particular oil with 8.8%, and gas with 14.9% of total world mineral-rich country
depend upon imports and foreign revenues (World Energy Yearbook 1996, output) Agazadeh, 1996 and Rafsanjani, 1996). In Iran real GDP in 1994 rose by 1.8%, while the expected future rise is 2% per annurn (EIU 1996, Amirahmadi 1996). Iran's future long term growth depends on the outcome of the privatisation programmes started in 1988. Intermediate domestic-related industries are also significant in international trade.

A sharpdevaluationof Iran's currency at the end of 1993 and its volatility in 1995led to the exchange control of the Rial. Iran's foreign aid and debt repaymentsare influencedby its political relations with the The USA continuesto impose USA and its oppositionto the Arab-Israeli peaceprocess. EU members and Japan against aid for macroeconomic sanctions and pressurises developmentprojects in Iran (EIU 1994b). Since 1995 the successfulreschedulingof Iran's debtshas clearedabout $4 bn per year up to 1998.The repaymentof the foreign debts was accomplishedwith allocating about 15%-20%of foreign exchangeearnings (EIU 1996, Amirahmadi 1996). However, insufficient and increasing demand for has led to frequentdevaluationof the Iranian Rial and has producedinflation, currency the which increases costs of domestic and internationalservicesincluding the transport provided in Iran. The fixed Rial exchangerate in Iran after the Islamic revolution, and in particular with the beginning of war, has been abolished and gradually multi-exchangerate systems havebeenadopted(EIU 1993b).The recognition of the single U.S dollar exchangerate

181

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

of 1993 of about 1600 Rials, and the policy of paying for imports directly with export revenues,led to the setting up of the Foreign ExchangeMarket Regulation Committee controlling all imports. This, to a great extent, indicatesthe instability of the Rial and discouragesforeign investors or Iranian citizens living abroad from investing (EIU 1994b).
4. Changes in economic growth of the CAC countries

The CAC countries,as part of the former USSR and easternbloc, have recovereda large amount of aid from international bodies (IMF and World Bank) and aid from individual countries(mainly Germanyand USA) in the form of credits and grants.The and assistthesecountriestowardsa market economy aims of this aid were to encourage and a democratic political system, and to enable reconstructionand developmentof their weak macroeconomicposition and infrastructure,but progresshas been limited (Michalak 1995). Among the CAC countries, the Central Asian Republics are much richer, mainly in gas and oil, but due to various domestic and regional factors their future prospects uncertain(Gharabaghi1994,Barylski, 1994). are Since the collapseof the former USSR, all CIS countriesincluding the CAC countries have taken significant measures leave the Rouble zone and adopt their own national to currency.Because the previously highly integratedeconomyand inter-republic trade, of currency changesaffect greatly the volume and value of the foreign trade of these countries(question2.4) and their balanceof payments(UNCTAD/UN 1994).
S. Changesin the domestic politics of Iran and the CAC countries

The economiesof both Iran and the CAC countries (question 2.5) dependon the outcome of presidential and parliamentary elections, although there is not a strong parliamentarytradition of political parties. In particular, the CAC republics are new to the experience of democratic systems. Consequently, this implies any change in

182

Chapter 6 government development or parliament

Scenario Approach and ISLB Conceptual Model Development may (Staff significantly 1996). Both affect Iran the ISLB operations countries and have

programmes

and the CAC liberalisation

experienced different types of economic and political privatisation (Esterhuyse 1992). According

by moving towards of 1991

to Hunter (1994), at the beginning

about 70% of Iranian industry privatisation policies

was owned by government.

The new and partial re-

of Iran, after the war with Iraq ended in 1988, resulted from burden of the massive expansion of the government in 1979 (Payam Emrooz 1995). The aim of

pressure to reduce the financial sector after the Islamic

revolution

the war in 1988was mainly to free the economyof the strict controls governmentafter that were imposedon it during the war (EuropaWorld Yearbook 1994a). The concept and processof privatisation has a long record in the history of Iran with differing levels of success and failure since the constitution in 1906.This has resulted from the poor legal basis and implementation, and also becauseof contradictions betweenreligious and national beliefs and capitalism (Razaghi 1994).For instance,the 1988 led to a concentrationof power with the bazaar new privatisation processafter individuals (traditional private marketers in Iran) or with some government-based to (Amirahmadi 1995).Following such experiences there is resistance the privatisation industries(EIU 1994b). of the major government-owned In the CAC countries, (question 2.4) privatisation at different levels has been taking the disintegrationof the FSU in 1991; in the Central Asian countries it was place since minor in the first year (Ledgerand Roe 1993,Wright 1994,Rashid, 1994). Georgia in the Caucasus and four of the Central Asian republics (excluding Turkmenistan)will be servedindirectly by ISLB services.This meansthat any serious local disagreements, especiallyalong transit corridors to other countries,would disrupt ISLB services.The CAC countriesmay in future have territorial conflicts similar to the

183

Chapter 6 war between Azerbaijan

Scenario Approach and ISLB Conceptual Model Development and Armenia over Naghomo Karabakh and South Ossetia, and

Abkhazia between Russia and Georgia. This has resulted in a stop to rail transport over between Iran and both of these countries as well as Georgia, and a disruption of road

inter-republic trade flows. There has been a ceasefire since 1994,but fighting and rail disputesis alwaysa possibility and thesecountriesare along from a due to deep-seated permanent settlement(ElU 1994a). in 6. Levelsof employment Iran and CAC countries New investmentaffects both quantity and quality of employmentand brings higher in for individuals and nations.The level and changes employmentof a country earnings in (question2.6) may reflect the changes total physical demandat national and industry levels. Employment is also affected by the education of the labour force and sector better deployedtechnology.Both Iran and CAC countries are developingnations with future programmes and investment which will extensive create employment.

Employmentin Iran after 1988was increaseddue reopeningof someindustries.In the CAC countriesthe employmentrates declined due to the structural changescausedby the collapseof the Former USSR and the loss of foreign markets. 7. Future revenuesof Iran and the CAC countries Foreign revenuesare key sourcesfor acquiring imports and the new investmentof Iran and the CAC countries. Iran and CAC countries (in particular Azerbaijan in the latter) are amongsome of the world's largestmineral rich countries(oil, gas, gold and uranium) (Starr 1996). The economy and foreign revenuesof these countries depend mainly upon the mineral resourcesof oil and natural gas, and agricultural products (question2.7). The former is under the influence of world demand,and their supply will be key drivers of demandin thesecountries.Therefore,while thesecountriesare rich in

184

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

they will need high employment and investment for some mineral resources, exploitation of raw materialsor agricultural products.
6.5.2.3 Organisational 1. Changesin sectoral human productivity According to Gillen et al. (1985, p. 67): "Perhapsthe most commonly usedmeasure the productiveperformanceof firms is of output per employee".

Therefore,at national levels the productivity of different economicsectorsis considered in the form of peopleper tonne output or peopleper unit revenueor cost (question4.1). One influence over productivity is systematiceducation and training programmesfor The relationshipof this theory is supportedby research(The EconomicReview people. 1995).While generaleducationlevels in the CAC countrieswere better during the time of the FSU, in Iran there are various programmesto systematicallyreduceand replace through training. With the availability of data, different productivity unskilled workers typescan be preparedand appliedto the analysis. measures the above-mentioned of

6.5.3 Iranian supply of transport


6.5.3.1 Geographical 1. CAC origin-destination distances

The CAC countriesmay use different types of modes(i. e. rail or road or both ) and ports and border crossings according to their options and availability of the ISLB corridors. Natural barriers like mountainsand rivers and also the importanceof urban areas as production centres influence the directness of routes. The present Iranian transport infrastructurewas designedand constructedto serve Iran and to facilitate a reasonableaccessfor the current 24 provinces. Potential exists, however, for further cuts in travel distance(question 1.5). The geographicaldistancebetweeneach of the 185

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

CAC countries and the ISLB landbridge ports greatly affects travel time and the consequently costsof freight transport.Therefore,the presentroad and rail distances CAC countriesreflect the level of demandin eachof thesecountries. of routes serving Consequently,any shortening of the origin-destination distance via the ISLB may stimulate additional freight volume through Iran (Jin 1993). For example, a major
reduction in distance is expected to be seen by rail for Central Asian countries between

Bafgh and Mashad through the desert. Improvementsin time and frequency,and the costs of a freight journey if necessary,in particular for the rail network that when is less developed.Improvementin ISLB servicesmight be a strong comparedwith road for more direct north-south rail links for both Central Asia and Caucasus reason comparedto regions.This can make theseservicesmore economicaland advantageous
other routes. such as the TSR and TCR.

6.5.3.2 Political
1. ISLB Network transport investment

The transport system of any country has vast areas of investment in the carriage of

freight. Transport investment is long-term and strategic in nature, as passengers and In tenns of geography, urban and inter-urban contexts are well as expensive. distinguishable.In physical terms, both the above-mentioned categorieshave their own special network infrastructure including modal and nodal, mobile vehicles and fixed facilities and technologies,and traffic system.Each of the modesand nodeshave their own physical elementsthat signify the capacity performanceof that mode or node. Therefore,in the transportsector,like the generaleconomy,investmentsare possiblein four different forms: capital (networks, vehicles, etc.), labour, entrepreneurship g. (e. educationand training), and land to improve the quantity and quality of the transport services (Bamford, 1995, The Economic Review, 1995). When transport investment 186

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

decisionsare made,all thesefour areas,either for freight or passengers, needto be seen weak links will negatively affect the expected performance. as a chain otherwise Among different elementsof an inter-urbantransportsystem,network, ports and border investment are long-term and very expensive (question 2.8). Therefore, any crossing future ISLB project needsa strongjustification to receive investmentfunds priorities other national transport,industrial, or agricultural, sectors. comparedwith Iran has mainly used its own financial resourcesfor transport investment.In particular Islamic revolution, it has used less foreign aid. The annual Central Bank of after the Iran reportsrevealthat investmentin transportinfrastructureand vehicles accountsonly for about 5.7% of the total national investmentduring 1989-1993(question2.8). In an to Kaufmann (1962), approximately one-third of public earlier period, according investmentwas allocatedto transport. After the collapseof the USSR and the new role of Iran as a transit land, UN agencies the Trans-AsianRailway and ECO becameinvolved in Iran's transport associated with infrastructure. Particularly, the opening of the ECO Trade and DevelopmentBank by transportprojects which can the three main membersis aimedpartially at Iranian-based Iran in its regional long-term plans (OPECBulletin 1996). assist The landbridgeconceptrequires Iran to look at its transport systemat a more regional rail than a national level (e.g. the 160 km Mashad-Sarakhs connectionto Central rather Asia during 1992-1994 was due to the new transport investment priorities, while renewalport projectsreceivedthe highestresourceallocations(PBO 1993)).
2. Governmental discounts for ISLB services

Transport infrastructure including ports, rail tracks and rolling stock are owned, operatedand maintainedby Iranian governmentalorganisations.While ISLB services have mainly two private operational costs and fixed government charges and dues,
187

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

similar to the KEDAM landbridge, the government may offer some generous discounts

to all user countries(question2.9). 6.5.3.3Technical


1. Future transport infrastructure and superstructure capacities The capacity determinants of landbridge supply include all physical elements of

the modes and interfaces as well as the carrying and handling technologies. An unexpectedlyhigh increasein demand for the ISLB transport systemwill produce an imbalanceand put the existing infrastructureunder pressure meet the demandwhich to may lead to malfunction and high costs.On the other hand,expansionand upgradingof the landbridge supply is time consumingand expensive.Therefore,as the ISLB offers international services,on the one hand, it must haveenoughshort term surpluscapacity
to be used and prevent excessive demand on the system and a systematic increase in the

ISLB physical capacities(question3.1). For Iran, road and rail networks have played an important role in the distribution of domesticand foreign trade of the country and, therefore,roadsare reasonablywide and well developedwith hard asphalt surfaces,while a single track railway connectsthe capital Tehran to different partsof the country. According to their aim, various studieshave investigateddifferent featuresin terms of transportcapacity.Sheffi et al. (1982, p. 212) noted that:
"rhe capacityof a transportationfacility is the maximum flow that can go through the facility".

ESCAP (1967, p. 16) definedport capacityas:


"ne capability of the port to handlea determinedquantity of cargoeswithin given " periods of time, usually one year.

ESCAP (1967) suggested following five measures port capacity: the of


* Length of berths

188

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

* Amount of surface water

* Depth of water * Shedand storagespace * Equipment Imakita (1978) consideredports as transit gatewaysand not end points. Their capacity cannot therefore, be measuredseparatelyfrom other links on land or sea. Imakita as that factors affecting port operationscan be categorised follows: considered * Navigational aids system * Quay handling and transfersystem * Storagesystem,and * Co-ordinationwith inland transport. Sinclare (1977) suggestedthat quay crane capacity, transfer capacity, and storage capacity in cargo handling operations determine the total performance or practical capacityof a port. Kondratowicz (1992) in a study about modelling seaportsand inland freight terminals of observedthe five following broadclasses variables:
* Terminal resources * Storage facilities

* Cargoes * Meansof transportationand * Servicesystems. Tongzon (1995) has employedthe six following principal quantitativemeasures port of perfonnance: * Total throughput * Number of commercialship visits

189

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

* Vessel size and cargo exchange * Nature and role of the port

* Port functions and * Infrastructureprovided.


2. Double track rail

The presentISLB rail modecapacityis restrictedmainly by the numberand capacity of rolling stock as well as by the single rail track. Gradients, signalling systems,the numberof stationsand daily availabletrains are also important issues.The rail network of Iran is mainly single track but doubling and renewingthe lines hasbeenstartedsince the revolution (question 3.2) although progressis slow . The presentdouble line rail project can effectively increasethe annual perfonnanceof the ISLB by increasingthe speedof trains, reducingcosts,increasingthe availability of block trains, etc. as long as it is paid for by further investment.
3. Rolling stock

The supply of train serviceswas studiedby RamamohanRao and Sriarnan(1985) and the five following variables were identified as capacity determinantsof the rail mode: * Wagonsin usefor freight movements * Wagon loading * Speedof freight trains in train kilometresper enginehour
* Turnaround time of wagons in days

* Empty wagon movements (percentage of empty wagon kilometres to total kilometres). Lansdowne(1981) in a study of the rail freight traffic assignmentfor the FederalRail
Administration (FRA) of USA has employed three types of variables:

* Distanceand ownershipof eachlink 190

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

* Carrier accessibility to each node

* Origin-destinationdemandmatrix (number of carloadsor tonnes of freight between eachpair of O/D nodes). The inadequacyin the number of rolling stock (locomotives and wagons)has played a in the dissatisfactionof the TSR landbridgeservicesdespitethe fact that significant role
the FSU was among one of the main manufacturers of these vehicles. With the present

and renewing policies the ISLB services seem to be inefficient and rolling stock deployment of the CAC countries and other train-owned operatorstogether requires the imports and developmentof the domesticrolling stock plants (question3.3). with
4. Road freight vehicIes

Heavy road freight vehicles are effective in providing the ISLB servicesrequired particularly by the three CAC countries close to Iran and are also the most important
Iran's foreign and domestic trade. The present fleet is over 15 carrying elements of

years of age. Therefore, an effective and comprehensiveheavy road freight vehicle renewingsystemmust supportISLB operationsand reducethe cost of dervicesservices (question3.4).
5. Transport technology

The transport technology of the ISLB covers a range of activities from port installations and equipment to the road and rail fleet and traffic control systems. Changes transporttechnologyin general,and innovations in handling operationsand in in particular, have taken place extensively during the last three decades rolling stock have beenobtainedin after the developmentof containerisation.The best achievements the developedworld by applying thesechangesto all modes and interfaces.Iran, as a developingnation, has not beensuccessfulin keepingpacewith the supply of transport technology and in particular for foreign trade demand in the major ports and border crossings(question3.5). In this context, primarily quay cranesfor generalcargoesand 191

Chapter 6

Scenario Approach and ISLB Conceptual Model Development

installations (e.g. for different types of other effective and efficient electromechanical

bulk cargoes)train signalling and communicationsystems,etc. are of significance for increasingpresentcapacitiesand reductingcosts.
6. ISLB competitors

under different scenario In chapterfive it was seenthat ISLB hascertain advantages


in distance, transit time, ports infrastructure, container facilities, etc.. Both main themes ISLB (TSR and TCR) suffer from long distance rail legs and supply competitors of the levels of services but they can offer more co-ordinated and skilled of rolling stock and management or politically freight rates (question 3.6). By prioritising the reduced

Bafgh-Mashad by-pass rail link the ISLB transit time and costs can be improved greatly against similar competitor services.

6.5.3.4Organisational
1. Increase in the productivity of transport system

How much the present Iranian labour intensive transport sector is capableof transit services for CAC countries dependson the providing efficient national and future conditions of ownership,managerialstyles,the quantity and sophisticationof the (question4.2). While the road mode and transporttechnologyand governmentpolicies domestic and foreign trade of the country, rail has the most ports carry most of the The quantity and quality of ISLB servicescan be promoted poorly educatedworkforce. by systematicinvestmentin the educationand training of labour and management all at levels of sector(Wollf 1981, The Economic Review 1995). 2. Role of the Economic Co-operation Organisation The Economic Co-operation Organisation(ECO) as a regional and international body since the collapse of the FSU in 1991, has played a significant role in the in particular for new memberstatesand observers of promotion of regional affairs, and 192

Chapter 6

Scenario Approach and ISLB Conceptual Model Development The functional system of the ECO is based

the CAC countries as well as Afghanistan.

the three early members and Iran is the centre for transport matters. It works to on ECO transport policy for the 10 present member state, but due to establish a common

the physical and regulatory differencesbetweenthe three main membersand the CAC there is massivework to be done.Iran hasbeenkeen in promoting the ECO in countries in particular in transport,by providing funds and other physical aids like a general,and for the shipping division of the organisation(question3.6). cargoship

6.5.4 Outputs The output of the ISLB conceptualmodel is the Iranian transport supply in the form of annualtonnage(and also, where available,costsandprices for users).

6.6 Conclusions
This chapterstartedby consideringplanning underuncertainty.Then as a means for gaining more insight, the scenarioapproachwas reviewed in detail. It appearsto be techniquefor the further analysisof the ISLB. Certain themesas an a suitableanalytical initial part of a scenariostudy were developed,and essentialkey areasfor future ISLB decisionmaking were identified, examinedandjustified. The ISLB conceptualmodel was then investigatedunder the four basic geographical, themesand technical and organisationalthemes.The scenariocharacteristics, political, key decisionfeaturesnow will be developedinto an operationalmodel capableof being analysed.

193

Chapter 7

Operationalisation

of the ISLB Model

7. Operationalisation of the Iranian Sea-landbridge model


7.1 Introduction
The conceptualmodel needsto be convertedinto a working model for the later stages of Therefore, it is necessaryto specify and define all the variables required for the work. and further phasesof the study, the broad variableswhich were conceptualised developed in previous chapters. The operationalisation determines the final step in preparing, data (Madison 1980,Smith 1991,Neuman 1994). and splitting up processing The Iranian Sea-landbridge(ISLB) scenario modelling and analysis will consider the (basedon Mercer 1995a): following four main headingsseparately
* ISLB study characteristics

* Time period and level of aggregation * Descriptionof ISLB methodology,datacollection and sources * Selectionand identification of ISLB scenariovariables.

7.2 ISLB study characteristics


7.2.1 Time period and level of aggregation
The modem conceptof the ISLB both in theory and practice is very new and was only initiated after the collapse of the former USSR. It is intendedto cover the study as a long term policy project up to the year 2005 and,basedon 1979-1993data,it will be considered longitudinal time seriesscenariostudy (Babbie 1995,Neuman 1994). as a The future time frame of a researchstudy dependsupon its objectives.Long term projects fall within 5 to 20 years,but it was decidedto limit the time frame of the ISLB commonly between 1994and 2005 and allow reasonable application of the scenariotechnique(Porter

194

Chapter 7

Operationalisation

of the ISLB Model

et al. 1991). The research period must also be taken into account (end of 1993 to the

beginningof 1997). The study is based on the use of data for Iran during 1979-1993and the most recent (1985-1993).While it datafor CAC countriesbefore and after their independence available havebeendesirableto have had data from a longer period for estimatingthe models, would
the data for earlier years were either not available or too incomplete to include in the ISLB

Consideration of the time period in the study of scenario developmentsand models. forecasting is fundamental, and it is planned to construct the base year scenario with forecastsinto the future as a "single-period model" instead of breaking the horizon into fractions (multiperiod) (Mitchell et al. 1979). According to Kanfanani(1983,p. 90):
"ne purposefor which demandanalysisis performeddeterminesthe level of detail to which it needsto be carried out".

Goods traffic modelling can be analysedat different levels. Some models analysecertain links while others apply to a region or country with microeconomicand macroeconomic natureshaving disaggregate aggregatelevels of detail (Quinet, Reynaud, and Marche or 1983). The level of detail selectedfor the ISLB model is aggregatefor several reasons. They are the lack of relevantstudiesabout Iran and CAC countries;the long time period of the study; application at national and international levels; and broad geographical dimensionsof the study (Cole and Chichilnisky 1978). The ISLJ3is both a qualitative and quantitative study. The qualitative nature of the model will help to understand and develop different aspects and impacts of the landbridgeon the transport system of Iran, while the quantitative model attempts to provide greater
integration and internal consistency.

195

Chapter

Operationalisation

of the ISLB Model

7.2.2 Data measurement and level of accuracy by or Data,according its nature, be measured differentscales units(Neuman to can 1994).According Wilsonet al. (1979): to
"Measurementis a procedurefor classifying individuals, groupsor other units and them into previously defined categories". putting

through for Demand transport the performance a transport modecan be measured of or indicatorsin terms of time, value, volume (tonnage),distance,cost or a various According Quinetet al. (Hicks 1977). to tonne/km. these units.e.g. of of combination some (1983,p. 141):
"As far as transportoperatorsare concerned,the quantity of freight is naturally measured by weight, or more exactly mass,in tonnes,quintals or kilograms,but this measurement to what is involved: net or grossweight; actual or variesconsiderablyaccording tonnage". chargeable

is to Thetradepassing variable the of throughISLB corridors considered bethedependent in functionof capacities services the transport networkof Iran.The metric and studyanda the is the main unit which is commonly usedto determine infrastructure trade volumeof
of transportnetworksand interfaces.Therefore,the unit of measurement and superstructure demand is the total metric tonnes handled and transported annually. Different of factors using different units of measurements(mainly costs) are the explanatory independent the ISLB demand. variablesof Consequently, output of the ISLB demandmodel usesthe following two indicators:. the * Freight demandin tonnesby mode and interfaces * Freight transportsupply in termsof infrastructure- capacitiesand frequency. A regressiontechnique is selected for the quantitative analysis of the ISLB study. To developthe operationaldefinitions for ISLB variablesthey were listed and classified under Their numerical valuescan technicaland organisationalthemes. the geographical, political, be seenin Appendix 2.

196

Chapter

Operationalisation

of the ISLB Model

7.3 Description of ISLB methodology and data sources


7.3.1 Data sources Someof the data was collectedbefore the beginningof the study but most of it was during the research period through direct correspondencewith different received in on of administrativedepartments Iran. Existing research the operationalaspects transport in Iran is poor. Published data mainly comprisesthe annual performancereports of the different ministries and governmentalorganisations.Some organisationslike the Central Bank of Iran, the Statistical Centre of Iran, and the Programming and Budgeting Organisationwere more forthcoming with datathan others. The nature of the researchmeansthat the ISLB study is heavily reliant on recordeddata included in the publications of international organisationssuch as the United Nations, the World Bank, and the various Iranian agencies.The newly independentCAC countries, formerly part of a centralised Communist administration, had the most limited data (Fischer,Sahayand Vegh 1996). There were also some limitations in the data available from Iran. Sourcesfor the datacollection can be seenin Table 7.1 .

7.4 Selection of variables


This sectionis concernedwith the determinationof the variablesfor the ISLB models and multiple regressionanalysiscarried out in the next chapter.The identification of the key quantitativevariablesdue to the subjectivenatureof the scenario-generation technique is of significant importance(Schoferand Stopher 1979).

7.4.1 The dependent variables


For the purposeof this researchthe demandfor transportin Iran is basedon three broadcategories:
* General cargo and oil products (foreign trade)

197

Chapter

Operationalisation

of the ISLB Model

* Domestic trade

* CAC foreign trade 7.4.1.1 General cargo and oil products (foreign trade) Iran is a semi-market-economy with a high degree of government control over foreign trade.The economyof Iran combinestraditional economicvaluesbasedon culture and religion with a blend of both capitalist and socialist economies.Economic health, including, therefore,foreign trade, is largely dependenton a relatively high population, a vast land areaand crudeoil revenues. The total volume of Iranian generalcargo and oil products is a major flow through ports bordersas shown in Table 7.2 (seealso section2.3.4.6).The methodologyused and across to calculatethe annualtonnageof the generalcargo and oil productsof the foreign tradeof Iran for the period of 1979-1993is shown in Table 7.2. The export of oil is a completely issue,not directly relevant to this study, since it differs in the type of production separate
management, the nature of the transport (domestic consumption, production and the

exporting is carriedout completely by pipeline) and the export terminals. The volume of Iranian generalcargo and oil productsis shown in Figure 7.1 as a function time. It indicates sharp reductions in the volume of trade at three different times. The of first drop was due to the Islamic revolution in the period 1979-1980,which was followed by a recoverywith the start of the war in the middle of 1981. This growth continuedup to 1983when oil prices fell significantly causingthe seconddownturn in this trade. Because of the war, this situation continued until the end of the war in 1988. After the war, reconstruction resultedin continuousgrowth exceptfor a minor decline in 1992.

198

>
:: r cr : a. Q. 80 (D 0 +b 1 0 9 = 0 0 CJ. g .g 0 0 0

g G'n
rA b

n-

5,1 0=
0 P 0

:l
:j

tj 0

<8

P,

91

t;.

(D :i fp

0
,j

Vz :10 .

e
0
(D 0 -h :1 0

2 :i Lh

0 9 Cr CL Z. CD > n

C-D 4 n.

09
0

I ci 0 9, -

9
. 0

2Z(A ci.>CJ
C, (D g E:
(D ,0 CD-

PD r

3-

e :j
0

P
0

me

r.

OQ
(D (D

In .
c2,0

(D

Gn

0 A
9p -

5 c2.
0 (D

E
(D 80 ;g :1

2 J, dn* V :2 , (,p
0 00 3

9 ,<

0 Gn

<

Z:

CD

ii

CD (D

0A2<o

E3* = :i c2>-

A)
c) ;m9205 -h

> cz

cx
--1
(A

e E.55 e 19, 25. 5:


1 00
>d

21

c2.
Z

CA cO

n-% 91

9 :E 'ZJ

1', '
,i -4

R.

9 V

8, 9

(blo
txf

t ..
(Z

0 eb
(D $., N

p
0

CD -. , v'

0
1

a =r CJ - c: 2;-. %0
000 -b kb -p 0

r .

(D r

50 rn cn e" 0( ti

:i E
D D Gn 0

C)
0 Z,

't"o3

1.
C-2.

0
00

501

E 0 Ici
cr (D

ei>
oi : 0

<

0
0

>
ri
Ici

.a

2 P

En

0 K9

cn

(0,0

nt

(1) uj

43 c02

2
F;

:3, p; m n GM
926 0

1.
9
w

'0 -, 1
0 a0

12 z
0
0

2 po 0' >
0 c2. u',
0

0 0% %< r: o, m ci '

.1in' m! jj1 %0 C) z
b
(

00%..o
..

0 e K0

cr

iE
cn 0 :i

(Z
2=
:iggF

a. 2 c z

e 7. rA

pp
?

cn

%m

0
.

`,

Zr 5%00

V) ,

ti 0

199

(D < 0 0 (D

CD-

R.

:2.

in

CZ.

926

Zr

.,.

Z:
cz

w- - 6-- s. %0 -.
5 4 0Z 0 2 Z 0. 5 Z Z --, cz Z- l< 'Z ; 0 i

mm c: CZ

('

m 21

p 0

0-w

cn

c02 >Z2 e

j 0

91 ZI.8 0

0, 0.0,
td 0

%0%0ZZ ZZ Z c7. w0 0 gr. %DW lt"i0 0 %0 te 00 tA P r- "". , .0

2 5o
ll

0 z u- ci zz o ,0,0 Ei: 0 00 -

N) 44.1

?
0.

EL

%<
X0

-9 51 9

000

rZ EI

E; zi
m'

< 0

5f 0 e

ti 0
51
CD

5 -E: -c 9 j*
21

C %', Co)

c C, Zg

a p

92.
g

ff P- P EL trZ 0 Cb cr, 2, 01 0 0 0 00
pe hpe

0-;

C)

(D

m- 5,%0 ti e-r, oo -Z A:0 0 4 C


eb 00 CJ 0' 1--4 ,

r0 cr CD 0 % <2 O= C.

.e

%A-0

!c-

-9 0 -, 1
2

GM

GM

0 g
%D %0 1> 926 0 g

0,2 c3
(.,

1
,'

< Z o . 0cC, .0
120 ZO %0 Co 0
D

2 -4 0 rr

Ei. 2 cn iri ci

m
A,

po

Ci C.;

n "D

0
3

0 G4 .

te c3,

-b2

9 ?o o
<

N)

< CD 0

2 1

c02
9

>
9

tr

<

5'

c) 0
ti
0

C]m 0

cn me. -. ) 0
ci
(A

,.,

?, %7
0

R5:

g.
(A

Ici 0

c,

9 e Gn
0

Ild
c02 , L.

e
Gn
0,

;5
mz. "!
2. rr

f'i> m cn r
0

-M

9 0. %

200

EIci co
06

cn

ti Si

s"

1 h"

1 h-t N- %c o

-4 %0 o

. s%0 o

s" c

>- 0 140 %.

oo0022000000og-a
D o
t-i ,0 ta C CY, 0, 0, 1.0 t, ) LA ,) 0 t, ) tJ -j %0 (A (A C> (A

0, r.
co

-4
c (> CY% tA w W (> CD 00 CD -4 %0 O\ 1.0 1 _j

-pt' W

N" lu

. (D -1

CDNC> C% UJ 1

12

b-A

zr. '0 02

00O\00%0 -a (: s ! 4 00(A LA 00 : 00tJ LA U)00 ,0 00% -j LA% -4 W LA% W C2N-4 -4 -i _] CD, CY w 0 -j 10 0 U -j a


00
s
YD W o N-

Gr

12.
26

CD

m ib

rQ
>-,

0 p

1 -

00
(. A W

0 %0 , 2 h . 1

2 -4 w

8 c 0
m-

W (., (

t, ) c L, 1 -4

-3 w tJ t-i

t 0-IZ

LA 8

W 0 LA w

! W 00

w W o (>

cl C=

eD

P- 0 Z

tz 9-

S, 1
tJ t, ) N-

ci n 11 mZ

-CJ %0 rcr
CA p 19 00 p

00

>

+b c2. MG <

-3

c> (A '-.

t A 2 w (A tij

t)

(>

ta

t)

A Z

$--

tA

(A

Ab

CJ 0- 00

CJOCI ro

-2

L')

bgK

:t

2. 0 10
I ch tj o

ti

-4

L', .4 z c7, -3 00 %0 CY, (> tA 00 WI c p. L e


0 tj -2 10 c> A1

(, j

N-

N-- b-.

L,

m-

41.

vi

'-,

9c

92

0 fi

2 0

cj

L4 %0 %D Ch W CY% W C7N >-

>V

:Z

gg

ch

rd
0

20C Q. c c2.0
rn
. --, c,

IPO

t-i t4
A.

00

a tA

CD (D

CD -4

-3 -2

-j

C -j
U) CD

00 W

Ftit

NO %0 00 U) -4 -j W -j

CO rg w

tA 00 .A.

-a tj CYNC:) (D -P- Q b--

o' I (7,w %0 ,w t', J 2,

t)

E
0

i %D Z (D 8 00 C% (A 00 00 w cy% w c2, %0 w G aCYNm 00% - 0, (A l CD


b-

Cs

sr
Z

IL

3 w c2 YD 00

-2

%0 (A

ON v 00

00 (DN 0 . -ZJ 0 9-

9. =0
%D IP 2%00--

1. 1 E11 oo aw LA -> l' f , , 9 . C t2 Ili t2 A 4 lIJ 511 l


U) k. 1

,ej0

1-

j w 0, , m- .
t4

Lft 1u3 4) ti

ch %0 C

(> 1w

Im
( -4

ti

151
(> 12 %0

0-w

b-. w

120

1 >Z 0--. >" t. 0.-

1 b-. tot

11 wb. tA

1 9.(3

1 s. c, b-. s..

to:

0, 1w w w

-4 j2 --x

" 'A -Zg --7


F 0 a=

I
9
n

e M

201

Chapter

Operationalisation

of the ISLB Model

Figure 7.1 The volume of general cargo and oil products (foreign trade) for Iran
-*-General 4 Domestic Cargoes and Oil Product (Foreign Trade) Trade of Iran

32000 30000 28000 26000 24000 22000 20000 18000 16000 14000 12000

300000 280000 260000


240000 E 0 Q

220000 200000 180000 a 9

160000
140000

197919801981198219831984198519861987198819891990199119921993
Years

Source:Various PSO,SCI and Iranian CustomsAdministration for the period of 1979-1993.

7.4.1.2Domestic trade of Iran In order to forecastthe future volume of the domestictrade of Iran, attemptswere made to collect and group all available important commodity data. Although the ISLB project is the level of national planning, it is developedfor the domestictrade of Iran in a a study at bottom-up manner.The total statistical composition of the domestic trade of Iran includes 112 commodities and goods. All detailed and available data from 1979 to 1993 about then sortedinto sevenmajor classes cargoes,commoditiesand productswere collected and as shownin Table 7.3. The main differencesbetweentheseclassificationsis the different handling, transport and packaging requirements.The volume of total domestic production, as well as the total volume and sharesof these seven categoriesof Iranian domestic trade, are presentedin tonnesin Figures7.1,7.2 and 7.3 and in Table 7.4.

202

cn 7Q

N cr
(D CD

K I 4 O co
N 0

0
0

& CD 0
GEn
4

s
IL

co

<

f]Q

!.

0 43 0 .5 a

fo

E;
co

Z C02 a Gn

Flo CA

a
L < 0

$0

p 0E
0Wco
V3 QQ cil

N
ff

Ln 1 0 F. k 9
r
co

2
r.
0. r,

.1 0

R.

c
I , CD rjQ co rA

(V CD
0

tA

C ID
OTI 00

E co

0
0

203

Chapter

Operationalisation

of the ISLB Model

Tn)Nli-'7 'A rr%ntilniipri

1 Cotton seed Tobacco(green) Tea (leaf) Log Timber Rail-sleepers Roundwood Mine timber & beam Fuel wood Charcoal

2 Bentonite& fuller earth Magnesit& whiting Dolomite Silica & quartzite Feldspar Barytes Flourspar Sulphur Salt Slag Talcum Kaolin She]]

3, Water & electric metres Soap& detergents Tyre Plastic Plastic p pes Fibber glass Glass Bricks Cement Plasterquick lime Plywood Paper Cardboard Stationary Wall paper Household furniture Batteries Lamps Gascylinder Cables Fertiliser

Figure 7.2 Scattergram of the actual volume of the domestic production of Iran.

280000' 260000" 2400000 2 200000" 180000 160000 140000 1979 1980 1981 1982 1983 1984 198S 1986 1987 1988 1989 1990 1991 1992 1993 Years ...........

Sources:Various SCI annual and occasional (Statistical Reflection of Iran) reports for 1979-1993.U.N Statistical Yearbook 1993, International Trade Statistics Yearbook, and UNCTAD (1993a) Handbook of InternationalTrade and DevelopmentStatisticsfor 1992,variouspages.

Unlike generalcargo and oil products (foreign trade), the domestic trade of Iran has two significant points in 1981 and 1989; it gradually declined from 1979 until 1981and again from 1985, a critical year in the war with Iraq, until the end of the war in 1988. The fluctuation in the forms of Iranian production, causedby revolution or war, has meantthat the importanceof somesectors,the volume and the type of road and rail vehiclesutilised, 204

Chapter

Operationalisation

of the ISLB Model

has varied especiallyin termsof requiredcapacities.The contributionsof the sevenproduct to the total domestic trade of Iran is shown in Table 7.4. The averageshare in classes domesticproductionof eachof the sevenproduction categories be seenin Figure 7.3. can Table 7.4 Volume of seven groups of domestic production and trade of Iran (million
Year Heavy Volumeof Light Industries Industries agricultural production 3 2 1 Minerals 4 Reefer 5 Food Oil production products 6 7 Iran's total domestic p oduction

34.64 1979 34.02 1980 1981 29.03 39.14 1982 1983 48.81 1984 38.26 1985 42.17 1986 47.07 1987 48.06 1988 44.02 1989 40.84 1990 52.22 1991 51.41 56.10 1992 1993', `, ',j7.88 66.36 TOW 21.58 1% 1

28.11 29.88 35.03 40.85 49.16 46.50 50.49 47.67 27.73 27.15 29.09 34.40 38.08 38.59 41.61 564.32 18.35

3.65 59.56 56.72 2.04 1.78 46.45 64.37 2.47 71.99 2.89 3.16 74.91 3.73 77.86 69.07 2.28 71.43 2.43 2.60 71.00 2.83 63.25 4.47 73.45 6.19 96.17 6.43 101.59 7.82 t- 99.13 54.75 109695 * 35.671 1.78

4.78 26.14 2.85 2.89 4.851 26.27 3.00 4.24 24.38 3.85 7.85 28.52 3.98 5.12 32.55 4.051 5.56 34.98 4.18 5.72 36.90 5.45 33.39 4.37 4.39 4.68 33.51 4.61 4.51 33.75 4.531 4.10 41.01 5.95 5.14 42.31 6.32 5.87 48.55 5.98 6.38 47.33 6.75 -,: ,, `6.70 57.34 67 681 80.96 547.01 * 2.63 17.79 2.201

159.71 156.66 143.91 187.05 214.48 207.42 221.04 209.30 192.22 187.63 185.74 217.94 252.59 262.40 1277.22 3075.31 100

Sources: Various SO annual and intermittent (Statistical Reflection of Iran) reports for 19-19-1993. U.N (1995), Statistical Yearbook 1993, International Trade Statistics Yearbook, and UNCTAD (1993a) Handbook of International Trade and Development Statistics for 1992, various pages.

Figure 7.3 Shareof different sectorsin the volumeof the domestic production of Iran.

Reefer 2.2%

Oil 17.8%

Food 2.63% AgricLdtural 21.6%

Mineral 3S. 67% Light Industries 1835%

Heavy Industries 1.78%

Source: Various SCI annual and occasional (Statistical Reflection of Iran) reports for 1979-1993. U.N Statistical Yearbook 1993, International Trade Statistics Yearbook, and UNCTAD (1993a) Handbook of InternationalTrade and DevelopmentStatisticsfor 1992,variouspages.

205

Chapter

Operationalisation

of the ISLB Model

7.4.1.3 Central Asian and Caucasus foreign trade

The statisticsof the former republicsof the USSR on tradearecompiled"as a whole" since,a long and continuousseriesof detaileddatais not available.There are somerelevant dataafter 1985for different republics,but they differ in structureandcontinuity.This paucity datais because the natureof the former structureof the USSRsystem,and also because of of " of the "absenceof customsborders between Russia and other former Soviet republics. (UNCTAD/UN 1994,p. 69). Due to the limited published data the dependent variable for tonnageof eachof the trade of the CAC countriesthrough Iran is as the volume aggregate the CAC countries' tradewith Iran (Fischeret al. 1996). The annualvalueof tradewith Iran into volume (Leontief 1979)by using the value for eachof the CAC countrieswas converted andvolumeof tradewith Iran in 1992and 1993asthe basis.
The methodology used to compute the volume of trade of the CAC countries is summarised

as: 1. The conversionof all CAC foreign trade values (excluding inter-republic trade) from Roublesto US $ wherenecessary. 2. Applying a share(%) of the Asian foreign trade of eachof the CAC countriesfor import to separately the total internationaltrade (mainly on the basisof availableCAC and export directionof tradedatafor 1990)to acquirethe probableinflow andoutflow tradeof the CAC countriesthroughthe ISLB. 3. Summingthe latestavailablefigures for Iran's foreign tradevolume and value with each of the CAC countries, and categorisingthis total volume and value to different product groups.The shareswhich result are usedas the basis for determiningthe cargo types of all CAC tradewith the rest of the world. for from using the 4. Data for the whole period 1987-1997 eachCAC country is generated
product type categorisation obtained in step 3 for Iran (for the latest observed years 1992 and 1993). The ratio between total volume and total value is assumed to persist through time, as 206

Chapter

Operationalisation

of the ISLB Model

the split of total volume and value into product categorisation and, thereby data (by product

is derivedfor the period 1987-1993. categorisation) 5. Extrapolationbetweeneach available value and volume of Iran's trade with each of the CAC countriesand the value of the proportional foreign trade of eachof thesecountriesto import andexporttonnagefor eachyearastotal Asian internationaltrade. obtain
It is assumed that the future international trade of the CAC countries will have the same

asthe tradewith Iran in 1992and 1993.The compositionof the tradewith Iran composition by between4 and46 commoditiesand products,all of which are transported road and varies Therefore,it will be assumedthat the future potential for CAC trade with other rail only. the rangeof the ISLB servicesare the sameasthoseduring 1992and 1993 countrieswithin Iran. The result of thesecomputationsare shownin Table 7.5 and Appendix 2. with Table 7.5 Description of the converted volume of international trade of the CAC countries for Asia (000 tonnes).
1989 1988 1987 293 292 272 Azerbaijan 134 168 116 Armenia 33 26 26 Georgia 118 133 145 Gyrkyzstan 931 966 891 Kazakhstan 103 68 85 Tadjikistan 112 143 72 Turkmenistan 2066 1658 Uzbekistan 1 2066, 1 3629 1 3779 1 3508 Total CAC SameasFigure7.4. Sources: 1990 299 184 29 124 965 117 139 870 2726 1991 1992 112 169 12 78 29 29 8 70, 668 1 819 7 43 157 48 502 791 1895 1646 4993 UL 61 10 -29 '40 , 592 42 --150 - 426 ;, '1350 j ,, Total (%) 1499 (8.1) 701(3.8) 202(l. 1) 637(3.4) 5831(31.5) 465(2.5) 820 (4.4) 8378(45.2) 18532(100)

Figure 7.4 showsthe trend of the CAC countries' Asian foreign trade volume with a sharp decline with the beginning of political democracy increaseduring 1987-1988 and a gradual
and unrest in 1989.

207

Chapter

Operationalisation

of the ISLB Model

Figure 7.4 Total volume of the CAC Asian Foreign Trade.


4000000-*I 3000000-c

2000000-1000000-0-i
1987 1988 1989 1990 Years 1991 1992 1993

from: based andconverted Sources: on States,p. 1. Trade valuesfor 1992and 1993 in: Michalopoulos (1993), Trade Issuesin the New Independent 26, The World Bank, Washington. Statistics,p. 565 TradeandDevelopment 2. UNCTAD (1993a),Handbookof International 3. UNCTAD/U. N (1994), Economic Bulletin for Europe,p. 85.

The converted trade statistics of the CAC countries as shown in Table 7.5 has some for deficiencies; for Georgia,data for 1988was assumed 1987,and data for 1990assumed for 1991,1992 and 1993. The data of Georgia for 1990 seemsto be unreliable and very inconsistent yearsand alsowith other republics;while for Uzbekistanonly, the with previous dataof 1987wasappliedfor the year 1988.

7.4.2 Selection of independent variables


A numberof suitable independent variableswere testedduring different runs of the (see section 8.3) for the three sub-modelsof generalcargo and oil estimation procedures products (foreign trade) of Iran (GOPFI7),domestic trade of Iran (DOMTI), and CAC foreign trade (CACFT) (Sun and Bunarno 1973). The choice of independent countries directly to questionsraisedby the study. Miller and Wichen (1977, p. 29 1variablesrelates 292) according to Cox and Snell (1974) accounted for the following four factors influencing the natureand numberof independent variables: 1. Is the study intended to examine a rather specific hypothesis about the phenomenon under investigation, or is the study simply concerned with screening out important variables from a large number of potential variables, the important variables to be investigatedin a subsequent experiment?
208

Chapter 7

Operationalisation

of the ISLB Model

2. Can the response (dependent) variables be observed quickly, so that later parts of the study can be modified, if necessary, in view of the earlier results?

3. The time available,the easewith which the measurements be made,the availability can for of good official statistics,the amount of moneyappropriated the study,and so forth are often crucial in decidinghow many variablescanbe measured.
4. Will the study be used to establish comparisons with earlier, related studies?

They are describedlater in tenns of their strengthsand weaknesses undereachof the four themes:geographical, political, technical, and organisational(seeAppendices4,8 and 13 for the coefficient of correlation between independentvariables and three dependent
of GOPFr, DOMTI and CACFr respectively). variables

Based on the literature review, twenty variables were chosen as possible explanatory
for the GOPFT, 25 for DOMTI, and one for CACFT. These are listed and variables

the four themesin Table 7.6. classifiedunder


7.4.2.1 Geographical characteristics 1. Population

Population is an important indicator in most econometric modelling and in in long range planning and analysis as its uncontrolled growth concerns particular consumerdemandand unemploymentthat are major political issues(Amirahmadi 1996).It hashad a rapid growth (seeFigure 7.5) with an increase about22 n-dllionduring only 12 of years. Its general effects on both GOPF17and DOMTI during 1979-1993 was an approximatedoubling of demand.In the latter, it can be more influential as the result of the direct involvement of people in the production and consumptionprocess.The GOPFT is however,with capital and intermediategoods. more concerned,

209

Chapter

Operationalisation

of the ISLB Model

Table 7.6 Description of the selected variables for the scenario analysis of GOPFT, DOMTI and CACFT. Description variable, --1-- Variable,: ISLB Scenario Units theme' of . ..... . .... . , Dependent variables name CargoandOil Products General Foreign YGOPFr 000tonnes Trade Iran(GOPFT) of YDQMTl Domestic Tradeof Iran(DOMTI) 000tonnes YCACFT CACcountries Foreign Trade (CACFT) 000tonnes Description of variable Independent variables Units & areas of demand Geographical X20 Population GOPF17, DOMT1 million people 000 hectare, DOMT1 X6 Landunder cultivation 000 tonnes,DOMTI Fertiliser in X8 used agricultural Description variable 2 Political Nariable Units & areas of demand of name 000 Rials at 1982 fixed prices GDPperCapita X1 GDPperCapita $ /Rialsexchange rate Crude revenues oil Valueadded all majorindu; of
Fixed gross national Foreign Employment Final GDP Government expenditure goods (total input) investment trade goods value

X1
X2 X3 tries X4 X5 X6 X7 X14 X17 Xis

GOPFr & DOMTI million, US $ CACFr

Rials GOPFF & DOMTI million $ GOPFF & DOMT1 million Rials, GOPF17 & DOMTI million Rials, GOPFr & DOMT1 million Rials. GOPFM 000 GOPFr & DOMT1 million Rials, GOPFr & DOMTI million Rials, GOPFr & DOMTI million Rials. GOPFr DOMTI

Value of intermediate

3 Technical
Number

Description of variable
of locomotives Value added of agricultural Number of wagons in agricultural Fixed gross investment Number

Variable name
x1t Xg Xg X10 X10 XII X12 X13

Xnits & areas of demand,


GOP million Rials. DOMTI GOPFr million Rials, DOMTI GOPFT million Rials, DOMT1 million Rials. DOMTI million Rials, GOPFr & DOMTI million Rials, DOMTI million Rials. DOMTI million Rials. DOMT1 million Rials, GOPFr & DOMTI million Rials, DOMTI million Rials, DOMTI 'Units & areas `c of demand

of heavy goods vehicles Investment in agricultural maci7inery Value added of light industries Value added of transport Value sector machinery

added of basic metals, & non-metal industries Value added of food industries Investment Transport

Xis X16 X18 X19 X21 X22 Variable

in oil & gas machinery investment

Value added of oil sector Value added of crude oil sector 0, rganisationalt'-Description of variable

Ships service days in berths Berths productivity of non crude oil foreign trade of Iran Portstemployee productivity Rail/employee Productivity employee Productivity productivity of agricultural sector/

X11 X12 X15 X16 X23 X24 I I x2i

Ship service day/annual, GOPFr Tonnesiberth. GOPFr Tonnes/employce, GOPFT Tonnes /employee. GOPFr Torincs/employce, DOMTI Tortneskinployee, DOMTI

of industry, mine, and food scctor/employee 1 Productivity of oil sector/employee

Tonnes/employee, DOMT1

210

Chapter

Operationalisation

of the ISLB Model

Figure 7.5 Growth of the population of Iran.


7000060000-50000-40000-30000-20000-10000 0iii 1921 1931 1941 1951 1961 1971 Years 1976 1979 1986 1991 1993

Source:SCI (1982), A StatisticalReflectionof the Islamic Republic of Iran for the Third Anniversary of the Islamic Revolution, Bahman 1360, p. 6; Central Bank of Iran, for 1991,1992,1993,1994a, and 1994b.various national accounts.

2. Land under cultivation

Its growth (see Figure 7.6) is slow since it depends on national development by and programmes supervised the Ministries of Agriculture, JahadSazandagi, Power.
Figure 7.6 Land under cultivation and fertilizer used in Iran.
I 2589000 9 2089000 ;2 1589000 1089000 589000 89000 1iiiiiiiiiiaii 1979 Fertiliser UsedIn Agriculture Land Under cultivation -1111-I 20000 19000 18000 17000 10 00 6 15000 14000 13000 12000 1981 1993 1985 Years 1987 1989 1991 1993

--*-

Sources:SCI (1982), A StatisticalReflection of the Islamic Republic of Iran for the Third Anniversaryof the Islamic Revolution, Bahman 1360,p.33; SCI (1990a),A Statistical Reflection of the Islamic Republic of Iran for 1988,No. 8, p. 102.SCI, (1991b), A StatisticalReflection of the Islamic Republicof Iran for 1989,No. 9, p. 104; SCI (1993), Iran StatisticalYearbookfor 1992,p. 274; and SO (I 994a),Iran StatisticalYearbook for 1993,p. 96.

3. Fertiliser used in agriculture

Fertiliser has been one of the important componentsof the imports of Iran while domestic refineries also produce large amounts (8.1% of total GOPFT). The distribution and provision of different types of fertilisers can affect greatly the volume of agricultural

211

Chapter 7

Operationalisation

of the ISLB Model

commodities with a high correlation of about 0.70 with agricultural production crops and

during 1979-1993.(see Figure 7.6) It is therefore, proposed as a potential explanatory


variable in the DOMTI model.

7.4.2.2PoIitical characteristics The political theme covers all those macroenvironmentvariables with a national or
international flavour. Therefore, they are mainly composed of macroeconomic factors over

the governmenthasa major influence. which


1. GDP per capita

Real GDP per capita is a measureof real GDP to total population and relatively informative about the standardof living in a country (The Economic Review 1995).The variation in output of the four main GDP sectors(agriculture, industry and mines, oil, and services) affects GDP per capita. The extent of variation dependson the share of each in the total value of production and the population.Two major issuesin Iran after the sector revolution have been, on the one hand, population growth (Figure 7.6) and on the other, falling output. During the 1979-1993period, accordingto two official censuses, population doubled from about 30 to 60 million excluding roughly five million refugees from Real GDP, on the other hand, in 1993 comparedwith 1979 had neighbouring countries. 48 per cent growth which meansa decline in GDP per capita (An-drahmadi1996). only a Figure 7.7 shows the situation of real GDP and population patternsrelative to GDP per during of capita.The sharpdecline during 1987is due to a significant decrease oil revenues
the peak year of the war.

For CACF17 only available independentvariable is GDP per capita and its correlation the with the CACFr over time shows a strong negative relationship (-0.96) which is mainly due to the eventsafter 1989and in particular the collapseof the fonner USR at the end of 1991.Therefore,GDP per capita,as it reflectsboth GDP and populationvariations,was
212

Chapter

Operationalisation

of the ISLB Model

Figure 7.7 GDP per capita, population and final GDP In Iran (based on flxed prices 1982 Iranian Rials).

FP
70000000 60000000 EM-, 50000000 5 *m 32 40000000 c 12.30000000 20000000 10000000 0

Population

GDP per Capita, Ria

340000 310000 280000 250000 190000 160000 130000 197919801981198219831984198519861987198819891990199119921993 Years

i ig 2r
V

220000 L) I..

Source: Central Bank of Iran (Bank Markazi), (1991), National Accounts reports for 1974-1993. variouspages.

the only available informative independent variable for the analysis. It is chosen as commonly used to explain the distribution of the resourcesamong citizens. Appendix 2 shows the sum of the available CAC countries' GDP per capita during 1987-1993,all of in US $. which are measured
2. Rial exchangerates

Up to the Islamic revolution in 1979,Iran had a fixed exchangerate with very few More recently,the heavily by oil revenues. the national currencywas subsidised changes as new government adopted a more flexible exchangerate system where fixed rates are applied to different trade purposes.Finally, at the end of 1993, a floating system was to officially adopted.The fixed ratesera corresponded high oil prices,high production,and exportsof crudeoil, and also a relatively low population with moderatepopulation growth. After 1979, national income declined due to the deliberatereduction in crude oil supply and export by the government.At the same time, the war had gradually absorbedthe accumulatedsurpluses.Also, high demand for limited foreign currencies and the high population growth causeda great unofficial devaluationof the Iranian Rial. After the war, high levels of aggregate demandbrought aboutby reconstructionrequiredmassiveimports,
213

Chapter

Operationalisation

of the ISLB Model

led to a very high trade deficit requiring a devaluation (to the extent of about 300%) which in the Rial. It also led indirectly to the adoption of the single rate system as well as inflation the beginning of 1994 (Central Bank 1994a, Sanate-Hamlo-Naghl 1995a). Devaluation at during 1995 as a result of temporarily high debts but, to some extent, occurred once again was controlled by strict trade restrictions. The devaluation of the currency of one country, while facilitating improvements and

balanceof trade, also puts pressureon domesticproduction. On the one adjustmentto the hand, domestic producersin a free market economy engagingin foreign trade pay much higher equivalentnational currencyfor their intermediategoodsand materials.On the other hand,it may also lead to rise in the domesticproduction volume and value but, at the same higher costs of imports to all economic sectors dependenton imports. time, transfer Consequently,the higher cost of imports leads to higher wages, issuesof liquidity, and cheaperexport prices which may make domestic exports uncompetitive and reduce the foreign trade. If the value of the national currencyrises, while it makesimports volume of it also forcestrade partnersto pay higher prices for the exportsof the country. cheaper, The Rial exchangerate can affect DOMTI in the ways describedabove.Both thesecases in Iran, and in particular after the revolution in 1979and the war in 1981, were experienced to tighten control over the availability of currencywere imposed.There was when policies introduction of the multi-rate Rial exchangerate for different types of also the official imports and foreign paymentsafter the Islamic revolution in 1979(CentralBank 1992). 3. Iran's oil revenues The oil revenuesof Iran are a function of crude oil production, domesticconsumptionof the internationalprice of oil. A time seriesof oil revenuesis shown in Figure crudeoil and 7.8. As a memberof OPEC,Iran hasthe potential to sustainhigh volumesof oil production but, instead,has deliberatelyreducedits production since the beginning of the revolution. Sincethen, war hascausedboth supply and exporting facilities to be severelydamaged and
214

Chapter

Operationalisation

of the ISLB Model

of sanctionsmade by the US governmentin 1995 any increasein the oil with a new era becomes more difficult. production
Figure 7.8 Oil revenues In Iran.
20000 1800016000-14000-12000-loooo-.2 8000-6000-4000-2000-0 1979 1980 1981 1982 1983 1984 198S 1986 1987 1988 1989 1990 1991 1992 1993 Years

Source:Basedon CentralBank, 1991,1992,1993,1994a, and 1995b,various nationalaccounts, 191,30, pp. 13,36, and26 respectively.

The consumption of domestic oil (26.8% average of the annual production during the 1979-1993) with the high growth in population led to restrictions in crude oil period exportsoncethe governmentwas willing to export more (CentralBank 1994b).It would be that crude oil revenues have a strong positive relationship with GOPFr. A expected correlation coefficient of only 0.31, however, indicates a weak linear relationship with GOPFF. Different sectorsof Iran's economyare dependent short and long term oil revenuesas a on imports of capital and intermediary goods. In or the necessary major source of revenues terms of both revenuesand volume of production the oil industry is an important sectorin
DOMTI but shows a low positive coffelation of only about 0.18 with DOMTI. 4. Value added of all major industries

Value addedis the differencebetweenproduction costsand the selling prices of goods and services.According to Riahi and Fekrat (1994, p. 3):
"It is a measureof the total return generatedin a firm through the utilisation of its productive capacity. i.e. labour and capital in the broad classicalsense. "

215

Chapter

Operationalisation

of the ISLB Model

It may be a strong predictor of GOPFr as it represents the contributions of all major

industries to GDP and is also an indicator of the costs and selling prices. It has a high 0.91 with GOPFIC over the period of study and is shownin Figure 7.9. correlationof about
YIgure7.9Value addedof all major industries and volume of GOPFr of Iram

-W-GOPFr

32000000
w

Value Added of All Mqor Industries ------j

--1350000

QU

27000000 -22000000 -17000000

--1150000 -6 --950000
750000

12000000
Years

111
1979 1980 19811982 1983 1984 1985 1986 1987 1988 1989 1990 19911992 1993

550000

Source:GOPFr sameasTable 7.2. Value addedof all major industriesis basedon variousCentralBank of Iran 1991,1992,1993,1994a,and 1995b,annual reportsof nationalaccounts.

The value added of all major industries that are comprehensivelycovered by all seven
classes of DOMTI might be a strong explanatory variable as it has the highest correlation

of about0.90 (with DOMTI) amongall the explanatoryvariables.


5. Fixed gross national investment

According to The EconomicReview (1995,p. 17) investment:


"Involves expandingthe productivecapacityof the economy,thus leadingto economic growth, which is one of the fundamentalobjectivesof economicpolicy for any government. This meansthat the choicesthat societymakesabout allocating resources investment to to presentconsumptionarecrucial for determining the future path of the economy. relative Thus we can think of investmentas being concernedwith extendingthe quantity or quality of land) or with improving the the factors of production (capital, labour,entreprreneurship, technologyof production".

Fixed gross national investmentoccurs in terms of investment in machinery,construction and tools by both the public and private sectors involved in the four main sectors of agriculture, oil and gas, industry and mines, and services(Central Bank of Iran 1991). It has a relatively low correlation of about0.24 with GOPFT. The future shareof private and public investment is important (in particular the former) to both demand(increasingthe volume and changing in the composition) and supply of ISLB (by the involvement of the

216

Chapter

Operationalisation

of the ISLB Model

in renewing and owning the road fleet and rail rolling stock). Figure 7.10 private sector

the the trend of fixed grossnational investmentin the past fifteen yearsemphasising shows share developed by the private sector. The share of transport varies but considerable 10.8%for all freight and passenger urban and inter-urbantransportover generallystandsat the whole period.
FIgure7.10 Gross nationaIinvestmentof Iran: Private sector, and transport shareat 1982fixed prices. governmen4
ftivate smtor 16M 1400 VA r= Covernnwnt smtor -of tmsport %Wch

-a 1000 800 600 400 1-0

200 0
1979 1980 19811%2 1983 1984 1%5 IM L987 1M 1989 1990 1991 LQ921993 Years

4m 350 300 250 200 150 100 50 0

Source:Central Bank of Iran (1991, p. 295), (1992, p. 72), (1993, p. 31), (1994a, p. 55), (1994b, p. 45), variousNational Accounts.

New investment in annual and long term programmes varies at sectoral levels. This includesboth investmentin constructionand machineryfor all four sectorsof the variable Iranian economyand hasshown a weak correlationof about0.36 with DOMTI.
6. The value of foreign trade goods

In general,import and export valuesare composedof two product categoriesin the national accountsof Iran (seeFigure 7.11). The value of goodscovers the trade in oil and material commercial trade. The trade in services includes transport, gas, and all other insurance,travelling, training and health. The net value of foreign trade in servicesis a much smaller contribution and includes transport exports (Central Bank of Iran, various annual national accounts).The values which are consideredin this study are the total

217

Chapter 7

Operationalisation

of the ISLB Model

and imports and are broadly classified as the exports of crude oil, bulk values of exports cargoes, and general cargoes.

Figure 7.11 Goods and crude oil exports and Imports of Iran at 1982 fixed prices.
--*-Export 350 3000-2500 jf A 2000 C. 1500 value of all products --WImport value of all products

l(WO 500
0iiiiiiiA1 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Years

Source:Central Bank of Iran (1991, p. 210), (1992, p. 80), (1993, p. 34), (1994a, p. 58). (1994b, p. 48), variousNational Accounts,. 7. Employment

Employment is the third and secondmost highly correlatedvariable (0.88) with both GOPFr and DOMTI respectively.As shown in Figure 7.12 employmenthas followed an upward trend from 1979 to 1993. Figure 7.13 indicates the importanceof employmentin the serviceand agricultural sectors. in 7.12 IFIgure Employment Iran.

C C 0

1979

1981

1983

1985

19EW

1989

1991

1993

Years

Source:Central Bank of Iran (1991, p. 249), (1992, p. 57), (1993, p. 24). (1994a,p. 47), (1994b,p. 37), variousNational Accounts.

218

Chapter

Operationalisation

of the ISLB Model

Figure

7.13

Sectoral

contributions and mines

to employment [=Services --*-Construction

in Iran.

Agricultural CIndustries
7000 6000 5000 4000

1600 1400 1200 1000 Soo

-0

3000 2000 1000 0 1979 1985 Years 1990 1993

600 400 200 0

Source:Central Bank of Iran, (1991, p. 249), (1992, p. 57), (1993, p. 24), (1994a,p. 47), 1994bp. 37).

8. Final GDP

According to The EconomicReview (1995, p. 13):


being availableto a societyby observingchanges "We can monitor changesin the resources in real GrossDomestic Product (GDP)".

Final GDP is the third most highly correlatedvariable with DOMTI (0.87) and to a lesser extentwith GOPFr (0.77).
9. Government expenditure

Governmentexpenditurecoversthe cost incurred in the ministries,relatedorganisations and municipalities and mainly concerns government purchasing. It has only a weak negativecorrelation of -0.29 with DOMTI.
10. Value of intennediate goods

The inclusion of this variable in the model was due to its contribution to the total DOMTI. It is strongly under the influence of imported goods and, therefore, indirectly could be a strong indicator of GOPFr.
7.4.2.3 Technical characteristics

The foreign trade of Iran mainly takesplace through six major ports, four on the Persian Gulf in the south and two in the north on the CaspianSea.More than 85% of the foreign 219

Chapter 7

Operationalisation

of the ISLB Model

trade of Iran runs through the two southernports of Abass and Imam which also account The following fourteen for about 80% of the total port infrastructureand superstructure. in technical explanatoryvariables(five for GOPFr and twelve for DOMTI) presented this sectionare thosewhich directly affect the performanceof the modesand interfacesin terms of capacityand servicesat planning and macrolevels.
1. Number of locomotives

Diesel locomotives (variable for GOPFr) are the main motive power for rail transport in Iran. As a mountainouscountry, the availability of high power locomotives plays an important role in both freight and passenger movements.

Figure 7.14Locomotivenumbers in Iran.


320-300--

0 280-260240-220-200. 180-160 140 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993
Years

Source:Based on PBO (1989-1993),The Preliminary Results of the First Social, Cultural and Economical DevelopmentPlan of the Islamic Republic of Iran, TransportationSector,unpublished,IRIRC 1984-1993.

The variation in the number of locomotives during this period as shown in Figure 7.14 is locomotives under long term repair or a lack of spare mainly due either to war damages, parts. The production capacity of the country in locomotive manufacturing is about 0.3 locomotives annually (IRIRC 1984-1993)which is an indication of the dependence the of rail industry on imports. Given the size of Iran, the railway plays a significant role in

220

Chapter 7 supplying distributing cheap transport of iron ore to two giant

Operationalisation irons melting

of the ISLB Model and This also in

plants

the metal products Arabian

of these plants

to rail connected Gulf

destinations.

includes

exports

to the southern

States of the Persian variable

through

the ports of Imam GOPFI.

and

Abass. As such it may well be an important

explaining

DOMTI,

2. Number of wagons The number of different wagons (variable for GOPFI) provided in a freight train is a

key variable in the annualoutput of ports, crossborder terminals and all types of domestic and foreign trade of Iran. Figure 7.15 showsthe averageannualperformanceand available URC for foreign trade. The role of IRIRC in the GOPFT and DOMTI is the H; wagonsof two wagons can carry a load equal to 47 trucks, and it reduces vital since every the long distanceroad traffic and also costsof transport(Sanate-Hamlo-Naghl significantly 1993a).Figure 7.16 showsthe variation in the numberof wagonsin Iran during 1979-1993.
Figure
Soo 400 300
200 100 0 1986 1997 1989 1989 Years 1990 1992 1993

7.15 Train

wagon

utilization

in Iran.

11

37-2.6

2S8.4

297.2

34 172.7

Source:Basedon PBO (1989-1993),The Preliminary Resultsof the First Social, Cultural and EconomicalDevelopment Plan of the Islamic Republic of Iran, TransportationSector,unpublished,IRIRC 1984-1993.

3. Number of heavy goods vehicles

Up to the end of 1994,when the port of Abasslike Imam, was connectedto rail, heavy goodsvehicles (HGVs) (variable with 0.05 correlation with GOPFr) were the only means of freight transportout of the ports. The traffic of HGVs into major ports is a function of the type and availability of goods, and the length and quality of roads. Basedon Atrchian (1994), only those HGVs having a carrying capacity of between 13.5 and 22.5 tons are
practically involved in the long distance carriage of freight between ports and border

221

Chapter 7

Operationalisation

of the ISLB Model

crossingsto and from inland origin/destinationpoints. Therefore,for this study the number HGVs hasbeencomputedand deducted from the total road freight vehicles. of An analysisof the growth of HGVs asthe backbone Iran's foreigntradeshowstwo distinct of
the periods 1979-1986 and 1987-1993. There are two organisations involved situations over

and interestedin HGVs: the traffic and registration departmentof the disciplinary forces the Ministry of Roads and Transportation (MRT), which is mainly interestedin the and trucks for freight transport. Unfortunately, there are always differencesin the number of data of these two sources.As discussedin chapter two, every few years the published registration and traffic departmentpublishes the actual existing number of trucks and the deduction of those that have been scrappedor banned.At the end of vehicles after 1986,there was a censuson all types of vehicles, the partial effect of which is shown in Figure 7.16. The sharp decline in the number of HGVs in 1986 is, on the one hand, the destructionof HGVs in war zonesand, on the result of the significant role and consequent other hand, the result of the nation-wide vehicle censusand the deductionof scrappedand obsolescent vehiclesfrom the total registered. Figure 7.16 Number of wagonsand heavy goodsvehiclesof Irm I
175000 170000 165000 > 160000 155000 150000 145000 C61 140OW 135000 130000 125000 120000 1979 198019811982 19831984 19851986 1987 1988 1989 199019911"2 1993 --*-No. of heavy goodsvehides No. or wagons I 14500 14000 13500 13000 12500 12000 11500 11000

Years Sources:Basedon IRIRC (1984-1993),PerformanceReport, and various SCI publicationsfor 1979-1993.

222

Chapter 7

Operationalisation

of the ISLB Model

4. VaIue added of transport sector

The definition of private road transportaccording to the Central Bank of Iran, covers road freight vehicles and buses as well as urban transport and contributes the main proportion of the value addedof the transportsector (variable with correlation of 0.59 for GOPFT and 0.68 for DOMTI). In the public sector, the IRIRC is important. The second most important group of contributors to the value addedof transport are the governmenttransport services with two national domestic and international airlines. In sea run air transportthe function of the Ports and Shipping Organisation(PSO) as a governmententity is important. The transportsystemin Iran largely dependson imports for a high proportion of the motive power of all modes and interfaces. Available domestic production is not sufficient to match the requirements of even domestic trade, including passenger is movements.This meansthat the transport superstructure mainly dependenton imports from abroad and is considered an intermediate good as it produces returns and has a long-term life period. For instance,if Iran has more locomotives,wagonsand medium and HGVs of its own for internationaljourneys it will be possibleto take a much greatershare in the national trade which is carried out by international road or rail systemsvia different border crossings.At the sametime, by equipping ports with more container gantry cranes and other heavy lift equipmentthere has been a large reduction in the demurragepaid by governmentto ships during the recent years (PSO 1993). Consequentlythis variable can havea direct relationship with the GOPFr and also makesother associated resources more effective. Figure 7.17 shows a historical record of the variable with two main negative effects at the beginningof the revolution and during the critical yearsof the war.

223

Chapter 7

Operationalisation

of the ISLB Model

Figure 7.17 Value added of three modes of transport

in Iran.

--S-Sea

Modes

Air Modes

Land

Modes

m 80000 70000 60000 : 2 50000 .0 40000 30000 20000 10000

110000 100000 90000

1000000 800000 .2 1 600 000


400000 200000

01

5 F. ;6wa.
0% ON

36 r. ;6g
Years

g6 gg

;i

Source:Basedon Central Bank (1991, p. 214), (1992, p. 41), (1993, p. 17), (1994a,p. 40), (1994b p. 30) variousnational accounts.

The value addedcontribution of the transport sector is composedof land, air and marine the former taking the most important proportion in both private and public activities with sub-sectors.Transport service costs are a significant proportion of total commodity and product costs and have an effective role in the marketability of goods.Private road freight for urban and inter-urbanpurposeshas a significant role (seefigure 7.17) in the movement of GOPFT and DOMTI. Therefore, with a correlation of about 0.69 it may be a good explanatoryfactor in defining DOMTI.
5. Transport investment

Transport investment (variable with correlation of 0.33 for GOPFT and 0.44 for DOMTI) covers all urban and inter-urban modes and interfaces in the forms of infrastructure, facilities, installations and vehicles. The development of the transport infrastructure is basically for the movementof freight and, particularly in the long term, causes the flows of commodities and materials to be possible or increased.e.g. the exploitation of new mines or establishmentof major industries. The returns from the investmentin transportcan be seenas a long-term causeof increasein DOMTI (Gillen and Waters, 1996).Transportinvestmentis a significant part of the investmentin the total fixed

224

Chapter 7

Operationalisation

of the ISLB Model

investment and covers all urban and inter-urban modes and interfaces with, as was gross

(Bamford 1995). in Figure 10,an average shareof 13.1%during 1979-1993 shown


6. Value added of agricultural

Agricultural production is the most intensive employmentsectorand secondin terms of output volume. The products and commodities in this class affect other classeslike food and are generallysignificant for every individual in the form of daily and refrigeratedgoods The value added of the agricultural sector (see figure 7.18) (DOMTI consumption. has shown the secondhighest correlation with DOMTI (0.88) and is expectedto variable)
be a strong explanatory variable.

Figure 7.18 Value added of the agricultural sector in Iran.


3550000' 3350000-3150000-29500002750000-co 2550000"' 2350000" 2150000" 1950000 1750000 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Years

Source: Central Bank of Iran (1991 p. 162), (1992, p. 18), (1993, p. 10), (1994a, p. 33), (1994b, p. 23), variousNational Accounts. 7. Fixed gross investment in agriculture

This has (DOMTI variable) a negativecorrelation (-0.32) with DOMTI which appears
to contradict common theoretical assumptions. Iran has regular destructive earthquakes and

frequent floods in different parts of the country, which may be a reason for the above contradiction. Also, the dependenceof the country on unirrigated land means that the periods of drought influence harvestsmore effectively when there is a shortageof rainfall, and reducesthe effects of new investment. 225

Chapter 7

Operationalisation

of the ISLB Model

8. Investment in agricultural

machinery

As a part of the fixed gross investment in agriculture, this (DOMTI variable) has a strong positive correlation of 0.77 with DOMTI. It functions as a direct input to the agricultural sector as it can increasethe areasunder cultivation, minimise the final costs and facilitate performance.
9. Value added of light industries

This (DOMTI variable) is composedof values for the six industriesof textiles, wood, and miscellaneousindustries with a high correlation of 0.86. paper, chemicals,minerals, This category is the third most important contributor with 18.35% of the total average The most significant contribution is madeby textiles. DOMTI tonnageduring 1979-1993.
10. Value added of basic metal, machinery & non-metal industries

This (DOMTI variable) is composedof the two strategicand labour intensive areasof heavymetal and non-metalindustriesand has the highestcorrelationamongother variables DOMTI (0.92) but had the smallest averageshare (1.78%) in the total volume of with DOMTI during 1979-1993.
11. Value added of food industries

As one of the most important labour intensive industries with a direct impact on individuals, this (DOMTI variable) only hasa moderatecorrelationof 0.55 with DOMTI.
12. Investment in oil and gas machinery

As only a minor part of the total investmentin the sector,this (DOMTI variable) shows a weak correlation (0.24) with DOMTI. Indirectly, however, all industries are affected by the developmentof this sectorin terms of the changesin fuel consumptionpatterns.
13. Value added of oil sector

This (DOMTI variable) includes crude oil export value addedas well as oil products and hasa moderatecorrelationof 0.49 with DOMTI.

226

Chapter 7

Operationalisation

of the ISLB Model

14. Value added of crude oil sector

This (DOMTI variable) is the main part of the oil sector value added with a high (0.85) to DOMTI. In fact, it is the difference between the costs of crude oil correlation production and the selling price of domesticand internationalcrude oil.
7.4.2.4 Organisational characteristics

The organisationalthemescover thoseindependentvariableswhere the major element their output is humanskill and decisionmaking. of
1. Ship service days in seaports

Total ship turnaroundtime coversthe averagedays in port for a ship at eachof the six major ports. It is also called port staying days (PSD). It is the ratio of the total period of days all ships spendin ports to the total numberof ships visiting Iran, as shown in column
2 of Appendix 3. Each ship's PSD are composed of four elements: operating days (when

the ship is berthedand working), servicedays(when the ship is alongsideworking but may also have someidle time), waiting days(when, after arriving at a port, a ship has to wait to come alongside),and idle times (a part of the service time when ships are alongsidebut, due to technical or other reasons,are not engagedin loading or dischargingoperations). Each of these factors is indicative of port operating efficiency and the data for Iran is shownin Figures7.19 and 7.20. As shown in Figure 7.19 the data relating to these four main ship indicators reflect the capability of the major Iranian ports to serve ships and indicate that the averagestaying days of a ship during 1992 was equal to 9.6 (sum of service and waiting days) with the being: operationaldays69%; waiting time 26%; and idle days5%. threemajor components

227

Chapter 7

Operationalisation

of the ISLB Model

Figure7.19ConTonentsof annual average stayingdaysfor six major ports of Iran.


13Service days 16 14 12 10 8 6 4 2 0 19M 1981 1982 1983 1984 1985 1986 Years 1987 1988 1989 19M 1991 1992 15.25 0

Sources: Iran Statistic Yearbooks for 1981, and 1992 SCI (1982, p. 589) and SCI (1993, p. 392); PSO (undated probably 1991, p. 91), An Investigation in the Performanceof the Iranian Ports for 1320-; PSO (1992, p. 108),OperationalReport of the Ports and Shipping Organisationof Iran for 1371.

FIgure 7.20 Annual average service days components at berths for six major ports of IrAn.
E3 Operationdays
1189

0 Idle

14 12 10 8 6 4 2 0 1980 1981 1982 1983 1984 1985 Years 1986 1987 1988 1989 1992

Sources:Iran Statistic Yearbooks for 1981, and 1992 SCI (1982, p. 589) and SCI (1993, p. 392); PSO (undatedprobably 1991, p. 91), An Investigation in the Performanceof the Iranian Ports for 1320-; PSO (1992, p. 108),OperationalReportof the Ports and Shipping Organisationof Iran for 1371.

Total ship turnaroundtime is a very generalindicator in the caseof Iran as ratesin southern ports differ considerablyfrom Northern ports where much smaller coastalships make for more rapid turnaroundtimes.

228

Chapter 7 2. Average trade) berth productivity of Iran's

Operationalisation general cargo and oil products

of the ISLB Model

(foreign

The assessment

of Iran's seaport productivity

where there is centralised

government

is land and seatraffic, and management, a difficult control over cargooperation,equipment, task especiallygiven the paucity of data.Low port productivity in Iran is due to both internal inefficiency (previouscongestion, key role of the ship's gear,the lack of a comprehensive the training schemefor port workers) and also to inefficient road/rail transportmodes.There is in both quantitativeand qualitative shortages skilled labour and personnel, also evidenceof etc. which impair port productivity. Iran's port equipment and procedures,ships' cranes, basicallymust servethe non-oil cargohandling operation,but berthsof someof the seaports as well and complexity arisesdue to the mix of traffic commercialports serveoil products dataandutilisation of resources. Port performancecan be divided between ship, cargo and personnelindicators where the former five indicatorshavebeenexplainedalready.One major measure the productivity of of is its output in generalterms.It is the ratio of total port throughputto total working days a port in a year.According to Table 7.7, for example,for 1993about80,865tonnesof GOPFTdaily handledby the six major portsof Iran (Column I for 1993divided by 341 working days). was berth utilisation in (e. In other approaches g. Sinclare1979),someaggregate variablessuchas For into the calculations. example,for the same termsof ship servicedaysin ports areentered tonnageof each in Table 7.7 the ratio of column I to 3 gives (seecolumn 5) the average year When the average(column 5) is divided by 7.10 as the generalmeasure servicedays of ship. of eachship in major ports in 1993,2021.8tonneswill be the daily productivity of eachship. A more specificmethodlies in termsof the ratio of output to the numberand lengthof berths days. The berth (ship) productivity of six major Iranian seaportswas as well as working in column 6 of Table 7.7. This specific port productivity is basedon calculatedand shown

229

Chapter 7

Operationalisation

of the ISLB Model

total sea trade assuming 341 working days a year and a fixed number and length of berths for

(77 the period of 1988-1993 berthswith 14,592metresof oceangoing berths).


Table 7.7 Gene I productivi ty of six major Iranian seaports.
Year Total seatrade (000, tonnes) PSO employees No. ships Ship Service days NO Average shipment with oil products (tonne) (1/3) Berth (ship) Productivity with oil products (tonnes) (5/4) (X, 2) Productivity per PSO personnel tons1GOPFr (1/2) (X13)

2
1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 4993

6813 1495 7.740 7850.20 1014.24 11736 1722.59 1395 11.58 7498.20 6801 647.51 10460 1538.00 9511.74 623.72 1194 15.25 7014 1619.19 11357 9257.92 9.31 994.41 6708 1326 1 1830.05 12276 1652 10345.04 914.00 6491 11.31 2632.87 17090 9957.40 10.17 978.10 1291 1889.33 6804 12855 7.70 9744.98 1345 1265.58 7092 1848.14 13107 9.93 1032 12747.09 6774 1283.70 1941.98 13155 9.99 13486.60 1350.01 2491.30 6664 1231 16602 8.72 12704.08 1456.90 2120.95 6457 1078 13695 12436.07 11.11 1119.36 2476.66 6598 1314 16341 8.70 11385.04 1308.63 2737.60 6837 1644 18717 6757 1897 7.20 11468.64 1592.87 3219.77 21756 7.15 12969.06 1350.94 4092.27 6145 1 1939 25147 ', '2021.76 I ', 5956 1-'1921'1 -",, -, 7.10 1"'-",, 14354.50 27575 , -, 4629.79 -,, . Source:Basedon PSO (undated probably 1991, p. 5 1), An Investigation in the Performance of the Iranian Ports for 1320-1369; PSO (1992), Operational Report of the Ports and Shipping Organisation of Iran for 1371, various pages.

Productivity in ports according to the UN (1967) is calculated on the basis of tonnes/ but since such detailed data at shift and port operational levels are not man/hour output; for the ISLB, an alternative method is to consider macro variables e.g. berth available lengthsand annualworking days,averageannualtrade as shown in Figure 7.21 and Table 7.7 or using an important port employmentsector such as the PSO productivity (Table 7.7 column 7).

230

Chapter 7

Operationalisation

of the ISLB Model

Figure 7.21 Specific port productivity of the six Iranian major seaports. e (U .M
2.7 Non-oil cargo productivity (Tonnestmetresof berth per day) Total seatrade productivity (Tonnes/metresof berths per day)

S. 6
4.2 3.8 1

Pw

1988

1989

1990 Years

1991

1992

1993

Sources:Basedon the PSO (1992), OperationalReport of the Ports and Shipping Organisationfor 1371;PSO (undatedprobably 1991), An Investigation in the Performanceof the Iranian Ports for 1320-1369various pages;PSO (1993, p. 5 1).

3. Ports/employee productivity The calculation of the productivity of ports according to personnel must broadly cover all

port authorities,stevedoring,shipping,forwardersand shippersinvolved in direct port trade. Such data is not available.Cargohandling in all three stagesof port operations-ship, quay, and storage - is directly under the practical administration of port authorities and then Port authorities are not involved in the ship (hold) operation in stevedoringcompanies. is southern ports.Therefore,the role of PSOemployees importantin the productivityof ports. After the war, there has been a tendencyto reducethe number of employeesat a national level. In the PSO, attemptswere madeto increasethe productivity and reducethe number
of staff as shown in Figure 7.22.

231

Chapter 7

Operationalisation

of the ISLB Model

Figure 7.22Productivity of six inaJorports of Iran in ternis of tonnes for non-oil seatrade per PSOemployee.
69001 F640091,5900-a 5400-4900-4400-3900" 5.3400-2900-ILI 2400-1900. 1400

Productivity: tons non-oil seatrade per PSOemployee PSOemployees

.......... 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Years

Source:sameasFigure7.21.

4. Rail / employee productivity

The Islamic Republic Iranian Railway Company (IRIRC) is a particularly labourintensive entity with largely poorly educated employees. Since the last restructuring, the

IRIRC employeeshas shown a declining trend as shown in Table 7.8. On the number of basis of permanentand on-contracttemporaryemployeesof IRIRC during 1986-1993,rail for IRIRC is computed and shown in Figure 7.24 which generally reveals a productivity in decrease the numberof employeesand an increasein the volume of traffic and resulting According to PBO (1993), during the seconddevelopmentprogrammethe rail productivity. IRIRC must increasethe number of highly skilled staff from 4% of the total in 1993 to 10% in 1998 and also improve the skills of the presentpersonnelthrough IRIRC training centresand univesities.
Table 7.8 Number of emp oyees in IRIRC. I 1986 1988 1987 1989 1990 38480 1 37170 38960 41290 40790 1991 36940 1992 I 1993 35780 34450

IRIRC Employees Source:Basedon IRIRC (1986-1993),Factsand Figuresof the IRIRC for 1986-1990, and IRIRC Report of IRIRC for 1984-1990,variouspages. (1984-1990),Performance

232

Chapter 7

Operationalisation

of the ISLB Model

Figure 7.23 Freight tonnestper employee in IRIRC.


1..

114.42 m ep C, cu 0 45.19 Ir 2301 49 41 2 70 69 50.94 9&07 ...... 867 74.65

1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Years

and Source:Basedon IRIRC (1986-1993),Factsand Figuresof the IRIRC for 1986-1990, IRIRC (1984-1990),PerformanceReportof IRIRC for 1984-1990,various pages.

The organisationalthemeof DOMTI hasthree independent variablesrelatedto productivity by Hensher and Daniels (1995). These variables are the ratios of the total as advocated They to total employmentin the following sectorsduring 1979-1993. volume of production havea high correlation with DOMTI of about0.85,0.58, and0.87 respectively. " Productivity of agricultural sectoremployees " Productivity of industry, mine, and food sectoremployees " Productivity of oil sectoremployees.

7.5 Conclusions
The ISLB scenarioconceptualmodel developedin chaptersix was operationalisedin for this chapterunder four themesseparately generalcargo and oil productsforeign tradeof
v

foreign Iran (GOPFr), domestic trade of Iran (DOMTI) and Central Asian and Caucasus trade demand (CACF17).Sources of data and the related problems of collection and this data were identified and introduced. Key issuesand uncertainties administration of identified in earlier chapters were expanded and described under the four headings: political, technical, and organisational.Independentvariables were derived geographical, from thesefour areasto be usedfor estimation, scenariobuilding and forecastingpurposes
in the next chapter.

233

Chapter 8

Scenario Analysis of the ISLB Demand

8. Scenario analysis of the ISLB demand


8.1 Introduction
identified and operationalised, the In the precedingchapter,datawere prepared, and framework for the ISLB scenarioanalysisin termsof eventsand trendswas developed. The purposeof this chapteris to estimatethreeISLB demandmodels,developmentof scenarios interpret the relationships between dependentand independentvariables under the and three most probablescenarioconditions. The approachin this chapteris basedon four main headings: * Objectivesandregression analysis * Model estimation * Interpretationof the results * ISLB scenariodevelopment and model forecasts

8.2 Objectives and regression analysis


Many approaches scenarioanalysisuse someform of quantitative forecastingas to the basis for further study (Cole and Chichilnisky 1978, Zenter 1982). Time series and causalmodelslike regressionare widely usedtechniquesthat facilitate a scenarioanalysis (Taylor 1992,Tongzon 1991,Picard and Nguyen 1987). Regressionanalysis (RA) is used in many situations where one or more independent variables may explain the variation in a dependent variable and when "long range forecasts" are required (Kneafsey 1975). It also promotes better understandingof the relationships between different variables within the economic system. According to Abraham and Ledolter (1983), it is mainly basedon two distinctive conceptualstages, that of model-building and forecasting.This processis illustrated in Figure 8.1.

234

Chapter 8

Scenario Analysis of the ISLB Demand

Figure 8.1 Conceptual framework of a forecasting system. Model-building phase Forecasting phase

No Theory and previous studies Stability Diagnostic YesForecast 11 checking: Yes Forecast checking: is the __0 ---i generation isthe upgrading model model stable adequate? _j

No jspecificatjion Model

Model estimation

Source: Based on Abraham and LedoIter (1983, p. 4)

RA is a mathematical procedure which explains the functional linear and curvilinear relationships of a dependentvariable by one or more independentvariables through the development of an equation indicating the relationship of dependentand independent variablesto eachother (Barron and Targett 1985and Chatterjee& Price 1991).Regression analysisis often usedto provide a seriesof forecastsbasedon an interpretationof historical dataand forecastvaluescan be varied through changes the parameters the model. in of The degTeeof this variation can be measured by a value called the "coefficient of determination"of any independent variablesto the dependent variable; it is the ratio of the explainedvariation to the total variation (Wheewright andMakridakis 1985)and is denoted by rl. The effectivenessof the designed model is assessed using the r2 statistic which

235

Chapter 8 the goodness of fit and explanatory explains

Scenario Analysis of the ISLB Demand power of the model and through various t-

tests, the coefficients in the model are assessed.

The ISLB regression analysis (RA) study posed two questions: 1) to identify what are the by the most powerful internal and external independent effects and contributions made affect the three dependent variables (volume of the three ISLB demands) variables which to assesstheir relative importance, 2) to forecast future values of these independent and and then dependent variables (tonnage). Since the main measure of the performance of all modes and interfaces is the tonnage handled or tonne/km travelled, any assessment must be based on some measure of the

the trade handledby eachmode or interfacewhich, in turn, may annualphysical weight of dependon someother independentvariableslike the number of road vehicles,rail rolling averagedaily performance(MacNully, 1977). Specifically, the objective stock, or a ports is, with the available data, to determine the relationships between the dependentand independent so that future volumesof the three typesof demandcan be identified variables and predicted.Macroeconomicdata for 1979-1993are employedto analysethe effects of technical and organisationalfactors on the total tonnagedemand political, socio-economic, for the transport system of Iran. The use of available data from different parts of Iran's basisfor sucha study. a reasonable main national economicsectorsprovides The aim is to measurethe performanceof Iran's transport systemunder uncertaininternal that will affect the future volume of trade.Therefore,RA provides and externalconditions both an estimatedvalue of the trade and of the effects of the independent variableson the dependent variablegiven the scenariobeing tested. The use of multiple regressionanalysis is limited by a number of inherent assumptions be met if analysisis to work as hoped.The four following main assumptions which must basedon Wheelwright and Makridakis (1985): are

236

Chapter 8

Scenario Analysis of the ISLB Demand

1. The linearity of the relationship between dependent and independent variables. If

is breached,the variablesmust be transformedinto new variablese.g. this condition through the application for example,of logarithmic, squareroot, or other transformations. (constantvariance)of the regression 2. The homoscedascity errors or residuals
(the difference between actual and forecast values) must be present over the

entire set of observations. 3. A plot of the residualsindicatesthe independence consecutiveresiduals of If (non-autocorrelation). autocorrelationis present,however,then this indicates variable shouldbe ornitted so that a non-linear that one significant independent relationshipbetweenthe variablesexists.
4. Residuals must be normally distributed.

Model specification involves establishing the functional relationships between the dependentand independentvariables of the three models (generalcargo and oil products foreign trade of Iran (GOPFT), domestic trade of Iran (DOMTI), and Central Asian and Caucasusforeign trade (CACFT). This step is concernedwith the determination of the in the equationsand their functional form. According to Taneja (1978), there are variables in three aspects the specificationof a model: 1. the selectionof variables(seechapter7, section7.4 for ISLB) 2. the fonnat of the variables(seechapter7, section7.4 for ISLB) 3. the functional form of the model (seesection8.4.1 for ISLB). Scattergrams, each of the independentvariables against the dependentvariable give a of guide to the nature of this relationship. The sample correlation coefficient (r) is a rough that indicateshow closely the linear relationshipexists betweenrandomdependent measure and independentvariables and varies between-1 and +1. A value of -1 indicatesa perfect

237

Chapter 8

Scenario Analysis of the ISLB Demand

inverse relationship whereby when x (the independentvariable) increases, dependent the (y) decreases vice versa.When r is equal to +1, alternatively, a perfect direct and variable When it is zero, there is no relationshipbetweenx and y. relationshipexists. The linearity or any other probablefunctional form of the final model is determinedby an assessment the frequencyof the residualsas well as their graphicalrelationship with the of independent variablesin the final model (Taneja 1978). Two kinds of indicators are examined as an empirical basis for the ISLB model: factors related to the macroeconomicsituation of Iran and the CAC countries and factors which concerntransportperformance. Trade and in particular foreign trade forecastingis a common practicein scientific studies where volumes or values are regressedagainst some influential and explanatory socioeconomic variables or time (Pollins 1982, Giannopoulos 1984, Tongzon 1991, and Dagenaisand Martin 1987). A country's health largely dependson the purchasingor selling of goods and servicesto other countries, and also is a function of population and market centres, geographical location, industrialisation, economic wealth, etc. The total volume of foreign trade is related directly and indirectly to GDP, income from foreign sources, population, industrialisation, geographicalconditions, transport infrastructureand costs, etc. (Whiting 1986).In practice, it is affectedby national tariffs and restrictions and also by a variety of to other minor factors. It is necessary assumethat these variables are exogenousto the model being developed. In principle, many such independent variables might be simultaneouslydeterminedin such a generalmodel along with the tonnageof foreign trade (Giannopoulos1984).

238

Chapter 8

Scenario Analysis of the ISLB Demand

The ISLB model is investigated using regression analysis where it is assumed that the

dependentvariable Y (e.g. foreign trade tonnage) is dependenton several independent variablesX.

8.3 ISLB model estimation


According to Chatterjee and Hadi (1988) three factors which determine the regression

" and the model assumptions. According equationsare mainly "variables, the observations, to Mendenhalland Sincich (1989,p. 80):
"ro decidequantitatively how well a straight line fits a set of data is to determinethe extent to which the datapoints deviatefrom the line." To estimate the GOPFT and DOMTI models an elimination (backward regression) and

then stepwiseprocesses was investigatedwhere only those variables significant at 0.05%


lower are included in the preliminary or final estimated models. probability or

ISLB Model estimations were carried out for three demands (GOPFr, DOMTI, and CACFT). Thesewere introducedin sections7.4.1.1,7.4.1.2,and 7.4.1.3respectively.

8.3.1 General cargo and oil products foreign trade of Iran


A correlation analysis was performed as a preliminary assessment the selected of twenty independentvariables with GOPFT as the dependentvariable. The variableswere as ranked according to their correlation coefficients with GOPF17 shown in Appendix 4. Using the results of a correlation analysisthe first sevensignificant variables,all of which have positive correlation coefficients of higher than 0.70, were found to be: the value of
foreign trade goods (M. port productivity/employee (XI5), value added of all major

industries (X4), employment (XA value of the intermediate goods (Xjg), population (X2@9 and final GDP
(X20)-

To specify the GOPFT model, 20 independent variableswere testedin 32 runs and in five phases.Three of these runs specifically were only applied for the assessment each of
239

Chapter 8

Scenario Analysis of the ISLB Demand

the secondand third GOPFT estimatedmodels. The results of eachphaseand variable of in Appendix 6 consecutively. To estimate the GOPF17model an run are presented elimination process was investigated where only those variables significant at 0.05% probability or lower are included in the preliminary and final estimatedmodels It means that at each run the independentvariable with highest p-value was eliminated from the
subsequent run.

The independent variables were regressed against the GOPFT using the backward resulted in equation (1) where the p-values are (for X5, X6 and elimination procedureand X14) less than 0.05 on all coefficients in the model and there is a high r2 of 0.93. The empirical results of the final estimatedmodel are presentedin Appendix 5 (Abraham and Ledolter 1983).

0.0048 X5 + 0.0041 X6 + 0.0023 X14 (EQ.1) = -7365.866588 (4.3) (2.52) (-2.93) (-1.276) where X5 is fixed gross national investment, X6 foreign trade goods value, and X14 final GDP.

YGOPFr

From the empirical resultsof the regressionanalysisit can be seenthat X6 has had a strong individual correlation with its share in the r2 of 0.93 as high as 13.5%. It was found, however, that as a strong predictor there are not significant groundsfor it to be used as a suitable independentvariable for forecasting GOPFr. Also as independentvariable it is (X6) in fact the GOPFr but in terms of value; it producesa very direct logical correlation with its volume (i.e. it says the same) and consequentlywas excluded from further runs (Miller and Wichern 1977). Removing X6 resulted in the r2 (or explanatorypower of the independentvariables) of the first model dropping to about 0.80. The secondestimated model that resultedis shownin equation(2).
0.008 X5 + 0.0054 X14 -24277.46 is fixed gross national investment, and X14 final GDP. where X5
YGOPFF"":

(EQ.2)

240

Chapter 8

Scenario Analysis of the ISLB Demand

High multi-collinearity among independent variables of a model can give a false indication

for the value of the B coefficient and one or more variables with such behaviour should normally be droppedfrom the analysis. To increasethe explanatorypower of the estimatedGOPFT model in the secondstepof the GOPFr estimation procedure, the following five independent variables, as well as
the employment variable (X7)1, were included in the model with X5 and X14 one repeating

time (stepwiseregression)through runs 14 and up to run 20 as shown in Appendix 6 at a (Miller and Wichern 1977,Chatterjeeand Price 1991). 1. Port productivity/employee
(X15)

2. Rail productivity/employee(X16)
3. Government expenditure (X17) 4. Value of intermediate goods (XI8)

5. TransportInvestment(Xig) 6. Employment(X7)
7. Population (X20)

The result of eachof theseinclusions was evaluatedby p-values as shown from run 13 to 20 in Appendix 6. While all of them influenced the explanatory power of the second estimatedmodel found in run 13 to be higher than 0.80, they also increasedthe probability of a higher error than that adoptedby the 0.05% level of significance(high p-values). In assessing final two remaining independentvariables (XS and X14). the sign of the the
correlation analysis of X5 with GOPFr changes when assessedalone (-) and together with
X14

(+) which is not logical.

To find out the importance of these two independentvariables in the total r2, each was
regressed against GOPFr. It was found that X5 had a positive correlation of about r=0.23

I Thesevariablesin the first run were excludedpurposelydue to a perfect model with r2 I asthe numberof the variablesequalledthe numberof observations (Abrahamand Ledolter et al. (1983, P. 42). 241

Chapter 8

Scenario Analysis of the ISLB Demand

but could only explain GOPFIwith

an r2 = 0.056 which is a weak explanatory variable and

not significant, providing anotherreasonfor its deletion in addition to its incorrect sign. Thus,
X14

with population

(X20).

which was the only variable among the new set of

independentvariables which gave a p-value of lower than 0.05 (0-029), were regressed
against GOPFr to give the final model shown in equation (3).

The empirical resultscorresponding this final GOPFr model are given in Appendix 7. to aOO213X]4+0.00044X2o = -24552.8+ whereX14iSfinal GDP andX2opopulation.
YGOpFr

(EQ. 3)

The results of the GOPFr final model show a high value of r2 (0.87), with reasonablyhigh t-statistics corresponding to very low p-values for both independentvariables of the GOPFT model. They still do not indicate a good fit which can be assured analysisof the by residuals.Chattedeeand Price (1991, p. 8) pointed out that the:
"Results are valid and havemeaningonly insofar as the assumptions concerningthe " residual terms in the model are satisfied.

An examination of Appendix 7 shows that the estimatedparametersof all independent variables have the expectedsigns, with statistically significant impacts on GOPFr. The estimated form of the final model (equation 3) resulted in an r2 of about 0.87 which indicates a strong predictive ability. There exists moderate collinearity of about 0.49 between the two independentvariables (final GDP and population) which according to Wheelwright and Makridakis (1985, pp. 176-77) "is a frequent problem in economic and business data becauseof the high correlation between the different factors". This is acceptable given the long rangeaims of the project. Consequently,the analysis results in the fact that changesin the GOPFr are related to
changes in real GDP and population. In practice, to a great extent, there are cause and

effect relationships between them. The assumption is not unrealistic since GDP is a measureof the output of the economyof Iran whose efficiency is largely dependentupon the import of mainly intermediategoods. Also the growth of the population requiresmore 242

Chapter 8

Scenario Analysis of the ISLB Demand

employment and economic centres, which again calls for more foreign trade and greater aggregate demand for imports.

8.3.2 Domestic trade of Iran


There is not a comprehensivepublished data source about the geographyof interprovincial freight transport in Iran, nor for any of the foreign trade of CAC traffic through Iranian ports. The basic reason for the former is mainly the lack of research and comprehensivescientific studies in this particular field and also as there is no statutory reasonto do so. To further the analysis,at first the interrelationships availabledata must be analysedand of As processed. a secondstage,since the natureof transport servicesand networks vary, the processof data identification for links will be carried out separatelyfor road and rail modes.To specify the DOMTI model, 25 independent variableswere testedin 48 runs and in four phases. Four of theseruns specifically were only applied for the assessment each of variable of the secondand third estimatedmodels. The results of each phaseand run are in presented Appendix 9. The results of the correlation analysis(see Appendix 8) carried out betweenall variables involved in the DOMTI study, show that most of the independentvariables have a high positive correlationwith DOMTI, a preliminary indication of the linearity of the model. For the a properly understanding, resultsof the "net" correlation with DOMTI were categorised between0.70 and over (high), 0.50-0.70 (moderate),and lower than 0.50 (low). On this basis, 12 rankedhigh, 4 moderate,and 8 achieveda low score,two of which were negative
(Salvatore 1982).

To assess describethe future situation of DOMTI the actualdata for 1979-1993will be and usedand thereforethe mathematicalmodel must be established.

243

Chapter 8

Scenario Analysis of the ISLB Demand

The regressionresults, parameters,and residuals of the first estimatedmodel of DONITI used the backward elimination method as shown in Appendix 10 (Chatterjeeand Price 1991). Two independentvariables have a high explanatory power of about 0.96 with expectedpositive signs,significant low p-values and high t-statisticsas shownin (EQ.4).
0.038466755 X5 + 0.004440867 Xg (EQ. 4) 37740.6476+ (7.21) (15.083) (-2.392) where X5 is fixed gross national investment, and Xg value added of the agricultural sector.
=YDONM

One of the variables(Xq) is among those with a very high Pearsoncorrelation coefficient, while the other (X5) was weak. They were 0.88 and 0.36 respectively.
Therefore, in the second phase, the estimated model with X5 and Xg was run (runs 14-23)

and startedwith X2. which was intentionally droppedfrom the first phasedue to a perfect
model situation and presented a p-value higher than 0.05% level of confidence (0.06). Then, from the model, ten other explanatory variables (X15 to X25) were tested with a view

to modifying the model. One explanatory variable (X23) was added to the first DOMTI estimatedmodel in run 23 and the results are shown in (EQ. 5) and Appendices9 and 11. The results of the second estimated DOMTI model indicate an increasedr2 0.98, a t(2.85), but a very low p-value close statistic for X23which is higher than 2 and acceptable to 0.016. X5 Xg X23 YDOmn-54345.3444+0.034965907 +0.053414667 +42364.205387 = (9.032) (2.853) (-3.944) (7.95) (EQ. 5)

where X5 is fixed gross national investment, Xg value added of agricultural sector, and X23 productivity of agricultural sector/employee.

To forecastthe future volumes of the aggregate DOMTI, the final parameters the model of must be estimatedto provide the most reliable basisfor obtaining demandpredictionsof all
ISLB modes. The second estimated model includes three variables of which only one can

be considereda macroeconomic variable (X5) which has a weak correlation. Therefore, a

244

Chapter 8

Scenario Analysis of the ISLB Demand

for a strongermodel was continuedwith first and secondestimatedmodelsup to run search


27 with X24 and X25. Then since some of the explanatory variables have shown a

higher than 0.70 with DOMTI, they were added for the secondtime to the correlation of the secondestimatedmodel up to run 36. In runs 37 and 38, attemptswere madeto assess
individual contribution of the independent variables of the first estimated model (X5 and Xg). It was found that X5 with r2 = 0.13 had contributed much less than Xg to the

the first model and, therefore,it was dropped.The model with two explanatorypower of Xg and X23was run again with most correlatedindependent variableshigher than variables 0.70. In run 42, the third DOMTI estimatedmodel had 0.98% explanatorypower (? ) and the inclusion Of X14was obtained.The estimatedparameters other regression and resultsof this phase are shown in Appendix 12 and seem to be appropriate with a common explanatoryvariable like GDP in the final model (EQ.6).
YDOMTI -95088.6714+ 0.032627984 X9+0.01559261 X14+3311.95902 X23 = (8.301) (2.245) (5.856) (-5.886) (EQ. 6)

where Xg is value added of agricultural sector, X14 final GDP, and X23 productivity of

8.3.3 Central Asian and Caucasusforeign trade


The objective of this part of the study is to specify and estimatean aggregate model for forecastingthe freight of the CAC countries through Iran. The CACFT volume shows an high positive correlation of 0.79 with GDP per capita (Appendix 13) expected,relatively which indicates the possible linear functional relationship (but negative againsttime) and of also is supportedby the scattergram CACFr againstGDP per capita. In order to specify the parameters the CAC foreign trade model, the simple regression of techniquewas usedto estimatethe relationshipbetweenthe total CAC Asian foreign trade
through Iran and GDP per capita of all these eight countries. The number of observations is

limited to only sevendue to the non-availability of data which may undermineconfidence


245

Chapter 8

Scenario Analysis of the ISLB Demand

in the results(Fischeret al. 1996).The regressionmodel for CACFI' is shown in EQ. 7 and
results in Appendix 14.

YcAc= -1274653.287 + 125.6 X (2.862) (-0.913) where X is GDP per capita of CAC countries.

(EQ. 7)

8.4 Interpretation

of the results

8.4.1 General cargo and oil products foreign trade of Iran


To examinethe adequacy the model there are severalquestionsto be answeredabout of the value of the regressioncoefficient, confidenceinterval, and confidenceattachedto the forecast values (Wheelwright and Makridakis, 1985). The slopes of both explanatory
variables will be hypothesised as null i. e. (B = 0) which means those variables (X14 and X20) in the regression analysis with t-ratios higher than 2 will be deemed significantly

different from zero. The regressionanalysisfound that the coefficients for X14and X20are not close to zero but are approximatelyfour to five times higher than the related standard error of the explanatoryvariables(Wheelwright and Makriclakis 1985).Therefore,the null hypothesisis rejectedand both independentvariablesare considered have an impact on to GOPFr (seetable 8.1). In obtaining the value of the regressioncoefficients of the general variables,a 95% level of cargo and oil productsforeign trade of Iran (GOPFr) independent significancewas applied andp-values of 0.0007 and 0.00011 obtained,lower than the 0.05 level of confidence,which assured true valuesof the regression the coefficients. The results of the final GOPFT regressionanalysis are presentedin Appendix 7 with an explanatory power of about 0.87, which indicates most of the future variations of the GOPFr forecastsare explainedby the two selectedindependent variablesof final GDP and population. Both of these two variables have positive signs as expected. The t-ratios producedin the final GOPF17 model are about twice as great as the absoluteand required
246

Chapter 8 value of 2, confirming that the regression coefficients

Scenario Analysis of the ISLB Demand of this model are not zero. The

the model as shown in Appendix 6 was testedfor final GDP and population in stability of runs 21 and 30 by regressingeachagainstGOPFr which showedrelatively high r2(0.6 and 0.7 respectively).
Table 8.1 Impacts of B coefflcient on Change in dependent variables X14 Every one million Rials increase in GDP X20 Every one personincreasein population , flnal GOPFr model. B Coefficient (000 tonnes) 0.002127379 0.000442677 Impact on GOPFT tonnes 0.002173x 1000= 2.173 in tonnesincrease GOPFr 0.000442677 x 1000 = 0.4426 in tonnesincrease GOPF17

Population has a high correlation of 0.83 with GOPFr. A scattergram of this relationship,shown in Figure 8.2, also showsthis tight linear relationship.The regression for population indicatesthat there is an increaseof 0.443 tonnesof GOPFr for coefficient every one personincreasein population. When population was testedindividually against
GOPF17 the result is higher at about 0.617 tons. Population has a two-edged effect on

GOPFr. On one hand, it causesa greaterdemandfor consumablegoods which accounts for 19.5% of the total 30.5 million tonnesof Iranian foreign trade in 1993. On the other hand, the increasedpopulation can causeincreasedGDP requiring an expansionof the economythrough more investmentand,consequently, more capital and intermediategoods being imported.

Figure 8.2 Scattergram of the GOPFT of Iran by population during 1979-1993.


31000-29000-27000 3 25000-23000-21000 0 19000. 17000-15000. 13000-37000 40000 43000 46000 49000 52000 SSOOO S8000 61000

Population, 000

Source : Central Bank (Bank Markazi) of Iran, 1991,1992,1993,1994a, 1994b, various national accounts reports,variouspages.

247

Chapter 8

Scenario Analysis of the ISLB Demand

2. Real GDP also has a relatively high correlation of about 0.77 with GOPFT as shown in Figure 8.3. The regression coefficient indicates (Table 8.1) that there is an increase of 2.13 tonnes of GOPFF (based on seven sectors) for every one million Iranian Rials increase in the real GDP composed of four sectors (agriculture, industry and mine, oil and gas, and the One aspect of the effect of the GDP on GOPFT is that, through the main service sectors). different sectors to import capital and intermediate goods, there are qualitative and needs of improvements in the output of the GDP sectors. The other impact is that better quantitative higher volumes of domestic production can lead to increased exports. and
Figure 8.3 Scattergram of the GOPFT of Iran by final GDP 1979-1993.
3100029000 27000-2500023000-210(W 192 L 0 U 17 15 4 13000 9200 9700 10200 10700 11200 11700 12200 12700 13200

Final GDP, bn Iranian Rials

Source: Central Bank (Bank Markazi) of Iran, 1991,1992,1993,1994a, 1994b, various national accounts reports, variouspages. When using the GOPFT model for forecasting, the structure of the relationships between

is subject to change over time. Time could result in inaccurate forecasts due to variables the over- or under-prediction According and divergence from the mean (Saunders, et al. 1987).

to Barron and Targett (1985, p. 45):

The size of the "Ibe residuals determine the accuracy of the forecast . ..... residuals is therefore an indication of forecasting accuracy. An overall measure of the size of residuals is their standard error. " The other method is to plot the frequency of the residual histogram where the band of the residuals is presented against the frequency of the repetition of the band.

248

Chapter 8 The analysis of residuals provides a basis for assessing

Scenario Analysis of the ISLB Demand the accuracy of the GOPFr model.

It involves plotting the differencebetweenpredictedand actual GOPF17 againsteachof the independentvariablesin the final model. The residualsfor the GOPFT model are shown in Figure 8.4. The scatterplot of residualsshows the dispersionof the 15 points. The outliers or extreme
predicted values of GOPFr by the final model lie outside the range +2900 to -3600

in (Chatteeeand Price, 1991).The scattergram presentations Figures 8.2 and 8.3 for the GDP and population versus GOPFr show that the functional pattern of the model is appropriate. The plot of the estimated residuals against GOPFT tonnage indicates that residuals are randomly distributed about zero and lie between -3569 and 3247. Ten high value residualsare nearly balancedby 7 points (47%) located in the negativeband which
indicate that the predicted values of GOPFF by the final model are lower than the relative

observedones.Eight points lie in the positive bands,indicating that the predictedvaluesby the model are higher than actual volumes and can reasonably be acceptableas the systematicdistribution of the residuals.The almost counterbalancing residuals(+8 and -7) suggestthat there might be over forecastingby the model, which can be acceptedsince GOPFr involves long range prediction. According to Barron and Targett (1985, p. 43) "residuals should be random; without any pattern or order". There are few and in .... particular two extremeresidualswith different signs (belonging to data in 1983and 1985) and may indicate the existenceof some inadequacyin the assumptionsor in the GOPFr
model specification (Chatterjee and Price 199 1).

249

Chapter 8

Scenario Analysis of the ISLB Demand

Figure 8.4 Plot of the predictedresidualsagainstactual tonnageof the GeneralCargoand Oil Product ForeignTrade of Iran during 1979L93.

2900 240D 1900 1400 900 400 '100 QPO 1480D Redduals , -600 -1100 -1600 -2100 -26M -3100 -3600 GOM (000.tonnes) ; 16800 28800

100

244800

Therefore,accordingto this residualanalysisa first order regression model must be applied for forecastingthe GOPFT of Iran during the forecastperiod of 1993-2005.The summary to the first estimatedmodel is shownin Appendix 6. of eachrun up

8.4.2 Domestic trade of Iran


The results of the domestictrade of Iran (DOMTI) regressionanalysisare presentedin Appendix 14. The explanatorypower of the final model, with an r2 of 0.98, is higher than for GOPFT (0.87). This indicatesthat the proportion of DOMTI which is not explained by fluctuations in the three variables in the model is only 2%. All t-ratios (b/standarderror) are greater than the absolute value of 2, which indicates that the constant term and regressioncoefficients of all three independentvariablesare significantly different from 0. The accuracyof the final model as shown in Appendix 15, was expressed residuals as by

250

Chapter 8

Scenario Analysis of the ISLB Demand

the observed values. Since the largest residual is only -6.5%, it can be seen percentages of

that the final DOMTI regression equationis quite accurate. The final estimatedmodel (in run 42) resultedin three independentvariablesupon which the volume of the DOMTI will be assessed then forecast.Thesewere value addedin and
agriculture (Xq), final GDP (X14) and productivity of the agricultural sector MA all of

havethe expectedpositive signs.Their impact on DOMTI is shownin Table 8.2. which Tabl e 8.2 Impacts of B coefficient on final DOMTI model. B Coefficient Impact on DOMTI tonnes Change in dependent variables
(000 tonnes) Xg
X14 [L23

in Everyonemillion Rialsincrease the in valueadded agricultural in Everyonemillion Rialsincrease DP in increase Everyoneman/tonne agricultural productivity

0.03263 0.03263 1000 32.63 x = increase DOMTI in tonnes , 0.01559 1000 15.59 0.01559 x = in tonnes increase DOMT1 , 3311.96I 3311.96= 3311.960 increase DOMTI in tonnes

The coefficient of determination (? ) ranges from 0.96 in the first estimated equation to

0.97 and 0.98 in the secondand third estimatedmodels respectively.All three explanatory variables of the final model 1993). The stability of the model was testedby the omission of the variable X23 from the third it was found that both remaining variables (X9 and X14) still have very low pmodel and valuesof about 1.93x
10-06 (X9. X14,

and

X23)

are highly correlated with DOMTI (Jin

and 2.32746x

10-06

respectively(both with positive signs).Let

the three variablesin relation to DOMTI. us considereachof


1. Value added in agriculture (Xq)

This may be an indication of the impact of the sector activities on the total economic operation of the country. It accounts for a 0.88 correlation with DOMTI (Figure 8.5). Agriculture in Iran is an important sector.The bulk of Iran's exports are through this sector while imports of wheat,barley, and maize are significant items.

251

Chapter 8

Scenario Analysis of the ISLB Demand

Figure 8.5 Scattergram of the DOMTI of Iran by value added of agriculture sector during 1979-1993.
280000 260000 240000 220000 200000 180000 160000 140000 41800 2000 2200 2400 2600 2800 3000 3200 3400 3600

Value added of agriculture sector, bn Iranian Rials

Source: Central Bank (Bank Markazi) of Iran, 1991,1992,1993,1994a, 1994b, various national accounts reports,variouspages. 2. GDP (X14)

DOMTI also has0.87 correlation with final GDP excluding services(Figure 8.6).
Figure 8.6 Scattergram of the DOMTI of Iran by GDP during 1979-1993.
280000 260000 240000 220000

t 200000
180000 1600001 140000 49000 9500 10000 10500 11000 11SOO 12000 12500 13000 13500

GDP, bn Iranian Rials

Source: Central Bank (Bank Markazi) of Iran, 1991,1992,1993,1994a, 1994b, various national accounts reports,variouspages.

3. Productivity of the Agricultural Sector (X23)

This has a 0.85 correlation with DOMTI (Figure 8.7). Since this is a derived variable its (given by the ratio of agricultural volume/employees) high correlation can be explained directly by the impact of the humanskills and technologyon productionvolumes. The linearity of the single variable models can be determinedby assessing magnitudes the of the r2 close to + or -1 (which in the three estimatedmodels of the DOMTI were very high and varied between0.96 and 0.98).

252

Chapter 8

Scenario Analysis of the ISLB Demand

Figure 8.7 Scattergram of the DOMTI of Iran by the productivity of the agricultural sector during 1979-1993.
280000
260000 240000 220000 200000 180000 160000 140000

10

11

12

13

14

is

16

17

Productivity of the agricultural sector, man-tonne

Source: Central Bank (Bank Markazi) of Iran, 1991,1992,1993,1994a, 1994b, various national accounts reports,various pages.

The DOMTI linearity can be confirmed by plotting independentvariables against the dependent variable and inspectingthe relationship.In the caseof two or more independent variables, however, analysisof a plot against actual values of the dependentvariable is
more appropriate. In fact, this accounts for considering them as the dependent and

independentvariables in a simple regression (Taneja 1978). The plot of the residuals against predicted DOMTI in Figure 8.8 shows two high residuals with different signs respectively.One is equal to 6.5% of the total observeddata and the other to 4.2% in 1987 and 1982. The dispersionof the residuals shows a +8 and -7 distribution which suggest they are about normally distributed as shown in Figure 8.8 and can be acceptedas a reasonable model for DOMTI. The Final DOMTI regressioncan now be used to prepare a forecast.This requires that
Of X9s X14, values

and

X23

be estimatedand then be substitutedinto the regression equation

to compute a prediction for DOMTI in the year 2005.

253

Chapter 8

Scenario Analysis of the ISLB Demand

8.43 Central Asian and Caucasusforeign trade


The results of the Central Asian and Caucasus foreign trade (CACFT) ordinary least

in squares regressionmodel are presented Appendix 14. Equation 7 indicatesthat GDP per capita affectsCACFF positively. by It also showsthat about 38% of the future value of the CACFr cannotbe expressed the GDP per capita and is shown in terms of a high negative value intercept. The results in Appendix 14 indicate that with an increaseof one US $ in the GDP per capita of the whole CAC region, the CACFr increases about 125.6tonnes.Another result of the coefficient by in the model is the moderateexplanatorypower of the model (r2 of 0.62) which is probably for acceptable CAC countriesin their current transitional situation. The t-ratio of the GDP per capita is above 2 which confirms that the regression coefficient is not zero and With a p-value of about0.035, the future forecasts be further verified. acceptable. can

254

Chapter 8

Scenario Analysis of the ISLB Demand

8.5 ISLB scenario characteristics and descriptions


8.5.1 Justification and choice of ISLB scenarios
The application of just one scenario with any measure of probability is not of sufficient as an assessment different futures, while the use of many scenariosis not (1987), in sevensamplestudies(Becker 1983, possibleor practical. According to Schnaars Dekluyer 1980, Zentner 1975, Linnernan and Klein 1977, MacNully 1977, Wilson 1978, the Vanston 1977)all have used three to six scenarios.Consequently, developmentof and to investigatethe potential of Iran as a landbridgecountry three scenarios reasonable seems for CAC countries. (1987), six out of sevenstudiesusedthe baseyear'scenarios the According to Schnaars; as most probableanticipatedscenario.This approachis also adoptedin the ISLB study. The two scenarios labelled as optimistic and pessimisticviews (Mandel 1983). are other The contentsof thesethree scenariosare basedon researchand literature. Staff (1996) has three possibilities for the future of the Central Asian countries: proposed * they could come under the hegemonyof one or more outside powers, Russiabeing the most likely candidate. * they could lapse into chaos. Tadjikistan and Afghanistan have already done so, threateningthe security of adjoining regions. from within, through the emergence an * they could achieveequilibrium and coherence of anchorstateor states. On these basesthe description of the three ISLB scenariosas shown in Table 8.3 were developcd(Portcr 1985). The other aspectof the ISLB scenariosis to attempt to identify the effects of a changing
demand on the Iranian transport supply system. All the inputs of the ISLB scenarios will

examinethe future effects of the independent variableson volumesof demandand then

255

Chapter 8

Scenario Analysis of the ISLB Demand

Scenario No.

Table 8.3 Description of the three alternative ISLB scenarios of demand and supply to 2005 in Iran and CAC countries. of - Description scenarios Most probabIe scenario: It basically incorporates the most recent and available that economicinformation on Iran (1979-1993) and CAC countries (1985-1993)but assumes there will be no radical deviation from expectedpolicies of Iran, CAC republics, CIS and in the internationalenvironment.This scenario is used as the basis of comparisonof the other two probablealternatives.An increasein the real GDP of Iran and GDP per capita of the CAC to countries(on the basisof past historical data for thesecountries)is assumed affect the total flow of the foreign and domestictrade through the Iranian network and interface system.The will be greaterinvestmentand utilisation of existing transportcapacitiesand services response and and also someother macroeconomic political factors. Optin-dstic: A significant increasein oil prices and consequentlyoil revenuesof Iran and most CAC countriesas oil producing countries.This event results in much greaterinvestment for in transportinfrastructureand superstructure Iran and also higher transportdemandby both Iran and CAC republics. This optimistic situation is associated with significant changesin the foreign policies of Iran in particular USA. Pessimistic: This scenarioassumes radical changeand decay in the environmentof the a CAC and CIS countries and Iran. Re-emergenceof the former USSR leads to greater interaction of the CAC countries with CIS and Russianeconomic and political forces which a result in changesin the direction of CAC trade and consequently reduction of generallywill its flow through Iran. The objective of this scenariois to determineits effects on the internal and externalenvironmentof the transportsystemof Iran in termsof policies and strategies.

will evaluate their impact on the transport supply system (Ducot and Lubben 1980, Nijkamp and Blass 1994).Therefore, on the one hand, the ISLB scenariosare exploratory scenarioswith certain independentvariables as causeswith probable future situations as effects. On the other, the ISLB scenariosare descriptive as potential future events are selected and applied (Jungermann, 1985 and Neuman 1994). ISLB can also be an explanatoryresearchas it is partially quantitative and has identified relationshipsbetween GOPFr, DOMTI and CACFr and the six independent variables.

8.5.2 ISLB Scenario descriptions


For these scenarios a detailed Table 8.4 is developed showing the specific in quantifiable eventsor trends for each of the key issuespresented chaptersix (seeTable 6.3). Theseare in turn classified according to the four broad areas(geographical,political, technical, and organisational) used throughout this dissertation and first introduced in chapter four. The figures given in Table 8.4 are justified in more detail during the model forecastingundertaken the three scenariosin section8.6. for
256

-0 E9

t0
n C,

n 2. P-. 3, --1 c2.


,
=$ 1-

E r

Ei. E3
a

<

9-9-

51 0

0 Aaa. 0 :E
n

<

Ci. E;A n
-, 5.

Z a o

-2'c:
1-

- s.

Z- < 00< = 10
z ,

n-. tz q ng - r- ?000 C,

i,: E<2:'MSo 1

Er S

n =
v

n
A "

s. Er ju :rC2in0

CL 0 le 3

:E

o ', ,2 ', 0,0

* 0

1-ti c

cr

; E

80-,
nA :,. M-Z:

g0

:l
-CD

, g WZ >g

0. 0-

0 3
CDP. , 0

E'rg P-0, .4 0 1

00

n8=0 23. c2. , Pe

P-w <

R,

- t =-.

,2R

0g CS gF9.90
c. a

Ci.

D -, f. - Er

l2 1
0 -ti EE; C>-9E, -.

C->
n >
;- :3

ti

c*
<

77

E 1 0
53 0

&0-9 cr <w

bl --0 0 -0 cr
0
0

w-

log, :i%
c cI et
ti

ZD

0,0 12
R. ?. 0&j2,0

E;

5,

c31 0 s C",CD

0 -0

EL 0
F-

81

1'K> (
Fr.g rr i 00

0 o

OQ

[ 5.
A 0

ili

* F, ,,' Z, 4,.

' I ci

99

92.

c>

er E>1, wo 90

K,

:a
Eg 0 -v

ti
F-

0 (einI -ci
N'

I.
0

2,2 9 ., .

l, 9 g. '
ii

:u
* 2

1.
3E 9

-R w -

k -4

C,

je :1r.

e g

1 G.j ti l

(V

257

) C-. 120

to 'Z

PID

<

:3

E 0
rD

5< Z: ;3 F- m n 91 ", " F; E 00 o - b, OQ n ,"0,


C, gk a

,,

n0

e .
r

c> --

N9 p CI
31,0 <

Ea cr

r' -

n "Z
52 9

(b

ir -. ,
25-

'9gR
zr 2

c2.0* Z, ZA , 0 9e j -. cr Z.
n

Ei-

UQ "0
>"

5g 92 .9ll- 0 5, n' g.EL -=


R

E3 g9e-pZ; *j.
n

5' 'r ;22 ;A o 0 F; -

25, cL
c2.

'' f> EL =. .S"r, - (


0

N
cr

P2 22 15 90 ZE F -

. 10 -nm,

' E: 3 -. c2 nr2,, 0
-

4 -g 0'.
90
0.

-.

-1, -<
c

C p 0 10 E :r r:

0 c71
Un

5--

0,0 !',
:E

CZ- <0

rg
c-g 0

!j .m

2, :g
CD

*B
C, CL (1

EL

EL e Ei- .

c" < E; ZO r
Er. -

8 C,.3 (A 0
:2w

-RE a-> ApZ 2--. 0


I< g 5. eaX
z5 &

c21.

i ;15, m. . - u.
92. = E; -

`0

'0 @ ZL 0 .

E
fb

; _g
'2"

19E.

10

!A 00 !1

CD-

0
=-.09 g2.'
C,

00
A 92., 0, --. ." oo w-

c2.tA
CL r, 0-

E EH F '-

*2v
r 0"

=I,

g. 0 %. ;; < :i-. 0

Gn

C>20

>

F M7
c

eD

9, C0

'. r-

pl%

C -- 2. 00

JE 0 1 5. Sr 1

E; C1: < 2t A-

'.
--

E; E;9 2 . Ci .

<h

OQ

3 -5C, c,

;5- 2vE

' .e

2-

na

ZO
ID
g :j.

rr
Z;.;

.
9 E9

r,

F,

cr

-0

, Bw

F-Ei.

c
2j

ti

ge

2. H
kp
.>

0 9 F- X

<
00 5-

OQ 90 r', ;j

c g K ; s. 0 ,

jE

9. 2

51

Ei*

6a3.

12. U

. P'. C: i

.8

9. F
cn

03

0E>

0 ,

`V.

rr

E'r m rn r4 Ou

0.

258

--A
00

-ci.

--3
Z

'. > an

5n
ro

=<< 5 3 t r .

Cm
cB

:15 w 'IM
<,

e n
:n 0> Z

0
:i

0
L

1 n -v

, -a 0, i

c. "A.
=, E (A "

&

92.

Z: W f. 0- -,

%r'. '.

w,0.

CL 0
:E0

( c-,
*

c: 1 - -'-,
0

f. Zg z5i t c- - N
(C,

-- F 'n' "' g'. lp 0 -] 0


Z3
9 Z:, -

-i

-, !-ec <

t w 1- 7 <n a

1,21

w.

1g2- CD- .0-9&F. rw


n

22wn0F

g 3,2 c: tc 2
(D

r:

Z$
!-, <

--

b
n c.

oo

CD (D

:E , Ei 9 0
C)

9 E -< *i, C,
0--X :, g =

s 2 f, t^
Z,

.2 -1 1.00

<l 10 0 -

t-

M,

3-e8u Ei- ---ol


C)

o , ,

E.

7 er

"'
c

0q

92 0 9 E 23
'

9--

0 ;E
:1

g a--

' -

(-D
rt,

0 Ll: "

0n =, a. 0p
c> INIm

Ze
r *

AE*

qg ,

C) F r- E3 cr :i .

0,

E" -20wR9

go 0 C

-0

'>5R. E;

ga

: I.,

-0

<

i
(1

Ao

* p

9. r-

ig,

m0

-v F;

i2 jE

0 :er71iSt :8e
Ur

C,

ID

2,
9

0a ', -e G, 9 Z- r- nE. -a2-. g' 2 JE 0s0- g L


4

g-

ergE;

2> a.

g* = .

e 0
<
E0

79

n. a gl-.
E; cp:Z

0, &q &0

0 2
z 0' *2

'

0 .

E; '

2 IMW Er

eb
19

NI

Kr

tZ

259

00

00
0 J

CL

C)

0n

c71

C) GZ

00

cm : r"v
=. >0 t9 , (n , :j

x 01, n" 42, 0 L E


::

=..

50 ,8--

g0-29

,g3

Etn

-J
,

Ei

Ec pc2. a

E :i -. EL2. . -- g - ZE, ,- ;1 a'g 0- JAE ;, 2, . Fy


C, < ce

eZ. ] 3 CD

g0Z ac :ea e
00

F- , r_ ,55 p- c. i-5; -p, N E;" 0


:2E

9A Z"

-l- p :
<

EL

M ii 01; 5.

ci.

00 :i<R

2 <OW-.9 r,
QQ

, K- ZO 10 0 E; ,
::. 3

9 W9
zr

P2

:E"p, E0-

r.

<

r, E

&
t

P. l<
Ei Gl,Er.-0, -ti
0 m-

<E (12
-Z0. ro:

> :Z 155

0 0
9 10 e

E-=, 0
t', '

7;

e.

-0

5-9 E;
t

3 ci.
0

LA

7r

Ei

>

OQ

0 ,

0 ;5 lk
0 2=--9-

i.p -

2 2. < 5, l= Ei* L,

g.
,

a 0'

39
:Z

508

7rp

c-, n3Z,

0-0 . -

,-E --4

0 , ,-5:

Er c2. '. o e mm

55 E5z -2N E

8rg -,

,1. -

ei.

9
ZE,

FE
ZZ', B, e 10 <

:'

l 12'-.

-.

00

. 9 E,.EEZ.5t
i ri E; *

EF

li

c. JE c2.

asg0 ->

2K

1 r) n-

< Er. 12.

rE
t! F0 v. Z
(71

!2 10
CFQ C00.

4>2i
U'
P E',00
vi mH
0

Es9
= P,

rj

e2=
0,

z Er

H C, 0. 2. 092E; - g-<

H ,-

OQ

sk
A r-

b2
Gr < <
M-

s-

23

'. - 0-

%0

-0 . F,

g. Zn K. ", 2 E

>

g,

I
9: a
tv

!10.

Ei

gr

260

- -p Z: 53

vgl .gE,
OD ",

01C)cs 0-g; - EW 00 0, lp
12-.

9 C) dnr, '10 Vtod 0p --W -c,


ZZ,

0w p z- p ti ,
O>

00

ZC0 CJ

&"

5c0
; 501
10
1:

z3 >"' 2' 0
0

- z,n
.9X

'2 , cr :- 10ZO -08, -u -4 <>O - ;, w


%0

r.Z :Zc-

2w Z %0 c>
,.

-1

i- ti Z0A1

;.&-4 c
9. CCD

R)

s A U te0 --0

>O" 0

2f - E&

00 Cip
ON m S

2 Ei*,

Gn
LA

CD.

c12111 43 geN
1? -

-vZ2.
t^

r- *

00

9g

'ci
CA

ba

- 10

EF

5*

0
ll %o ,0-0 e (A eq
C)
5

r- 9

F
m.

ij 5*

t-< ," o gC% ig t.


%0

53

e', r-a `97 gE -9 S. r. PC26

Ch -2 0ww
9= cim
5; *

>o
W

-0 li,

vZCn
bu E. D

0" E

c--1.1.

to> ', 09
li :

1[.

2.0 Ei

e,

v--,
E
0*

F- L,
8, g, L,

p. ti %0 w - !ij

cr

D. C 'Z-' E-

w0N

%0 UJ%D ,,

>O Ici
C)

1110M -c 10 1-1

-0 E g

ZA

LE -

'a E '0 E

tr

2u. "E

191

!', 8

Ir_ r,

;6 ,a

c sE
W0
C.

g ic,

EO
ll

c,- E j*P
Z: 0

Er
'P
-.

"

oo, ', e---0 -

ti 0L'

%. I

NO th %0 tA .
15

z.

c2.

g %0

iz E 0 z,

C) Er CJ

e- p

ei.

t: -1 :c. 9 P., lp b--

-5 _"

-p 3 PZCO, w t

1 1-, e
rA

'P

ta

ww
22W

O. :3u 14 Z. ) , 5' z
ki 0"1-.

1.9 rr

zo ZZ

Iv
F=

11

il

2. rr ch

Glu

261

Chapter 8

Scenario Analysis of the ISLB Demand

8.5.2.1 ISLB most probable scenario

This scenariorepresents most likely situation for the environmentof the ISLB. the The trendsin this scenarioare basedon the independent variablesusedto forecastthe three volumesof demandfor GOPFF,DOMTI, and CACFT. The future growth for Iran is based on 1989-1993 a stableperiod. as
It is assumed there will be no significant changes in the domestic politics of Iran and the

CAC countries. Iran moves toward international economy and according to Amirahmadi (1996, p. 127):
'The Iranian governmentbelievedthat the break-upof the Soviet Union had resultedin a less ideological global order, which is more than ever centredon economicissuesand coin With suchconsiderations mind ... Iran embarkedon bringing its economic operation. in line with current trendsin the internationaleconomy.Seekingloansand policies investmentsfrom abroadand approaching IMF and the World Bank were initiatives taken the in this light. As a prerequisitefor IMF assistance, governmentwas compelledto acceptthe the for economicliberalisation and stabilisation". organisation'sguidelines

Therefore, the political issues in this scenario assumethe continuation of the present situation, with both Iran and the CAC countries attempting to pass the previous critical to overcomepresenteconomicinstabilities at national and international levels. period and This probablesituation can be supportedas Barylski (1994, p. 413) accordingto Kozyrev (1993) hasnoticed that:
"Me ideal internationalsystemfor Central Asia as a club of friends of regional stability " consistingof Russia,Iran, Afghanistan,Pakistanand the rive former Soviet republics.

During this period Iran would have the ability to cope with developmentprogrammesand to improve the ISLB, mainly by simplification of the transit trade and the transport regulationsrelating to its corridors for the CAC countries.The moderatescenarioassumes that internal opposition forcesin Iran have sufficient power to contain a significant adverse foreign relations development with USA, given that external forces are not effective against present Iranian policies (Roberts (1996). It is assumedthere will be no official taken againstIran. There are no internal or relationswith USA but also no seriousmeasures

262

Chapter 8

Scenario Analysis of the ISLB Demand

external threatsin terms of revolutions or war with other countriesfor both Iran and any of CAC countries. Therefore, the Iranian economy operatesunder present conditions with low economic growth. Further specific assumptionsfor this scenario central planning and are shownin Table 8.4. 8.5.2.2ISLB optimistic scenario This scenarioarguesthat Iran, in responseto the improved internal political reforms, settles its conflicts and improves relations with the USA and other countries. Oil price increases provide significant currencysourcesfor developmentprojects and completion of transport projects. This is a period of credit and high stable growth for Iran. Further for this scenarioare shownin Table 8.4. specific assumptions
8.5.2.3 ISLB pessimistic scenario This refers mainly to political changes in CIS and CAC countries, which causes ISLB

competitors to attract the Asian foreign trade of the CAC countries. These will still be opportunities for neighbouring and other countries to have trade with each of the fonner USSR republicsbut in a more tightly controlled way, which will createa more competitive situation for the ISLB. Further specific assumptionsfor this scenarioare shown in Table 8.4.

8.6 Model forecasts


in Table 8.4 for the three ISLB scenarios, three Under statedassumptions the made
final forms of the model will be used to forecast the total tonnage of the three ISLB

for tonnageof the ISLB policy scenarios will then scenarios the year 2005. Theseaggregate be distributed over the ports and border crossingsas well as onto the rail and road modes which together form the transport system of Iran under each scenario.The translation of these forecasts as the transport supply of Iran will provide the new level of service requirements which may require expansion of the physical transport capacities when 263

Chapter 8

Scenario Analysis of the ISLB Demand

comparedwith baseyear facilities and infrastructure(Zwarn and Valk 1984).Eachof these stepsmust be carried out by multiplying the future input values of each of the significant independent variables by their respective estimated parameters and then summing. Accordingly, future values of the independentvariables under most probable, optimistic will take place for eachof the three areas and pessimisticmust be forecast.Model forecasts
of study under three scenarios.

8.6.1 Most probable scenario


8.6.1.1GOPFT Three attemptswere madeto forecastthe future values of GDP (for the year 2005) by developing a model of its growth againsttime, by linear, exponential,and polynomial techniques,all of which exhibited a low coefficient of deterininationof 0.24,0.23, and0.29 respectively. ForecastingGDP is an uncertainexcerise(Bell et al. 1983).GDP is the main political and macroeconomicindicator of GOPFr. Its composition can be reflected by other different variableslike investment,value added,etc. The averageannualgrowth of GDP during the turbulent yearsof the revolution and the war was equal to 1.9%.The 1.9% averagegrowth in GDP is a reflection of the past behaviourof the Iranian economyduring 1979-1993and will be considered the basisfor the growth which will underpinthe real GDP forecasting as of future values of the most probable scenariofor GOPFr. While in the most probable that there will be no war and no unusualincidents for Iran, its low scenarioit is assumed value (1.9% base growth) may, to a great extent, reflect the instability of the economic systemduring the period over which data was collected (column 4 in Table 8.5 for 19791993).The coefficient of the GDP growth between1993-2005must be at such a value that keepsfuture GDP valuesfirstly within a logical boundaryand then, secondly,between,the optimistic and pessimisticGDP scenarios.During 1979-1993the highest positive growth 264

Chapter 8

Scenario Analysis of the ISLB Demand

in real GDP was 13.2 % in 1983 (column 4 in Table 8.5). It will be assumed that by 2005,

the most probable GDP with a base growth of 1.9% (in 1994) will expand to 8.5% (annually by an addition of 0.6% per annum from 1994). This was derived from the averagevalues of the GDP growth during 1989-1993as the more stable period after the war with Iraq. Under such assumptions, the values of real GDP for 1994-2005 are in calculatedand surnmarised Table 8.5 column 3. Forecastsof population are basedon a linear time seriesprojection with a coefficient of determination(r2) of about 0.9996which indicatesan almost perfect historical relationship given by the mathematicalmodel in equation (8) with the forecastsderived shown in column 2 of Table 8.5.
Table 8.5 Actual and forecast independent and dependent variables of the most probable GOPFT scenario by! rrowth factor and time series projection. 5 2 3 4 Year GOPFractual Actual forecast Actual and values of and GDP growth (0.6 rate in & forecast thepopulation Iran(X20)forecast of Iran (X14) GDP over1.9from1993) 1979 37991000 10551300 16247 1980 39646000 9555500 13027 -9.44 1981 41221000 9320700 14166 -2.46 1982 42800000 10539800 13.10 17289 O 1983 23756 44438000 11934700 13. 1984 46201000 12043800 0.91 19250 1985 47807000 18723 12072300 0.24 1986 49363000 10248900 17660 -15.10 1987 50995000 10368100 1.20 18747 1988 52672000 9468000 16489 -8.70 1989 54504000 22850 9781500 3.30 1990 24956 56401000 10930200 11.70 1991 57799000 12181200 28700 11.40 1992 12477800 59229000 2.40 277731 1993 60708000 13101000 305551 5.00 1 ForecastIucs: j, "- Most, '. - 'probable scenario,, ',,forecasts va ,-I,, 26.75/14=1.9' 1994 31426.30 62599400 13349919 1.90 1995 32860.20 64242800 13683667 2.50 1996 34486.80 65886200 14107860.7 3.10 1997 67529600 36321.80 14629851.6 3.50 1998 38384.80 69173000 15258935.2 4.30 1999 39977.40 70816400 16006623 4.90 2000 43298.80 72459800 16886987.3 5.50 2001 46216.00 74103200 17917093.5 6.10 49496.10 75746600 19117538.8 6.70 _2002 2003 53191.70 77390000 20513119.1 7.30 2004 57366.60 79033400 22133655.5 7.90 2005 62097.00 80676800 24015016.2 8.50

265

Chapter 8

Scenario Analysis of the ISLB Demand

Yp, =36305 + 1643.4 X p

(EQ. 8)

27 as future time periods move from 1994 to X is given values of 16,17,18 where . ..... 2005 and Y stands for population and forecast values as shown in column 2 of Table 8.7.

Using the future values of GDP and population and the estimatedfinal model shown in 5, and on the basisof the final model parameterestimations,the bestestimate equation future values of the GOPFT of Iran was predicted up to the year 2005 as shown in Table 8.5 column 1.
Table 8.6 Comparison between regression GOPFT forecast and three time series models (000 tonnes). Models' 4 3 2 0474. YTsTp eU. x Y 16082+77.417 X2-225.27 X (9) Y 13701 EQ. =12737+1101.2t (10) Polynomial Regression Exponential Linear trend ?=0.87 0.29 0.23 0.24 rz= rz= r-32296.43 31426.30 29053.28 30356.20 1994 34625.92 32860.20 30463.56 31457.40 1995 37110.25 34486.80 31942.31 32558.60 1996 39749.41 3632180 33659.80 33492.83 1997 1 . 35118.62 42543.40 38384. 34761.00 1998 45492.23 39977.40 36823.32 35862.20 1999 48595.89 43298.80 38610.77 36963.40 2000 51854.38 46216.00 40484.99 38064.60 2001 55267.71 4949910 39165.80 42450.19 2002 1 58835.88 53191.70 44510.78 40267.00 2003 62558.87 5736660 46671.39 2004 41368.20 66436.70, 62 48936.88, 43469.40, 2005

Time seriesmodels rely heavily on historical data and stable conditions. They attempt to find the trend, seasonal, cyclical variationsthat exist in the dataseries.There shouldbe and a long historical period of datawith which to test the model. Forecastingis a practice that should be groundedon more than just a good fit between simulated and historical values. Therefore, as an additional procedurefor forecastingthe GOPFr, ex-post forecasts of the GOPFr were derived from models based on trend, exponential, and polynomial time series techniques and then compared with the main
regression model in Table 8.6 (Doganis 1985).

266

Chapter 8

Scenario Analysis of the ISLB Demand

The coefficient of determination (? ) in Table 8.6 indicates that all three time series models

havelow explanatorypowerscomparedto the regressionmodel. According to the latest data available, a comparisonshows that the GOPFI7forecast for 1995 is only 6.6% over-estimatedwhen comparedwith actual foreign trade in the same year (PSO 1994).
8.6.1.2 DOMTI The aims of the DOMTI analysis were to assessthe impact of different variables on

DOMTI and also to derive a model capableof predicting the domestictrade of Iran in the long run between 1993 and 2005 by the application of multiple regression analysis. Obviously, the accuracyof any forecastingmodel will be conditional upon the accuracy with which the independent variablesare predicted.
The future values of DOMTI GDP are based on the most probable scenario GOPFr

assumptionas shown in column 4 of Table 8.7. The future values of the value addedof agriculture(Xq) was consideredto be 4.8% and productivity of the agricultural sector(X23) for 4.2% in 1994 and then were upgraded with the average growth rate of these two variables during 1989-1993(0.5 and 0.4 respectively). The future values of these three independent variablesare shown in column 2 and 3, and 6 and 7 of Table 8.7. By inserting the future valuesof thesethree independent variablesinto the final DOMTI model (EQ.5), future values of DOMTI for 1994-2005were estimatedand shown in column I of Table
8.7.

An assessment made betweenmultiple regressionand four other different time series was techniques (EQs. 12,13,14, and 15), for estimating DOMTI forecasts, all of which

estimatedin an r2 lower than the DOMTI derivedby multiple regression(seeTable 8.8).

267

Chapter 8 YDol,m = 261.25X2 -I 030652.77X + 1016654072 where x is 1994,1995,1996 ............... and 2005. YDol,m = 153277 e'*"" where x is 16,17,18 . ...........27 equivalent to 1994,1995,1996,.... YDOMT, 7033.51X -13763527.26 = where x is 1994,1995,1996 ............... and 2005. YDomn = 13966578.09 Ln (X) -105855434.19 where x is 1994,1995,1996 ............... and 2005.

Scenario Analysis of the ISLB Demand (EQ. 12) (EQ. 13) 2005. (EQ. 14) (EQ. 15)

There were significant differencesbetweenpredictedand actual valuesshown in Table 8.8.


Among other techniques the polynomial has the highest r2 while the results of the log and

exponentialmodelsarevery close.
Table 8.7 Actual and forecast of the DOMTI independent and dependent variables by growth ctor and time series proje tion (000 tonnes). PAS GDP VAA X X14 xg 23 Values Growth Values Growth Growth Values Actual Y 5 6 7 4 3 2 1 11.36 10551300 1851200 1979 159713.50 11.08 9555500 3.40 1914900 1980 156657.59 -9.44 -2.50 9.39 9320700 2.00 1952700 1981 143905.97 -2.46 -15.30 34.00 12.58 10539800 13.10 7.10 2091400 1982 187046.21 24.00 11934700 13.20 15.59 4.90 2193000 1983 214483.34 0.91 12.14 12043800 7.30 2353700 1984 207417.78 -22.10 0.24 13.3 9.60 12072300 7.80 2537600 1985 221041.09 14.75 10.90 10248900 -15.10 4.50 2650500 1986 209298.56 1.20 14.97 1.50 10368100 2.50 2715800 1987 192223.03 13.53 9468000 2648000 1988 187630.97 -8.70 -9.60 -2.50 3.30 12.33 9781500 3.70 2746000 1989 185742.17 -8.90 25.80 15.51 10930200 11.70 8.10 2967500 1990 217939.21 14.96 12181200 11.40 5.20 3120200 1991 252590.10 -3.60 2.40 16.49 10.20 12477800 7.40 3351600 1992 262396.35 5.00 15.85 13101000 5.50 3535700 1993 277223.84 -3.90 , Forecasts values ' "i, "ik,. J 11,, , - I I, "' 1,,",,, , I-,,, 4.2 1.9 16.5 13349919.00 4.80 1994 288618.90 3705413.60 4.6 2.5 17.3 13683667.00 5.30 1995 309172.90 3901800.50 5.0 3.1 18.2 14107860.70 5.80 1996 319859.10 4128105.00 3.7 5.4 19.2 14629851.60 6.30 1997 339795.80 4388175.60 5.8 4.3 20.3 6.80 15258935.20 1998 362984.10 4686571.50 4.9 21.5 6.2 7.30 16006623.00 1999 389779.50 5028691.20 5.5 22.9 6.6 16886987.30 7.80 2000 420941.40 5420929.10 6.1 24.5 7.0 2001 8.30 17917093.50 456983.10 5870866.20 6.7 26.3 19117538.80 7.4 8.80 2002 498519.50 6387502.40 7.3 28.4 7.8 20513119.10 9.30 2003 546617.60 6981540.10 7.9 30.7 22133655.50 8.2 9.80 2004 600688.30 7630823.30 8.5 33.3 24015016.20 8.6 2005 10.30 664279.50 8416798.10 X14 X23 Xq+0.01559261 + 3311.95902 (6) YI)OMTI=-95088.6714+0.032627984 VAA: Valueadded theagricultural PAS:Productivity theagricultural/employee of sector. sector. of

268

Chapter 8

Scenario Analysis of the ISLB Demand

Figure 8.9 DOMTI forecasts for Iran 1993-2005


680000 630000 0 , 580000 530000

0 cr 480000 0 430000 0 380000


280000 da 330000

230000 IF-1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

Years

Table 8.8 Comparison between regression of most probable DONITI forecast and three time series models for 1994-2005(000 tonnes). 'Models (13) Polynomial(12) Log (15) Regression Linear trend (14) Exponential 0.67 0.69 0.67 0.67 --: 265349.20 271854 261201.60 288618.90 1994 261291.68 1995 274608.60 283327 268198.80 309172.90 268325.19 1996 284191.10 275358.70 295324 275196.10 319859.10 1997 294107.90 307842 282392.21 282193.40 339795.80 304370.80 1998 320883 289190.60 362984.10 289425.72 314991.80 334446 296173.90 389779.50 1999 296459.23 303157.20 420941.40 2000 325983.50 348532 303492.74 2001 337358.70 393-141 310140.50 456983.10 310526.25 2002 349130.80 378272 317123.80 498519.50 317559.76 2003 393925 324093.10 546617.60 324593.271 361313.80 331062.40 600688.30 2004 410101 331626.781 373921.80, 664279.50 338031.70 2005 426800 338660.291 386969.801

8.6.1.3 CACFT The future valuesof the GDP per capita are required for forecastingCACFr during the 1994-2005.The latest available data concerning the foreign trade of most CAC period countries shows a high negativevalue for exports while imports are in a better situation. At the same time the "discrepancybetween the initial trade data and the adjusted trade values is in some casesquite substantial" (UNCTAD/UN 1994,p. 69). It is assumedthat they have had a fiegativegrowth in GDP per capita of about -2% in 1994(Michalak 1994). Then, it will grow by 0.6% per annum to produce about 1.85 million tonnesof trade and close to that which thesecountrieshad with Iran in 1993 (seeTable 3.39. On the basis of 269

Chapter 8

Scenario Analysis of the ISLB Demand

this assumption, by calculating the growth rate the future values of GDP per capita of CAC

countriesis insertedin the final CACFT model (EQ. 7). The CACFT valuesfor the period 1994-2005were estimatedand are shown in Table 8.9 where comparisonsof the CACFT forecastvolume for 1995with actual transit trade through Iran (about 1.35 million tonnes in 1995)showsabouta4% under-estimation (PSO 1994).
Table 8.9 CACFr independent and dependent variables by growth factor and time series projection (tonnes). GDP per capita CAC Asian foreign trade
YCACFT x

Volumes 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

GDP per capita Values 31800 31700 3630570.60 32700 3780642.90 35400 3509606.90 36800 2728208.40 36100 33400 1896981.90 1647787.90 23100 1351898.30 21200 Most probable scenarid forecast values (growth rate 0.6) 1334812.60 20776.00 1298275.30 20485.10 1277689.40 20321.20 1272590.10 20280.60 1282776.20 20361.70 1308348.40 20565.30 1349670.80 20894.30 1407409.10 21354.00 1482505.30 21951.90 1576253.20 22698.30 1690285.40 23606.201 1826674.50 24692.10 1

-0.30 3.20 8.30 4.00 -1.90 -7.50 80 -30, -8.20 -2.00 -1.40 -0.80 020 -. 0 40 . 1.00 1.60 2.20 2.80 3.40 4.00 4.60

8.6.2 Optimistic scenario


8.6.2.1GOPIFT It is assumedthat the population growth in the optimistic scenariois at the same rate as the most probable scenarioand only the future real GDP will be forecast for the period of study. A maximum value of 13.45%for the GDP of Iran in 2005 is consideredto be similar to what was experiencedin 1983 as the highest value in the last two decades. Therefore, the optimistic scenarioGDP varies between 1.9% and 13.45 % during 1994270

Chapter 8

Scenario Analysis of the ISLB Demand

2005 with an increase in the growth rate of 1.05% annually. On this basis, future values of

both independent variablesare shownin Table 8.10. Table 8.10 GOPFT independent and dependent variables by growth factor time series pr jection for the optimistic scenario (000tonnes). and 3 2 4 1 Year GOPFT values the Actualandforecast GDPgrowth of actual Actualandforecast Iran(X20) (000) Tran(X14) GDP of & forecast onin 10551300 37991 1979 1624 39646 9555500 19801 13027 -9.44 9320700 41221 1981 14166 -2.46 10539800 42800 13.10 1982 17289 11934700 13.20 44438 1983 23756 46201 12043800 0.91 1984 19250 47807 12072300 0.24 19851 18723 10248900 49363 1986 17660 -15.10 50995 10368100 1.20 18747 1987 9468000 52672 16489 1988 -8.70 9781500 54504 3.30 1989 22850 56401 10930200 11.70 24956 19901 12181200 11.40 57799 28700 1991 2.40 12477800, 59229 1992 27773 1 1 1 5.00 1 131010001 60708 1993 30555 Optimistic forecasvalues t scenario 1.90 13349919.00 1994 31426.26 62599.40 2.95 13743741.60 1995 32988.20 64242.80 4.00 14293491.30 65886.20 1996 34882.27 5.05 15015312.60 67529.60 1997 37142.84 6.10 15931246.70 39816.88 69173.00 1998 7.15 17070330.80 70816.40 1999 42966.22 8.20 18470097.90 72459.80 46670.82 2000 9.25 74103.20 20178582.00 2001 51032.99 10.30 22256975.90 75746.60 2002 56183.06 11.35 24783142.70 77390.00 2003 62286.89 12.40 27856252.40 79033.40 2004 69555.70 13.45, 31602918.30 80676.80 2005 78259.21

8.6.2.2DOMTI The future values of the optimistic scenario GDP are based on the most probable GOPFr assumptions, where the growth of 1994(1.9%) goesto 13.45%in 2005 with a rate of about 1.05%annually as shownin Table 8.11. The maximum rate of growth in the productivity of the agricultural sectoris assumed be to 26.2% in the year 2005. This was actually experiencedin 1990 as the best past situation

271

Chapter 8

Scenario Analysis of the ISLB Demand

during the period of study. A minimum value of 1.9% is similar to that for the most

DOMTI. The optin-dsticscenariovalue added of the agricultural sector ranges probable between a minimum of the most probable scenario DOMTI and a maximum value of 13.49%,similar to the optimistic scenarioGDP (13.45%)in 1983.
Table 8.11 DOMTI independent and dependent variables by growth factor and time series projecti n for the optimistic scenario (000 tonnes). X14 23 xg Values Growth Values Growth Values Growth Actual Y 11.36 10551300 1851200 159713.50 1979 1 11.08 9555500 3.40 1914900 1980 156657.59 -9.44 -2.50 9.39 9320700 2.00 1952700 1981 143905.97 -2.46 -15.30 12.58 34.00 7.10 10539800 13.10 2091400 1982 187046.21 4.90 11934700 13.20 15.59 24.00 2193000 214483.34 1983 0.91 12043800 12.14 7.30 2353700 1984 207417.78 -22.10 0.24 9.60 7.80 12072300 13.3 2537600 1985 221041.09 90 10248900 -15.10 14.75 10: 4.50 2650500 1986 209298.56 1.20 14.97 1.50 2.50 10368100 2715800 1987 192223.03 9468000 13.53 2648000 1988 187630.97 -8.70 . 9.60 -2.50 3.30 12.33 9781500 3.70 2746000 1989 185742.17 -8.90 15.51 25.80 11.70 8.10 10930200 2967500 217939.21 1990 11.40 14.96 5.20 12181200 3120200 252590.10 1991 -3.60 2.40 16.49 10.20 3351600 7.40 12477800 1992 262396.35 5.00 15.85 13101000 3535700 5.50 1993 277223.84 -3.90 Forecasts values 4.20 1.90 16.50 4.80 13349919.00 3705413.60 1994 288618.90 6.20 17.52 2.95 13743741.60 5.59 3912546.22 304906.09 1995 18.96 8.20 4.00 6.38 14293491.30 4162166.67 1996 326381.64 10.20 5.05 20.89 15015312.60 7.17 4460594.02 1997 353778.82 12.20 6.10 23.44 15931246.70 7.96 4815657.30 1998 388087.95 14.40 26.82 7.15 17070330.80 8.75 5237027.31 1999 430778.11 16.20 31.16 8.20 18470097.90 9.54 5736639.72 2000 483294.66 18.20 36.84 9.25 20178582.00 6329234.60 10.33 2001 548054.02 20.20 44.28 22256975.90 10.30 11.12 7033045.49 2002 628068.68 22.20 54.11 24783142.70 11.35 11.91 7870681.21 2003 727342.42 24.20 67.20 27856252.40 12.40 12.70 8870257.72 2004 851239.03 26.20 84.80 31602918.30 13.45 2005 1007011.94 10066855.48 13.49

8.6.2.3 CACFT It is assumed that IRIRC under the optimistic scenariowill have about 8 million tonnes of demandfor rail from the CAC countries (Sanate-Hamlo-Naghl1996c).To accomplish demand,it is assumedthat the CAC countries with a negative growth such an optimistic per capita GDP of about -2% in 1994, should have a considerablegrowth rate of about 32.48% annually attaining to $ 95004 GDP per capita in 2005. On this basis, the total

272

Chapter 8

Scenario Analysis of the ISLB Demand

optimistic scenario GDP per capita values of CAC countries were computed and inserted in

7 to producethe optimistic CACFr volumes shownin Table 8.12. equation


Table 8.12 CACFT independent and dependent variables by growth factor and time series projection for the optimistic scenario (000 tonnes). GDPper capi (X) CAC Asian foreigntrade Volumes(YCACFT) Values Growth 32700 3.20 3630570.60 35400 3780642.90 8.30 36800 3509606.90 4.00 36100 2728208.40 -1.90 33400 1896981.90 -7.50 23100 1647787.90 -30.80 21200 1351898.30 -8.20 Optimisticforecasts values 20776 1334812.60 27524 2182361.11 32.48 34272 32.48 3029909.91 41020 32.48 3877458.71 47768 32.48 4725007.51 54516 32.48 5572556.31 612;4 32.48 6420105.11 68012 32.48 7267653.91 74760 32.48 8115202.71 81508 32.48 8962751.52 1 88256 32.48 9810300.32 32.48 95004 10657849.12

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

8.6.3 Pessimistic scenario


8.6.3.1GOPFT The pessimisticforecastingscenarioof GOPFr keepspopulation growth as specified for the most probable scenario.However, the growth of the real GDP (X14)drops from 1.9% in 1994by an annualrate of 1.6% to -15.7% in 2005 which is close to the value of real GDP in 1986 as shown in Table 8.13. In fact, the independentvariables of the for GOPFr move in different directions for the forecastperiod. The pessimistic scenario downwardpressure the growth rate of the real GDP of Iran arisesfrom the effects of the on US embargo policy and the external political environment of Iran. In the pessimistic scenario,it is expectedthat the GOPFF comparedwith the baseyear (1993) drops by about 26.5% to 22.46 million tonnesin 2005.

273

Chapter 8

Scenario Analysis of the ISLB Demand

Table 8.13 GOPFIPindependent and dependent variables by growth factor and time ries projectio for the pessimistic scenario ((000 tonnes). 3 4 2 1 Growth Actual and forecast Actual and forecastGDP Year GOPF17 actual (X14) & forecastvaluesof the population of Iran in Iran (X2o) (000) 10551300 37991 16247 1979 39646 9555500 13027 1980 -9.44 9320700 41221 14166 1981 -2.46 42800 10539800 13.10 1982 17289 44438 11934700 13.20 23756 1983 46201 12043800 1984 19250 0.91 47807 12072300 0.24 18723 1985 10248900 49363 1986 17660 -15.10 10368100 50995 1.20 18747 1987 9468000 52672 1988 16489 -8.70 9781500 3.30 54504 1989 22850 10930200 11.70 56401 24956 1990 11.40 57799 12181200 1 1991 28700 2.40 12477800, 592291 1992 1 27773 607081 131010001 5.00 30555 1993 Forecast values 62599.40 1.90 13349919.00 1994 31426.26 13309869.30 0.30 64242.80 1995 32064.05 65886.20 13136841.00 1996 32418.63 -1.30 67529.60 12755872.60 1997 32330.22 -2.90 12181858.30 69173.00 1998 1 31830.68 -4.50 11438765.00 70816.40 1999 30970.99 -6.10 10557980.10 72459.80 2000 29818.01 -7.70 9576088.00 74103.20 2001 28449.68 -9.30 8532294.40. 75746.60, 2002 26949.49 -10.90. 7465757.60 77390.00 2003 25400.86 -12.50 6413085.80 79033.40 2004 23881.77 -14.10 5406231.30 80676.80 2005 22460.27 -15.70

8.6.3.2DOMTI The pessimisticscenariofor DOMTI assumes that both the DOMTI variableswill have a downward growth. Value added of agriculture sector moves downward with a rate of 0.7% to about -2.9% in 2005 which closely correspondsto the situation experiencedin 1988. Productivity of the agricultural sector will decline by -22.2% per annum a rate of growth of -2.4% in 2005, nearly matching the 1984 situation. The future values of these independentvariablesand their effects on volumes of the pessimisticscenariofor DOMTI are shown in Table 8.14.

274

Chapter 8

Scenario Analysis of the ISLB Demand

1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 F20EO: 4 12005

Table 8.14 DOMTI independent and dependent variables by growth factor and time series profection for the pessimis ic scenario (000 tonnes). Actual Y Values (X9) Growth Values (X, 4) Growth Values (X23) Growth 1851200 159713.50 10551300 11.36 156657.59 1914900 3.40 9555500 11.08 2.50 -9.44 . 2.00 143905.97 1952700 9320700 9.39 -2.46 -15.30 2091400 187046.21 7.10 10539800 13.10 12.58 34.00 2193000 214483.34 4.90 11934700 13.20 15.59 24.00 2353700 207417.78 7.30 12043800 0.91 12.14 22.10 2537600 221041.09 7.80 12072300 0.24 13.30 9.60 1 2650500 209298.56 4.50 10248900 14.75 10.90 -15.10 2715800 192223.03 2.50 10368100 1.20 14.97 1.50 187630.97 2648000 9468000 13.53 -2.50 -8.70 -9.60 2746000 9781500 3.70 185742.17 3.30 12.33 -8.90 2967500 8.10 217939.21 10930200 11.70 15.51 25.80 252590.1 3120200_, 5.20 12181200 11.40 14.96 3.60 _ 262396.35 3351600 7.40 12477800 2.40 16.49 10.20 277223.84 3535700 5.50 13101000 5.00 15.85 -3.90 Forecast values 288618.90 3705413.60 282043.90 3553491.60 275072.70 3432672.90 264452.40 3339990.70 250341.70 3273190.90 233723.40 3230639.40 214720.50 3211255.60 3208044.40 194337.50 171925.50 3182380.10 1 3134644.40 165068.90 124806.90 3065682.20 1 101901.00 2976777.40 4.80 4.10 3.40 2.70 2.00 1.30 0.60 -0.10 -0.80 -1.50 -2.20 2.90 . 13349919.00 13309869.30 13136841.00 12755872.60 12181858.30 11438765.00 10557980.10 9576088.00 8532294.40 7465757.60 6413085.80 5406231.30 1.90 0.30 -1.30 -2.90 -4.50 -6.10 -7.70 -9.30 -10.90 -12.50 -14.10 -15.70 16.50 16.20 16.10 15.60 14.70 13.60 12.20 10.70 9.10 7.50 6.00 4.70 4.20 1.80 -0.60 -3.00 -5.40 -7.80 -10.20 -12.60 -15.00 -17.40 -19.80 -22.20

8.6.3.3 CACFT The pessimisticscenariofor CACFT assumes that all political and economiceventsat domesticand international levels affect directly and indirectly the volume of foreign trade that the presentinternal situation of the CAC countries.In the worst situation it is assumed of CIS countriesbecomes worse and GDP per capita of all the CAC countriesdeclines dramatically to -8.6% in 2005 (with a rate of reduction of -0.6%) which closely in to corresponds the situationexperienced 1993).This is as shownin table 8.15 .

275

Chapter 8

Scenario Analysis of the ISLB Demand Table 8.15 CACFT independent and dependent variables by growth factor and time series proiection r the pessimistic scenario (tonnes). GDP per capita CAC Asian foreign trade Volumes (YcAcFT) Values (X) Growth 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 3630570.60 3780642.90 3509606.90 2728208.40 1896981.90 1647787.90 1351898.30 Pessimistic scenario 1334812.60 1266966.21 1185634.38 1092143.45 988004.40 874871.51 754498.13 628690.74 499263.35 367993.52 236581.77 106615.56 32700 35400 36800 36100 33400 23100 21200 forecast values'' 20776.00 20235.82 19588.28 18843.92 18014.79 17114.05 16155.66 15154.01. 14123.54 13078.40 12032.13 10997.36 3.20 8.30 4.00 -1.90 -7.50 -30.80 -8.20 -2.00 -2.60 -3.20 -3.80 -4.40 -5.00 -5.60 -6.20 -6.80 7.40 -8.00 -8.60

8.7 Conclusions
The purposeof this chapterof the study is to developthe three scenariosfor the general cargo and oil productsforeign and domestictrade of Iran, the Central Asian and Caucasus foreign trade. Three types of scenario(most probable,optimistic, and pessimistic)with a time spanof up to the year 2005 were developedusing different trends and eventsto show were developed the future of the ISLB under three distinct future environments.Forecasts independentvariables for each of the demandmodels. These three models using selected of flows along the ISLB modes and foreign trade and provide an aggregatevolume domestic interfaceswithin the ISLB network. The framework has now been provided to the assess potential of the Iranian transportsystemby undertakingthe scenariodistribution and assignment.

276

Chapter 9

ISLB Scenarios: Modal Split and Distribution

9. ISLB Scenarios: modal split and distribution


9.1 Introduction
The objective of the modal split and the distribution modelling and analysisis, for each ISLB scenario,to assignall three types of forecastdemandonto the Iranian rail and road (ports and border crossings)in order to emphasise the modes.It also models interchanges importance of modal choice and distribution for the CAC countries. It then assesses the impactson the landbridgesupply systemof Iran with an emphasison capacities. According to Nijkamp andBlaas (1994, pp. 35-36):
"Rational and consistentpolicy analysispresupposes reliable assessment balanced a and and consequences choice possibilities in relation to policy evaluationof all foreseeable is initiatives. The aim of generatingandjudging alternativeframeworksof policy measures a far from easytask for mainly two reasons...... of generatingmeaningfulchoice options the assessment expectedimpactsof policy measures". of and

Each of thesethree types of demandhas some special features,all of which call for some kind of additional considerationof eachtype of demandin later stages.Generalcargo and oil productsforeign trade of Iran (GOPFr) usesthe capacitiesand servicesof all six major the three modes of road, rail, and pipelines. In terms of vehicle capacity only ports and certain types are used. In terms of the direction of trade, it mainly starts from or ends at
southern ports and north-west border crossings towards the central provinces of the country

foreign trade (CACFT) is similar to the Tehran. The Central Asian and Caucasus around GOPFT but is expectedto be much smaller in volume and in priority and through the two ports in the south and the ten border crossingson both sidesof Iran in the north. Domestic trade of Iran (DOMTI) constitutesthe major part of the demandfor Iranian transportsupply and employsnot only all modesbut all typesof vehicles. The objective of the optimistic scenariois to show a possiblelong-term future for Iranian transport demandand supply. It is basedon data from 1979-1993for Iran and 1987-1993 for the CAC countriesand also upon the presentsituation of thesecountriesand the world 277

Chapter 9 political

Modal Split and Distribution ISLB Scenarios: Its purpose is to examine transport supply by and economic environment.

the maximum probable annual freight movements resulting from Iran and CAC analysing trade and to assess the implications for infrastructure capacities and investment. In the short term, freight transport demand and supply are unlikely to have very significant variations. Forecasting and assessment of the maximum probable volume and output for demand and supply of a transport system is necessary, even if there is not full capacity utilisation (Matthews 1995).

9.2 ISLB demand scenario distribution and modal split


According to Quinet,Reyaud,and Marche (1983,p. 161): "The influence of distanceon the rail/road modal split certainly reflects the influence of transportcost". To assess potential of the Iranian transport systemas the supplier of future landbridge the for CAC countries,a model must be developedto distribute the forecastvolume of services the three scenarios GOPFT and CACFr to interfacesand then, togetherwith DOMTI to of them among road and rail modes.This model is composedof the sequentialsteps assign shown in Figure 9.L In chaptereight, three forecastsunder most probable,optimistic, and pessimisticscenarios were developedfor GOPFr, DOMTI, and CACFr to serve as a basis for future estimates to the ISLB transportdemand.It is now necessary assignand distribute theseforecaststo of Iranian modesand interfaces(Williams 1980).It is necessary evaluateIran's presentand to future macroeconomic policies which will affect transportsupply and demand.Thus, where possible, the impacts of the distribution arising from thesethree scenarioswere compared with the national policies of the second developmentprogrammeof Iran for 1994-1999 provided by the programmingand budgetingorganisation(PBO).

278

Chapter 9

ISLB Scenarios: Modal Split and Distribution

Figure 9.1 ISLB model of distribution


Domestictradeof Iran

and assignment.
CentralAsian and Caucasian foreign trade

Generalcargo and oil product foreign trade of Iran

Optimistic scenario

Most probable scenario

Pcssimisticsccnario

Total demandfor Iranian transport

GOPFr and CACFr demand for Iranian transport

Assignment to modes

Distribution on interfaces

Comparisons baseand horizon years of freight supply capacity factors

Roadmode II

Rail mode II

Ports

The ISLB network systemis composedof rail, road and ports. Detailed assessment the of border crossings were excluded due to the lack of information concerning their infrastructureand facilities. Ports,road and rail modeseachhavea sub-system with special featuresand involvement in the total processof the Iranian landbridgesupply, all of which refer back in one way or anotherto the total transportdemand. According to the assumptionsmade earlier in chapters two and five, six ports were considered as the major ports of Iran. Due to their infrastructure and geographical

279

Chapter 9

ISLB Scenarios: Modal Split and Distribution

proximity, the two largest ports of Abass and Imam were included in the ISLB supply for CACFr distribution. For these two ports the three modes of rail, road, and model pipeline, were consideredas the supplier of services for combined GOPFr and CACFr. Due to its nature, DOMTI only utilises three modes. For other ports, only road was considered. Port capacities,which are calculatedby extrapolation betweenbaseand horizon years,are those required to match the level of demand for each scenario for each port. They are consistentwith the level of throughputin terms of length of berth, and numberand areasof facilities. In the caseof transportmodes,those factors which primarily have an immediate impact on carrying capacity, such as the number of wagons,locomotive, train weight and trucks, were consideredto be significant in describing the physical capacity (Shneerson
1981).

9.2.1 Individual and regional shares of CAC countries in CACFT


The impact of CACFr on Iranian transportsupply under the most probablescenariois not significant. It totals about 1.8 million tonnes, equivalent to 3% of the most probable scenarioGOPFr, but in practiceit is extremelyimportant for Iran. The transit and handling of CACFr requiresother than ancillary and back up services(freight forwarding, shipping agents,finance, and insurance,etc.). It also requires berth utilisation, warehousing,truck and rail transportservices.The initial step for the analysisof the forecastresultsof CACFT is to determineregional and countrywide sharesof the CAC countries, since these have different distribution and modal split implications. The regional demandof the CAC countries is basedon the averageshare of total 19871993trade for eachcountry. Thesesharesfor the most probable,optimistic, and pessimistic scenariosare derived and shown in Table 9.1 and indicate that five Central Asian countries haveabout 87.2% of total CACFT comparedwith the Caucasus figure of 12.8%.The CAC 280

Chapter 9

ISLB Scenarios: Modal Split and Distribution

kept constantfor all three scenarioswith Azerbaijan, in the Caucasus area,and sharesare Uzbekistan in the Central Asia region having highest demand with 8.1%, and 45.2% respectively. The shares of other countries are Armenia (3.8%), Georgia (1.1%), Gyrkyzstan(3.4%), Kazakhstan(31.5%),Tadjikistan (2.5%), andTurkmenistan(4.4%). Table 9.1 Comparisons of the three scenario forecasts at regional and individual levels for eight CAC countries (000 tonnes). 1994 1995 1996 , 1997 1 '4998 1999
1334.81 173.53 1334.81 1161.29 173.53 1334.81 1161.29 173.53 108.12 108.12 108.12 50.72 50.72 50.72 14.68 14.68 14.68 45.38 45.38 45.38 420.47 420.47 420.47 33.37 33.37 33.37 58.73 58.73 58.73 603.34 603.34 1 603.34 1 1298.28 168.78 2182.36 1898.65 283.71 1266.97 1102.26 A64.71 105.16 176.77 102.62 49.33 82.93 48.15 14.28 24.01 13.94 44.14 74.20 43.08 408.96 687.44 399.09 32.46 54.56 31.67 57.12 96.02 55.75 586.82 986.43 572.67 1 1161.29 1129.50 1277.69 1111.59 166.10 3029.91 2636.02 393.89 1185.63 1031.50 154.13 103.49 245.42 96.04 48.55 115.14 45.054 14.06 33.33 13.04 43.44 103.02 40.31 402.47 954.42 373.48 31.94 75.75 29.64 56.22 133.32 52.17 577.52 1369.52 535.91 1272.59 165.44 3877.46 3373.39 1 504.70 1092.14 950.16 141.98 103.08 314.07 88.46 48.36 147.34 41.50 14.00 42.65 12.01 43.27 131.83 37.13 400.87 1 1221.40 344.03 31.82 96.94 27.30 55.99 170.61 48.05 575.21 1752.61 493.65 1107.15 1282.78 1116.02 166.76 4725.01 4110.76 614.25 988.00 859.56 128.44 103.91 382.73 80.03 48.75 179.55 37.54 14.11 51.98 10.87 43.61 160.65 33.59 404.08 1488.38 311.22 32.07 118.13 24.70 56.44 207.90 170.09 5572.56 4848.12 724.43 874.87 761.14 113.73 105.98 451.38 70.87 49.72 211.76 33.25 14.39 61.30 9.62 44.48 189.47 29.75 412-13 1755.36 275.59 32.71 139.31 21.87 57.57 245.19

most: -

CACIFT

1308.35 1138.26

probable
scenario 9ptimistic scenano Pessimistic, scenario

Central Asia Caucasus CACFT Central Asia Caucasus CACFT Central Asia , Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic 1 Pessimistic Caucasus,

Azerbaijan

Armenia

Georgia

Gyrkyzstan

Kazakhstan

Tadjikistan

Turkmenistan

Uzbekistan

43.47 1 38.49 591.37 579.82 2135.70 2518.80 446.578 395.44

281

Chapter 9 Tnhlt-, Q-I r-nntinned.

ISLB Scenarios: Modal Split and Distribution

2000 "" :'2001


Most probable scenario Optimistic scenario Pessimistic scenario CACFT' Central Asia Caucasus CACET Central Asia Caucasus CACFT Central Asia Caucasus Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic 1349.67 1174.21 175.46 6420.11 5585.49 834.61 754.50 -656.41 98.09 109.32 520.03 61.11 51.29 243.96 28.67 14.85 70.62 8.30 45.89 218.28 25.65 425.15 2022.33 237.67 33.74 160.50 18.86 59.39 282.49 33.20 610.05 2901.89 341.03 1407.41 1224.45 177.04 7267.65 6322.86 944.80 628.69 546.96 81.73 114.00 588.68 50.92 53.48 276.17 23.89 15.48 79.94 6.92 47.85 247.10 21.38 443.33 2289.31 198.04 35.19 181.69 15.72 61.93 319.78 27.66 636.15 3284.98 284.17

2002
1482.51 1289.78 192.73 8115.20

2003'C
1576.25 1371.34 204.91 8962.75

2004,

2005,,

Azerbaijan

Armenia

Georgia

Gyrkyzstan

Kazakhstan

Tadjikistan

Turkmenistan

Uzbekistan

1690.29 1826.68 1470.55 1589.21 219.74 237.47 9810.30 10657.8 5 7060.23 7797.59 8534.96 9272.33 1054.98 1165.16 1275.34 1385.52 367.99 499.26 236.58 106.62 320.15 434.36 205.83 92.76 64.90 47.84 30.76 13.86 120.08 127.68 136.91 147.96 657.33 725.98 794.63 863.29 29.81 40.44 19.16 8.64 56.34 59.90 64.23 69.41 308.38 340.59 372.79 405.00 13.98 8.99 4.05 18.97 16.31 17.34 18.59 20.09 89.27 98.59 107.91 117.24 5.49 4.05 1 2.60 1.17 53.59 57.47 50.41 62.11 304.73 333.55 362.37 275.92 16.98 12.51 8.04 3.63 466.99 496.52 532.44 575.40 2556.29 2823.27 3090.20 3357.22 115.90 74.52 33.58 157.77 37.06 39.41 42.26 45.67 224.07 245.26 266.45 202.88 5.92 9.20 2.67 12.48 80.37 69.36 74.37 65.23 394.36 431.65 468.95 357.07 4.69 10.41 21.97 16.19 712.47 764.01 825.66 670.09 J 3668.07 4051.16 4434.261 48 7.35 :8:. 9 48.19 l 166.33 106.94 225.671 .9

9.2.2 Justification of ISLB interfaces and distribution for combined GOPFT and CACFI7 Foreign trade distribution models link both GOPFF and CACFr to the origin and destination ends as international outlets of the transport systemof Iran. Both exogenous and endogenous factors affect the choice of ports. Exogenous factors at different levels are concerned with port hinterland factors and include relevant geographical factors or supply of port servicescover those variables.The endogenous macroeconomic

282

Chapter 9

ISLB Scenarios: Modal Split and Distribution

which affect directly the "ability of the ports to perfonn their basic function of variables the transferof goods.e.g. adequacy and physical condition of terminal facilities, managing terminal ratesand charges"(Sun and Bunamo 1973). Sun and Bunamo (1973) and Ffrench (1979), in their studiesof the competition between US and Canadian ports for foreign cargoes, focused on exogenousand endogenous influential factors at regional, national and international levels. The exogenousvariables factors, over which there is little or no and port management which cover macroeconomic control, are considered: * the commodity effect (typesand volumesof trade) * the trading partneror overseas tradeorientation, and * the hinterlandeconomy
The five following endogenous or supply factors were also accounted for:

* the adequacy physical plant condition of terminal facilities and * terminal ratesand charges * frequencyand geographiccoverageof steamship and freight services * the ability to ensurethat inland rail ratesare equalised,and * solicitation of tradeby port developmentofficers. McCalla (1994) in discussingCanadianport container perfonnancecountedthe following five important influencing factors: * port facilitics * inland transportconnections * shippinglines servingthe ports * demandfor containershipping, and * diversion of containersbetweenCanadaand the United States.

283

Chapter9

Modal Split and Distribution ISLB Scenarios:

On such a basis, four different ports in the south of Iran arc either fully or partially capable the CAC countries, while two other major ports in the north along the Caspian of serving Sea are technically suitable as second landbridge ports between the Persian Gulf and the Indian Ocean coast of Iran. For reasons of economy, however, it seems that CAC

would be handledby road and rail modesto/from major potential southernports. shipments This is despitethe fact that northernports havethe potential to function as landbridgeports for large volumes of shipments. Among four other southern ports, Bushehr and ChahBahar have several disadvantages compared with the larger ports of Abass and Imam, in because someessentialareasthey are completely lacking. Theseinclude the distanceto the CAC countries, the lack of rail connections,considerablelimitations in port entry by the number and types of berths, warehousing,mobile and fixed cargo handling ships, equipment,back-upports and shipping servicesand city infrastructure. According to the Departmentof Transport(1982,p. vi): feasible theuseof modelling "Allocatingtraffic to broadgeographic seems and groups furtherseriuos deserves considerable techniques of consideration this level.However, in to traffic to approach allocating an problems foreseen developing econometric are individualportsand...." Therefore, the two modes of rail and road, and the two largest ports of Abass and Imam (since they have many more facilities and infrastructure) were considered as serving the CAC countriesand Iran aslandbridgemodesand interfaces. simultaneously The method used to distribute all CACFT to Iranian ports and border crossings is to the closest ports and border crossings for the two regions of the Caucasusand allocate CentralAsia. Therefore,the port of Abassis expectedto function for the five Central Asian countries, and the port of Imam for the three Caucasusrepublics. In practice, all these countriesuseboth ports for their shipments,but sincethere is not any publisheddata about theseoperations,it will be assumed that CAC shippersconsiderthe cost of freight in terms of shorter road and rail distances.As a result of the assumptionsmade, to distribute the

284

Chapter 9

ISLB Scenarios: Modal Split and Distribution

CACFr, first it was disaggregated in terms of the CAC countries (see Table 9.1) and then

theseparts were addedto the port of Abassand Imam volumesof trade.Total ISLB traffic for the combined GOPFr and CAM,
shown later in Table 9.4.

and the landbridge ports of Abass and Imam, is

The allocation of the CAC countries trade flows to border crossings was based on the location of thesecountriesnear north-eastand north-westIran. Three of the CAC countries (Armenia, Azerbaijan and Turkmenistan)have a direct border with Iran, while Azerbaijan and Turkmenistan as well as Kazakhstan can be served through Caspian Sea ports. Turkmenistanand Azerbaijan eachhave four border crossingswith Iran. The foreign trade with Turkmenistan is assumedto be shipped only through the Bajgiran and Lotf Abad border crossingswhile the total trade of Armenia and Georgiawill be handledthrough the Noor Dooz border crossingon the Iranian-Armenianborders.The trade for the republic of Azerbaijan is divided so that all trade for the AutonomousRepublic of NakhJavan moves to/from Djulfa and only by rail, while, for the mainland of Azerbaijan, all trade is handled to/from Astara,by road. The trade of the four more distant Central Asian countries is provided by the rail-connectedborder crossing at Sarakhs(see Figure 2.1 for ports and border crossings).

9.2.3 Distribution of GOPFT and CACFT to ports and border crossings


For GOPF17 major ports and ten border crossingsand for CACFr two ports as well six as sevenborder crossing have been consideredas origin-destinationends of the transport systemof Iran. Both the war with Iraq and the collapse of the former USSR affected the operational aspectsof Iranian ports and border crossingssuch that there are large variations in the tonnage of the GOPFr. The distribution and modal split of the three forecasts bring

285

Chapter 9

ISLB Scenarios: Modal Split and Distribution

different insightsinto the capacityand servicesupply of the transportsystemof Iran and, in particular,in relation to interfaces(all ports and border crossings)and heavyroad vehicles. The distribution of the GOPFT regressionforecastsfor the time horizon of 2005 is based on the 1993 distribution and modal split for Iran. This year provides a stable situation for port operationsas it had the highestvolume of Iranian foreign trade. It was also important since it was the end of the critical first developmentprogrammeof the country since 1989. The methodologyfor the distribution of GOPFF to interfacesis basedon the 1993 foreign trade distribution, where ports accountedfor 90.2% and border crossings9.8% by tonnage (PSO 1993,SO 1994a).The sharesof the six ports and border crossingswere calculated relative to the total as shownin Table 9.2 and Figures9.2 and 9.3 for the combinedGOPFT and CACFr models,and individually for eachscenario(seeAppendixes16,17, and 18).
Table 9.2 Volume and distribution of base year and multiple regression forecast of three ISLB scenarios for GOPFT and combined GOPFT & CACFT for year 2005 at interfaces (000 tonnes). Interfaces (ports & border c ossings) Type of forecast Volume of Ports Border demand crossings Lase ear 1993 30555 27575 2980 Vic 1 100 9.80 0 of Total base year ",90.20 6085.50 62097 56011.50 Most probable GOPFT scenario 9.80 % 100 90.20 6.68 1 Most probable CACFT scenario 1826.68 182 1826.68 , Most probable GOPFT & CACFT scenario 63923.67 57838.18 7912.18 % of total ports & border crossings 12.03 87.97 operations 7669.40 Optimistic GOPFT scenario 78259.20 70589.80 Optimistic CACFT scenario 10657.85 10657.85 10657.85 2 Optimistic GOPFT & CACFT scenario 88917.05 81247.65 18327.25 % of total ports & border crossings 81.59 18.41 operations Pessimistic GOPFT scenario 22460.26 20259.16 2201.10 TO-6.62 Pessimistic CACFT scenario 106.62 106.62 3 Pessimistic GOPFT & CACFT scenario 22566.88 20365.78 2307.72 % of total ports & border crossings 89.82 L2... 18J operations

286

Chapter

ISLII

Scenarios:

NI-1,11 Split and Distribution

As can be seen from Table 9.2 the combined GOPFF and CACF-Finost probable sccnario forecast is about 39.1% lower than the optimistic scenario, and 64.7 % higher than the pessimistic scenario in 2005. The distribution for all three follows the sanie procedures.

Figure 9.2 Distribution of the threescemiriosfor conbined GOPFr and CACFr forecastsover ports.

100000
80000
9)
4)

1n Nlasl nmhnhit- 3 Onfini. eie n1

60000

(D CD

quuuu
20000 0

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

Years

Figure 9.3 Distribution of the three scenarios for combined C0111,71' and CACFT forecasts over border crossings.

20000 15000 .2 10000 5000

10 Most

I probable E3Optimistic 0 Pessimistic

0
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Years

9.2.3.1 Distribution

of the combined GOPFTand

CACI; Tto ports

Shneerson (1983) has suggested three phases in the port planning process: * Forecasting the demand for port services * Allocating demand to individual ports 287

Chapter

ISLB

Scenarios:

Modal

Split

and

Dis(ribution

* Determining resultant cost performance characteristics. By extending these stages into the ISLB process, the GOPIq and combined GOPII' and CACFIF forecasts can now be allocated to regional and individual ports. Therefore, the total GOPFIF forecast, as the sum of both imports and exports, was allocatcd to the six niajor the base year 1993 for port shares. Only the two landbridgc ports of'Abass and ports using Imam receive CACFF. These are shown for the three scenarios in Table 9.3. The future forecasts of total GOPFF is mainly composed of that part which runs through ports, and constitutes the main basis for capacity and services required along ISLB corridors as shown for the three scenarios in Figure 9.4.
Figure 9.4 Comparison of three combined GOPFT and CACFT scenario forecasts for the southern ports of Iran
80000 70000 60000 - 50000 - r 040000 - 30000 - 20000 - 10000 - 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 M Pessimistic 0 Most probable 0 Optimistic

Years

The volume of trade forecast under the most probable and optimistic scenarios for 2005 for both southern and northern ports is respectively about double and triple that of 1993. The combined GOPFIF and CACFIF optimistic scenario forecast share for all ports is about 40.5% higher than the most probable scenario, while the most probable scenario is 1840/0 higher than the pessimistic scenario in 2005. The regional distribution for 1994-2005 GOPFIF forecasts is based on 85.4 % for the southern and 4.8 % fOr the northcrn ports of'
the total GOPFF in 1993. These are shown in Tablc 9.3 and Figures 9.4,9.5,9.6 regional levels and for southern and northern ports (see also Appendix and 9.7 at

16 for 1994-2005).

288

Chapter

ISIAI

Scenarios:

Modal

Split and Distribution

Figure 9.5 Comparisons of three G01IFT scenario forecasts for two northern ports of Iran.
4000 3500-3000 C 2000 .2 1500--1000-500-C 2500--

FCI Optimistic

0 Most probable 0 Pessim=istic

0--1994 1995

-III
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Years

Figure 9.6 Comparisons of four southern ports of Iran for the three combined GOPFT and CACFT scenario forecasts in 2005.

130ptimistic

OMostprobable

Ellessil

25000 20000

Imam

Abass

Bushehr Southern ports

Chah Bahar

Figure 9.7 Comparisons of the three GOPFI'scenarios the two northern ports of Iran in 2005.
I 00ptimistic OMost probable

for

Ellessimistic

2000 1500

1000 500
Anzali

--I

FNorthern

qIr
Nooshatit ports

289

Chapter 9

ISLB Scenarios: Modal Split and Distribution

Type of forecast 1 2 Baseyear % of total GOPFT Most probable GOPFT scenario Most probable GOPFr & CACFT scenario Optimistic GOPFT scenario Optimistic GOPFT & CACFT scenario Pessimistic GOPFT I scenario Pessimistic GOPFT & CACFT scenario

Table 9.3 Volume distribution of base year and three ISLB scenario forecasts for GOPFT and combined GOPFT & CACIFT for year 2005 (000 tonnes). M ! or ports Anzali Abass Bushehr Chah NooImam Total Bahar shahr ports 1783.00 818.00 1078.00 396.00 27575.00 10047.00 13453.00 5.80 2.70 1.30 ;,, 95'20 ' 44.00 1,32.90 .13.50 20429.91 27322.68 3601.63 1676.62 2173.40 807.26 56011.50 20667.38 28911.89 3601.63 1676.62 2173.40 807.26 57838.18

25747.28 34434.05 27132.80 43706.38 7389.43 I 7403.29 9882.52 9975.27

4539.03 4539.03 1302.70 1302.70

2113 2113 606.43 606.43 I

2739.07 1017.37 70589.81 2739.07 1017.37 81247.66 786.11 786.11 291.98 291.98 20259.16 I 20365.77

According to Allera et al. (1981, p. 33): it "In order to provide a link betweenthe trade and modeof forecasts, is helpful to tradein termsof the berth modecharacteristics associated with particular categorise commodities". The foreign trade of Iran has been reviewed in terms of detailed commodity trades in chaptertwo (Table 9.4). Although data about the composition of CACFT through Iran is for detailed distribution and assignmentpurposes,it is assumedthat all not available CACFr scenariosconsist of general cargo and may therefore, be added to the same categoryof Iranianport cargo.
Table 9.4 Distribution of GOPFT & CACF1Pthree scenario forecasts for 2005 at ports and national levels for eight types of trade (000 tonnes). 4 3 1; 1., 2 Dry bulk Reefer Bagged Scenarios General Mostprobable 6652.46 4106.41 245.16 1940.84 Optimistic 9470.17 5175.20 308.97 2445.99 Pessimistic 2334.14 1485.28 88.67 702.00 Mostprobable 9266.88 2978.17 437.16 1967.23 Optimisc 18948.30 3753.31 550.95 2479.25 Pessimistic 1077.19 158.12 2869.74 711.54 Mostprobable 605.07 108.05 104.44 302.53 optimistic 762.56 136.17 131.63 381.28 Pessimistic 39.08 218.85 37.78 109.43

Imam with Caucasus Abass with Central Asia Bushehr'

290

Chapter 9

ISLB Scenarios: Modal Split and Distribution

Table 9.4 continued. Scenarios Chah Bahar


Most probable

Optimistic
Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic Most probable Optimistic Pessimistic

2 General Dry bulk 184.43 519.75 655.03 232.43


66.71 456.41 575.21 165.08 169.53 213.65 61.33 17334.78 30202.30. 5715.84 187.99 000 000 000 000 000 000 7712.39 9719.721 2789.541

3 Reefer 000 000


000 000 000 000 000 000 000 786.76 991.541 284.57 1

4 Bagged 318.56 401.47


115.22 21.73 27.39 7.86 40.36 50.87 14.60 4591.26 5786.25 1660.651

Anzali

Nooshahr Total Ports (GOPFT & CACFT)

Table 9.4 continued.


1
Scenarios Most probable

5 Oil products
000

Imam with Caucasus Abass with Central


Asia

Optimistic
Pessin-dstic Most probable Optimistic Pessimistic Most probable Optin-tistic Pessimistic Most probable

000
000 9918.13 12499.56 3587.35 2416.70 3045.69 874.11 653.88 824.07 236.51 1151.90 1451.71 416.64 468.21 590.07 169.35 14608.82 18411.10 5283.96 .

'6 Metallic 7477.35 9423.50


2704.53 1202.20 1515.10 434.83 000 000 000 000 000 000 543.35 684.77 196.53 129.16 162.78 46.72 9352.06 11786.15 3382.61 ,

7
Mineral

8
Vegetable oil

9
Total

245.16 308.97
88.67 1885.27 2375.95 681.89 58.71 73.99 21.23 000 000 000 000 000 000 000 000 000 2189.13 2758.90 791.80,

000 000
000 1256.84 1583.97 454.60 6.12 7.72 2.22 000 000 000 000 000 000 000 000 000 1262.97 1591.68 456.81 1

20667.38 27132.80
7403.29 28911.89 43706.38 9975.27 3601.63 4539.03 1302.70 1676.62 2113.00 606.43 2173.40 2739.07 786.11 807.26 1017.37 291.9 57838.17 81247.65 20365.77

Bushehr

Chah

Bahar

Optimistic Pessin-dstic Most probable

Anzali

Optimistic Pessimistic Most probable

Nooshahr

Optimistic Pessimistic

Total

Ports I

Most probable Optimistic Pessimistic

(GOPFT & CACFr)

The eight commodities which comprise GOPFT and CACF17 will further help the allocation of berths and storage facilities and highlight the new requirements under different scenarios. the six major ports of Iran the distribution of the total ISLB foreign For trade for horizon year 2005, in terms of theseeight main categories trade,is basedon the of 291

Chapter 9 shares of the total composition

ISLB Scenarios: Modal Split and Distribution of Iranian imports and exports in the base year of 1993 for trade (Central Asia for the port of Abass and the

each port, and for all CAC countries'

Caucasus for the port of Imam). The results of this process are shown in Table 9.4 for the three scenarios.

9.2.3.2Distribution of the GOPFT and CACFT to border crossings The methodologyused for the distribution of the border crossingshareof the GOPFT (9.8% of total GOPFI) is to use the baseand closestyear data where available for Iranian foreign trade through these crossingsfor all three scenarios.For the combined CACFT distribution, rail and road access thesecrossings,and also their proximity to other CAC to formed the basis for all scenarios.All trade therefore,for four countriesand capital cities, Central Asian countries (except Turkmenistan, see below) was allocated to the border crossing at Sarakhs as it is rail connectedand is the border crossing to the other four CentralAsian Republics. Bazarganis the busiestborder crossingof Iran since it bordersTurkey, one of Iran's major trade partners, has major trade links, and is also Iran's closest gateway to European markets.To determine the share(%) of all border crossingsin the base year, Bazargan's deducted from the total volume of GOPFr moving through the ten border share was in Table 9.5. Some of the border crossingsof Iran such as Taybad and crossingsshown MirJaveh mainly serve only one neighbouring country. Therefore, the total trade of Iran with Afghanistan in 1993 was assumedto go through Taybad, and the same assumption was made for Pakistanthrough MirJaveh.The distribution of the 1993 baseyear trade of Iran with Turkmenistanwas basedon the proximity of the BaJgiranborder crossingto the that two-thirds of all the foreign trade capital city of Turkmenistan(40 km). It was assumed of Iran with Turkmenistan moves through BaJgiranand one-third through Lotfabad. The Caspian sea trade with Turkmenistan, Azerbaijan and Kazakhstanwas excluded as this
292

Chapter 9

ISLB Scenarios: Modal Split and Distribution

trade is mainly carried out with Russia and other CIS countries. Since Iran does not have a

direct border with Georgia it was assumedthat all Georgia's trade is handled through
Armenia at the Noor Dooz border crossing. Table 9.5 Volume, share and distribution of base year and forecast of three ISLB scenarios for GOPFT and combined GOPFT & CACFT for the year 2005 at ten border crossings (000 tonnes).
1 Astara Type of trade and forecast Base year --,''-, " - , , :, 1257.16 -, - % of Total GOPFT 4.11 2552.19 Most probable GOPFT scenario M ost pr( stprobable GOPFT & CA.CFr scenarios Optimistic GOPFT scenario Optimistic GOPFT & CACFT scenarios 4 Pessimistic GOPFT scenario Pessimistic GOPFT & CACFT scenarios 2647.57 3216.45 3993.41 923.12 930.89 Djulfa Bazargan Mirjaveh 172.04' 0.56 349.61 349.61 440.60 440.60 126.45 126.45 Razi 1 5.78 0.02 12.42 12.42 15.65 15.65 4.49 4.49 139.65 '', ' 1146.77 , 3.75 0.46 285.65 2330.50 300.44 359.99 446.32 103.32 104.18 2330.50 2937.07 2937.07 842.93 842.93 SaraEW 67.87 0.22 137.86 1646.69 173.74 8977.12 49.86 137.93

1 2 3 -4
,

Type of trade and forecast Base year, "%of Total GOPFT', Most probable GOPFT scenario Most probable GOPFT & CACFT scenarios Optimistic GOPFT scenarios Optimistic GOPFT & CACIFT scenarios Pessimistic GOPFT scenario Pessimistic GOPFT & CACFT scenarios

Bajgiran 67.62 0.22 137.23 190.82 172.95 485.58 49.64 52.76

Taybad 30.35 0.10 62.10 62.10 78.26 78.26 22.46 22.46

Lotfabad . 33.81 0.11 68.31 95.10 86.09 242.40 24.71 26.27

Noor Dooz '58.95 0.20 124.19 213.70 156.52 678.74 1 44.92 50.15

Source:Data for CAC countries and Afghanistan foreign trade with Iran is basedon Sanate-Hamlo-Naghl, 1994b,pp. 15-18

Even undernormal conditions before the collapseof the USSR the traffic flow through the major Iranian border crossings of Bazargan and Djulfa was usually congested,due to Customs formalities and procedures.After 1992, with the emergenceof the new CAC countriesand the suddenexpansionof trade through thesecrossings,the system,including infrastructure,was not able to respondefficiently. The Astara border crossingbecamevery active after the independence the Republic of of Azerbaijan in 1991, and especially as a result of the decline of the Djulfa rail terminal because conflict betweenArmenia and Azerbaijan.It now only servesthe Autonomous of

293

Chapter 9

ISLB Scenarios: Modal Split and Distribution

Republic of Nakhjavan. There is not yet any published data about the distribution of CAC

countries' foreign trade with other countries through Iranian border crossings.Therefore, since statisticsabout the trade of all CAC countries (including autonomousrepublics) with Iran was available for 1992 and 1993 for all border crossings, it was assumedthat all foreign trade of thesecountries also follows the samepattern as with Iran. Accordingly, the distribution of Azerbaijan trade is basedon 90% for the mainland of Azerbaijan through Astara and 10% for the Autonomous Republic of Nakhjavan through Djulfa. All these calculations are shown in Appendix 19 for 1994-2005for all three scenariosand all ten border crossings.

9.2.4 Total ISLB demand modal split


Three types of modal split were undertaken for the three ISLB scenarios.These
reflect the different natures of the GOPFr, DOMTI, and CACFr demand. GOPFr uses

primarily heavygoodsvehicles (HGVs) 13.5 - 22.5 tonnescarrying capacity.DONM uses all types of vehicles. The CACFr uses a new range of modal services due to the availability of road and rail modes. Modal sharedata were not directly available for all three modes.Therefore, they must be derived by calculation. The methodology used to obtain the modal sharesfor the three scenarioforecastsduring the period 1994-2005(Appendices20,21, and 22) was basedon the fact that the oil products trade is only handled through ports and that imports have decreaseddue to the reconstruction programme of Iran's oil industries after 1988. Therefore, the shareof GOPFr accountedfor by oil productswas assumedto drop from a % maximum value of 24.1% in 1993to a minimum value of 13.3 (the averagegrowth of oil productsthrough ports during 1979-1993)in 2005 for all three scenarios. Oil products are not carried by rail even from rail connectedports. The two modesof road and pipeline transportoil products from ports (by road from special terminals belonging to 294

Chapter 9

ISLB Scenarios: Modal Split and Distribution

the Ministry of Oil and outside port limits). The distribution of oil products by pipeline mode is based on the policies of the second development programme (1994-1998) intended to increase the pipeline share from 59.8% to 61.6% by 1998. Therefore, the original share for 1994 (59.8%) was assumed for the base year of 1993 and then increased at a rate of 0.36% to 1998 (PBO 1993) and at the same rate to 2005 for all scenarios. The GOPFT volume through non-rail connected ports was based on the available annual freight journeys and the total trade of these ports issued by the PSO. Port number of road Imam and two border crossings were the only interfaces involved. The next step modelled the average volume of tonnage carried per journey by a HGV for the port of Abass. This be used as the basis for the HGV calculation for all ports and border crossings. would Using this average tonnage per j ourney, the volume handled by road was computed for the Imam and for the two border crossings of Razi and Djulfa. This was then deducted port of from the total volume of the port of Imam's trade in 1993 to find the rail share, as it is the

connectedport (up to 1994). only rail and road After the war, oil productswere imported heavily and carried by pipeline and road. Due to improvementsin the Abadan refinery production and the operation of other new refineries in the port of Abass and Arak, it is assumed that the shareof oil productsin the GOPFT up to 2005 falls from 24.1% in the baseyear to 13.3%, as the averagegrowth over 15 yearsis only 0.9% eachyear (10.8/12= 0.9). A comparisonof the pipeline sharebetween 1994 and 2005 suggeststhere will be about a 3.96% increase, which will cause a decline in the contribution of the road mode in transportingthe total GOPFT. This conforms with the policy of the governmentto lighten road traffic and reducethe costsof imported oil products. An important issuein the future modal split for the ISLB study is the connectionof the port of Abass to the Trans-Iranian Railways at the end of 1994. This can significantly affect

295

Chapter

ISLB

Scenarios:

Modal

Split and Distribution

modal shares and increase the rail contribution

to foreign trade, provided that additional

rolling stock is made available. On the other hand, any increase in rail cargo from the port of Abass may cause a similar decreasein rail operations at the port of Imam or other border crossings. This latter alternative seemsmore probable. The share of rail freight in GOPFIF was based on a 10.9% annual increase from 1994 (PBO 1993) for all scenarios. The rail operations of the port of Abass at the end of 1994 (with an output of about 1.2 million tonnes in the first year of operation) is deducted from the rail share in 1994 (Sanate-Hamlo-Naghl 1996b). The modal split of the GOPFT forecast for three modes (8.5%, 10.6%, and 80.9% for pipeline, rail, and road modes respectively) has resulted in the situation shown in Figure 9.8 and later in Table 9.6 for the three scenarios in 2005. Figure 9.8 Modal split of three scenarios for GOPFT in 2005.

9)

Pesminktic

Most probable

optinlLsfic

Scenarios

Oil products in the total pessimistic GOPFr follow the same trend as the most probable and optimistic scenarios and drops from 24.1% in 1993 to 13.3% in 2005. The pipeline share from ports drops from 59.8% in 1993 to 50.2% in 2005 (by an annual rate of 0.8%) as shown in column 7 of Appendix 22.

296

Chapter 9

ISLB Scenarios: Modal Split and Distribution

9.2.4.1 Modal split of the combined CACFT & GOPFT demand The objective of the study is as an aggregate planning project at national level, and the

model aims to forecast the likely traffic of each basic mode. Therefore, the CACFr was
split only between the two modes of rail and road. It takes into account the long distances

betweenthe CAC capital cities and the two southernports of Iran and also the observation
of the past few years behaviour of the CAC shippers in preferring rail to road. The rail trade of the Caucasus region is disrupted by Armenia and the Autonomous

Republic of Nakhjavanfrom both the eastand west routes to Djulfa in Iran. Therefore,the CAC trade through each border crossing connectedwith both modes (rail and road) is to assumed be handledonly by rail. Consequently, shipmentsto the four distant Central all Asian countries which are assumedto end or originate from the port of Abass will be
transported through the Iranian rail network through the Sarakhs border crossing; while Djulfa handles only the trade of the Autonomous Republic of Nakhjavan. The share of the CAC countries in the forecast trade is based on the average share of each country's trade

during 1987-1993. The split of the foreign trade of Turkmenistanfor the threescenarios was basedon the ratio of the bordercrossingtrade of Lotfabadand Bajgiran in the bilateral tradewith Iran. Due to its proximity to the ISLB ports, it is assumedto be transportedby road only. Similarly, sharesof the Azerbaijan Republic and its AutonomousRepublic of Nakhjavanwere based
respectively on the proportion of 10% (through Djulfa) and 90% (through Astara) of trade wit Iran. All combined GOPFr and CACFr modal splits at individual and combined

levels are shown in Table 9.6 for 2005, and for all the three scenarios and three modes for the period 1994-2005 in Appendix 23.

297

Chapter 9

ISLB Scenarios: Modal Split and Distribution

The ratio of combined GOPFT and CACFT assumed road traffic for 2005 compared with the base year contributions is 290% for the optimistic scenario, 2 10% for the most probable

78% for the pessimisticscenario(seeTable 9.6). scenario,and


Table 9.6 Volume of modal split for the GOPFT and combined GOPFT & CACFT base y ar and regression forecasts of three ISLB scenarios for he year 2005 (000 tonnes). Modes Modal split GOPFT& Road Rail Pipelines Type of forecast CACFT 30555 23579.3 2581 4394.7 I Baseyear 993 77.1 8.5 100 14.4 % of total 62097 50250.8 6550.6 5295.6 2 Most probableGOPFr scenario 100 80.9 10.6 8.5 % of total 63923.67 50550.56 8073.87 5295.6 3 Most prob ble GOPFr & CACIFT scenarios 79.1 100 12.6 8.5 % of total 63055 8530.25 6673.9 78259.2 OptimisticGOPFTscenario 80.57 10.9 8.53 100 4 % of Total 6673.9 88917.49 68778.22 17417.83 OptimisticGOPFT& CACFT scenarios 3.05 100 77.35 19.6 % of Total ] 1499.58 22460.27 18512.51 2448.17 5 Pessimistic GOPFTscenario 1499.58 22566.88 18530.01 2537.08 Pessimistic GOPFr & CACFr scenario; L . 9.2.4.2 Real DOMTI modal split DOMTI is a gross forecast and includes the volume of exports of Iran during 1994-2005 which must be excluded when estimating domestic modal split. Therefore, it was necessary to deduct exports from all three scenario forecasts of DOMTI. This was based on the

stableyearsof the as total national export volume during 1988-1993 reasonably averageof
the country. The average of 11.35% was deducted from DOMTL Real DOMTI economy of

is the main body of demandfor the transportsupply of Iran. For example,the threeGOPFr forecastsare about 9.1%, 7.7%, and 25% of the three real DOMTI forecastsin 2005 for
most probable, optimistic and pessimistic scenarios respectively. Domestic transport at the national level in any country is a complex issue. On one hand, it is an essential part of production and on the other, it provides essential and other services for individuals. It covers many production centres and uses extensive routes both for receiving materials and the distribution of products. Significant factors, include the number and composition of

298

Chapter 9

ISLB Scenarios: Modal Split and Distribution

vehicles, traffic control, prices, provision and range of services required, and fuel

consumption(Sampson and Farris, 1979). It is assumedthat, for 1994-2005,the impact of different transport supply investment For creates policies under the three scenarios such modal capabilities similar to 1979-1993.
example, there had been a growth of about 315% in 1993 for real DOMTI rail mode as shown in column 9 of Table 2.55, which means a rate of growth of about 23% or 819,000

tonnes annually during 1979-1993.The sharesof the three modes for actual DOMTI in 1994 was based on the average share of these modes during 1979-1993.This was accountedfor by rail (4.6%) and by pipeline (10.3%) and the rest for road (84.74%) with different annual growth rates from 1994 for all three scenariosas shown in appendix24. The assumptionfor the three scenarioswill be changedwith an increaseof 0.3% as the
annual increase for rail (overall 4.6% from 1995) due to the new rail line to the port of Abass and an increase of 0.36% in pipeline share from 1994 to 10.3%. The share of the three modes in transporting real DOMTI for the three scenarios is shown in Table 9.7 for 2005 and Appendix 24 for the period 1994-2005.
Table 9.7 Mod I split of three real DOMTI in 2005 (000 tonnes) Modes 'Scenarios 2005 -Base Year : 509,222.97 7""', Mostprobable . 773,350.75 Road 220,566.6 optimistic Pessimistic 77,176.45 51,921.29 Most probable 78,852.23 16,802.0 optimistic Rail 7,650.09 Pessimistic 96,087* Most probable 145,926.54 Pipelines 2 117' optimistic 1 14,525 pessimistic -:

9.2.4.3 Total demand and modal split for the Iranian transport system The combined forecasts of the GOPFr, CACFr and DOMTI in all three scenarios represent the demand on the Iranian transport system. This is composed of the sum of the

299

Chapter 9

ISLB Scenarios: Modal Split and Distribution

three demands for each of the road, rail and pipelines modes. The general modal trends of the three scenarios are shown in Figure 9.9 and Appendices 25 and 26.

Figure 9.9 Total future forecast volume (GOPFT, DOMTI, and CACFT) for Iranian transport networks and interfaces. Pessimistic 0 Optimistic --*-Most probable
1200000. 11000000-800000..

600000-400000
200000

0ii 1994

1995

1996

1997

1998

1999 2000 Years

2001

2002

2003

2004

2005

The differencebetweenthe total demandon the transportsystemof Iran as shownin Table 9.8 is relatively high. The baseyear demandcomparedwith the optimistic, most probable
and pessimistic scenarios accounts for 27%, 42% and 248% respectively. Table 9.8 Total three demand of scenarios for Iranian transport system in 2005 (000 tonnes). Base -, year: Volume Demands
GOPF17
CACFr

78259.21 N-3-."'30555.0'
10657.85 --1"' ', MO. O* -,

Real DOMTI Total optimistic GOPFr CACF17 Real DOMTl Total most probable GOPFr CACF17 Real DOMT7 Total pessimistic

271660.3' 998129.52 '-,, 302,215.3: 1,087,046.58 ' 62097 -'30555.0, 1826.68 - 000.0 -271660.31 657231.50 721,155.19 30555.0 22460. j 000.0 106.62 271660.31 99351.76 -, ", 121,918.64 ,,, '"302,215.3, -',,

In order to determinethe future activities for the Iranian network, modal split models for
total demand must be designed to see the adequacy and shortcomings of the ISLB supply

systemin the handling of the total forecastswhich are shown in Appendix 27 and Table
9.9.

300

Chapter 9

ISLB Scenarios: Modal Split and Distribution

Table 9.9 Total forecast demand for Iranian road and rail transDort in 2005 (000 tonnes). Road Rail. Scenarios Pipelints 559777 59996 Most probable 101382 Optimistic 842173 96272 152600 95707 Pessimistic 10187 16024 As the ISLB study is at the national level, total demand was limited only to the

"assignment" of the modes of road, rail and pipelines and not to every individual link (Werner 1985). The composition of the total demand for the three modes of the Iranian is transportsystemundereachof the three scenarios shown in Table 9.8 whereDOMTI has The future demandfor the Iranian interfaceshas the highestand CACFr the lowest shares. beenassessed GOPFr covering all six major ports and ten border crossings.It has also for beenassessed the combinedGOPFF and CACFr which includes two ports in the south for
and six border crossings. The GOPFr and DOMTI forecasts are related to Iran and were treated as affecting internal

decision making, mainly following the samemodal and interface sharesas they had during
1979-1993 (and in particular 1988-1993). The choice of modes for freight flows for the CAC countries is under the controI of foreign decision makers and shippers and, therefore,

shouldbe treatedseparately.

9.3 Future Iranian transport supply requirements


Accroding to Baumgartner(1989, p. 35):
The capacityof a transportinfrastructuremay be defined as the maximum numberof vehiclesthat can usethe infrastructureover a given unit of time under specifiedconditions. In principle, infrastructurecapacity is established plannedon the basisof projectedfuture or demandwith a pre-determined " shortfall in performance.

The presentcapacity of Iran's transport supply in terms of both modes and interfacesis limited mainly by fleet size of heavy road vehicles,rolling stock and single rail networks.
Labour productivity in general is poor throughout the transport sector.

301

Chapter 9

ISLB Scenarios: Modal Split and Distribution

At some point in the ISLB scenario study time period, the level of demand for transport is

likely to exceed the maximum available existing capacity. Therefore, the need for new by capacity must be detern-iined analysingthe different types of future demandand base year supply figures.

9.3.1 Port traffic and capacity


9.3.1.1 Maritime traffic of GOPFr and CACFr

According to Bruun (1990, p. 49):


'Ile data neededfor traffic analysisare information on the distribution of the vesselsto be " expectedby size and numberand their cargoconsideringthe future development. In terms of maritime traffic (number of visiting ships) the combined GOPFr and CACFr

through Iranian ports under the three scenariosin 2005 were basedon the 1993 average in trade of eachship (seeAppendix 28 and Table 9.10). Increases maritime traffic directly
affects the berths and cargo facilities, equipment utilisation and the deployment of marine vessels (e.g. tugs) and personnel. The daily rate of maritime traffic into the six major

Iranian ports changesfrom 5.4 in the baseyear to 12.9 (most probablescenario)and 14.9
(optimistic scenario). This would require a substantial increase in pilotage and tug services

which will needlong-term planning as they require a skilled labour force. Also new marine vessels will be needed to support the expected extra traffic. This will require foreign currency,legal formalities and time. To someextent, this amountof maritime traffic can be reducedby the employmentof larger ships and better utilisation of ship spacefor vessels
calling at Iran's major ports. Table 9.10 Comparisons of three ISLB scenarios with baseyear situation and their maritime effects on norts (unit.- shin).
Maritime Maritime I Daily traffic traffic rate kl. ABaseyear Most pro V 1201 A. -584 -;, '60' 639, ". 2149 "180 -12 ajO.441 323 V*; -, A3- 'R)AV; 88 " 4410) 748 16`0.84 24'! 253 40 4762 ),j 02 1.,1"AS. j I Daily rate able 3.30 5.80 0.89 0.24 2.00 1 0.69 1 12.9 Maritime I Daily traffic rate Optimistic 1577 4.30 2076 5.70 407 1.10 111 0.30 943 2.60 319 0.87 5433 14.90 Maritime Daily traffic rate Pessimistic 430 1.26" 474 1.30 117 0.32 32 0.09 271 0.74 91 0.25 1415 3.9

Imam Abass _Bushehr Chah Bahar _Anzali Nooshahr 1T tal

302

Chapter 9 9.3.1.2 Future port capacity of Iran

ISLB Scenarios: Modal Split and Distribution

Bruun (1990, p. 49) considers that port capacity: "Dependupon the exterior physical factors including depth, channeland basin " characteristics. Bruun then lists four following factors determining the port capacity on the land side: * Berth capacity (length of berth in relation to depth) * Area available for loading/uloading * Storage area, and * Traffic area.

Shneerson (1983, p. 226) definesport capacity"by the number of tonnesthat passthrough the port in a given time" and argues that port capacity is a function of the direct (loading/unloadingto rail and truck) and indirect (into transit sheds)delivery performance. This determinesthe level of traffic and servicesand also the depth of berths and entrance channels,length of berths, warehousingareasand numbersand types of cranagesystems, etc. To identify the required port capacityand bottlenecksfor five different types of trade and the also for the six major ports underthe threescenarios, distributedcombinedGOPFT and CACFT trade in 2005 was comparedand assessed national and port levels. This was at by the use of combined generalcargo trade ports (excluding oil productsof 14.6 achieved million tonnes,21.8 million tonnes,5.9 million tonnes for the most probable,optimistic and pessimisticscenariosrespectively).Extrapolationswere made basedon 1993 and are shownin Tables 9.11 and 9.12. The existing port capacityof Iran in the baseyear of 1993 in total and for five types of trade (generalpurposeberths, container,iron ore, grain and liquid food oil) were investigatedbasedon PSO estimationsfor the period 1979-1993(see Figure 9.10).

303

Chapter 9

ISLB Scenarios: Modal Split and Distribution

Figure 9.10 Development

of the nominal (design) port capacity In

Iran.
3633'30'27 24 21-is12 916'3-01979

iaimAisia 1980 1981 1982 1983 1984 Years 1985 1986 1987 1988 1993

Source:Basedon variousPSOannualperformance reportsfor 1979-1993.

The total capacityof 30.15 million tonnesindicated in column 3 of Table 9.11 is derived for from the PSO and was increased this studyby including the food oil unloadingterminal the port of Abassand the port of Imarn silo, asthey play a significant role in the capacityof performance volume of both ports.
Table 9.11 Capacity and trade in 1993 and 2005 under three scenarios for all six major ports at different t, pes of berths (million tonnes). 2005 1993 4 3 2 1 '-`-, 'Base'year'c'apacities,, Forecast trade for ports and ' Type of Berth 'No'. 4 I. berths re uired capacities Optimistic Pessindstic Estimates" Estimated Most 3, by PSO capacities probable (3.1) 62.84 15.08 77 Ocean going: of 26.65 '31 9, P, 43.23 , which 11.04 48.77 5 32.07 ' 19.4 6 1, -19.45' General purpose 1.95 0.49 4 1.39 40 Containe? -4 0.79 2.76 2.19 3.20 2 Iron ore '72.74' 9.72 2.79 7.71 2. Grain 1.59 0.46 2.60 1.26 Liquid food oil 62.841 31.99 15.08 26.65 43.23 Total 0')A,, ",150 3,50 23 Barge harbours 30.1, 35A9 Total with _', _`Y, barge harbour

1 1.1 1.2 1-3 1A 1-5 2 3 4

Source:Basedon PSO 1993and 1992annualreports. 1: PSOestimates. Scenariocapacitiesfor containerterminalswere basedon the volumesof containertrade 2: of about2.4% of total tradethroughports in 1993baseyear and not includedin the total. 3. for silo and liquid food oil berths.4. excludingoil products.

304

Chapter 9

ISLB Scenarios: Modal Split and Distribution

The carrying capacity of the port Imam silo is 70,000 tonnes (PSO 1993) and for two ships, and was approved by the Transport Co-ordination Committee of Iran in 1992 to operate at a daily rate of 7500 tonnes (2.7 million tonnes per annum) (Payam Darya 1992). For the port of Abass food oil installations, 300 tonnes per hour output gives a nominal capacity of 2.6 million tonnes. On this basis the ports have been expanded and improved from 12.5 million tonnes in 1979 to 35.5 in 1993. The assessment and comparison of the three scenarios shows that, for the four types of trade, the pessimistic forecast (15.04 million tonnes) is well below the base year capacity (35.5 million tonnes) for total ocean going berths. For both the most probable and

optimistic scenarios, the port capacity issue is significant and indicates the need for an increase as shown in Table 9.11. Further expansion is required by expanding port productivity. The shortfall in performance of the ports could be filled by the increase and development of port technology and changes in daily shifts for some particular categories of trade, since the output of Iranian berths is low due to the dependenceof cargo operations on ships' gear and on the many untrained port workers, both of which are below

internationalaverages 1981). (Shneerson The assessment indicatesthat under the most probablescenario,if the ports are expanded then at the sameaverageannualincrease 1.54million tonnesper year during 1979-1993, of in 2005, the combined ports in Iran will have a 53.92 million tonnes capacity which is higher than the combined GOPFT & CACFT for the combined ports most probable scenario(43.2 million tonnes).Using 53.92 million tonnesprobabletrade "as a preferred capacityutilisation" (Chang 1978)therewill be about 10.69million tonnesexcesscapacity for the most probablescenariofor all cargoes(exceptoil products).

305

Chapter 9

ISLB Scenarios: Modal Split and Distribution

For the most probable scenario containers, iron ore, and food oil will still have surplus port and grain in deficit. Under the optimistic capacity with general cargo food oil installations will have a shortage of capacity. scenario all except

According to the PBO (1993), it is planned to increase the port utilisation coefficient from 90% in 1994 with 28 million tonnes port capacity to 95% of 34 million tonnes in 1998. This is in addition to an increase of 8.5 million tonnes in the foreign trade of minerals, 42.5 million tonnes. On this basis, roughly another 5 million tonnes capacity (with making 100% utilisation) is needed in 2003, providing ports with 47.5 million tonnes capacity, the most probable scenario growth rate, but is still less than the which should cover

It should also be bome in mind that, according to the optimistic scenariorequirements. PSO (1993), mineral installations, even under the optimistic scenario forecasts, have but cannot be used for general cargo purposes. This is a critical surplus capacity requirementfor the optimistic scenariowhere,in the caseof northern ports, even if all the 4 million tonnes of port capacitydevelopmentis implemented,in the optimistic expected scenarioGOPFr alonerequires about3.8 million tonnes(Torkan 1994). The critical shortageis in generalcargo berths for refrigerated,bagged,metal and grain iron ore and vegetableoil installations still have spare capacity. For the cargoes,while the sametrend (1.54 million tonne annual expansionfrom 1994) optimistic scenariowith demandfor therewill be about a 5.53 million tonne shortageof capacityand a consequent in the policies of port development improvement. and changes On an individual port basis, the required capacitiesfor 2005 were also comparedwith the baseyear of 1993and are shownin Table 9.12. Assessment indicatesthat almost all six ports have had a high margin of sparecapacityof about 16.17 million tonnes in 1993, but there will be shortagesunder the most probable if and optimistic scenarios port capacitiesremain frozen at the baseyear levels. Under both

306

Chapter 9

ISLB Scenarios: Modal Split and Distribution

the most probable and optimistic scenarios the two largest landbridge ports of Abass and Imam will have serious shortages of capacity, while all others are expected to be

the most probable scenario, except for the port of Anzali which reasonablysafe under
needs about 50% extra capacity. With only slight improvements, the others also can cope with the optimistic demand. TabIe 9.12 Capacity and trade in 1993 and 2005 under three ISLB scenarios for every major port of Iran (million tonnes). Actual in 1993(milli n tonnes) Trade for 2005(million tonnes) GOPFIP& CACFT forecast scenarios* Capacity Trade* Baseyear Most probable Optimistic Pessimistic 3 4 5 2 20.67 27.13 10.05 7.40 12+ (Silo 2.74) 18.99 8.43 31.21 6.39 12.9+ (Food oil 2.6) 1.19 1.49 0.59 0.43 1.35 1.02 1.29 0.37 0.50 1.05 61.12 41.87 14.59 19.56 31.60 1.02 1.29 0.37 OA6 0.80 0.34 0.43 0.12 0.16 0.45 0.62 1.36 1.72 0.49 1.25 62.84 43.23 15.08 20.18,1 30.15 (35.49) 1 1 Ports' trade without oil products

Ports

Imam _Abass Bushehr Chah Bahar Totalsouth Anzali Nooshahr Total north Total Iran

Source: Data for 1993 is based on PSO annual port performance for 1993. *: Trade excluding oil products.

9.3.13 Required length of berths for combined GOPFT and CACFT

The argumentthat an increasein total port demandand traffic will producemedium in and long term needfor increases the numberof suitableberths,restson the assumption that existing facilities are utilised appropriatelyto handlethe required demand.Therefore, changesin past and presentworking practices(e.g. the number and composition of shifts
the quantity and performance capability of the cargo handling on quays and ships) and technology (e.g. the number and types of shore cranes) can play a significant role. However, an analysis of cost and project time is required before any judgement should be passedon to the further extensions of port infrastructure.

307

Chapter 9

ISLB Scenarios: Modal Split and Distribution

According to PSO practices, each berth has about 189.5 metres length for each ship (the

berths(14,592metres)to the numberof berths(77) ratio of the total length of ocean-going in 1993.The nominal maximum numberof ships which could havebeenaccepted these in berths six major ports in 1993was found from the following equationsfor 77 ocean-going
for two situations. First with 7.10 service days per ship as derived for 1993 (see column 3 in Table 7.7) and second for best situations if ports were able to provide a one service day

per ship during the sameyear (seeTable 9.13): Y= 77 x 365 / 7.1 3958.5 (EQ. 1) = where Y is the maximum number of ships which could be acceptedin 1993 with one serviceday.
This is the ideal estimated number of ships in 1993 with 365 days a year working and 77 berths for ocean-going ships.

Y=77x365/1 (EQ.2) =28105 where Y is the estimatednumber of ships in 1993provided by 77 berths and with 7.1 servicedays.
The estimated number of ships in 2005 for all scenarios is shown and compared with the above two estimations in Table 9.13. It shows that under the same working standards of

in 1993(7.1 servicedaysper ship) therewould be a seriousshortage the numberof berths. It can be arguedthat, with the samenumberof berths,the servicedaysfor eachship must
be reduced to 5.9 and 5.2 days respectively for the most probable and the optimistic

scenarios. Similarly, on the basisof 189.5metreslength of berth and 7.1 ship servicedaysof 7.1 (as in 1993), the required number of ocean-goingberths under the most probable and the is optimistic scenarios 93 and 106respectively. This can be achievedthroughimprovements the productivity of port manpower,the use in of larger ships with higher averagecarrying capacitythan in 1993,the applicationof new in technology,or changes the patternof foreign trade,e.g. containerisation.
308

Chapter 9

ISLB Scenarios: Modal Split and Distribution

Table 9.13 Estimated length and number of berths for Iranian ports under three ISLB scenarios. N umber of possible ships 1993 2005 I service 7.1 service Most Optimistic Scenarios Pessimistic day day probable 76 3958.5 5432 Maritime traffic 28105 4762 14 No. of berths 77 93 106 lower than base year Length of 14592 I 20087 lower than base year 17624 berths/ml I I
Note: 189.5 metres per post of berth in 1993.

9.3.1.4Required storage areas for combined GOPFT and CACFT

One aspectof port operationsis the form of direct and indirect delivery. The former relatesto the road and rail transport link and vehicles calling at ports, whereasthe latter
to the availability of open and covered storage facilities within port boundaries. relates Therefore, another aspect of the GOPFT and CACFT scenarios is the impact these forecasts have on the storage facilities of ports which directly affect future port throughput. According to the PSO (1993), the volume of direct and indirect deliveries of the six major Iranian ports in 1993 is shown in Table 9.13 and indicates that direct delivery accounts for about 68.1% of total trade through ports, whereas only 3 1.9% moves to inland destinations from storage areas. Assuming that these proportions remain fixed in the future, the

delivery operationsin 2005 for thesescenarios shownin Table 9.14. are expected
Table 9.14 Direct and Indirect delivery in base year 1993 and three ISLB scenarios for combined GOPFT & CACFT (excluding oil products) In 2005 through Iranian ports (000 tonn S). Total Indirect Direct 5730.00 17951.00 Base, Yea 12221.00 , 100-00 68.10 31.90 %'of total delivery to inland z" -, 43229.36 13790.17 Most probable 29439.19 62836.55 Optimistic 20044.86 4279169 15081.82 Pessimistic 4811.10 10270.72 . The analysis indicates that the two delivery systems for the pessimistic scenario are slightly under the base year figures. The critical figures for the most probable and optimistic scenarios, requiring the road trucks and wagons to carry 29.4 and 42.8 million

309

Chapter 9

ISLB Scenarios: Modal Split and Distribution

tonnes in direct delivery from/to ships in limited port and quay areas, or storage of 13.8 million tonnes and 20.04 million tonnes in an indirect system, is highly questionable.

Among all eight broad types of trade identified through Iranian ports only generalcargo, metals,refrigerated,baggedcargoesand containerare using both typesof open and closed
storage areas. Therefore, excluding oil products, dry bulk, vegetable oil and minerals from the total forecast, the combined trade through Iranian ports in 2005, which needs storage, is derived on the basis of the base year's existing open and covered storage. It is shown in

column 4 of Table 9.15. Since data about the composition and volume and tarnsit time of is cargoesstowed in warehouses not available,it is not possibleto estimatefuture storage placesby cargo types. Therefore, the required area and the number of coveredand open
storage facilities under the most probable and the optimistic scenarios were determined by extrapolating the total indirect delivery volumes. This was based on each type of storage in the base year. Having considered 45 covered warehouses in 1993, each will have about 9786 m2 (440351/45) average area. The number of warehouses required under the three

is scenarios shownin Table 9.15.


Table 9.15 Covered and open storage areas require under three ISLB scenarios. Trade (000 tonnes) St rage (sqJm 5 4 3 2 1 Scenarios Total Open Indirect warehousing No. covered Covered delivery* `,`5730 17,95 ,, Z"F, 1' 'BaseYear,,, 45 ", '-r440,35l', , '4229775:%, ! , ,,,, ,V, of 1,31.90 80.60 "-4-1-9.40 32,064.87 Most probable 10228.69 786,076 7,550,624 80 (35 new) Optimistic 15556.43 48,766.25 122(77 new) 1,195,513 11,483,457 Pessimistic 3522.93 11,043. 28 of 45 270,737 2,600,558 1 Note: *: GOPFr& cAcFr whichrequires in vegetable grain(dry oil oil, storage portsexcluding products, bulk),andminerals.

Under the most probablescenario,a total of 80 warehouses e. 35 new warehouses) (i. are
required within the Customs limits, and 122 under the optimistic scenario. The estimated figures for warehouses under the most probable and optimistic scenarios are significant indicators in terms of numbers required (about I and 2.7 times the base year figure), 310

Chapter 9 construction

ISLB Scenarios: Modal Split and Distribution time and cost, if alternative policies, such as increasing direct delivery, cannot The development of open areas are cheaper and faster to make available, and legal conflicts.

be implemented.

but in many ports, may bring social, administrative,

9.3.1.5Required inland road and rail equipment Inland road and rail transportplays a significant role in the success performance and of overall transport supply. This is especiallythe case for heavy goods vehicles (HGVs) which are the primary mode servingports. The road and rail transportare not controlled by the ports. Therefore, their utilisation and acquisition for the transport of foreign trade is important at national policy levels. Combined GOPFr and CACFr use the biggest proportion of the common HGVs of between 13.5 and 22.5 tonnescarrying capacitywith 48% of the total numberof vehiclesemployedfor transportingforeign tradefrom ports and border crossingsin 1994(Atrchian 1995).Assuming that the same48% shareof HGVs of this size is applied to the total number of vehicles in 1993 (198,957)then, the number of HGVs for the three scenarioscan be calculated (see Table 9.16). The provision of this number of HVGs is far away from the expected figure in 2005 for both foreign and domestictrade trucks (PBO 1993). For rolling stock in 2005, on the basis of total rail trade and the number of wagonsand locomotivesin servicein 1993,there will be 44,572 wagonsand 644 locomotivesrequired for the most probablescenario.According to the secondfive year development programme of Iran during 1994-1998,(PBO 1993) there is an annual renewal programmeof 740 wagonsfor the period 1994-1998.If this is continued at the samegrowth rate until 2005 there will be a total addition of another 8,880 freight wagons.This makesa total 23,280 rail wagons,which is about52% of the requirednumberfor the most probablescenarioand 33% for the optimistic scenario.For locomotives there is a significant differencebetween the number of locomotives in 1993 accordingto IMC
311

(208) and as quotedby the PBO

Chapter 9

ISLB Scenarios: Modal Split and Distribution

(373). It is planned to increase the number by 14 per annum (based on PBO 1993) so that

therewill be another 168 new locomotivesby 2005. The rail motive power fleet for IRIRC and PBO locomotive numbers in 2005 will then be 376 and 541 respectively,but both
estimates are far lower than that estimated under the most probable and optimistic scenarios as shown in Table 9.16. Table 9.16 Number of vehicles & rolling stock required in base and horizon years under three scenarios.
1993-, ,,, Baseyear Most probable 244145.90 559773.52 2005 Optimistic 842128.97 Pessimistic 1011729.79

2 3

4 5 6 7

8 [9

Volume of total road demand (combined GOPIFT, DOXITI & CACFr byroad) Total no. of HGVs No. of total HGVs with 13.5,18, & 22.5 tonnes; carrying capacity (48% of 2 for base year) No. HGVs with 13.5 tonnes; carrying capacity (18 % of 2) No. HGVs with 18 tormes carrying apacity (7% of 2) No. trucks with 22.5 tonnes: carryin capacity (23% of 2) Volume of total rail demand (combined GOPFT, DOMTI & CACIFT by rail) Total no. of wagons used & I requir d 1 Total locomotives

198975 95508

456207 218979

686322 lessthanbaseyear 329435 lessthanbaseyear

'35816 13928 45764' 19383

82117 31935 104928 59995.15

123538 lessthan baseyear 48043 lessthan baseyear 157854 lessthan baseyear 96270.07 10187.16

14400, 208

44572 644

71521 lessthan baseyear 1033 lessthan baseyear--]

The weight of a train and the averagespeedof a freight wagon are important capacity issues,and, if Iran is to provide a landbridgesystem,theseare two sensitivecharacteristics, by which national and internationalservicesfor CAC countrieswill be assessed. According to the seconddevelopmentprogramme,Iran is trying to improve the weight of trains and
the average speed of wagons by renewing the old rail lines and other construction projects. The total number of required trains (including light, moderate and heavy weight criteria) is

basedon the shareof rail in the import/exporttrade for the four centralAsian countriesand
the Autonomous Republic of Nakhjavan (see Table 9.17). For Iran, where rail makes very little contribution to exports, it is assumedthat all rail trade under the three scenarios consists of imports. On the basis of the rail GOPFr and CACFr 312

Chapter 9

ISLB Scenarios: Modal Split and Distribution

in Table 9.17, the annual and daily number of required trains was scenarios shown allocated to the import and exports of CAC countries and Iranian imports as shown in

Tables9.18 and 9.19.


Table 9.17 Import/export rail freight of Iran and CAC countries in 2005 under three ISLB scenarios. Im port Ex ort Total Tonnes % (of 1993 Tonnes % (of 1993 Tonnesfor for (000) value value CAQ CAQ 6550.60 Most 6550.60 probable 8530.25 Optimistic 8530.25 2448.17 Pessimistic 2448.17 6066.36 59 8729.64 14796.00 41 Most probable 35394.70 41 59 50933.80 86328.5.00 Optimistic 354.08 41 59 509.50 863.58 Optimistic 22 13663.50 78 48443.50 62.11 Most probable 79720.70 78 282646.30 22 362.37 Optimistic 797.43 22 78 2827.50 3.63 Optimistic 287701.50 50 287701.50 50 575.40 Most probable F250 1678611.00 50 1678611.00 3357. Optimistic 16791.95 50 16791.95 33.58 50 Optimistic 41 18723.50 26943.50 Most 59 45.67 probable 41 109251.10 59 157214.90 Optimistic 266.47 41 1092.79 59 1572.60 Optimistic 2.67 57 470624.50 43 355032.50 825.66 Most probable 57 2745888.40 43 2071459.60 4817.35 Optimistic 43 20721.80 48.19 27468.40 57 Optimistic Import Export Most 796779.37 726850.64 probable 4648865.90 4240865.60 Optimistic 42423.35 46504.65 I Optimistic

Countries

Iran GOPFT rail for ports Azerbaijan (10% for Nakhjavan) Gyrkyzstan

Kazakhstan

Tadjikstan

Uzbekistan

Total CAC countriesby rail

Table 9.18 Annual and daily number of block train required in horizon years for imports under three demand scenarios for GOPFT and CACFT. 2005 Pessimistic Optimistic Most probable Freight train weight GOPFT CACFT GOPFT CACFT GOPFT CACFT 10 544 1456 1033 1896 Light freight train of 177 4500tonnes 1.5 0.03 0.5 2.8 5 Daily number 4 6 306 581 1066 Moderatefreight 100 819 train of 8000tonnes 2 1.6 3 0.02 0.8 Daily number 0.3 3 Heavyfreight train of 50 291 533 153 409 16000 tonnes 0.8 1 1.5 0.01 1 0.4 Daily number 0.14 1 11 1

The effectiveness of the block train can be seen in these tables. It requires sufficient

locomotives and wagonsand, most importantly, strong foundations for rail tracks to allow 313

Chapter 9

ISLB Scenarios: Modal Split and Distribut'On

the movement of moderate and heavy trains compared with the currently operating light trains.

Table 9.19 Annual and daily number of block trains required in 2005 for exports under three demand scenarios for CACFT. 2005 Most probable Optimistic Freight train weight Pessimistic CACFT CACFT CACFT 162.00 Light freight train of 4500 942.00 9.00 tonnes: 0.44 2.60 Daily number 0.03 91.00 530.00 Moderatefreight train of 5.00 8000tonnes 0.25 1.50 Daily number 0.02 45.00 265.00 Heavyfreight train of 3.00 16000 tonnes: 0.12 0.72 Daily number 0.01

9.4 Conclusions
This chapter described and distributed the forecasts for the scenarios for two foreign trade interfaces(six major ports, ten border crossings)and the two modesof road and rail. An estimation was then madeof physical capacity for transportsupply that would be required in 2005. The pessimistic forecasts gave values that were mainly less than base year figures and therefore, it was assumedthat the existing transport supply of Iran can match physical demands. Maximum estimated throughputs were seen in the optimistic forecasts. These cannot physically be handled by the Iranian transport system under present development
programmes. But there are significant changes in the present national policies under the

seconddevelopmentprogrammesof 1994-1998.Thesechangesconcernextendingexisting facilities. Most of the most probable throughputs are achievable by improvements in
working procedures and by renewing in particular the road and rail modes.

314

Chapter 10

Conclusions and Recommendations

10. Conclusions and recommendations


10.1Introduction
The conceptof a landbridgerefers to different types of integratedorigin-destination internationalmovementsof shipments(in variouscombinationsof sea,land and air) under a single waybill. There are different examplesof landbridges with different characteristics to transport supply facilities, organisationalstructure and managerialskills. Certain related lin-dtationsof existing landbridge studies were discussed.They include the failure of a to accountjointly for both demandand supply of landbridge comprehensive academicstudy Most articles on landbridgesdiscussthe Trans-SiberianRailway or east-westcoast services. landbridgesof the United Statesof America. Comprehensiveand valuable studies include Hayuth (1982) and (1987), Miller (1978), Mahony (1985), Damas(1992a)and (1992b), and Raguraman academicstudiesof landbridgesin general and Chan (1994). No comprehensive werefound. The main featuresof the research be summarised as: can literaturerelatedto landbridges *A comprehensive review of * An investigation and analysis of Iranian transport supply and demand including both domesticand foreign trade. * An investigationand analysisof the demandof the Central Asian and Caucasus countries (former USSR republics)for transport * The developmentof a demand and supply model related to an Iranian Sea-landbridge (ISLB) for eight CentralAsian andCaucasus countriesand Iran. * Evaluationof the impactsof demandon landbridgesupply. *A comprehensive review of the scenarioapproachand its application to the Iranian Sea landbridgestudy using a regression technique.

315

Chapter 10

Conclusions and Recommendations

This study seeks to be an original contribution both to the specific Iranian context and to the landbridges in general. Although data in general was difficult to obtain (particularly study of for the CAC republics), the study provides a comprehensive and systematic analysis. Further complications in the analysis relate to the uncertainty of currency exchange rates, the shipping fleet and border crossing statistics.

10.2 Changes required for the Iranian transport system


In chapterfour it was found that a landbridgecompeteswith all-water transportand can in terms of distanceand transit time and costs.A landbridgeoperationdue offer advantages depends requireshighly effective coto its multi-modal and multi-country characteristics and Therefore,its performance efficiency is dependent the level, economyand on and ordination.
reliability of the services provided.

The changes for the transportsystemof Iran to function effectively asa landbridgeare needed The following providesa summaryof the major issues. substantial.
If Iranian ports are to increase their landbridge traffic for the CAC republics there will need to be changes in port behaviour, marketing and performance. In chapter two it was seen that both the ports and the rail service will need to replace their undereducated or poorly trained labour with greater expertise. In the ports there is no effective controlling organisation

for training of dockersand alsono extensivefacilities for training of port workers responsible
in general.

In chapter two it was found that transport modes and interfacesin Iran (excluding road vehicles) are owned and operated by the government and thus highly centralised. The Ministry of Roads and Transport (MRT) functions as the highest authority for transport matters, but there are eight ministries involved directly in transport issues of freight, MRT hasports andrailwaysunderits supervision construction, and supervision. management but not the merchantfleet, border crossings,pipelines and oil tankers.The exclusion of the 316

Chapter 10

Conclusions and Recommendations

merchant fleet and border crossings from integrated control may cause practical problems in

the developmentof a single waybill landbridgeconceptas investigatedin chaptersfour and five. On the other hand,greaterintegrationmay lead to much greaterbureaucracy. One important characteristic of an effective landbridge is its close dependenceon however,was seenin chaptertwo to havea very low share Containerisation, containerisation.
of the foreign trade of Iran (2.4% in 1993 of total foreign trade of general cargo and oil

In this respect it seems that there is no effective co-ordination between the products). Programmingand Budgeting Organisation(PBO), ports, Customs,Ministry of Economics Commerce,andboth private and governmentcargo owners.Given a growing volume of and foreign trade of Iran and the CAC countries,there will be a greaterneedfor improved and for faster port services.It will becomeincreasinglynecessary the maritime fleet, ports, road
and rail modes to develop their infrastructure to handle container trade and operations.

The border crossingshave becomemuch more important in the transportchain since the collapseof the former USSR. Analysis of Iranian trade with the CAC countriesfor the two following 1991showeda significant increasein the volume passingthroughinadequate years terminals. It requires a further expansionof the infrastructure and superstructure these of crossing points to reducelong waiting times and increasethe efficiency of nodal supply services. The literature review showedthat the landbridge concept is an internationalservice where land transportplays a significant role. Therefore,joint venture transportservicecompanies from both Iran and the CAC countries could effectively implement the landbridge single waybill along appropriate corridors.

10.3 Landbridge demand and supply


The ISLB study provided an operationalmethod for the investigationof landbridge
demand and supply at the macroeconomic level. The scenario forecasts were initially based

317

Chapter 10

Conclusions and Recommendations

on the relationship to a number of explanatory predictors of tonnage resulting from three types of demand in the eight CAC countries as well as Iran. The three types of demand were

the generalcargoandoil productforeign trade(GOPFr), the domestictradeof Iran (DONM) foreign trade (CACFT). The forecasttonnagewas then and the Central Asian and Caucasus
assigned to the transport system of Iran to assessthe impact and the required strategies. A limitation of this approach was that the demand data for the CAC countries was restricted

only to 1987-1993 for Iran to 1979-1993. and Forty-six independent variablesare used in the empirical analysisfor three ISLB demands. Six of thesevariableswere shown to havesignificancefor estimatingthesevolumesof trade as they were higher than the requiredlevel of accuracy(0.95%). Two independent variables (GDP and population)were specific for the generalcargo and oil product of foreign trade of
Iran, three (GDP, value added of agriculture, and productivity of the agricultural

for sector/employee) the domestictradeof Iran, and total GDP per capita for the foreign trade of the CentralAsian andCaucasus countries. In chaptertwo it was found that demandfor foreign and domestictransportin Iran is focused on the main centre, consumptionin Tehran.In order to operatean effective landbridgeit of to will be necessary operatea north-southsystem,including the upgrading of the Astara, Bajgiran and Sarakhs bordercrossings. other words, if freight transportis to be an integral In part of foreign and domestictrade,it hasto be developedto matchthe demandand needsof a changingpatternof a landbridgedemand.

10.4 Transport structure in Iran


The single rail network is distributed unevenly throughout the country and covers only 14 of 24 provinceswhich has createdan unevenbalancebetweenroad and rail in Iran. Rail aswell asroad mainly carriesdomesticand foreign tradefrom only the two largestmajor southernports to Tehran and central provincessuch as Arak and Esfahan,while heavyroad 318

Chapter 10

Conclusions and Recommendations

the only means to carry foreign trade from all the other four major ports vehicles are

(including the two northern ports of Iran along the Caspiansea) to the above areas.This structural imbalanceof traffic flows leadsto empty running and, possibly, higher costs for freight. Two northernports aswell asthreekey bordercrossings Astara,Bajgiran andNoor of Dooz do not have rail connections despite their significant recent operation for CAC countriesandRussiaafter 1991. by the private sector,consistingof individuals, companies Roadfreight is operated mainly or co-operatives, and has the highest shareof both foreign and domestictrade. Heavy goods to vehiclesin Iran are old. Therefore,to have an effective landbridgeserviceit is necessary renewalpolicy in Iran. adoptandimplementa more productiveandcomprehensive Manufacturingof the rolling stock and heavygoodsvehiclesin Iran is low and the renewalis
below that required. Therefore, future landbridge services in terms of transit time and cost are affected.

In chaptertwo it was found that the containerfleet accountsfor only about2.8% of the total landbridgesystem,it is likely that IRISL fleet in numberand 2.2% of the dwt. In a successful fleet (nationalor otherwise)and effective containerports play a significant role in a container termsof transittime andcost. The government Iran is the main sourcefor transportfinancewhile the private sector,such of as banks, has becomemore important since 1988 with the end of the war with Iraq. Some organisationssuch as ports are mainly self-supporting,but the railways are supportedby
govemment credits.

Iranian roads are reasonablywell developedand connect all provincesand more than 470 towns but expressways with four or more lanes form a very low shareof total roads (about 2.4%). Because the internationallycompetitive natureof the landbridgeservice,there is a of needfor more direct routes.

319

Chapter 10 An increase in transport productivity

Conclusions and Recommendations in Iran probably requires more training and education in

different general transport sectors. Currently in Iran such study is given by one university and several high schools for maritime and navigation, while somp universities deal only with road construction and other engineering aspects, but not with commercial transport planning and operational areas.

10.5 Main findings of the scenario analysis


Threetypesof forecastdeveloped up underthreescenarios to the year2005 in chapter distributedand assigned Iranianinterfacesand land modesusing 1993asa base to eight were The evaluationwas carriedout and whereavailable,impactswerecompared with Iran's year. presentandfuture policies for its freight transportsystem. In chaptertwo it was found that the transportsystemof Iran has beenfaced since the early in ports which endedin 1993.To someextentthis inefficiency 1970s with seriouscongestion due to the inadequate port infrastructureand later to the war with Iraq, which resultedin was the closureof the largestcommercialport in the country (Khoramshahr) and a reduction in the operationof port Imam. However,the road fleet, in termsof both numberand frequency journeys,may be considered the most importantinfluenceon this insufficiency. as of The volume of total combined trade for Iran and the CAC countries, estimatedto pass through the six major Iranian ports under the most probable and optimistic scenarios,is 2.1 and 2.9 times that of the baseyear 1993 (about 30.6 million tonnes).The respectively in implication of suchincreases the volume of port traffic is considerable. The resultssuggestthat maritime traffic will rise from 1921ships in 1993by 148% (most 183% (optimistic scenario)but drop by 26% under the pessimistic probablescenario)and scenario.

320

Chapter 10

Conclusions and Recommendations

increasein maritime traffic will causeseriousshortfallsin infrastructuresuch The estimated (e. asberthsandmaritime supportservices g. tugs,pilots, andother key personnel) also in and facilities. land cargohandlingequipmentandtechnology, warehousing and in The impactsof the most probableand optimistic scenarios termsof port capacitiesshows in by that thereis a significantneedfor improvements performance an expansion ports and of productivity projects.Theseare requiredto achievethe estimatedcargo trade for the above (about 43.2 million tonnes and 62.8 million tonnes respectively,excluding oil scenarios to appears be in ImarnandAbass,asthe products).For ports,the critical capacityrequirement largestports in the country, and in port Anzali in the north. Other ports should be able to capacity. absorbthe increased by in Increases port capacitycan be implemented the expansionof port infrastructure,cargo
handling technology and also through an increase in human and facilities productivity. The extent of maritime traffic under the most probable and optimistic scenarios (applying ships'

servicedaysof the baseyear 1993(7.1 servicedays))revealsthe insufficiencyof the lengthof berths.This suggests requirementfor port infrastructureexpansion(17.6 km and 20.1 krn a for the most probableand optimistic scenarios respectively,comparedwith 14.6 km in the baseyear). In chaptereight it was found that about40% of the total port trade(excludingoil products)is and carried out as an indirect delivery systemusing closed warehouses open storageareas therewill within the ports' Customsareas.Under the most probableand optimistic scenarios be a considerable needfor expansionof storage to about35 and 77 new warehouses, to and 7.6 million sq. m and 11.5million sq. m of openstoragerespectivelyfor thesetwo scenarios in 2005. It was seenthat heavygoodsvehicleshave a dominantrole in foreign trade,but the average journey per vehicleis low (about9 per annum)due to the ageof the fleet andconsequent high

321

Chapter 10

Conclusions and Recommendations

cost of repairs. It was found that the road fleet needed to be expanded by 129% (most

and probablescenario) 245% (optimistic scenario)in 2005 comparedwith 1993.This could have a seriouslong term policy implication due to the lack of currencyfor imports and low domesticproduction. Under both the most probableand optimistic scenarios rail mode of Iran will be highly the
utilised and effective. In chapter two it was found that the rate of renewal of rolling stock is

low due to war effects,import restrictionsand low domestic output, while rail tracks still The present freight train weight (4500 tonnes) for the three ISLB need extensiverepair. with two other heavytrain types(8000 tonnesand 16000tonnes).It scenarios was compared was seen that it is an effective way of optimum utilisation, of the presentfacilities when the block train concept,especiallyfor the CAC countries,given that the rail combinedwith
lines in 2005 provide such infrastructure.

10.6 Recommendations
has approached landbridgeconcept and its application to Iran in This research the both demandand supplyareas,at nationalpolicy levels and for eight customercountries.But further research eachof theseareas in needsto be carriedout to identify further insights much and implications. On the basis of the findings presentedearlier, a number of steps are for recommended summarised improving the existing situation: and

10.6.1 Recommendations to the Iranian government

The ISLB like other landbridgeservicesin the world requiresthe implementationof the
single waybill. Therefore, among the eight ministries involved in Iranian transport matters, three (MRT, Ministry of Commerce, and Ministry of Finance and Economic) should be in charge of the ISLB service operated by or through the PSO, IRIRC, IRISL, and Customs.

Theseall needa nationalco-ordinationcentreto facilitate servicesinternally with eachother 322

Chapter 10 and the private and reliability ISLB sector, and also with of the landbridge for the CAC CAC shippers

Conclusions and Recommendations to cover and secure the level, economy

services provided. countries, as with international services, require joint venture flows of

operations

transport shipments economic Funding

and stevedoring along levels. transport during borders internal

companies. and external

This should ISLB

greatly

facilitate

new investment, transit times

routes

and maintain

and costs at

infrastructure the war with

and superstructure Iraq. The new CAC

in Iran is mainly countries along

a government

issue, in

particular Iranian

and close to northern

and the existence

of the ECO, has brought This makes foreign

about a need for more economic, inevitable, in particular rail in the

political, field of

and social integration. transport. Specific

investment

areas

for

foreign

investment

are prioritised if ISLB

networks,

expansion

of the rolling

stock and road freight

vehicle

manufacturing,

services are to

be economic.

The transport sector in Iran is mainly a government owned and operated sector, although there have been changes since 1988. To prevent the harmful consequences of rush into there should be a systematic approach to management and ownership in privatisation, transport, and in particular for the ISLB as an international service. Even under private there should be a degree of integration of freight services, modes and interfaces. ownership Any private entity in the transport service must be strongly established, able to operate for a long term and backed by parliament.

10.6.1.1Recommendations Programming and Budgeting Organisation to The Programn-dng Budgeting Organisation(PBO) is a national centre for policy and implementationaffiliated to the Ministry of Financeand Economics,and, also, making and The collection and preparation data hasthe StatisticalCentreof Iran underits supervision. of for the ISLB study took a considerable time. Information given in different PBO annualand
323

Chapter 10

Conclusions and Recommendations

occasional reports, although valuable, were not comprehensive and consistent in the transport

One reasonfor the poor level of academicresearchinto the operationaltransport sector. Iran might be the lack and unavailability of thesedata for the public. In Iran, as a systemof centralisedand plannedeconomy,all private and public sectorsshouldreport their statistics
annually to the PBO. This would enable the PBO to enforce a comprehensive national policy towards the collection of different levels of data in both demand for and supply of transport at

a more modal andprovincial level. The PBO and the Central Bank of Iran (Bank Markazi) should be responsiblefor the and implementationof policies to improve and increasethe present low level provision trade in Iranian and CAC foreign trade, in particular for Iran as it is mainly containerised based trade. govemment 10.6.1.2Recommendations Ministry of Road and TransPort to The ISLB road links must play an essential role in the success the proposed of services,in particular for the three CAC countries connected to Iran (Azerbaijan, Annenia, and Turkmenistan).New transit trunk routesto the eastand west bordersmust be provided and equipped by the Ministry of Road and Transport (MRT). The south-west north-west by in provincial route shouldbe given priority as a transit route and expanded express-ways the longer term, as it also can play a significant role in any future landbridgebetweenthe PersianGulf and the Black Seaport of Batumi, and as an alternativeto the Suez Canal in tradebetweenEuropeandAsia andAustralia. 10.6.1.3Recommendations Ports and Shipping Organisation to The Ports and Shipping Organisation(PSO),which functionsas the port statebody and operatorof the major ports, plays a central role in the potential ISLJ3services.The present and traditional structureand working practicesshould be changedand improved in such a

324

Chapter 10 way as to allow requirements. The training relevant programmes the PSO should include of the port functions to perform in accordance

Conclusions and Recommendations with international port marketing

new courses in landbridge

operations

to

port workers

and agents within

port areas to assure the improvement and other services. trade will improve

of the key role

of the private

sector in the ISLB

port cargo operations that containerised

The PSO should necessary clearance

take into account the present

foreign

trade. It is container

to expand and handling

container-related

infrastructure

and in particular,

in the yards at ports. can be greatly increased if the PSO uses quay cranes for berths. before any infrastructure

The present port capacities This change

in the port operational

technology

is recommended

expansion.

10.6.1.4 Recommendations to Islamic Republic of Iran Railway Company The rail network in Iran should be expanded (by passing Tehran) along south-west/ northwhich will facilitate more direct and econornic services west and south-east/north east routes for GOPFr, DOMTI, and in particular CACF17through higher speedand frequency of freight trains and railcars.

The present projectsof the two key ports (Anazali andNooshahr)andonebordercrossing rail (Astara) must be given higher priorities for trade from CAC republics and, probably trade from RussiaandotherCIS landbridgetradein the future. The renewal of the presentold rolling stock to support economic and competitive ISLB is vital. Application of the moderateand heavy train conceptshould increasethe services rail the ISLB rail servicesbut it needsappropriate track and much more heavyduty output of locomotives.

325

Chapter 10

Conclusions and Recommendations

10.6.1.5 Recommendations to Islamic Republic of the Iranian Shipping Line

IRISL is one of the major merchantfleets in the middle eastand the main carrier of the foreign trade. IRISL, in joint venture with the CAC countries and with the ECO national as shippingcompany,shouldbe considered a main carrier along the ISLB sealegs,its inland transport and in the implementationof the ISLB single waybill. The presentnumber and capacityof the containerfleet of the IRISL is very low to cover the marketsof the CAC trade in the world andto keepISLB services viable. Therefore,they needto be expanded. partners to 10.6.1.6Recommendations the Iranian Customsadministration The approachof the Iranian Customsneed,to be revised,in particular in containerised trade. The importanceof landbridge services for CAC and other countries needs to be
recognised. Border crossings in Iran need to be treated like seaports and their superstructure developed to

bottlenecks. handlecontainertradeeffectively andefficiently, andto preventoccasional The Iranian Customsadministrationwithin the ISLB systemmust be consideredas a vital transport organisationsimilar to the PSO for ports, and not simply as a revenuemaker. Border crossingsshould be upgradedand expandedfor through transportservicesand the implementation the singlewaybill conceptof the ISLB. of

10.6.2Recommendationsto Iranian transport operators


After the collapse of the fonner Soviet Union, Iranian road freight operatorshave expandedinto new areas,including the CAC countries.The importanceof the Iranian road freight private sectorasa major carryingmodeof the ISLB service(in particularfor the three land-lockedcountries)must be recognisedby the centralisedsystemof Iran which mainly Roadfreight operators the owns and operates transportinfrastructureand services. needmore government support and stability of working conditions to contribute reliably and
in the new ISLB working environment. This means that there is a need for competitively

326

Chapter 10

Conclusions and Recommendations

different Iranian ministries to facilitate road freight movements in terms of transit traffic with all the CAC countries (e.g. insurance, passport issues, travel expenditures, etc.).

10.6.3 Recommendations to Economic Co-operation Organisation


ECO as a regional organisation has been given a new international role. The significance

by of the transportrole of Iran has alreadybeenrecognised ECO. ECO should and centrality toward the developmentof common transportand trade policies for all member now move will greatlybenefit ISLB services. states which consequently

10.6.4Recommendationsfor further research


10.6.4.1Recommendationsfor researchinto the Iranian Sea-landbridge The ISLB study focuseson a landbridgestudy at an internationaland macroenvironmental level. This is appropriatefor the objectivesof the study, but may be too broad to observe level. Suchproblemsinclude extensivelandbridge at a disaggregate properly other problems by different ports, border crossings,or modeswhose functions require a servicesprovided high degree co-ordination. of In some landbridgesboth rail and road modes are effectively involved in the operations. There is a needfor further and detailedstudiesat the level of individual road, rail link, and border crossing.The treatmentof landbridgetransit time, cost, and frequencyof port and levels requires further elaboration,refinement and service at modal and origin-destination development the model for eachCAC country. of Iran asa new landbridgecountryneedsthe fonnation of a nationallandbridgepolicy basedon internationaltransportservicecharacteristics. g. a singlewaybill). (e. Due to the lack of data, forecastsof the CACFr tonnagewas basedon convertedfigures using valuesandvolumesof Iran's foreign tradewith eachof the CAC countriesin 1992and 1993 as the basis of conversion.Future researchshould considerdifferent CAC shippers'

327

Chapter 10

Conclusions and Recommendations

options with real and longer period data, and more explanatory variables. In particular, recent

through Iran may lead to more accurateforecastsof demand for ISLB trade experiences services. The Iranian foreign trade forecastsin the ISLB study include a considerable volume of oil productsthroughports,insteadof throughpipelines,mainly due to the war andreconstruction period. Future studiesmay considermuch lower movementsof oil productsthrough ports, for the volumeof demand the Iraniantransportsupply. which shouldaffect greatly for 10.6.4.2Recommendations researchinto landbridges in general Although modem landbridgeshave beenin operationsince the 1970s,they are poorly in the literature.Thereis a needfor a more systematic to approach the collection and covered data related to landbridges, and for a more comprehensive publication of operational
definition and conceptual approach to landbridges. At present, published data is generally

toward singlemodes. orientated Infonnation and educationon the landbridgeconceptand practicesshould focus on critical integrated transportsystems, modal movements, of and all physicalsupplyand serviceissues, which canbe perceivedaslandbridgepracticeandnot astraditionalmodalcarryingof goods.

328

References
for Abeelen,B. Hoekert,M. and Puylaert, H. (1984) "The Developmentof Scenarios Traffic and Transport", PTRC, 14thSummerAnnual Meeting, July 1984,pp. 221-230. Abraham,B. and Ledolter, J. (1983) "Statistical Methodsfor Forecasting", John Wiley & Sons, New York, PA. Ackoff, L. R. and Sasieni,W. M. (1968) "Fundamentalsof OperationsResearch",John Wiley & Sons,London, p. 61. Afshar, S. (1995) "IIRR Readiesto Play Key Transit Role", Rail BusinessReport,Railway Gazette Yearbook,Pp.58-9. Afshar, S. (1996) "Interview with Minister and Vice Minister for RoadsandTransportin No. 15Q p. 19. Sarakhs",Sanate-Hamlo-Naghl, Agazadeh,V. (1994) "CaspianShipping Company",An Interview with Presidentof the Republic Azerbaijan,Maritime Transport International, Walsh, C. Ward, D. (eds),Sterling Publications of Limited, London. Agazadeh,G. (1996) "Iranian Minister AssuresInvestorsof Stability", OPECBulletin, January,p. 31. Akbarzadeh,S. (1996) "Why Did Nationalism Fail in Tadjikistan", Euro-Asia Studies,Vol. 48, No. 7, Carfax, Glasgow,pp. 1105-1129. Akiner, S. (1995) "The World ofInformation " Middle EastReview,20th ed, Unwin Brothers, Woking, p. 9 J. Allera, S. V. and Parsons, E. Fenyoe,R. H (1981) "United Kingdom International Freight Forecaststo 1981: Shipping Routesand Modes", Vol. II, Publishedby: National PortsCouncil, London, p.33. Allport, R.J. (1986) "The Use of ScenarioTechniquesto FormulateTransportStrategyfor an Urban Area", PTRC, 14th SummerAnnual meeting,July, pp. 229-240. Amara, R. and Lipinski, A. (1984) "BusinessPlanningfor Uncertain Future Scenarios& Strategies", Pergamon Press,Oxford. Amir-Ahmadi, H. (1995) "Evaluation of the Performance the First DevelopmentalProgramme of (Persian Politicaland Deficienciesof the SecondProgramme",Etela Aat Siyasi-Eghtesadi economicalJournal), Farvardin & Ordibehsht,No. 7&8, pp. 83-92. (1996) "Iran's Development:Evaluation and Challenges",Third World Quarterly, Vol. 17,No. 1. pp. 123-147. in Arlt, H. W. (1987) "Information Requirements StrategicPlanning in the Ports Industry: Specificationand Managementof a Data Base",Maritime Policy and Management, Vol. 14, No. 1, pp. 51-2. Ashar, A. (1995) "Coastal Shipping for Domestic Freight", World WideShipping, December/January, 28. p. Askari, H. (1994) "It's Time to Make PeaceWith Iran", Harvard BusinessReview,JanuaryFebruary,p. 50. Atrchian, M. J. (1995)"Road TransportFleet of the Country:244 Idle Days", Sanate-Hamlo-Naghl, No. 140,p. 23

329

in (1996) "Role of Transportation Employment:RoadTransportation: Two Million , No. Employment",Sanate-Hamlo-Naghl, 148,pp. 24-25. for Baba,N. A. (1994) "Remaining Stresses Prospects Stability and Developmentin Central and Asia", Proceedings a Seminaron the investigationof resources capacitiesof Central Asia and of Studies(A PersianSeasonal in: Journal),No. 8, Year and Caucasus- Central Asia and Caucasus Three, Winter, p. 128. Babbie, E. (1995) "The Practice of Social Research",7th edition, Wadsworth,London, p. 50. Badiyee,R. (1993a) "A Detailed Geography Iran ", Vol. I. EghbalPublications,Tehran,p. 15. of (I 993b), "Detail Geography Iran", Vol. M. EghbalPublications.Tehran,230. of Bamford, C. G. (1995) "Transport Economics", Bryan, H. (ed.),University of Huddersfield, HeinemannEducationalPublishers,Oxford, p. 36. Baron, M. and Targett, D. (1985) "The Manager's Guide to BusinessForecasting-How to Results", Basil Blackwell, Oxford. Understandand UseForecastsfor Better Business Barylski, V. R. (1994) "The RussianFederationand Eurasia's Islamic Crescent",Europe-Asia Studies,Vol. 46, No. 3. Links: InfrastructureCapacity,Saturation J. Baumgartner, P. (1989)"The Main European Levelsand in: Conference Modal Complementary" Researchfor Tomorrow'sTransport,The European of Ministers of Transport(ECMT), I Ith InternationalSymposium Theory andPracticein Transport on Economics, Brussels,12th-14thSeptember, 35. p. Becker,S. H. (1983)"Scenarios: A Tool of Growing Importance Policy Analystsin Government to Forecastingand Social Change, pp. 95-120. 23, and Industry", Technological Bell, G. J. and Blackledge,D. A. and Bowen, P. J. (1983) "The Economicsand Planning of Transport", William Heineman,London, p. 25. Blauwens,G. and Van De Voorde, E. (1988) "The Impact of Port Changeon Inland Transportation",Maritime Policy and Management,Vol. 15,No. 2, p. 132. Benson,D. Pugg,R. and Whithead, G. (1994) "Transporl and Logistics", RewoodBooks, Trowbridge, Wiltshire, p. 22. Bonney,J. (1991) "Sea-LandTrans-SiberianContainerFranchise",Journal ofAmerican Shipper, Vol. 59, November,p. 59. Beresford,A. K. C. and Dobson,H. W. (1987) "Lloyd's Maritime Atlas", Lloyd's of London Press,Essex. Chatterjee,S. and Hadi, S. A. (1988) "SensitivityAnalysis in Linear Regression John Wiley & ". Sons,New York. Cahtterjee,S. and Price, B. (1991) "RegressionAnalysis by Example", John Wiley & Sons,New York. Caney,R. W. andReynolds,J. W. (1981) "ReedsMarine DistanceTables". ThomasReeds Publishers, London. CentralBank (Bank Markazi) of the Islamic Republicof Iran, (1991), "National Accountsof Iranjor 1974-1987",Department EconomicAccounts,Tehran. of (1992) "National Accountsof Iran for 1988-1990 Department ", of . EconomicAccounts,Tehran.

330

EconomicAccounts,Tehran. of EconomicAccounts,Tehran.

(1993) "National Accountsof Iranfor 1991", Department of (1994a)"National Accountsof Iranfor 1992". Department

(1994b) "National Accountsof Iranfor 1993", , 1, Department EconomicAccounts,FerdoosiAve. P.O. Box: 11365/853 Tehran. of Chang,S. (1978) "ProductionFunction, Productivity, and Capacity Utilisation of the Port of Mobile", Maritime Policy and Management,Vol. 5, p. 302. Technology in North-South", Cole, S. and Chichilnisky, G. (1978) "Modelling with Scenarios Futures, August, pp. 308-313. International Yearbook(1991) National MagazineCompany,London. Containerisation (1995)National MagazineCompany,London. International Yearbook, Containerisation Transpacific, ContainerisationInternational, (1994) "Regional Ports Hope for Mainline Success: RegionalFocus", December,pp. XI-XIL RegionalFocus,Central Containerisation International,(1995) "Central EuropeIntermodalism, Europe,AdvertisingSection",February. Studies",Applied Cox, D. R. andSnell,E. D. (1974) "Me Choiceof Variablesin Observational Statistics,Vol. 23, No. 5 1. Cunningham,J. H. W. (1982), "Freight Modal Choice and Competition in Transport:A Critique Categorisationof Analysis Techniques", TransportationJournal, Vol. 21, No. 4, p. 66. and Custom! Administrationof Iran (1990), " Annual Reportfor- Imports". (1990), "Annual Reportfor-Exports". (1992 ), " Annual Reportfor- Impotis ". (1992 ), "Annual Reportfor- Exports ". Traffic for the Port of Montreal Dagenais,M. G and Martin, F. (1987) "ForecastingContainerised (1981-1995)", TransportationResearch,A, Vol. 21A: Policy and Practice,No. 1, pp.1-16. Damas,P. (1992a)"TSR's SecondWind with Sea-Land",ContainerisationInternational, January 1992. Damas,P. (1992b) "Unlocking the TSR's Potential", Containerisation International, July. Department Transport(1982) "Port Choiceand the Routingof UK Trade:A Studyof the 1978 of Surveyof the Inland Origin and Destinationof Trade". Final Report,EconomicConsultant,London, P. vi. Darzentas, and Spyrou,T. (1996) "Ferry Traff ic in the AegeanIslands: A SimulationStudy", J. Society,47, pp. 203-216. Journal of the OperationalResearch Davies,J. (1991) "Sea-lands'sTrans-SiberianPlan Gets Boost from Soviet Decision", Journal of Commerce,18thSeptember, 389. p. Day, M. (1980) "International Trade in a Nutshell". Sweetand Maxwell, London, pp. 21-22. K. Delehanry,J. and Rasmussen, (1995) "Land Reform and Farm Restructuringin the Kyrgyz

331

Republic", Post-SovietGeography,November,Vol. XXXVI, No. 9. p. 565. Dewhurst,L. H. J. and West, R.G. (1990) "Closing InterregionalInput-Output Models with EconometricallyDeterminedRelationships",in: New Directions in RegionalAnalysis:Integrated Multi-Regional Approaches,Anselin, L. and Madden,M. (eds),BelhavenPress,London. pp. and 171-186. Doganis,R. (1985) "Flying off Course". GeorgeAllen & Unwin, London, pp. 185-194. in Ducot, C. and Lubben, G. J. (1980) "Methodology: Developments TechniquesEmployed in Forecasting-A Typology for Scenarios",Futures, February,pp. 51-54. EconomistIntelligent Unit (1992) "Country Profile for Georgia,Armenia,Azerbaijan,CentralAsian Republics,Annual Surveyof Political andEconomicBackground",World Microfilms, London. (1993a),"Country Profile for Georgia,Armenia,Azerbaijan,Central Asian Republics,World Microfilms, London. (1993b) "Country Report Iran", World Microfilms, London.. (1993c)"Country Report for Georgia,Armenia,Azerbaijan,CentralAsian Republics",4th quarter.World Microfilms, London.. (1994a)"Country Report4th quarter 1994for Georgia,Armenia,Azerbaijan, Kazakhstan, CentralAsian Republics,Uzbekistan",World Microfilms, London. (1994b) "Country Report Iran", World Microfilms, London. (1996) "Country Forecasts Iran", I st quarter,World Microfilms, London.. of Eller, D. (1987) "Marriage a la (sea/air)Mode". ContainerizationInternational, April, pp. 42-47. (1992) "Intercontainer's New-ageChallenge",ContainerizationInternational, September, 72-75. pp. ESCAP(1967) "The Administrationof TransportationModelling Projects", HarvardUniversity, 33 Lexington Books,Toronto, pages and 100. Ehteshami,A. (1995) "After Khomaini: TheIranian SecondRepublie', Routledge,London, p. 12. Esterhuyse, Willie. (1992) "Scenariosfor South Africa-Instability and Violence or Negotiated Transition ?", Long RangePlanning, Vol. 25, No. 3. pp. 21-26. Europa World YearBook, (1994a)"A-I: ', basedon IMF, EconomicReview.World Bank, Statistical Handbook:Statesof the former USSR.World Bank: The Transitionto a Market Economyand USSR.Different pagesfor Azerbajan.Armenia,Georgia, StatisticalHandbook:Statesof the Fon-ner Gyrgizstan,Kazakhstan, Tadjikstan,Turkmenistan,Uzbekistan Iran. Vol. 1,EuropaPublications, and London. (1994b)"K-Z! '. basedon IMF, Econon-Lic Review.World Bank, Statistical Handbook:Statesof the former USSR.World Bank: The Transitionto a Market Economyand for StatisticalHandbook: Statesof the FormerUSSR.Different pages Azerbajan,Armenia, Georgia, Gyrgizstan,Kazakhstan, Tadjikstan,Turkmanistan,Uzbekistan Iran. Vol. 11, EuropaPublications, and London. (1995a) "A-I: ', basedon IMF, EconomicReview.World Bank, Statistical Handbook:Statesof the former USSR.World Bank: The Transitionto a Market Economyand for StatisticalHandbook: Statesof the FormerUSSR.Different pages Azerbajan,Armenia, Georgia, Gyrgizstan,Kazakhstan, Tadjikstan,Turkmanistan,Uzbekistan Iran. Vol. 1,EuropaPublications, and London.

332

(1995b)"K-T', basedon IMF, EconomicReview.World Bank, Statistical Handbook:States the former USSR.World Bank: The Transition to a Market Economyand of StatisticalHandbook: Statesof the FormerUSSR.Different pagesfor Azerbajan,Armenia,Georgia, Uzbekistanand Iran. Vol. II, EuropaPublications, Tadjikstan,Turkmanistan, Gyrgizstan,Kazakhstan, London. V. Fahay,L. and Narayanan, K. (1986) "MacroenvironmentalAnalysisfor Strategic Management",West Publishing Company.St. Paul, USA. p. 216. Ffrench,A. R. (1979) " Competition Among SelectedEasternCanadianPortsfor Foreign Cargoes",Maritime Policy and Management,Vol. 6, No. 1, pp. 7-8. ", Fischer,S. and Sahay,R. and Vegh, A. C. (1996) "How Far Has the Transition Progressed? The Vol. 10,No. 2, pp. 45-66. Journal of EconomicPerspectives, Long RangePlanning, Vol. 26, Foster,M. J. (1993) "ScenarioPlanning for Small Businesses", No. 1. pp. 123-124. FraserJ. R, (1993) "The Emergenceof Vladivostook as a Commercial Port: US VersusJapanese Development",Maritime Policy and Management.Vol. 20, No. 2. pp. 101-107. FraserJ. R. (1994) "The Ports of the RussianFar East: A ManagementStrategy",Maritime Policy Vol. 21, No. 1, pp. 37-44. and Management, Frenkiel, N. F. and Goodal,W. D. (1978) "Simulation Modelling of EnvironmentalProblems", (eds),John Wiley & Sons,Chichester.NY. p. 72. Galer, G. and Kasper,W. (1982) "ScenarioPlanning for Australia!', Long Rang Planing, Vol. 15, No. 4, pp. 50-55. Gaven,B. (1987) "The ShipsAtlas 1987", Reigate,Surrey,England. Ghaem-Maghami,(1994) "SCAP: Multimodal Transportationand Role of Iran", Sanat-e-Haml-oNaghl, No. 131, p. 30. for K. Gharabaghi, (1994)"DevelopmentStrategies CentralAsia in the 1990:in Searchfor Alternatives",Third World Quarterly, Vol. 15,No. 1, pages.105and 117. Giannopulos,A. G. (1984) "Modelling of InternationalFreight Transportfor Greece", International Journal of Transportationand Economics,April, Vol. XI, No. 1, pp. 64-72. Gillen, W. D. and Waters,G. W. (1996) "TransportationInfrastructureInvestmentand Economic Development:Investment",TheLogistic and Transportation Review,Vol. 32, No. 1, p. 67. Ginter, M. P. and Duncan,J. W. (1990) "MacroenvironmentalAnalysis for Strategic Management",Long RangePlanning, Vol. 23, No. 6, pp. 91- 100. for Goldfarb,L. D. andHuss,R. W. (1988) "Building Scenarios an Electric Utility", Long Range Planning,Vol. 21, No. 2, pp. 78-85. Gordon,J. T. Becker,S. H. and Gerjuoy, H. (1974) "Trend-Impact Analysis: A New Forecasting Tool", The Future Group, Galstonbury,Connecticut. Gray, R. (1982) "Behavioural Approachesto Freight TransportModal Choice", Transport review, vol. 2, No. 2. pp. 161-184. Approach to Surveying the Demandfor Freight (1983) "A Socio-Organisational Transport", in: New surveyMethodsin Transport, 2nd International Conference,Elizabeth, S. Richarson,A J. (eds),FRG. VNI SciencePress,Utrecht, The Netherlands,pp. 243-253. Griffin M. (1996)"Uzbekistan".TheMiddle EastReview,21st edition, Unwin Brothers,Woking, , 333

U. K. p. 52. Haghi, A. M. (1995) "Economic Co-operations", Sanate Hamlo-Naghl, Vol. 145, pp. 10-13. Haglund, G. D. (1986) "The New Geopolitics of Minerals: An Inquiry into the Changing International Significance of Strategic Minerals", Political Geography, Quarterly, Vol. 5, No. 3, July, p.224. Hanter, S. (1995) "Eastern Europe and the Commonwealth of Independent States", in Central Asia Quarterly, 2nd, edition 1994, Europa Publications, p. 8 1. Harris, C. L. (1993) "Xinjiang, Central Asia and the Implications for China's Policy in the Islamic World", The China Quarterly, March, No. 133. p. 114. Hayuth, Y. (1982) "Intermodal Transportation and Hinterland Concept", Tijdsschrift voor Economische en Sociale Geografle, 73.

(1987) "Intermodality: Conceptand Practice-Structural Changesin the Ocean , Freight TransportIndustry", Lloyd of London Press,London. (1992) "Multimodal Freight Transport" In: Modem Transport Geography:B ehaven , PressLondon, N.Y. pp. 199-214. in Hensherand Daniels (1995) "Productivity Measurement the Urban Bus Sector", Transport Policy, Vol. 2, No. 3. p. 179. Herzig, E. (1995) "Iran and Former SovietSouth". Former Soviet South Project,The royal Institute of InternationalAffairs, London. Hicks, K. S. (1977) "Urban Freight", in: Urban Freight Economics,Hensher,David. A. (ed), CambridgeUniversity Press,Cambridge,pp. 100-130. ", Hicks, P. (1994) " Jane's ContainerisationDirectory 1994-1995 (ed), Twenty-sixth Edition, JanesInformation Group, Alexandria, USA Higgs, A. (1995) "Turkish Bottlenecks", ContainerisationInternational, February,pp. 80-81. Holt, J. (1993) "Transportation Strategiesfor the RussianFederation: StudiesofEconomies in Transformation", PaperNo. 9,71beWorld Bank, WashingtonD.C. Honeycutt,A. (1989) "Maximising the EmployeeProductivity Factor", InternationalJournal of Manpower, Vol. 10,No. 4, p. 24. Howard, N. (1990) "The Manageras Politician and General:The MetagameApproach to Analysing Co-operationand Conflict" in: Rational Analysisfor a Problematic World, Rosenhead, J. (ed), Wiley & Sons,England,pp. 239-265. Hunter, B. (1996) "StatesMan Year-Book., statistical, Political and EconomicalAccount of the A Statesof the Worldfor the Year 1996-1997", The Macmillan Press, London. p. 712. Huss,R. W. andHonton,J. E. (1987) "ScenarioPlanning-What Style ShouldYou Use?". Long RangePlanning,Vol. 20, No. 4, pp. 21-29. Imakita, J. (1978) "A Techno-Economic Analysis of the Port Transport System SaxonHouse, ", Peakfield,England.pp. 12-14. Irecha,M. C. (1997) "Reforming CanadianPorts", Maritime Policy and Management,Vol. 24, No. 2, pp. 123-124. IRIRC, (1984-1993)"PerformanceReport of the Islamic RepublicIranian Railway Companyfor 1984-1993 Statistics& MachineServicesand Public RelationsBureauof the IRIRC, English ",

334

Version, Tehran. IRIRC (1986-1990) "Facts & Figures of the Islamic Republic Iranian Railway Company of Iranjor 1986-1990". Statistics & Machine Services and Public Relations Bureau of the IRIRC, English Version, Tehran. IRISL (1990) "Annual of the IRISL Reportfor 1990 ", Ghaem Magham Farahani Ave, IRISL, English and Persian version, Tehran. Jane's World Railway (1990) "Iran and Turkey", Allen, G. F. (ed), Thirty-second edition, Jane's Transport Data, pp. 613-4 and 75 1. Jenkins, A. 1. (1994) "All Change-New Directions for the Road Transport Industries of Russia, Ukraine, Kazakhstan and Belarus", Transport Reviews, Vol. 14, No. 4, pp. 289-320. Jin, DI. (1993) "Supply and Demand of New Oil Tankers", Maritime policy and Management, Vol. 20, No. 3. pp. 213-219. Johnson, G. and Scholes, K. (1988) "Exploring Corporate Strategy", Piench Hall International, London. pp. 58-62. Jungermann, H. (1985) "Psychological Aspects of Scenarios", in: Environmental Impact Assessment, and Risk Analysis, Vincent, T. Covello, J. L. Mumpower, P. J. M. Stallen, V. R. R. Uppuluri, (eds), Spring Verlag Berlin Heidelberg, New York, NATO Scientific Affairs Division. pp. 325-346. Jungermann, H. and Thuring, M. (1987) "Ibe Use of Mental Models for Generating Scenarios: A Psychological Analysis and Experiment" in: Judgmental Forecasting, Wright, G. and Ayton, P. (eds) John Wiley & Sons, Chichester, pp. 245-266. Junior Atlas WHSMITH, (1979) John Barthlomew & Son Ltd, Edinbourugh and Esselte Map Service, Stockholm. Kanafani, A. (1983) "Transportation Demand Analysis", Mcgmw-HiII Book, New York. Kaufmann, H. J. (1962) "Planning for Transport Investment in the Developing of Iran", Papers and Proceedings, American Economic Review, 52. P. 396. Khan, M. A. (1989) "Realistic Planning for Transportation-A Flexible Approach", Long Range Planning, Vol. 22, No. 5, pp. 128-136. King, W. (1986) "Handbook of Strategic Planning", John Wiley & Sons, New York. pp. 6.146.16. Kleiner, A. (1995) "Creating Scenarios", in: The Fifth Discipline Fieldbook- The Art and Practice the Learning Organisation, Scenge, M. P. and Roberts, Charlotte and others (eds), Nicholas of Brealey Publishing, London. p. 275. Kneafsy, T. J. (1975) "Transportation Economics Analysis", Lexington Books, Lexington, pp. 7379. Konratowicz, J. L. (1992) "Methodological Solutions for Increased Efficiency of Modelling and Simulation of Seaports and Inland Freight Terminals", Maritime Policy and Management, Vol. 19, No. 2, pp. 157-164. Korea Maritime Institute (1992) "Sea Transportation Between Russia and Korea, Prospectsfor the Development of Trans-Siberian Container Service ", Joint Research Report Between Korea Maritime Institute and Russia Soyuzmomiiproekt, Busan.

Lane, A. (1984) "Lloyd's ofLondon ", PenguinBooks, London, p. 174.

335

Lansdowne,Z. (1981) "Rail Freight Traffic Assignment", TransportationResearch,A. Vol. 15A, p. 183. Ledger,G. and Roe,M. (1993) "East EuropeanShipping and EconomicChanges:A Conceptual Model", Maritime Policy and Management,Vol. 20, No. 3, p. 233. Leemhuis, P. J. (1990) "Using Scenariosto Develop Strategiesat Shell", in: TheManager's Strateg", Taylor, B. and Harrison, J. (eds),HeinemannProfessional Casebook Business of Publishing,London, pp. 5-18. Legler, B. J. and Robertson,D. T. (1976) "National and RegionalEconometricModels", in: ForecastingTransportationImpacts Upon Land Use", Wendt. P.F. (ed), Martinus Nijhoff Social Sciences Division, Leiden, Interprint, Malta, pp. 16-31. Lensink, R. (1995) "Foreign ExchangeConstraintsand Developing Countries", Economic Modelling, Vol. 12,No. 2. pp. 179-191. Leontief, W. Gray, C. and Kleinberg, R.(1979) "The Growth of Maritime Traffic and the Future of World Ports", International Journal of TransportEconomics,Vol. VI, No. 3, December.p. 249. Len-nan, and Garcia-Garcia,J. and Wichelns, D. (1996) "Land and Water Policies in Z. Uzbekistan",Post-SovietGeographyand Economics,Vol. XXXVII, No. 3, March. p. 149. MacNully, C. A. R. (1997) "Scenario Developmentfor CorporatePlanning", Futures, April, pp. 128-133. Mc Donell, G. (1995) "The Euro-Asian Corridor Freight and Energy Transportfor Central Asia The Royal the CaspianRegion", Post-SovietBusinessForum, Russiaand CIS Programme, and Institute of InternationalAffairs, London. Macomber,F. (1969) "Land Bridge StepsToward Reality", Distribution Manager, January,pp. 30-31. Mahoney,H. J. (1985) "Intermodal Freight Transportation", Eno Foundationfor Transportation, Westport,Connecticut,pages271 and 322. Maddison,R. (1980) "A. ConceptualModelling, B. Logical Modelling". The Open University Walton Hall, Milton Keynes,pp. 7-11 Press,M352 CourseTeam, Mandel, F. T. (1983) "Futures Scenariosand Their Usesin Corporate Strategy", in: The Strategic Management Handboo,Albert, J. K. (ed), MCGraw-HIII Book, New York. Martino, P. J. (1978) "rechnological Forecasting",in: Handbookof Futures Research,Fowles, J. (ed), GreenwoodPress.London, pp. 389-390. Flows at Airports", Transportation,Vol. 22, Matthews,L. (1995) "ForecastingPeakPassenger No. 1, p. 55. Mayer, M. H. (1973) "Some GeographicAspectsof TechnologicalChangein Maritime Transportation",Economic Geography,Vol. 49, p. 154. McCalla, J. R.(I 994) "CanadianContainerPorts: How Have They Fared?How Will They Do?", Maritime Policy and Management,Vol. 21, No. 3, pp. 207-217. Mendenhall,W. and Sincich, T. (1989) "A SecondCoursein BusinessStatistics: Regression Analysis", 3rd edition. Dellen Publishing, Sanfrancisco, 80. p. Mercer, D. (I 995a) "ScenariosMade Easy", Long RangePlanning, Vol. 28,4, pp. 81-86. (I 995b) "Simpler Scenarios:ScenarioForecastingas a Techniquefor Long-Range , PlanningMust Be Kept Simple to Be Effective". ManagementDecision, Vol. 33, No. 4, pp. 32-40.

336

States",The World Bank, Michalopoulos,C. (1993) "Trade Issuesin the New Independent D. Washington, C. Michalak, W. (1995) "Foreign Aid and EasternEuropein the New World Order", Tijdschrift voorEconomische Sociale Geografte,Vol. 86, No. 3. p. 264. en Miller, K. R. (1977) "Land Bridge, Mini-Bridge, and Micro- Bridge: A Questionof Getting It Togethce', Transportation Journal, Vol. 17, No. 1, pp. 64-66. Miller, B. R. and Wichem, W. D. (1977) "IntermediateBusinessStatistics:Analysis of Variance, Regression, Time Series", Holt, Rinehort and Winston, New York, pp. 290-291. and Miller, B. E. (1978) "Thc Trans-SiberianLandbridge,A New Trade Route BetweenJapanand Europe:Issuesand Prospects".Soviet Geography,Vol. XIX, No.4, pp. 229-237. Ministry of Financeand Economic (1992) "UnpublishedReport", Tehran. (1993) "Assessment the Economic Co-operationof the Islamic Republic of Iran of CIS countriesin 1992", A Publication of Public Relationsof the Export Promotion Centreof with Iran, Safir Publications,Tehran. Ministry of Roadsand Transportation,(1983) "Handbook of the Iranian Roads", Statisticsand Machinery Services, February,Narmak Publications,Tehran. MRT (1994) "CompleteText of the Proceedingsand Approvals of the Transportation Coordination High Council ofIran- sessions:1-58", Vol. I& II, TransportationDeputy, Nas Publication,Tehran. MRT (1995) "Annual Performancesof the Foreign Trade Outlets of the Country", Transportation and TerminalsOrganisationof the MRT, Plan and ProgrammingDepartment,Letter: 24/2/1374, Tehran. Millett, M. S. (1988) "How ScenariosTrigger StrategicThinking", Long RangePlanning, Vol. 21, No. 5, pp. 61-62. J. Mitchell, B. R. Tydemand,J. and Georgiades, (1979) "Structuring the Future-Application of a Scenario-Generation Procedure",TechnologicalForecastingand Social Change,Vol. 14, pp. 409428. Vol. 10, No. " Murrel, P. (1996) "How Far Has the Transition Progressed? EconomicPerspectives, 2, pp. 25-44.1 Naylor, E. M. (1983) "Planning for Uncertainty-theScenario-StrategyMatrix", in: "The Strategic Management Handbook", Albert, J. K. (ed), McGraw-Hill Book, New York, pp. 22.1-22.11. Neuman,L. W. (1994) "Social ResearchMethods: Qualitative and QuantitativeApproaches 2nd edition., Allyn and Bacon,London, pp. 18-27. Nijkamp, P. and Blass,E. (1994) "Impact Assessment Evaluation in Transportation and Planning", Kluwer Academic Publications,Dordrecht,The Netherlands. Noritake, M and Kimura, S. (1983) "Optimum Number and Capacity of SeaportBerth", American Society of Civil Engineering (ASCE), Journal of Waterways,Harbours and Coastal Engineering, Vol. 109,No. 3, pp. 323-339. Norse, D. (1979) "Methodology: Developmentin the TechniquesEmployed in ForecastingScenarioAnalysis in Interfutures", Futures, October,pp. 412-422. A Noble, J. and King, J. (1991) " USSR- Travel Survival Kit ", Lonely Planet, London.

337

Nooshahr (1982) "Proceedings of Marine Organisations Seminar of Iran in Nooshahr", IRISL Publication, Tehran. Norris, J. T., (1972) "Minibridge in Focus,", Railway Management Review, Vol. 72, part 4. p. 141. Nutt, C. P. and Backoff, W. R. (1987) "A Strategic Management Process for Public and ThirdSector Organisation", Journal of the American Planning Association, Vol. 53, No. 1. p. 54. Ogden, K. W. (1992) "Urban Goods Movement", Ashgate Publishing, Gower House, Aldershot, England. Ogutcu, M. (1995) "Eurasian Energy and Politics", Futures, Vol. 27, No. 1. pp. 37-63. OPEC bulletin (1996) "Iran's Gas Oil Use in Power Plants Fall by Some 60%", January, p. 29. Padelford, N. J. and Gibbs, S.R. (1975) "Maritime Commerce and the Future of the Panama Canal,. Cambridge: Cornell Maritime Press. New York. pp. 1-2 1. Parente, F. J. and Anderson-Parente, j. k. (1987) "Delphi Inquiry Systems: History and Defining Characteristics of the Delphi Method" in: Judgmental Forecasting, Wright, G. and Ayton, P. (eds), John Wiley & Sons, Chichester, pp. 149-150. Payarn Darya, (1992) "New Norms for Unloading of Goods from Ships" No. 3, pp. 53-54. (1994) "Annual Performance of the IRISL and Affiliated Companies in 1993 ", A Monthly Journal of the Islamic Republic of Iranian Shipping Line (IRISL), No. 22, Persian and English, Tehran. Payam Emrooz, (1995) A Persian Monthly Social, Economical Journal, Shahrivar, Tehran, p. 7. Peters, J. H. (1993) "Russia's Waterborne Trade and Transport: Issues in Market Transformation", Maritime Policy and Management, Vol. 20, No. 4, p. 290. Picard, and Nguyen (1988) "Freight Transport Planning and logistics: Lecture Notes In Economics and Mathematical Systems", Beckman, M. and Krelle, W. (eds), Conference Proceeding of an International Seminar on Freight Transport Planning and Logistics Held in Bressanone, Italy, July, pp. 186-209. Pike, J. and Gandham, B. (1981) "Review of the Commodity Flow Studies 1975-1979", Transport Operations Research Group, Research Report No. 36, University of New Castle upon Tyne, p. 32. Pollins, B. (1982) "Modelling International Trade Flows: A Survey and Comparison of Simulation Approaches", International Political Review, Vol. 3, No. 4, p.504 Ports Development International, (1993) "Area Survey: Middle East, Oasis Trading: Iranian Ports Are Looking to Capitalise on Trade with Former CIS States", Pill, December, p. 2 1. Porter, M. (1985) "Competitive Advantage", Free Press, New York, pp. 455. Porter, L. A. Roper, A. T. Mason, W. Thomas, B. Rossini, A. F. Banks, J. and Wiederholt, B. (199 1) "Forecasting and Management of Technology ", John wiley & Sons, New York. pp. 259270. PSO (1994) "Annual Performance of the Ports and Shipping Organisation of the Islamic Republic Iranfor 1373", Plan, Programming and Budgeting Department of PSO, Tehran, pp. 14-20. of PSO (1993) "Annual Performance of the Ports and Shipping Organisation of the Islamic Republic of Iran for 1372 ", Plan, Programming and Budgeting Department of PSO, Tehran. (1992) "Operational Report of the Ports and Shipping Organisation of the Islamic , Republic of Iranfor 1371 ", Operations Department of PSO, Tehran.

338

(Undated,probably 1991) "An Investigation in the Performanceof the Iranian Ports , for 1320-1369", Ports Shipping Organisationof the Islamic Republic of Iran, Plan, Programming Tehran. BudgetingDepartmentof PSO, (1320-1369equivalentto 1941-1990), and (1990)"Annual Performanceof the Ports and Shipping Organisation of the Islamic , Republic of Iranfor 1370"0Ports Shipping Organisationof the Islamic Republic of Iran. Tehran. (1989)"Annual Performanceof the Ports and Shipping Organisation of the Islamic Republic of Iranfor 1369", Ports Shipping Organisationof the Islamic Republic of Iran. Tehran. (1988)"Annual Performanceof the Ports and Shipping Organisation of the Islamic Republicof Iranfor 1368", Ports Shipping Organisationof the Islamic Republic of Iran. Tehran. "Annual Performanceof the Ports and Shipping Organisation of the Islamic Republic of Iranfor 1367", Ports Shipping Organisationof the Islamic Republic of Iran. Tehran. PBO (1989-1993)"The Preliminary Resultsof the First Social, Cultural and Economical Development the Islamic RepublicofIran, Transportsector". unpublished. of Development, Social,and Cultural Plan of the Islamic Republic (1993) "Annexto the Second _, Iran, 1993-1998 Documents Publication,September, (proposal)". Centrefor Economic-Social and of Zohal Publications, Tehran. Quinet, E. and Reynaud,C and Marche, R. (1983) "Goods Transport", in: Ninth International Symposium Theoryand Practice in TransportEconomics,Transport isfor People,Madrid, on 2nd-4th November,ECM Publications,Paris,pp. 141-163. Raeece Dana,F. (1995) "A Potential Ground Waiting for Appropriate Seed",Sanate-HamloNaghl, No. 145,p. 16. Rafsanjani,A. A. (1996) "Rafsanjani Presents'96-97' Budget to Parliament", OPEC Bulletin, January 1996,p. 48. Ramarnoban Rao, S. V. T. and Sriaman,S. (1985) "Efficiency of Railway Transportation:the Indian Experience",in: International Railway Economics,K. J. Button and D. E. Pitfield, (eds), Gower publishing, England.pp. 297-317. Razaghi,E. (1994) "A Commenton the SecondDevelopmentalProgramme:Privatisation (Political-Economic Journal in Persian), No. 85-86 and Policies", Etela Aat Seeyasi-Eghtesadi 87-88. Rashid,A. (1994) "7he Resurgence Central Asia: Islam or Nationalism?", zed Books, London. of Reguraman, and Chan, C. (1994) "Me Developmentof SeaAir IntermodalTransportation:An K. Assessment Global Trends", TheLogistics and TransportationReview,Vol. 30, No. 4, pp.380of 81 ReitsmaJ.A. H. (1980) "Africa!s Land-LockedCountries:A Studyof Dependent Relations",Tijdschrifi voor Econ.en Soc.Geografte,7 1, No.3, p.131. Riahi-Belkaoui, A. and Fekret, M. A. (1994) "Merits of Derived Accounting Indicator Numbers", Managerial Finance, Vol. 20, No. 9.p. 3. Rijn, V. H. (1981) "The Trans-SiberianRailway: Already a Looming Threat, Aims of Industry", in: The Challengeof SovietShipping, National StrategyInformation Centre,London, pp. 30-37. Roberts,J. (1996) "Caspian Pipelines", Former Soviet South Projects,The Royal Institute of InternationalAffairs, London. Rutland,P. (1994) "Democracy and Nationalism in Armenia", Europe-AsiaStudiesVol. 46, No. 5, 339

Carfax Publishing,Glasgow,pp. 839-861. Investigation of the Level of Sajadpour,S. M. K. (1994) "Iran, CentralAsia and Caucasus: Analysis in the Economic and Political Relations",Studiesof the Central Asia and Caucasus, Studies,Year three, Seasonal Journal (Persian)for the Centreof the Central Asia and Caucasus Secondseries,No. 7, Winter, Tehran,pp. 7-16. ". Salvatore,D. (1982) "Theory and Problemsof Statisticsand Econometrics Mcgraw-Hill Book, New York. p. 147. Sampson, R. and Farris, T. M. (1979) "Domestic Transportation Practice, Theory,and Policy". J. HoughtonMifflin, Boston,pp. 4-5. (1982) January,p. 45 Sanate-Haml-o-Naghl, (1983) August, p. 105. 9 27-28 pp. 9- 10. . for (I 989a) "New Procedures Elimination of TransportationFormalities", No. 83, pp. (I 989b) "40 ThousandLorries, LargestNeed for InternationalTransport", No. 83,

(1993a)"EuropeanCommunity and TransportationIssue". No. 116,p. 20. (I 993b) "Railway, Main Componentof Transport in the SecondDevelopmental . Programme",No. 123.pp. 44-45. (I 993c) "Issue of Train Delays in Iranian Railway: Assessment Causes of and Some . Suggestions", No. 127,p. 35. (I 993d) "809 ThousandVehicles in Servicefor RoadTransport of Iran", No. 116,p. 19. (1994a)"Central Asian Transit Shipments,Lack of TIR Convention", No. 133,p. 10. (I 994b) "Commercial Balanceof Iran in 1993" No. 135,p. pp. 15-18. . (1995a)"Gold, Dollar, Deposit: What Banking SystemDoes ", Vol. 136,p. 13 (1995b) "Iranian Lorries in Armenian Roads",Vol. 141,p.22 . (I 995b) "Chancellor of the Central Bank of Iran: Dollar ExchangeRateWould , Not Change",Vol. 141, p. 56. (1995c) "Transit ChanceKnocks Only Once: Waiting for the Facilitation of the Ways -, Transit of Goods". No. 144, p. 23 Vol. 145,p. 59. (1995d) "Performanceof CommercialPorts of the Country in Eight Months",

(1995e)"Contradiction in Duties, a Matter for Railway: A Glanceto the Rail Transportin Iran", No. 144, pp. 43-44. (I 995f) "ECO Rail Capabilities,Critical Points: A Glance at the Iranian Rail , Requirements Within ECO Transport", Vol. 145,p. 24. BorderCrossingBridge BetweenIran and Armenia", (1996a)"Utilisation of the Conu-non -, 146, 19. No. p. (1996b) "InadequateStructure,Low Output: An Investigationof the Road _, Transportationof the Country From MacroeconomicsPoint of View", No. 148, p. 28.

340

(I 996c) "Railway, the Linkage of the InternationalWatersof Asia and Europe: , Prospects Transit Railway in Region and Iran", No. 150,pp. 6-13. of Saunders, J. and Sharp,A. J. and Witt, F. S. (1987) "Practical BusinessForecasting", Gower A. Publishing,England.pp. 314-326. Sbirley, G. E. (1995) "Is Iran's PresentAlgeria's Future", Foreign Affairs, May/June.p. 36. S. Schnaars, (1987) "How to Develop and Use Scenarios",Long range Planning, Vol. 20, No. 1, pp. 105-114,pp. 108-109. Schofer,J. and Stopher,P. (1979) "Specification for a New Long-RangeUrban Transportation PlanningProcess",Transportation, 8, p. 205. Shahbazi, (1996) "Homa Soon or Late ShouldJoin to GATT: GATT Agreementand its Impact S. Civil Aviation of Iran", Sanate-Haml-o-Nagl, No. 151, pp. 34-44. on Shearer, (1994) "BusinessForecastingand Planning", PrenticeHall, London. pp. II 1-118. P. A. Shemshad, (1994) "Iran After Completion of Mashad-Sarakhs Bafgh-BandarAbassWill Be and High Way of the Region", Special Interview with ECO Secretary,Kayhan Havai, Persian the (An Iranian daily journal), August 3 1, No. 1096, p. 2. version, (1995) "ECO: A Unique Organisation,a Vision of Common Destiny", -ECO, , No. 145,English news,p.3 SecretaryGeneralin an Interview, Sanate-Haml-o-Naghl, Sheffi, Y. and Mahmassani,H. and Powell, B. W. (1982) "A TransportationNetwork Evacuation Model", TransportationResearch,16A, No. 3, p. 212. M. Shisheh-chiha, (1994) "TransportationRegimesof Central Asian Countries",Translator, Review,Vol. 3, No. 8, The Publishing Houseof the Foreign Ministry, Central Asian and Caucasus Tehran.pp. 365-385. Shneerson, (1981) "Investment in Port Systems",Journal of Transport Economicsand Policy, D. September, 201-216. pp. (1983) "Short Term Planning for a Port System", Maritime Policy andManagement, , Vol. 10,No. 4, pp. 217-250. J. Shoemaker, H. P. (1992) "How to Link StrategicVision to Core Capabilities", Sloan Review,Fall, pp. 69-76. Management (1995) "ScenarioPlanning:A Tool for StrategicThinking". Sloan * Management Review,Winter 1995.p. 29. Sinclare,P. H. (1979) "Port Studiesin Developing Countries", Foxlow Publications, London. pp. 16-17. Sinclare,H. (1995)'Tadjikstan", Middle East Review,p. 121. Slack, B. (1994) "Domestic Containerisationand the Load CentreConcept",Maritime Policy and Management, Vol. 21, No. 3, pp. 229-236. Slater,G. A. (1978) "Choice of the TransportMode", International Journal of Physical Distribution and Management,Vol. 9, No. 4, p. 183. SCI, (1982) "A Statistical Reflection of the Islamic Republic of Iran for the Third Aniversary of the Islamic Revolution,Bahman 1360" StatististicalCentreof Iran (SCI), Plan & BudgetOrganisation (PBO).Tehran.

341

(1982) "Iran Statistical Yearbookfor 1981", Statistical Centreof Iran: Programmingand BudgetingOrganisationof the Islamic Republicof Iran, February,Tehran. (1990) "A Statistical Reflectionof the Islamic Republic of Iran -, Persianversions, Tehran. -, Tehran. (1989) "A Statistical Reflectionof the Islamic Republic of Iran English versionNo. 6, No. 7, Persianversions,

(1991) "A StatisticalReflectionof TheIslamic Republicof Iran" August,No. 8, Persian -, version,Tehran. (1990)) "A StatisticalReflectionof TheIslamic Republicof Iran ", November No. 8, English version,Tehran. No. (1991) "A StatisticalReflectionof TheIslamic Republicof Iran" January, 9, Persianversion,Ter. (1992) "A StatisticalReflectionof TheIslamic Republicof Iran " May, No. 9, English version,Tehran.pp. 140-150. (1992) "A StatisticalReflectionof TheIslamic Republicof Iran " October,No. 10, . Persianversions, Tehran. (I 994b) "A StatisticalReflectionof 7heIslamic Republicof Iran", July, No. 12,Persian i Tehran. versions, (I 993) "Iran Statistical Yearbookfor 1992", Statistical Centreof Iran: Programmingand _, Budgeting Organisationof the Islan-dc republic of Iran, January,Tehran. (I 994a) "Iran Statistical Yearbookfor 1993", Statistical Centreof Iran: Programming , Budgeting Organisationof the Islamic republic of Iran, March, Tehran. and (1995) "A StatisticalReflectionof TheIslamic Republicof Iran ", July, No. 13,Persian Tehran. versions, Smith, W. H. (1991) "Strategiesfor Social Research", University of Missouri-St Louis, Publisher: Ted Buchholz, 3rd ed. pp. 92-131. Speece, M. and Kawahara,Y. (1994) "Transportationin China in the 1990s",International W. Journal of PhysicalDistribution and Logistic Management,Vol. 25, No. 8. pp. 53-71. Starr, F. S. (1996) "Making EurasiaStable", Foreign Affairs, January/February. 80-92. pp. Stokke,P. R. and Ralston,W. K. and Boyce, T. A. and Wilson, 1. H. (1990) "ScenarioPlanningfor Norwegian Oil and Gas", Long RangePlanning, Vol. 23, No. 2. pp. 17-26. Stone,R. (1993) "Area Survey: Middle East-QasisTrading, Iranian Ports are Looking to Capitaliseon Trade with Former CIS States",Journal of Port Developmentand International, December,p. 21. Stover,G. J. and Gordon,J. T. (1983) "Cross- Impact Analysis", in: Handbook of Futures Research,Fowles,J. (ed), GreenwoodPress,London. p. 300-328. Strauss,R. (1993) "The Trans-SiberianRail Guide". 3rd. edition, CompassPublications,London. pp.254-55. Sun, N. C. and Bunarno,C. M. (1973) "Competition for Handling U.S Foreign Trade Cargoes:The Port of New York's Experience",Economic Geography,Vol. 49, p. 156-157. Talley, W. K. (1988) "Me Role of US OceanPorts in Promoting an Efficient Ocean

342

TransportationSystem",Maritime Policy and Management, Vol. 15,No. 2, pp. 147-155 Taneja, K. N. (1978) " Airline Traffic Forecasting:A Regression analysisApproach ", Lexington books, Lexington. Tarjornan,F. M. (1994) "Exports of CentralAsian ProductsThrough Iran", Kayhan Havai, No. 1096,August 31. p. 5. (1995) "ECO Shipping, Hoisting Sails, Waiting for Positive Wind", Sanate, Haml-o-Naghl, Vol. 145,pp. 18-19. ", Tarock, A. (1996) "US-Iran Relations:Headingfor Confrontations? Third World Quarterly, Vol. 17, No. 1. pp. 149-167. Tat, G. (1995) "A ThreeDay Seminaron Central Asian and CaucasianEconomy"," Sanate-HamloNaghl, No. 136.p. 19.Tehran. " Taylor, B. (1992) "Strategic Planning-WhichStyle Do You Need? in: Managing the External Environment,Mercer, D. (ed), The Open University, SAGE Publications,London. pp. 284-286. Railway: A Spring Excitemenf', An Interview with Managing Tehrani(1996) "Mashad-Sarakhs Director of the Roads' DevelopmentOrganisationof Iran, Sanate-Haml-o-Naghl,No. 147,pp. 3334. TheEconomist(1997) "Taking on the Mullahs", January18th,pp. 3-16. TheEconomistReview(1995) "What is Investment:Data Supplemene',Vol. 13, No. 1. pp. 11-21. World: World in Figures (1993) p. 56. 7he Economist-Pocket Timetablefor Railway,Roadand ShippingServicesOutside ThomasCook (1995) "Overseas England. Europe", Tren-dett, 1.(ed),March-April. ThomasCook Publishing,Peterborough, P. ThomasCook (1995) "OverseasTimetablefor Railway,Roadand ShippingServicesOutside England.Brenddan,F. (ed). Europe", March-April. ThomasCook Publishing,Peterborough, TNT Express, (1992) "Transport and Distribution in the SingleMarket". Gold Arrow Publications,London, pp. 65-68. Todd, p. (1990) "Modem Bills of Lading", Blackwell Scientific Publications,Oxford, pp. 243-266. Tongzon,J. (1991) "A Model for ForecastingFuture Supply of Shipping Servicesat Australian Ports", Maritime Policy and Management,Vol. 18,No. I. pp. 55-68. From 28 millions to 40 millions Tonnes", Torkan, A. (1994) "Port CapacitiesWill Be Increased Bandar-o-Darya, No. 47-48, Tehran,pp. 4-5. Torkan, A. (1995) "An Speechof the Minister for Roadsand Transportationof Islamic Republic of Iran in the Seminaron the Investigationof the Resources Capacitiesof CAC", Central Asia and Vol. 3, No. 8, SeminarProceedings. 332-336. The CaucasusStudies(Persian/English), pp. and Turkish StateRailway (1994) "Handbookof the TurkishPorts Operatedby TCDD " Istanbul. Twiss, C. B. (1992) "Forecastingfor Technologists and Engineers-A Practical Guidefor Better Decisions", PeterPeregrinus, IEE Managementof TechnologySeries, 15, London. pp. 53-77 UN (1967) "The Tum-Around Time of Shipsin Port". Departmentof Economic and Social Affairs, March, St/ECA/97. New York, pp. 15-19. UN (1981) "United Nations Conferenceon a Conventionon International Multimodal Transport", 343

TD/MT/CONF116,art.4 (2), United NationsPublication.New York. UN (1985) "Industrial Statistics Yearbook1982: CommodityProduction Statistics 1973-1982 Departmentof the International Economicand Social Affairs, Statistical Office of the UN. Volume 11. New York. ", UN (1992), "Statistical Yearbook1988189 Departmentfor Economic and Social Information and Policy Analysis, Statistical Division, Thirty-sevenissue,New York. UN (1993) "Industrial Statistics Yearbook1991: CommodityProduction Statistics 1982-1991 Statistical Office of the UN, Volume II. New York. ", UN (I 993a) "Statistical Yearbook1990191 StatisticalDivision. Thirty-eight Issue,New York. UN (1995) "Statistical Yearbook1993", Fortieth issue,New York. MethodsJor Improving General Cargo UNCTAD (1973) "Berth Throughput:Systematic Operations", TD/B/C.4/109 and Add. 1, Geneva,p. 27. (1989) "Mutimodal TransportandTechnological Development:Developing in , Freight Forwarding", ReviewofMaritime Transport,p.42. (1991) "Mutimodal Transportand Technological Development".Reviewof Maritime , Transport, New York, pp. 60-61. (1992) "Mutimodal Transportand Technological Development",Reviewof Maritime . Transport, New York, p. 60. (I 993)"Multimodal Transportand Technological Developments",Reviewof , Maritime Transport, New York, pp. 69-73. (I 993a) "Handbook of International Trade and DevelopmentStatistics 1992 . TD/STAT. 20, United Nations Publications,SalesNo. EIF. 93.11.D.9, ISBN 92-1-012032-9. (I 993b) " UNCTAD CommodityYearbook1993", United Nations Publications, , T`D/B/CN. I/STAT/1, New York. York. _, (1994) "Review ofMaritime Transport 1993", TD/B/CN/37, Genevaand New

UNCTAD/UN, (1994) "Economic Bulletin for Europe", Vol. 46, New York. Upshal,M. and Home, D. and Gian and Jenkins,C. and Harkness,E. (1994) "The Hutchinson Guide to the World', Helicon Publishing, Oxford, pp. 197-198. Transport-A Verroen,J. E. and Jansen,R. M. G. (1993) "rhe ScenarioExplorer for Passenger StrategicModel for Long Term Travel DemandForecasting",21st PTRC,SummerAnnual University of ManchesterInstitute of Meeting, Proceedingof SeminarD, 13-17September, Scienceand Technology,England. pp. 351-365. VIA International,(1993) "NAFrA, An Overview of Co-operation"Port of New York-New jersey Journal, January,Publishedby: The Port Authority of NY & NJ, New York, p. 9. Vlek, C. and Wilma, 0. (1987) "JudgmentalHandling of Energy Scenarios: Psychological A Analysis and Experiment" in: "JudgmentalForecasting", Wright, G. and Ayton, P. (eds),John Wiley & Sons.Chichester,pp. 266-273. Yearbookof International Organisations(1995) "International Organisations1994195", Vol. 1, Union of InternationalAssociation: (ed), publishedby K. G. SaurMunchen. p. 361. Wack, P. (1982) "Corporate Planning for an Uncertain Future", Long RangePlanning, Vol. 15,

344

No. 4, pp. 12-2 1. (1985, b) "Scenarios: Shooting the Rapids", Harvard Business Review, November. December, Vol. 63, No. 6, p. 139-150. Warf, B. and Cox, J. (1989) "The Changes Economic Impacts of the Port of New York", Maritime Policy and Management, Vol. 16, No. 1, p. 5. Werener, C. (1985) "Spatial Transportation Modelling", SAGE Publications, London, p. 45. Wheelwright, S. C. and Makridakis, S. (1985) "Forecasting Methodsfor Management", John Wiley & Sons, New York. Whiting, P. D. (1986) "Finance of Foreign Trade ", Pitman Publishing, London. pp. I- 11. Whitttaker, J. R. (1975) "Containerization: Press, New York, pp. 86-90,223-23 1. The Role of the Railways". Second Edition. Hasld

Wigg, M. K. (1982) "Planning for Uncertainly in Large Projects", in: New Dimensions of Project Management, Kelley, J. A. (ed), Lexington Books, Lexington, Toronto, pp. 77-95. Williams, W. E. (1980) "The national Transportation Policy Study: Commission and its Final Report", Transportation Journal, Vol. 3, pp. 5-19. Wilson, 1. (1978) "Scenarios", in: Handbook of Futures Research, Fowles, J. (ed), Greenwood Press, London. pp. 225-247. Wilson, J. M. and Bynner, J. (1979) "Research Methods in Education and Social Sciences- Block 1: Variety, in: Social Science Research", The Open University, Priory Press, U. K. p. 56. Wilson, F. John. (1993) "Carriage of Goods By Sea". 2nd editon, Pitman Publishing, London. pp. 918-935. Wolf, N. R. (198 1) "Forecasting the State of the Canadian Freight Transportation system to the Year 2000", in : Transportation Researchfor Social and Economic Progress, World Conference on Transport Research, Yerrell, J. S. (ed), Vol. 2, Gower publishing, Aldershots, Hants, England. Wor/dAtlas ofRailways (1978) Intercontinental Book Productions, Rand Mc Nally, London. World Energy Yearbook (1996) "Statistics", Petroleum Economist, London. p. 164.

Wright, J. (1994) "Eastern Europe and the Commonwealth of Independent States 1994 ", Europa Publications, London, p. 77. Wronski, M. (1975) "Containerisation-its Corporate Impact", in: Grandfield Research Papers, Marketing & Logistics, 197311974, p. 13. Zenter, D. R. (1982) "Scenarios, Past, Present and Future", Long Range Planning, Vol. 15, No. 3. pp. 14-16. Zwam, V. H. H. P. and Valk, V. D. J. (1984) "Long Term Policy and National Model", PTRC, Summer Annual Meeting, Seminar H. Transportation Planning Methods. p.243.

345

Appendices
Appendices of chapter two
Appendix 1. 1: Ports Directorates of Iran in the Persian Gulf and the Caspian Sea

In the PersianGulf, the Ports'Authority Directorateof KhosistanProvinceincludes

mainly the Ports of Khoramshahr,Abadan, and Imam which is the head port for the province. 2. The Ports'Authority Directorateof BushehrProvince,which basicallycoversthe port of BushehrandKharg Islandport authority. 3. The Ports'Authority of HormozganProvince,includesmainly the port of Abass,the

Rajayee, someminor ports. port of Shaheed and


Indian Ocean ports

4. The Ports' AuthoritiesDirectorate SistanandBaluchestan. of provinceincludesmainly the port of ChahBaharin OmanSea. Caspian Seaports 5. In the CaspianSeatherearethe Ports' AuthoritiesDirectorateof Gilan Provincewith the main port of Anzali and, 6. The ports' Authority Directorateof Mazandaran with the principal port of Nooshahr the CaspianSea.The locationsof all seaports shownin Figure the southern are on coastof
2.3.

7. Inlandwaterportsof the lake Urumiyeh Traditionally and economically,eachof the said commercialports mainly servescertain provincesas hinterland.Before the war with Iraq, the ports of Khoramshahrand Imam usedto be the largestports of Iran, but after theseports closed,there was a greateffort to Rajayee;the ports of Imam and Abass are the completethe port complex of Shaheed
largest in the country.

346

For the oil ports and terminals PSO ports act as a legal national port authority, giving

servicessuch as pilotage and tugs, buoyageservicesand the collection of dues and charges.In conunercialports where PSO branchesare not active and where there is no compulsorypilotage, Iran's Customscarry out the entire operationand act on behalf of
PSO wherever necessaryand when provided.

347

>- i >-* t- i b" ,- c %0 0 to %C %0 %0 %C %0


U

>%0

Co 22 (n i

h-h >" to to

;pwo
l=
rD

t-i :. LA 00 LA

0 c>0 C)
,-

W b-

0
D N,

L,0,0,0 CD 0 ch C>
>

1 0 CD 0 ) c>c>0 C> C) 0 c>0 0 C> c) c>0 8 0 C)C:(DC> 0. n


0
C> 0
0-00

t-J I NJ -3 -a a 2 -a 0 C-7i0 CD 0 , %Z 00 00 p (A i

>

too

t) tA b-. 00 C: c> )

-4

cl
m

-4 00 C>I CDI O
?0

b-

>

>> >
?0 5D 9

t5 (A gi a CY, ()0
Lfx LA b
LA C: ) 0 %,

?0 -

%0 00 p% !il - p

c 00

-4 CD b-- w tA 0% LA S

\ w a CD tJ 1-w 00 tA LA

fb ci

00
a c) C)
00

o C>

w c> C)

00 c>

W -

(D %0

t-J O\

LA D

w (A LA

w W (A w

b. b- b- D b-b- 2> W (: ,0 w 1. -4w (A


JE

00
ch

00
(

00
w

%D o

(:

c7, %0 ' m

!>a 0

0 .

00

00 M

00

CY k< % 00

(D C)
L') a tJ 00 L') %'m 00 00 (:
8
ON w

(Z) 8 t. CA(A w w u (A Ch b-ll C)


(> (D
1-1

C>

818 c:,81919118818 8. 9 1 1ta w w w (EI', (c31 ti b1 V 1.0 LA tli ci w !2 w tg -j t i, P.


a. "j t,. ) L4 , CZ -0 <2 tli Zo ,A ua %0 2 t, (D t' A CD c:, w oo b 0 'w 2 I Z
t-i r :, l l I e-

C)

! 0000 LA
o

00 b- e

'l' fb 10 fD ci fb ci
<

1
e

9 1
rD

f-N
EL r.

oo

CY%

(D (D

0 (D

(D C)

0 (D 0 (D 1C> (D
I

(D (D

(D

CD 0 g 0 0 (D0 C> (D 0 8 8 (D
00

00

LA
(D

> c 23 c> O\ w
C> iJ t li %0

; 1 00 - 2 8 2t 00 '0, 1C) w C: %0 ) t2
21

tA

tA

-4

(A

m1 X 0oRowo A 0. b- b- b- b.- b-. b- %Q b--b- bb-. w- ti W b- b- b- 8 (Z 0 a 0 b- %0 %0 14 >- m.> C> -PW > U) 00

,00 1
(
L. )

rn

NO :
4

(A w
" ti N) C:) b- 1.woo CYN-

C>

>- -a DO oO

A a

c7, tl, l

>

(11

! i

tu t) LN) C) a', O\

w CD p.

1 ..
8

E 0

wo N) tli 43 c:, 1 .4 bV b" (A (D

0 le

00 00 2
%0

CD

1 , ( IMc> ux . . .
(A
00

%0 t-i w MD p,

D ,
0 %.

th

6. -

tz

%0 L.) tli

L') -4 ffJ

>

LA

(A (D

(D

ti t) N)
N) 0-- t,

t^

CY%

CD (A

V- a 2 . cy, oo (Aa 1 43w .>

(A %v %D LA -p. LA 9 )
-

348

b- b.
%c c f.ii w

>- l -

>- 0- 1.

w- w- N- N00 00 Ul .4 = w 00 00 00 --1 w _e CD \D

Na %Z w W- (= Ic

00 00 = = -2 c\

w Z , 7,1w 2 w w ! -a 2 c%g .4 :b C:) t -I c .4, 0 w w u w (A e --3 (A c -, j 00 ,0 $.-j w 00 -4 %M 0 w 0 w 0 00 u %0 > 00 00


(A 00

(A

(A

t-i

22 LA

00

(A

00

c
(Z) 00 w C> 00 -j 00 00 w 00 (:

X Q

CO Cl.

8 C)CD CD 0 0 0) 0 C:CD CD 0 0 0 0 0 0 C:C)8 8 ) 0 10 C)C) -0 . .C)

(A

p %m .4

>

0 0 , CD -4 o LA Nz -4 -,

0 %. -P-

(D LA 0, % C)

(D 00

W 00

t-J 0

7: 9

1 -11 pl - :4 o o :, - i L) i ZA Zi ;- ;- w -i -4 0 0 ] LA 00 tli %0 w CD -a CD

"o

' S CD,

8 I 0 8

(A ch -4 0 C) 0 a ,0 ,v

C: ) CYN C: ) 0 0 0 , ,

C) 0 0 .

(A 0 C).

t, )

W CD

tJ

b--

w-

N4 th

N- Nei% 00 K) tJ

b00 2

Nc> c

b. -

4-1 (D oo 0,

t') -C

;4 :4 CD 3 :4 :p. w w uj ti w ZA LA a-Z w -0 15 00 t
%0 ti >-4

k4

vi a>', (A (A > p. > w t-i 0 O\ 0 c\ w (> o 9 --3 %0 -1 -4 (A :2 e- Z w (A w t, ) fit N)

CD C> C> C: ) CM CD CD CD C) c> .

t -1 0

e-

. . .
tA %0 w %0 8 tA -4 w IZ CD

" >O 8 8 8 , 4 !' 8


t2 -4 0 tA P. 0 00 2 vt o C> 00 CD c, 00 8 O\ -4 (: oo c tA %D a tj 00 8

i 0 00 8 0
00

%ID 00 -4 0 (A C-11W b-tA o c kli t-i 00 c\ 0 CD c > C)

th cy\ 0 , -4 CD

00 -2 C7N a w

wo

(D

0 C)CD

O\00 C> 00w (O c00 00 00


>. A C> t-J %0 tli 4) o t-i -4 c% tJ 2

10, DO

9 t-AA I U-3

Z) zo z4 1ia 1 1b, g' o j ; 100 1Z- 4) , s z g LA U) C) o\ git C>


%D -4 -4

N- b- b- ta 00 w (: %0 00 %0 44 N- 0-- 00 %0 W

w- b- N- 0% (A W" wo -2 w w
0 bl%D 00 N)

111 81 8

Lg" LI (A w s a a

N) II91

w LA LA 00 9 tJ al

1 CY, o - t) W 1-- 0 -4 P. %I0 > -4 s

--3 t^

s18 18181fi 8 1 L-1

349

00 00

'm 00

5 00

%0

cm k4

tz % 0 G `0 w

%0 %0

%D %D %0 %D m 00 CD e 00

iI t',
00
:2 .4i CY% 0, " ot w tA c)

YD 00 .a

%0 00 C,

%0 C

%0 00 g

1 No 43 %0 %0 %0 CD %0

00
13 A.
o " C> :Z

, 571 91

-4,9 0, t.

1 -1 -

8
Z;;

c: = p P

i > P ,0
2 ,

"*" 4. t 9 p
, 42 , w

> u
3

t!
-

(A
LA LA

-3
-1 (

D
LA 0

(A (21

w 10

0% CY,

- i " (A

'"Z: L i#
LA

4 1.i Z 0 .
w

IW

c>

-4

22 10 2 LD-,

-4

9
-4 0, 1, -I

I.

z.
!2 ::;

2 !1
w
0 .i

00
w

P. c>

(A g

Z";
e. Z

!-, pl,
., ,

bi
0

%0
tA
(>

00
.4,

00

", 1120 1211 '0


0,

A (A %0 74

t.

tw. )

UA

u %D CD

Wi

1 ta 00 1 ,>O 4 1 1 i 9 a 8 8 8 8 a ti t

2l ' il

<

u 9 ri

e>

0 % tA *

-0 c>

* CS,

, 12.91Z !2*2 ZO > , %0


m i

l > c> c>


A ,

CO u

LA

%0

8 8 8
s

cD

w -i
c> el) %0

50

; ','i , s Ii i x 2 '
1
C: )

'. i

9 -
8

10

2
, S s ' s

LA
-

'8 9 9
c,
( NO 00 c> 1 1

ti

be w ;0 iA u W !e %0 l> -

1
tA

91 ll

1u -1r-1 t:zl z:
W -j %. A 0 %0
t cm

E , Z: >< :: Z EZ t. . ,

Y, 10ue 0.
(M W

1 t- G 5 ti (0 )
9
u

' l
CD
LA t. tA

'

1V. ! u e
8 LA
CY, C: )

-1
%0

;..

a u

%0

i 5 i 3 V, U 1 u; 10
b u
u :Z

23 2 Z - a th

l ;l z5 gu
C) CY, .

u
9

8
,

5D

(A

00
350

-i
CD

Gn

N-L

W" NC 10

h-Im b-o YD h-%0

%0 No \o

GM

Ici

cn

Iti C.

t-i

L. ) 10

\o

D -4 1

00

w
8->

3 00 b--

t') ui

w t. )

a w W t-J UX W 10 o
-4 Lei b-t-J c\ -4

L')

%0 (ji

"0 LA

>

Gn

rA

10
%0

im
e

: i
0

Q i C\

-. j

(A

F ,
o

-i oo 00 -4 w
0

1.0

8 1 me

fb

D eb
M.

:. 3 rj

-4 00

00

-4

N-- %D (A >- -4

LA C: ) CD

t')

%0

--1

%0 c CZ.

e A al
; Ulk (_A CD w %0 00 (>

CD 0 -4
b-- W (

121 b-. W
(: W 1

g. UQ 2

%0

5D Z Z

i-.

L' ) > LA p b b :-1 L. :> )

ul

eg 0 qg 0
" 0 110

1
-

'? 0
g--e 2

p pta
%0

CD 10 00 a
t-J t-J

%D

a\ (X 00 th w
2
t^

00 > b->-. %0 (> 00

%0 t-J %D
w 0 !

%D bCY%W

00 00

(A

>-

%0 w -. CY%(A

CD
0

%M
0
w

g>

2
p

>-

t, 3 ta >Ut b--. 00
-4 -4 b--

6.O\ w 44

-4

>- f-W CY\ c> 0% %0 C) Ut 1W

>w >(A LA

0-- Nw t)
00 LA

0 %0

b'

t^ ICD

2 a w tA Ji CD1c> 0% 1 -2
-0

)'

b.' w (A

b-

b--

(D

S CA CD E'.

o 00 (A o

u3 2 %D PL -4 00 :! : ch 00 (D %10 00

%D %0 LA -4 0

0 w -4 V

%D %D -4 u (A -4 b- %0 t - - a s-w -4

-4 - -1 -4 Cl% %0

5 1-01 0-% . c2 9 rr

LA CD
0 0%
?0

%D :4 p bo A. CY% -4 IM

bU (A 00 %D c> p LI-3 Nb%CY%o 0 CD 43 1

:.a LA - a CD 00 0

6-. >W kIJ 2 00 C2N w (A 00

%0 %D

D %M C5 ' -,

2.09,
1 "' &

:-, 1*oo W %0 . %0 .4

351

5D

z
w

x ; x x x k< x
A el : A.

t4

6 0

g< x x
= c m

(A

k4

5D 5D 5D
C> CYN

fo Ci

;p'Z 00 rD :Z Pd

CD CD ui

ti

w "
(>

Ch CD cy, ti 00 23
W 0 -J (>

W -4 w
w
-4 .

C
0 8

1
ti

CD W

L4

'6 0 :p
,5 L4

6 6 6 6 6 c> 00
( :> " ( -3 c% -J Lh i-. :p 0 j> %0 ti

X'

P-N m bli Ll. cm

CD

p 00 C,
,
(A

O cy, 52

I P 16 6 1 5b c)1 p P CD Q 00 L (D C)
9 p p 00 6 1 0 p p1 p1 ti , 00 e cr

5D 5D 5D p

4- 00 00

00 CD -2 00

10 -p- w %0 (7, u Z. LA c> 0 121Co "0 P. w w C)


ON

16 16 l <3 t2 Z-

%0

w 1 16 1 1(D 6 9 p p
-j N-

M p g: 9 ' .tg .LA 0 (D 0

cl - a ll w .4 , 6 6 . . . .
1 m f 1 1 so

4 .e < 0

(> c, c> 0 ti Z0
tit CM -4 UJ) b--

b-

%0 6. -

tj

6 6 6 p 6 6 6 16 6
CD C) (D 6 b
,

0 CD LA
(A)

ZQ CD 00 (> r-) p 0 0 -j 2

352

ch

(ft

tz

E; fo

E. "mo

W ip. -

00 00 1-JI

1 4, .

0- 1w1

Cs

1CA

-0- W U3 00

F
w C) a (b -4 w 11)

P,

LA

LA

00

00

00

Cs -P6

Cal

I'll

co
0

IO fl CA

t ll oo

100

00

t-j tj 00

(1)

Lei :.a

]
1 0

C) ON
N)

41 .

to CD
LA

v M

C)

6 -4
00.

6 oN
,z "Cv,

b C)
SI

C)
b B

A ?D 8 w1 0 PO 11 OO

LA L" ITZ U)

0
1 1

LA

= 10

6 w coI%D-1

353

Appendix 6 General feature of the estimation procedure Searchingfor best fit of the GOPFT of Iran. Description of Independent variables Runs Model Intercept the run and r2 variables X1 X2 X3 X4 X5 X6 X7 X8 X9 XIO XI I Searching for best I - 117093.49
X12 X13 X14

predictive variables
X11 .......... ........ X11 X12

2 3 4 5 6 7 8 91
10 11

0.99 0.99 0.99 0.97 0.96 0.96 0.95 0.94 0.94 0.93 0.92 0.80 0.89 0.80 0.84 0.85 0.80 0.87 0.85 0.60 0.05 0.87 0.87 0.90 0.88 0.87 0.88 0.87 0.70 0.88 0.881

XI

X2 X3 X4 X5

X6 X8 X9 XIO X14

-113443.19 115135.29 -111438.60 -51981.52 -13659.02 -9658.17 -21130.09 -18726.59 -25251.57 -18032.47 -7365.87 -24277A6 -8074.45 -24168.39 -12321.19 -9691.20 -24749.13 -24493.36 -22008.88 -16815.49 15257.62 -224552.80 -24610.92 -9684.54 -23288.91 -25052-35 -17164.76 -26156.59 -9809.60 -18543.90 -27942.801
X2 X3 X4 X5

X12 X13

X6 Xg X9 XIO X13 X14

X2 X3 X4 X5

X6 X8 X9 XIO X14

X11 X12

X2 X3 X4 X5 X3 X4 X5 X3 X4 X5 X4 X5 X5 X5 Xf

X6 X8 X9 X10 X6 X3 X9 XIO X6 X11 X9 X6 X11 X9 X6 X8 X9

X11 X11

X14 X14

.......... . ..... . ........ MVI'V09

X11 X11

X14 X14

X11

X14

X6 X8 X11 X6X11 X5 X5 X5 X5 X5 X5 X5 X5 X5 X6X14 X14 XIS X14

X14

X14 First estimated model Second estimated model and exeltWon of X.6 Searching WIMP...

12 13 14 15 16 17 18
19

X16 X14 X17 X14 X18 X14 X19 X14 X20 X14 X7 X14 X14

20 21 22 23 24 25 26 27 28 29 30 31 32

Assessment Assessment of independent variable Final model the

X
X20 X14 X20 X7X14 X20XISX14 X20X16Xl4 X20Xl7xl4 X20XIRX14 X20 X19X14 X20 X14 X20 X4 X14 X20XII ___

for Searching more independen variables

Assessment Searching

354

x N-4

b b 1-11 CD 2

8 Lft

eb

A-

li .

ag

>g
1

10 e. W

b-w

Lft

C)

b.)

te
-

(> CD ON 00 rA 00 >- CY%r'n c> %0 2

(A w

u2
ba

(A . ul

00. cb c) 0

bN O c:. -4 . -3 o.
00

b-

V-N
c52

.> L')

c2% w

C:) t ki

rA a M

%, %0. ND. 0,

P- Po.

t^ mg

Nu
C) 0

0 rA P-N

5-

c:, 00
(D

a. L4
tA.

W.
m

CD

>IN

CD

CIN wi

ml

(! = :A 0
e c

g
1

CD CD 1M CD ZA c5 1.0 i I CY%

!-A ti

355

XXX
P

X X x x X x X x
l

V, t Coth
9 P P

C t
5> p

V
P

5)

0 %A
6

P 1 5>
Clo

' w P . ',> P, ' F, P . s> )


A 10 a -

%0

CY%

PI P U W
-4 it

'-'li WC>
p &
-

G9)9 ' .
g w

!: 2 t. ) Z LA
t t t

FD P

-. A

P P PI LAI P P P - P : > O * * *
w

C)

c>

0% tA 00
Co

c2% 00 w 00

%Dm %0 -3 -4 %0 W a', 00 C
00

Za

b',

;.-

-i

Zi

::

t3

-9
tA

t9

i') w u C) 13 >
p P
-a 43 km w

2, ,

9 C) 00

o o 5,, p p t. l :,. 6 0 k 43 1 t^1 w c bi a 2 2


P 5> C: P 1 P (c) >

LA
Z, C.

9 J 2
P 1

p #A '.

Z, J .
1

-P

c:,

L^ w

F)

P 1 -

CD

%0 4

k> j> Z 0. E

%0 w

CD 0
%0
u) im

0
lu*

(D

!2, , 0 Lb . c" , s it

4:) P P L.) i 0% o' LA


LA %0 (A

- L- 2 1
i lm -. Z
0 P
W

Co 2 g Z; 2" t t -.>-.>- - ) -.?>-.> P P 9 !b 6 ) P ?> > !b %A !i %. t! 9 CY, ei :> 0 9 > g A 1, Z, 2 J !:t wi %A Co CD -. 5D (A (A P p o CD CD (71 P !- P , 9 z4 00 00 z5 2 , %01
Q -

-)

- t tt Z s 1
9 1

1 lllP- 1 . 00

2<
1

eb pe

1 9

:g

!c
%0

(> ki.

LA

t^

"

z4 ch

CD
tm

CD:4

:4

21 0 w
c> Z

3 c>
CD

t"" 2 k c CD CD

1 -

g : , 00 Z 00 6 !b

e9 S

11

CDcl, Zc> c2. 2 :

9 -

1 1 1

1 1

# M>< 1

356

Appendix 9 General feature of the estimation procedure searching for best fit of the DOMTI of Iran. Description of Independent variables Runs Model Intercept r2 the run and variables X1 X2 X3 X4 X5 X6 X7 Xg X9 XIO Searching for 1 1.00 758205.20 X11 X12 X13 X14 bestpredictive variables '6............................ X1 X3 X4 X5 X6 X7 Xg X9 XIO X1 1 2 0.99 202955.30
X12 X13X14

3 4 5 6 7 8 9
10 11

0.99 218127.20 0.991 212872.90 0.99 0.99 0.98 0.98 0.99 0.98 0.97 0.97 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.961 0.97
0.98 0.98 0.98

X1 X3X4XSX6X7XS X13X14 X1 X3 X4 X5 X7X8 X13X14

X9 X11 X12

..............................

X9 X11X12

.......... ........ 9

147386.20 88121.10 -1982.40 -5536.20 -31402.10 -62060.20 -56773.00 -40401-00 -37740.70 -42595.80 -52758.80 -36997.60 -41736.10 -38444.40 -48809.30 -35316.30 -37094.90 -21796.80
-54345.30 -59087.30 -56212.10

X1 X3 XAS X1 X3 XAS

X7X8

X9 X12 X13X14

... 9 .......................... ....................

X8 X9 X12 X13X14

X1 X3 X4 X5 X9 X12X13X14 X1 X3 X4 X5 X9 X12 X14 X1 X3X4 X5 X9 X12

X1 X3X5X9X12 X3 X5 X9 X12 X3 X5 X9 X5 Xg X5 X9 X2 X15X5 X16X5 X17X5 XJBX5 X19X5 X20 X5 X9 X9 X9 Xg X9 First estimated model

12 13 14 15 16 17 18 19 20 21 22
23 24 25

Searching

X9 Assessment Assessment
, Secondestimated modeF

X21 X5 X9 X22 X5 X9 X23 X$ X9

X24 X23 X5 X9 X25 X23XS X9 X24 Xf X9 X25 X5 X9 X4 X23 X5 X9 X7 X23 X5 X9

Searching

26 27
28 29

0.97 0.96
0.98 0.98

-52886.80 -37765.00
-50999.10 -51214.20

...............

357

A----,

I; 0 f-nntiniii-A

30 31 32 33 34 35 36
37 38 39 40 41 42 43 44 45 46 47 48 1

0.991 0.98 0.98 0.98 0.98 0.981 0.98


0.13 0.76 0.91 0.87 0.86 0.98 0.98 0.98 0.98 0.98 0.72 1 0.76 1

X8 X23 X5 X9

146712.80 39218.30 9943.70

-55972.30 -49296.90 -56166.40 -78996.60 -51357.20 -54760.50 -43448.70

Assessment

XI

I X12 X23 X5 X9 X14 X23 X5 X9 X15 X23 X5 X9 X20 X23 X5 X9 X22 X23 X5 X9

X23 X5 X9

X5 X9
X4X23 X7X23 X9 X9 X9. X9 X9 X9

Assessment Assessment
Xs was eliminated due tolowcorrtiation

-94238.60 -1850.10 -95088.70 -88361.50 -98145.90 -80510.80 -94256.90 -7775.60 1 -93084.90

X12X23 X14X23 X14 X15X23 X20xl4X23 X22XIIJ

Final DOMTI model

X23 X9

X25XI14 X23 X9 X23

X14

Assessment jAssessment

358

E u p*
e =
Od =

0 C%

0 lz

>

.> w

CD 0

(X
ri

0 ep C B

m
ci

92. St lb> Gn

C) C)
00 o\ LA
rA s-Ut -4 l w 3 e ID Z %0 0-, %D

tal wI tAl w p. CD .4. -

:z1 W1
ta

%0

' cn

00

CD ul

0 w -4 00 CD CD
c3" 00

cx, u2 tA C7N 00 --3 > u2 42. m 9


N e..
92d

CD 0

,0 Le u2 w wx %0 (.A CYN t;; u2 Co: 2 :4, t-a 00. C:) 00 -

rA
p 10 (IN

1 ,2
x

c o. P.

c2, c>

.4 o

62

-4 t,.) q 00
22

00 10 1%01%0 tA
s

5D 10

9
N"

o
00

!-, 0
%D b-"0 p

00
'40

N-I

%O5

1 4

oo Z 00
w4 :i

0
CD

Z
G

1 40

EL
1

LA

C)

C>

bo

%D

2. :5

99
(AI

"- 0 W
-4 C)
a

rp

EL r.
3

u3 0-Z., u2

11 (D 0

1 92.
SA

%iD CD 0

ch

52 1 %0

Co = (D 5D

th p

(D 0

4. CD CY%

CD 9

Ei
eD

-> :>i * i 1
0-- (:
i

Ei

C)

JE>

tA t.

A cy,

Q N)

C: ) Irb el

CD

CD

43 CE,

359

r x t0

CD

n 0 eD
U3 U> 0 (A 0 00 00 bN

4 -cs

E.
o

LA

ON I ON (ON -4

%0

tl)

%0 00

t4

0
v

C -ci
:Z
(.A
o LA -4 01 ON I

0 00 0 ;: . 1.
tA 10

%0

C7, p CA
LA C) 00 > W

oo

190 v

p. V-41

a m9
C)
q -j

Od

00
63

bo bo 1 i, 3 I I ON o C: ) LA

tj

LA t! n
I aI FI

CD
>;

LA 00
10 -.4 CIS

.P.

%0 = %Q0

iR

b 6 Z G 9 b. VX
C) %10 0

p :5

6 F o

C)

E>

CD

>
bo U3 0 (D Os

fD

(D

g c)

%Z

to

. t: $

g*
:g
0Lot LA 00 'D tA ki fl) Ow

!A 2

(D =4

= :9

$A

0 2 00

= %0

a,

9'%

360

>O m ,ti M

m 1

>
w m

CI c 'Z , fD M
ul Z

- ' ja tj

%0

b-

01

PI a
C>

f-N

LA ! J -, tli
10

cy%

Z,

zr--;
2. eb :Z
C7N 1(AI R

E
o

E 39 . Cl. ts
&.

E. 0.
1

t, J tJ J C> t')
00

t-J LA

00

00

C)

00

EL

5k>
w

10 10 w

-4- w -4

t-i (A 1CD 1

5D!ii %0 0
-3 w

r-i .4 10 , 00 -4 . tA 0 -1 O\ 1o 100

.4 >
00 W

-4 2

L.3

4,

1.0 LA p

01 "D 0

tj

c>
1 14 o C
,.0

00
00

w r-

01 " Imw

: !V

a\ LA c7,

-i (A oO W

u2 u2

cn
wo

T v

-1

b- N-

fb

0 r.

>
62
c2. 0 5 2

t, J i -i 2

tJ t, ) cy% z
w e

t-i ti

oo

oo

" 2

CY\ w -i w w LA 00 -1 w 0' tA -4 , w

-4 0,
-4

" 0 cz) c) (A

t-J 0 %. %0 0 %

tJ (D p N -J 0 23

ta > %0

s-3 % ,0 tA

-4

%0 -4

00 >-

C) (it 00 w 00
C: ) t-i

0 (A .

- - p. (A ! LA L.) w

92. -.

CD -4 00 CA .

00

f md

t,i bo o
Nu

LA

LA 1

0 <
>

b G L L) :p. C2Z i

e<

Ch oo

cn

00

2i"m 2 ::. 'CY% *w

Li 00

-I

WI C>i -

Ln %10 .4 't A

00 LA -4 -4 %0 tA

oo

(A

* ]I .

i, . i ii P0 -4 1 ) C> LA :> i 00 Li ii o Z. - c71 -4

tJ o t, )

00 00 IZ

tA -

(A -i (A

CDO w m 00 > i 92-n


cm (A
eD

rr

M'

5-

.9

00

:4

9 -4 ID %0 th

1,

w CYNI(Al -ii

"

10 Q

c> c> C-11

00

1
n

E
Od

00

>
a\ CD:%
5D 5D p p

1 ,I b al c

:4 t-i 00 wl o CD c> CD.(D c> CD -a CD -4 C) 0 o

lij

! => 3 ll, %D w 1
-

, =0 - QM

361

GOPFT Border crossings All ports Imam Abass Bushehr ChahBahar Total Southernports Anzali Nooshahr Total Northern ports All ports GOPFT CACFT CombinedGOPFTand CACFT ports Imam Caucasus Imam GOPFT& Caucasus Abass Central Asia AbassGOPFT & Central Asia

Appendix 16: Share of regional ports of Iran In the forecast of the most probable GOPFT and CACIF, T scenario. 1994 1995 1996 1997 1998 1999 32860 34486.80 36321.80 38384-80 39977.40 31426.30 3079.80 3220.3 3379.70 3559.50 3761.70 3917.80 28346.50 29639.72 31107.10 32762.27 34623.09 36059.62 10339.25 10810.94 11346.16 11949.87 12628.60 13152.57 13827.57 14458.4 15174.19 15981.59 16889.31 17590.06 1822.73 1905.88 2000.23 2106.66 2226.32 2318.69 848.51 887.22 931.14 980.69 1036.39 1079.39 28172.87 29360.72 30729.42 32291.41 34 063.40 35449.05 1099.92 1150.1 1207.04 1271.26 1343.47 1399.21 448.33 472.18 408.54 427.18 499 519.71 1508.46 1577.28 1655.37 1743.45 1842.47 1918.9F 28346.50 29639.72 31107.1 32762.27 34623.09 36059.62 1334.81 1298.28 1277.69 1272.59 1282.78 1309.35 29681.31 30938 32384.79 34034.86 35905.87 37367.97 10339.25 10810.94 11346.16 11949.87 12628.60 13152.57 166.10 168.78 165.44 173.53 166.76 170.09 10512.78 10979.72 11512.26 12115.31 12795.36 13322.65 13827.57 14458.4 15174.19 15981.59 16889.31 17590.06 1161.29 1129.50 1111.59 1107.15 1116.02 1138.26 14988.86 15587.90 16285.78 17088.75 18005.33 18728.32 2003 2000 2001 2002 2004 2005 43298.80 462160 49496.10 53191.70 57366.60 62097 4243.30 4529.20 4850.60 5212.80 5621.90 6085.50 39055.52 41686.83 44645.48 47978.91 51744.68 56011.5 14245.31 15205.06 16284.22 17500.07 18873.61 20429.91 19051.47 20335.04 21778.28 23404.35 25241.30 27322.68 2511.33 2680.53 2870.77 3085.12 3327.63 3601.63 1169.07 1247.83 1336.40 1436.18 1548.90 1676.62 38326.85 40869.95 43749.21 45425.71 50681.73 54857.51 1515.46 1617.56 1732.36 1861.71 2007.83 2173.40 691.49 643.45 745.77 600.81 807.26 562.88 2078.34 2218.37 2375.81 2553.20 2753.60 2980.66 39055.52 41686.83 44645.48 47978.91 51744.68 56011.50 1349.67 1407.41 1482.51 1576.25 1690.29 1826.68 40405.19 43094.24 46127.99 49555.16 53434.97 57838.18 15205.06 16284.22 17500.07 192.73 204.91 177.04 15382.10 16473.98 17704.98 I 20335.04 21778.28 23404.35 1224.45 1289.78 1371.34 21559.49 23068.06 24775.69 18873.61 20429.5-1 219.74 237.47 19093.35 20667.38 25241.30 27322.68 I 1470.55 1589.21 26711.85 289159

GOPFr Border crossings All ports Imam Abass Bushehr ChahBahar Total Southernports Anzali Nooshahr Total Northern ports All ports GOPFr CACFT CombinedMost probableGOPFTand CACFr ports scenario -31 Imam 1424F. Caucasus 175.46 Imam GOPFT& 14420.76 Caucasus Abass 19051.47 I Central Asia 1174.21 I 20225.69 AbassGOPFT& Central Asia

362

Appendix 17: Share of regional ports of Iran In the forecast of the optimistic scenario in the GOPFT and CACFT. 1994 1995 1996 1997 1998 1999 GOPFr (1) Border crossings (2) All ports GOPFr (3) Imam (32.9% of 1) Abass (44% of 1) Bushehr (5.8% of 1j Chah Bahar (2.7%of 1) Total Southern ports Anzali Nooshahr Total Northern ports All ports GOPFr CACFT Combined GOPFT ports and CACFT (3+13) Imam Caucasus Imam GOPFT & Caucasus (15+16) Abass Central Asia Abass GOPFT & Central Asia 31426.2 32988.2534882.2 37142.8 39816.88 6 6 4 3079.7 3232.8 3418.46 3640.00 3902.05 7 4 28346.4 29755.3 31463.80 33502.84 35914.8 9 6 2 10339.2410853.1211476.27 112219.9913099.7 5 1T827.56 14514.8115348.20 16342.85 17519.43 1822.72 1913.32 2023.17 2154.28 2309.38 848.51 890.68 941.82 1002.86 1075.06 28172.8430354.2932819.36 35598.07 38728.62 , 1099.92 1154.59 1220.88 1300.00 1393.59 , 408.54 428.85 453.47 482.86 517.62 1508.46 1583.43 1674.35 1782.86 1911.21 28346.4929755.3631463.80 33502.84 35914.82 1334.81 2182.36 3029.91 3 77.46 4725.01 , 29681.3031937.7234493.71 37380.30 40639.83 42966.221 4210.69 38755.53 14135.89 18905.14 2492.041 1160.091 42265.71 1503.82 558.56 2062.38 38755.53. 5572.56 44328.09

10339.2410853.1211476.27 12219.99 13099.75 14135.89 , 173.53 283.71 393.89 504.70 614.25 724.43, 10512.7711136.8311870.15 12724.69 13714.00 14860.32 13827.5614514.8115348.20 16342.85 17519.43 18905.14 . 1161.29 1898.65 2636.02 3373.39, 4110.76 4848.12 14988.8416413.4617984.22 19716.2421630.18 23753.26

2000 2002 2001 2003 2004 200S GOPFr 46670,8251032.9956183.06 62286.89 69555.70 78259.21 Border crossings 4573.74 5001.23 5505.94 6104.12 6816.46 7669.40 All ports GOPFr 42097.0846031.75 50677.12 56182.78 62739.24 70589.81 Imam (32.9% of 1) 15354.7016789.851 18484.2320492.39 22883.83 25747.28 Abass (44% of 1) 20535.1622454.5124720.55 27406.23,30604.51 34434.05 Bushehr (5.8% of 1) 2706.91 2959.91 3258.62 3612.64 4034.23, 4539.03 Chah Bahar (2.7% of 1) 1260.11 1377.89 1516.94 1681.75 1878.00 2113.00 Total Southern ports 46276.9950849.83 56095.54 62155.76 69210.87 77491.21 Anzali 1633.48 1786.15, 1966.41 2180.04 2434.45 2739.07 Nooshahr 606.72 663.43 730.38 809.73 904.22 1017.37 Total Northern ports 2240.20 2449.58 2696.79 2989.77 3338.67 3756.44 All ports GOPFT 42097.08 46031.75 50677.12 5618i. 78 62739.24 70589.81 CACFr 6420.11 7267.65 8115.20, 8962.75 9810.30 10657.85 Combined GOPFT ports and 48517.19 53299.41 58792.33 65145.53 72549.54 81247.65 CA CF1P (3+13) Imam 15354.70 16789.85 18484.2320492.39 22883.831 25747.28 Caucasus 834.61 944.80 1054.98 1165.16 1275.34 1385.52 Imam GOPFT & Caucasus 16189.31 17734.65 19539.2021657.55 24159.16 27132.80 (15+16) Abass 20535 16 22454.51 24720.55 27406.23 30604.51 34434.05 Central Asia 5585.49 6322.86 7060.23 77T7.59 8534.96 9272.33 Abass GOPFT & Central 26120.65 28777.37 31780.77 35203.83 39139.47 43706.38 Asia , 1 -1

363

Appendix 18: Share of regional ports of Iran In the forecast of the combined pessimistic GOPFT and CACFT. 1998 1994 1995 1996 1997 1999 GOPFr 32064.05 32418.6332330.2231830.68 31426.26 30970.99 Border crossings %) 3079.77 3142.28 3177.03 3168.36 3119.41 3035.16 (9.8 All ports Iran's demand 28346.49 28921.77 29241.60 29161.8628711.27 27935.83 Imam (32.9%of total 110665.73 10549.07 10636.6410472.29 110189.45 10339.24 GOPFI) Abass(44%) 14264.2014335.3014005.50 13827.56 14108.18 13627.23 Bushehr(5.8%) 1822.72 1859.72 1880.28 1875.15 1846.18 1796.32 ChahBahar (2.7%) 848.51 865.73 875.30 872.92 859.43 836.22 . 28871.1428812.1528171.40 27324.09 Total Southernports 28172.84 28649.67 X1099.92 1122.24 1134.65 1131.56 1114.07 1083.98, nzali (3.5%) Nooshahr(1.3%) 408-54 416.83 421.44 420.29 413.80 402.621 Total Northern ports 1508.46 1539.07 1556.09 1551.85 1527.87 1486.61 29241.6029161.8628711.27 27935.83 All ports GOPF17 28921.77 28346.49 , 1 CACFT 1334.81 1266.97 1185.63 1092.14 988.00 874.87 130427.24 30254.01 29699.28 28810.70 Combined 30188.74 GOPIFT 29681.30 ports and CACF1P 10636.6410472.29 10189.4 Imam 10665.73 10549.07 10339.24 Caucasus 173.53 164.71 154.13 141.98 128.44 113.73 Imam GOPFT & Caucasus 10512.77 10819.86 10778.6210600.73 10303.19 10713.78 Abass 14264.2014335.30 14005.50 13627.23 13827.56 14108.18 Central Asia 1161.29 1102.26 1031.50 950.16 859.56 761.14 15285.4614865.06 143887 AbassGOPFT & Central 14988.84 15295.70 15210.44 Asia 2004 2003 2005 2002 2000 2001 23881.77 22460.26 GOPFT 26949.4925400.86 29818.01 28449.68, Border crossings 2922.16 2788.07 2641.05 2489.28 2340.43 2201.111 21541.36 20259.16 22911.58 All ports Iran's demand 26895.84 24308.44 25661.61 Imam (32.9%of total 9810.13 9359.94 8866.38 8356.88 7857.10 7389.43 GOPM 10507.98 9882.52 11176.38 Abass(44%) 11857.78 12517.86 13119.92 Bushehr(5.8%) 1729.44 1650.08 1563.07 1473.25 1385.14 1302.70 ChahBahar (2.7%) 805.09 768.14 727.64 685.82 644.81 606.43 20631.6219287.68 22060.33 Total Southernports 23514.13 26219.08 24924.71 AnzaIi (3.5%) 1043.63 995.741 943.23 889.03 835.86 786.11 Nooshahr(1.3%) 387.63 369.851 350.34 330.21 310.46 291.98 Total Northern ports 1431.26 1365.5811293.58 1219.24 1146.33 10778.09 21541.36 25259.16 22911.58 All ports GOPFT 24308.44 25661.611 26895.84 CACFT 754.50 628.691 499.26 367.99 236.58, 106.62 20365.77 CombinedGOPFTports 21777.94 26290.301 24807.7023279.57 27650.34 and CACIFT Imam 9810.13 9359.9418866.38 8356.88 7857.10 7389.43 Caucasus 13.86 98.08 81.73 64.90 47.84 30.76 Imam GOPFT & Caucasus 9908.21 9441.67 8931.29 8404.72 7887.8 7403.29 11176.3810507.9i 9882.52 Abass 12517.8611857.78 13119.92 Central Asia I 656.41, 546.96 434.36 320.15 205.83 92.76 I 13064.8212292.1311496.53 AbassGOPFT & Central 13776.34 10713.8 9975.27 I I Asia I

364

Appendix 19: Distribution of three GOPFT and CACFT scenario forecasts on ten border cro, ings (000 tonnes). Scenarios 1994 1995 1996 1997 1998 1999 Astara. 1388.93 1445.19 1510.55 1585.60 optimbaic 1388.93 1514.91 1654.54 1809.24 ressingslic 1388.93 1410.19 1418.84 1408.39 Most probable 1 155.371 161.671 168.991 177.391 Opti"Asdc 155.3711-69.421 18-5.00 199.121
ressintslic Misdo
11 probable

Djulfa

1671.13 1738.45 1980.93 2172.15 1380.27 1336.69 186.961 194.49 217.60 242.78
1SA d') I Q qg I

155371

1157761

I-SR-731

1157571

Bazargan

Midaveh

Razi

Taybad

Noordooz

Bajgiran

Lotfabad

Sarakhs

1179.43 1233.2 1294.2 1363.1 1440.5 1500.35 4 9 6 8 Gpdndstic 1179.43 1238.05 1309.13 1393.97 1494.33 1612.52 Posiumuc1179.43 1203.3 1216.67 1213.35 1194.61 1162.34 6 MoM pr0able 176.93 185.00 194.16 204.49 216.11 225.07 opdmbdc 176.9 3 185.72 196-39 209.11 224.171 241.90 1 rewmisdc 176.9 3 180.52 182.52 182.02 179.21 174.37 Mod 6.29 6.57 6.90 7.68 prob"k 7.26 8.00 Opumisdc 6.29 6.98 6.60 7.43 7.96 8.59 ressin"dC 6.48 6.47 6.29 6.41 6.37 6.19 Moo prOmble 34.49 31.43 32.86, 36.32 38.381 39.98 Opdndstic 34.88 37.14 39.821 42.97 31.43 32.99 FWWO95dC 32.42 32.33 31.83 31.43 32.06 30.97 MoM probabk 128.26 129.34 131.58 135.00 139.63 144.06 Optimi0c 128.26 172.91 218.23 264.28 311.16 358.99 reswnisdc 128.26 126.21 122.93 118.18 112.07 9.62 Most probabk 108.61 110.70 113.69 117.60 122.46 126.73 1 Opdmbdc 108.61 136.92 165.97 195.82 226.60 258.42 resmads* 108.61 108.03 106.42 103.49 99.33 94.11 Most pn"ble 54.15 56.67 61.04 55.19 58.62 63.16 optimisdc 54.15 82.81 97.73 113.10 128.99 68.30 ressin"ft 53.05 51.58 54.15 49.50, 53.85 46.90 I" prole 1172.32 1145.32 1131.93 1131.79 1144.79 1169.45 Opd"d5dC 1172.32 1875.861 2580.14 3285.24 3991.25 4698.32 rawmistic 1172.32 1117.70 1051.30 973.88 886.76 791.40 Scenarios 2000 2003 2004 200S 2002 2001 "" pr*" 1876.89 1994.12 2127.43 Optimwc 2386.20 2627.27 2900.72 Pusindsde 1280.52 1215.11 1144.02 mod rvb"k 197.12 223.99 239.69 optimilde 266.69 293.62 324.18 resdu*"ic 143.27 135.96 128.01 M` ",*"b, 1519.03 1734.491 1857.59 Optladstic1751.56 1915.27 2108.55 rewmiwc 1119.07 1067.72 1011.41 MoU prob" 243.77 260.20 278.66 Optindsdc 262.76 287.32. 316.31 Pesdndsdc 167.88 160.17 151.73 Me-stprobabk L 9.90 8.66 9.24 Opdndsdc 1 10.21 11.241 9.33 ,-"1 5.69 5.391 5.96 2278.95 2451.281 2647.57 3213.38 3573.911 3993.41 1070.80 998.79 930.89 257.45 277.58 300.44 359.12 391.47 446.32 119.821 111.77 104.18 1996.28 2152.97 2330.50 2337.63 2610.43 2937.07 953.29 896.28 842.93 299.47 322.97 349.61 350.68 391.601 440.60 143.01 134.45 126.45 10.64 11.47 12.42 12.46 13.91 15.65 5.081 4.78 4.49

M, 1",

Astara

Djulfa

Bazargan

Mirjaveh

Razi -

365

A e
q n 2

CD

1 YD %0 YD IM No %CD s Im CD 110 00 Ch
CD

y>
o

t4

Im

Ul

s.

10

LA (Z5 CY, r ,2 w i ui o u3 \o i 00 UJ oo (7, tJ 2 oo

- 00 00
1, C,

tA

tit "

>

t tj, C,

t, -

Ui 00

o,
00
fjQ C, C> 005> (J, s

00 -

uj

A. 43 -. 3 -J

W U

w U)

w -

2 %D %0

t, ) 00

bj -1

1 -

:,.
r n i 91 m
.

CD :
5A

vt
.

> 43 : -, ji,

w uj o -4
p p

U) tli t^ w
(A

e >
X

CM 0 C:m

u N. Z 12.

JE .

7 9

--e

eb

%0

0.

00 j t..

-2

00

Q\

C> 43

00 00 iii

q 74 91

:P,

E;
00 43 w " C)

%0 00 %D %0

4 5

= ' t c"I ZA ui

1,0 0 WO

t'

F h

14
c 5Z

, ' 2

c2%

2 'ti 00p 2 c7,p. s


i CO Ch m

u3

(>

b, u 0 oo

bi ch b%
LA
f. A

c2" .-

o tA og t-J 0%

l? s " s.
g
c7,

4 10

, v' le ,*l
?0
1- 1 1rit
u 00

F% %0 (A 0

00

*Z

<

3 (Y, 00 -

4C5%0

P t2e 00 .4

q
.
N

0
o

i
;0

00
. (Y,

%D (A

LA bo

r. 8 p -

LA

tJ

lII ; 311

.0

&
0*

366

Appendix 21: Modal split

f optimistic scenario for GO PFT. Modal share and total Tonnage share of three forecast volume of oil in GOPFT modes
products in GopFwr2

GOPFr (1) 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 31426.26 32988.20 34882.27 37142.84 39816.88 42966.22 46670.82 51032.99 56183.06 62286.89 69555.70 78259.21

Roadand rail GOPFr (2=1-5) 27040.06 28536.10 30337.67 32479.84 35009.58 37987.92 41493.62 45627.09 50516.16 56323.89 63258.20 71585.29-

Rail (3) 10-9% xI

Road (4=2-3)

2225.46 24814.60 3595.71 24940.39 3802.17 26535.50 4048.57 28431.27 4340.04 30669.54 4683.32 33304.60 5087.12 36406.50 5562.60 40064.49 6123.95 44392.21 6789.27 49534.62 7158157 55676.63 2 3530.2563055.04

Pipeline Pipelines modal Volume in (5) (7) share GOPFF' (6) 4386.2 60.16 7290.9 4452.1 60.52 7356.4 4544.6 60.88 7464.8 4663.0 61.24 7614.3 4807.3 61.60 7804.1 61.96 8034.7 4978.3 62.32 8307.4 5177.2 62.68 8624.6 5405.9 8989.3 63.04 5666.9 63.40 5963.0 9405.3 6297.5 63.76 9876.9 64.12 10408.5 6673.9

Note: 1: (59.8+0.36t) % of 7 where t is year,59.8% shareof pipeline mode in total oil productsfrom ports. 2: drops from 24.1 of baseyearsto 13.3in 2005 as the average growth of 15 yearsduring 1979-1993 (24.1 -13.3/12--0.9 eachyear backward).

Appendix 22: M dal split of pessimistic for G OPFT. Tonnage share of three modes in GOPFT GOPFr (1) Rail (2)
Road (3=1-2+4)

Total forecast volume of oil products in GOPFT Volume Share oil of (6) in products GoPFr 1 (7) 7290.89 7150.28 6937.59 6627.70 6238.81 5791.57 5307.61 4808.00 4311.92 3835.53 3391.21J 2987.221 23.2 22.3 21.4 20.5 19.6 18.7 17.8 16.9 16 15. 1 14.2 13.3

10.9%xI

Pipelines Pipeline (4) modal sharein GOPFr (59.8-0.8 t) (5)

1994 31426.26 3425.46 23699.17 4301.63 1995 32064.05 3494.98 24407.61 4161.47 1996 32418.63 3533.63 24902.82 3982.18 1997 32330.22 3523.99 25054.95 3751.28 1998 31830.68 3469.54 24879.88 3481.26 1999 30970.99 3375.84 24409.78 3185.37 2000 29818.01 3250.16 23691.12 2876.72 2001 28449.68 3101.02 22781.19 2567.47 2002 26949.49 2937.50 21743.93 2268.07 2003 1 25400.86 2768.691 20645.37 1986.81 . 1 2004 23881.77 2603.11 19549.14 1729.52 2005 1 22460.27 2448.17 18512.51 1499.58

59 58.2 57.4 56.6 55.8 55 54.2 53.4 52.6 51.8 1 51 50.2

INote:1: drops from 24.1 of baseyearsto 13.3as the averagegrowth of 15 years with a rate of 0.9 annually (24.1 -13.3 /12--0.9eachyear backward).

367

p '4

Wi

'

E et- E E

F -

E
0. 0. 92.

0> 0 92
E r

>

- n-8-cO> "' ',


N-

9'EI .
eD

C>

ut ti

tA) 0-

:-0%

" 0

!, :4 00 w a bo i43 p. f44 4) 0% oo -J (71 0 W


_]
4 (A (A 0"

hW LA 90 :
th
4 Ch

N
( (A %0

toi %0

N-

bw

*4

hD u
'4
*-

4 CD
tA g\

:4 u

b :> ;.- (11 (A --3


(A
w a w
0%

%D o

b 9 vi
%0

:> tA

8 '

'2, ',,

t: -

E. I

C> %A
c2%

00 ta

-4 ch 2

00

t)

:4

CD

0, %

00

00

bo o
c w ti %0 C !2 l oo Z. Co g (D (: tA uj -a X. A %0

1h 2 "1 j 0 ", jI
00 b -

(A

-4

(A .

w %0 00 p

P. tA o (D

&. ul

00 00

%0

LA w w b

th CD IC 10

%0 (:

WX (D U ?0 b-4 c>

tA

1
tA

v' ti , >-

:_ Lei C> -a

Zm A.

w t)

l w

zi ki

;- bl zi 0 CD k4
t #i Ul

bo W

t^ b% CD c>
V

-4

Ei e

tA

.a

c, (
(DN

.c,

t t. !0

!i 0 c p '. g za 2 c 'D 0%
ui

b-

00

b (A

CY% 110 "l lz> (A w 00 F%

CD

>o

0- (A .

ti 00

0 Nu

00

00 ,

CD

:4
CD

o be %D

0-3 " vi %0 w

-4 00 t 03

s b.- v %0 w

%0

-x

t4 00 (A U -t-

1 -j 1 (D

Li

wl W

wg Z p 1 "' t.. 5: j h- 0 0 W u G W t u 3 u %0 a zu

Q wCD

bou 1 t-1 1 4 .
C>

:-i b

368

; 77

1-3

b-i 0

8
.

S " 9- 9. 1 .

9 0

-0
0 C) 1,

00

-1

%0 -3

00 0

00

po
0

u3 o
> CD e

-1

0% "

' Li

' ; - wc71 w

K. A w

tA CD

tA

w Co

CD (A

c,

(A

;
ti

-4

0%
CD% c71 0

U CD

-2

bi

CD

;.. Q

b LA h" C)

CD

1 CD 13 C 8 !, :2
0 0

t.
s b %o 22

tA tA

Ch w

U) bi 0

0. %Q c>

0. c 0 v

b-%0 v

00 0

CD CD

t4

';

..
%0 w (D %0 ut v 00 %0 bi tit M C w

%0 %D 0 CD

tJ tn W

ta
0%

%0 t4
0

CM
ch

00
W

-4
u3

c c

ui

CD

o w

t LA

c>

X. A
8 00

L4 C, --, !-

FK c2

t-J a

c% -i %0 bi

" W ON O % 00 ul > ; %0

:t tit b

.i . %0 2 . th 2 0 c5 w

%D W 00 ch

t13

t2 => Q
IA %0 C: )

ti

tA

LA

2 -i

A.

l= l>

Ul
t^ ti

w - ti

a (A c%

(A

00
4.

j tA 00

-4 ON 0%

%0 th

%0 tA

a w

tA

Ch w

t, 3

Ul

: -a

00 1 -4

LA

c
-4

o
s

toi

%0

%0

9 1

L4 %0

;- ch t^ >.

(A La
1

g3 >, s i' 's, w-

%0 %0

Cz121 2 giII I , 1

1 ' 2 t2 Ei -c> 1
00

t^

s w1

pt

cm Co 2 kt
p

a c21

C, 0

'A ' i' l55 , . iA


CD

369

I 2'

-I z .. 00 El F;

a.
92

r6

0.

.0

-0

(D

EI

E'r
NI

Z Z;

0 9 Z8 .9

%D

tA

(:

CD 00 CYN4 Vei A. w

c
: Ult

(43 o b%

oo

Z
LA) IC

23

% (A

CD 00

bj

0%

00

*. 00 CY%

bm Q ,

c> 00

-1 ch
C>

tA 9

t ' %

5 w

00 ;-

ui w LA

%Q

2 12 ;--2 1 17, --,


0%
Z *. u3 ta p.

(:

:> *D C
43 > 00

;- bo t-J -j 00
%0 %0 (D (: %0 a -3

CD 1.. N, %0
i w c2 oo !

, 5?8 - oo c v %0 p a bo bo
%A w
A. w LA %0 tA -b-

:r : z4 b-, 0
Z 2

%D C) IC, 2h,

oo

00 % m
Ch w

%0 %0

lil f t

00 LA

t9

4) %0 Z

t^ 00 CY%%0

(:

.. A. %D

. 0-

-3

CY%

w Lix iJ 4''

e (A w oo "

%ACAi w

0-thiA- Z, % tu
008 KA

ie

0.03

(D G

lo

0,1 -31
t 2 v
9

CD W w4) %0 -4 w . ! x , -# 00 5e %0 w 4) c2% %0 tli 100 Co 0% L3 -4 1 1

C)

tA

-2

11 ,

u 23 rj t^ c311N12 e , 1

im
-

fb
0-%

- -a s A.

9 4 . %0 im i
p

t-A -i

a %0 _] CD CD C> OlDO w 00 2
9
, l

(A

1 4. 0 1 -

f.

%5 9 ,3 .

t 8 v ),00

toi0-0
9

00 le .

LA u

0%

oo

%0 ti

CD Co CD

00 th

ti

w -j (A c71 (D

-3

91
-i

"" CK 00 u3

(7, -4

43 -A- bj

C, 0
CD

*-

-pl

- 4 CD .

212!5 eLt M0 c2. . 0 'U C ,


% 0 00

%Q

370

Pd
[
0 -o -0

Z cr

0 .,

ME .
4A

0 13

0 ", 1 [ u' , - :
Z:

OV

bo 0 1

> po*

&j

1,0

Ch %D %0

Co

-3 A-

tA

ll
(>I C:)i 0

ki

-2

oo

-4 tA

Z.

ei
tA

'Av

t) ,m

-3 0, M 0- 00

Z wIll,
e -3
LA
t. A ON * tA 0,

3
C: ) LA

CD
-o' -4

CD
c>

0 % (A

(A :4

w ZP-

8 :>

%C cy% !s YD 2
J

A"
CD

U 0 -2
ti
t^ CD

:K CD

2 CD

--, w CD

Y, 00 Ch

-N CYN 00

-A-

-2

tA

ta

CD

00 c> w

tj (ji

>-

%0

(D

>

-3

>

Z. 2
N 9 -i t^ A

2 c7,

00

p ke c> (7, b b a bo 2 --(: 4 1;

tA

u3

t^

o :4

a
bi

)..

Od

C%

9
CD W b% 0

00
tj bo CY,

CY,

: ' ( A g 00 ;0

! -

b.
,0 c 2, b%

4 i ;0 E c.

iha
-i ti

t^ ; -

:_

00 CYN $-bo

po

%0

tA CD
t^

, cz

i w .
' %2 %

(A

>

w-

U)
9 .

m oo
bi ei

00
ti

CD
w

00
bi

GQ

tA W

C>

. CD

%D CY%

tA %0

%D (A

u3 00

(A

ki c>

w -

th

u3

oo
u 00

C) %0 -4 -3 to w w CD

bi 2

%0

-4

CD

00

2 j2 2 2g, tk 00

LA w

bIll oo

% tA

t-i

6-

C, 0 P. 09

eg

t p

1 1

14. 1

c2%

2 R 2 t DN i , tj
w (:
00

%D %0 Q

371

200 j,,

Cltv O-S 0 tvti 00 1. 0 0 * E -19L 0-3V 0 00 : O o . Ln 0-3 E M E a 50- E-O E 1 H 0*9


4 -4v
0 (D co eo

- V :4:3E. C :5 V CD
0

01 eD m

;:

M 2: 9
M'

I F- L Ek

. 0.

'E

2 0 -.

eD

ro

CD

91.

M.

0 0 0

Xx n 9 0 0
:r Ch

"a *a 0 0 0 0

K n n ne.0 9 0
a

.0

r
0

K- k. ff. g. K.

t4
-9Ch w tA

2 00
.1

p"

oo (A P% ON LA No LA 00 w

W W P. 4 0
-1 k4 -Ph w ()% p

I., -4 Ch
'A 0!, j A

*6 A) as -

W W

", 0-6

N 1

00(7,
P, P, " .1 CD 00

t'

L" i 4 W a, w 4 t-4 j .I o

q'I- l 1 --I'= Pi'I I I E r-, 5 ; . -- Q -4 0 4 421,

%0 fA

tA -4

tA

LA

"w 0 LA -4 b 0 1 :' :P,- 00 tl ; -4 t" .-

01% - 0

rr

w
01,

La

K 00 :4 G = %0 fA

00
W

I l$ t (71
LA C) I.A

00

0pl $h 91 : Lft -3 o 00 ta 6) L4 t; 10. bo bo j lo0.4 cr, 00 00 Q 00 00 w .9 OA %D CD tj ut CD %A

o w

:,

tA -$ p

Iw l (',,

.. 4

c.
tA .

1 w
V

ZZ

" w

t"

19

%0

' r
tA
-

tA
LA

1 5 Go

0 00 & =
0 w ka

CA

"0
1. -

9
Li La 00

%0 0%

; w

b %0 h Cy% P-. 4) LA

ch

%0 00

00 LIft w

00
C) P %C P%

ch Lft -pi
w

%0 "o

00

0%

tA w
W 0

'.

Lft

Co

00
.6 W tft LA 42% %D

5 '. . .
CP%tA -1

bo ;o
fA

f. ^

11,
4
tlj

bo

ww
CD

Ul L^ Z Go CA

w &I *A

-3 (A
tj

tA w

Q 00 t)

;o

Ch

Ch
La

fA

1 2 t a ;o 0
%0 LA %0 PO (A 0 P"

t"
j

P CD Go %0 tA

g
t

s ", o" so

!3
IA

' ;:
t)

9 w
w 00 . 9 00

ca

ON
w

%0 40
ILA

r.
i13

31 3

rl

90

va

' 'S. . 5 v . '5

it

;o

N
bo ca

00

i4

w
oN

%60,

14 00

Ch

00 Ol

00 t'j Wi

IA ' 00 to) La b% %A " 4 6) 0 S !A

CJ

372-

i Q 0

Er

0 tl 00

c)0 u
eD

c)

PO E?.

om

ti

1-3 CG
92. -

e.

Co) -1 0 F CD-

gl

--i

in. eD

5 9 ci .
93.

= Cb f

(b C%eb

,0

li 92. 1

2- 1 - 92P. 11 K Z

92.

1.5. 5 .
ti P-

0 0

1 R. .

2 c>
tr

0 0 0 0 2: e 0 e Z K z x

0 0

0 0 0 . . . 0 n

tr C, 9 a

rr

1: 1 F j:

th -4 t, 3 10 fli
l 0 b go

(A

b--

.1 CD
Q

-i
i (D

tA w
43 00

Z:: ' 'j' c%


LA -

P. w l

t^

ui w CD k4

c 0

0% CD 43 (X CD w

Q -

'-p- 11 --, bi (7' c c


tA w ca

2 4 (71 t^ 0% (Abi P 0 tA 1 1 1 -i w :. 1 9 i t

!i

%D -2

23

g
w 0

_) ti

9
-1
4

u l(Y, 44
;c *. 0

CY,
G, -

CD

ZZ cy, '. - -1

tA (D

w -4 CD 0-1 " CD CY, b-

t', I 1 '

z2 " t^ CD w CD "

c 0% 0 w CD "

i, ) b 2 ;. %0 s
%0 bi , ,

121
u %C
o

CD

toc t 0-- r,

L-1

CO) -,

bi

00

2
C>

1
2 ti

1 ,:: " l l 1,i i> 1 1i


4 b%0

Ch

g>

(A %D w
t3 ti

?0 00

-4

CD

%0 CD

CD

5D _A 50 Z C 0
IA

;0 %0 t^

!J3 00
-

00

CD

ij

Lm

P !13 La
1 --a

21 21 1
lp CY,
%0 l f. A

Ch

CD 3 (A

z c '-,

(A

_] (^

b-A > 0%

Z -a CY%

t^

cm% -

LA

!1

A i

' 00

*.

CD c

Ei 0%D

CD w oo b%

g V, s Ch t. A CD't 00A. M tj i
"I

w
tA

D C

cl,

1 1 e111 w ll leS -1 CD 1, Isi w A CD0% 81 No


"0 2 "0 ' ' ' 1' 1 ' i w

00

0%

LA
.4

(A

bi

7. 1 0 1 .

W zi 1 ,0 th t4 4. th Co ta Q ti to k3 `0 _A 4A

tj,

%0-1 ui YD c:, i2 i -,
w t

! Na , 00 -

CD

:4

Ch

%0

10

CD

6-

6.6 te

%0

b c> cm Vo

8
(A

E -

3e

00

t, 3

2
;0

-a t7, A p Cy% C: ) E LA ei

%0
Ch 00 c A 'CD 00 % %^ AN

NO ] 00 ll ;jt %A b.

u %0
c%
t^

e
1V
.

a :2

E
-l' w e

l'n
ci, b, vi

c>

(>O '

p 5

1
' N)

1
vi 1 ,A 1

--

b%

LA

a ,

e 22

373

;FFF v
-

!2 .-

0 ;;. 0

:1 00 m
1

I.
CZ

C.
9 -98

;;.

-0 -0
00
9 ,

1-3 n4xC, 0-.2 MCL 1


0

C,
1

FF
cr.
9-

5, -0

-1

01, 1160

ff1

14
W

:3E,
, 1? 9s

0.

..

:1

lz n

21 -0 P-

u
C) 00

P6
o t -, -3

t4 W
p
U) -4 N 4: > 00 O PO p 1 0-1 00 0 0 -

00

*A

00

. 0%
ch

w %J 00 U.) (A 0% . . f.A 0 ON C) 1

00

P. 90 .-

zA 0 00 -4 -10,

10 I'D

01% 17, 1 0 btjo

ul
! ,

Ut 0 ILA p 0

0Uj Uj P. 9% 00 1t)

3 ILA 00 Q W

p
I.-

t,.) -

14: 1 il) t; ! bo p, Ili 4 L4 bo C"71 0

-4.

00

foloE 0

% C, 0 1 1
1 -4

E
00 C)s LA i = -3 . o o LA , =
=

I-A 00 0%

WN)"
0OA 00

U3

cr,

bo :9- 00 kj :-j tA o b 7%il 00 *m %0 o

C) t) 00 00 1%0

C% w

ON -

W w

I-

tj 00

tA

L13 -4

No

.t

00 o La i13 ') (7 0

%o 00

fAL4 tj
w
No

U)

%0

tA

a, t OA w

Eft 41 00 j m C> 00

00 (it

41 "1 ..
to
-A %C
Od

ON

%0 ON CIN 00 W 00 4

qK 00 c7N

(7%

th

.P.

O. cy, ul -A (7% 'o


. 3 ig

o ,

%0 C7 .

00

= 0--2 0 Q

00 !" I
0%

60 o 00 = : t) L ) Ut I.110

-o tj D %. -P.

p.. p.,

(=

-4

%C W

w (>O

0tA -2 00

190 ,

Ln , 1, w w 00 , L't . ') i7 ;- b LI) C) . %0 U)

C) %. 00 -

00

C, t

%Q 14 C) %0 C) ,1 - bp,00 o

L
%Q

QN ,

! 00 0%

t1i
Lft

tA Li
C)

alk

%0

P.

tA LA - .3 Ch 0

0CA !3

!! C) tj bo tot00 EA ,a
C) %0 C)% CD tA

00 %D C)%00 A 01% 4 tj
tA

N
bo o

U.) =
-

(z

%0 tj

(A

tj 4) ij LA t A C) tA C)

IL*, 0.
C)

L4 (= 00 %0 U3 C> I t, ) tj

"

CA

1 10 U3 EA LA 00 00 (7*, ch

1 w 1 LA

v M W t, ) -

w aN ' 4
L.

*4 9
Cad

C)

1..& W OA pb
:

%0 b. ) w 0 ta %0 (A 4 9 % C7,

w 00 LA tA

401%tA

tA CA uj C7 ps -.-

Ch

.4

chw

w 00 -4

00

I O-A t =

t"

0 0 CNI C74w : -j

%Q 0

-j
ON
N)

14! (7, U, -j 'D 00 0 t,


4 w v w tj po W 00 00 w Q I.CD M 00 .

U)

%0

LA co

U)

No

00
bo 00 00

00 t9 5 0
00
4

%0 0

00

0-4 w %D
0" %0

=
e CD Olt 7, j En

%0
tj tA t-)

NO - -4 I I 00 1 LA bo G

.4 t4
O-A

I-a
L..

tA

7% "

OA

tA ij '-A 00

ul

00 CY, 00

, :; "

,"

R,

"
C=)

tit

C) 00

t4 -4 co -: -

L" 0

1,

th

p t, i tj, 0 V 9 .C) 0', p (,0 5 i -3 11 1 % -3 .o


Q
"

tj w

4 R
", it

1.0

ul

"' 0 A qO . . ! U3
L. ) -,

"0

-P. bo

0 o

C) L4

C7% N

tg

C,

rs

374

CO) 0

Z > n to > = = 9 cr 0 t4 rr

tz

N CD

0 Zm
,

=.

rA

E r"

: 7,

= >p EL
N
-

rr m I 6.

> g4 cr

0 00 VV
m

>.

o pv to

A)

00

*0 cn 0

-Pb L4

%D

p.

LA -4

1-1

o %, tl> 11-

i-

ol, -3
tv

! -

W -4

11- ON LA 00

(D

U. C) I'D )
-

"o f".
W -,

tv
w

4F
R,;
a

14A 0

:t- Lo CD -;

p. 00 ON 4. 0 0 . 1 1

ff

-, 0 ,

U,

91. P : 00 00 -4

!J

A 0 0

1,

(7N w
C)

C> C)

C) -3 1%0 1-3 tA C 00

'j w (A

00 1-00

-j 00 w

L. P.(A W

c)

2 CA t: , , ID -. Ps , ,I 1,11

I "Ir

cr

00

0. -

0. -

"

t1a

A.- -till. t1)

I-A 0

E J. -00
,--. --1 - --

ZW w " 0%D 0t) 10-j tj -,

Ill .0" (D O-A -,

,0

!-11 :Q -4 o C5 D 91,j 01 w ILA) 0 o LI) 00 00 tA -4


-Pl.

.P. i-

i-A %0 C) W

0
LA CD o
0

'I . , , I ,1

bo il) 00 0 0 Q

(v ; , %0 11 vI -" 1 1111
:

g,, n

(t,

--

O-h
k4 O-A

tv L.

2-. -

.-

1 1; -C 1 t
C, LA

8 j

tj w

oo 00 W I

-P, 00

.,

" !:, ,

lw

Ch

tz

o1z 4--d
w

p p p P P rW W
LA 14. t'.) o

ON

a t4 ---. -

-, --,-

4I
V)

w 0 w C a tA p

4 ..

;; -l t4

L#3

tj 00

tj C)

C71o 00 C) 0 !A ON 01% 1 00
ll-w 0

t1) -3

(7% w

P.. %o Q

-j LA

0.0 1 -1

--

W '0

1
W po

00 00
Q

'l

lly:

00 8 h 17, "1
w %0 Oo

.- 1 " , .

,I

375

""' , 1*1

List of the Iranian institutions used in this project:


Institution
StatisticalCentreof the Islamic Republic of Iran CentralBank of Iran

Description
Research provision of dataat national, and provincial and local levels. Affairs relatedto the nationalcurrencyand foreign revenues, tradeand national accountsof the country.

Persian naine
MarkazcAmare Jomhooriych IslamiyehIran Bank Markaziych Jomhooriyeh Islamiyeh Iran

Ministry of Roadand Transport

Administration for the management, and maintenance constructionof the transport infrastructureof the country in particular for rail, nationaland provincial and inter-provincial roads,ports, civil aviation and national airline.

VezarateRah va Tarabari JomhooriychIslamiyeh Iran

Ministry of Constructive Jehad Ministry of Oil

for Responsible rural roadsand developments and fishing. Management exploitation,exploration and, and refining, nationaland internationalmarketingof oil fields, terminalsand regionsof the country.

Vezarate JahadSazandagi Jomhooriyeh Islamiyeh Iran VezarateNaft Jomhooriych IslamiyehIran

Ministry of Interior

In chargeof all municipalitiesand governorships administrationat provincial and local levels of the urbanand inter-urbanareas. for Responsible borderialroadsand connections and CoastGuardand road traffic systemof the country.

VezarateKeshvar JomhooriyehIslamiyeh Iran

Ministry of Defence

In chargeof the threemilitary forcesand defenceof the country including military air passenger cargo transport. and

VezarateDcfa Jomhooriyeh Islamiych Iran

Ministry of I; d-ustry and Mining

for Responsible the developmentand management the industrial and mining sector of of the country and roadsin the mining areas. In chargeof the foreign trade issuesand the nationalshippingcompanyof the country. Responsible the collection of the national for revenues,expenditure,programmingand Budgetingand Customsof the country. In chargeof the Justiceand legal issuesand private companies In chargeof the management, operation,

Vezarate Sanayeva MA A Den JomhooriychIslamiyeh Iran

Ministry of Commerce Ministry of Financeand Economic

VezarateBazargani JomhooriyehIslamiyeh Iran VezarateDara Yee and Egh To SaadJomhooriych IslamiyehIran VezarateDad gostarl JornhoodychIslamiyeh Iran Sazemane Banaderva Kashti

Ministry of Justice Portsand Shipping

376
1
.0

Organisation

constructionof the ports and maritime administrationof the country.

RaniJomhooriyeh Islamiyeh Iran Sazcmane Namchva Bar BoodjchJomhooriyeh IslamiychIran

Planningand Budgeting Organisation

In chargeof the nationalplanningand development programmes the country and of also StatisticalCentreof Iran.

Islamic Republic of the Iran Railway Company Islamic Republicof the Iran ShippingLine Company National Iranian Tanker Company

In chargeof the rail networksand rail passenger RahAhanchJomhooriych and freight operationsandtraffic. In chargeof the nationalmerchantshipping fleet and most part of the foreign tradeof the country. In chargeof the national tankerfleet and most part of the crudeoil transporttrade of the country. I ShcrkateKashtiRani Mcli Naft KesheJornhoodych IslamiyehIran IslamiyehIran Kashti RaniJomhooriych IslamiyehIran

377

Anda mungkin juga menyukai