Mapping Local Production Systems in The UK

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN.
EL CASO BRITNICO
Lisa de Propis *
RESUMEN
Este ensayo presenta una reflexin sobre la presencia y el papel de los distritos industriales en Gran Bretaa. Cien aos despus de los escritos de Marshall, los distritos industriales han sido redescubiertos por Becattini (1987), y se han convertido en el objeto de una dramtica revuelta en el estudio sobre las dinmicas de los lugares, de las empresas y de las comunidades. Los ros de contribuciones que se han llevado a cabo en el debate anglosajn han participado an ms de la literatura americana y filo-porteriana, que ha introducido el concepto de cluster y lo ha propuesto como el motor vital de la competitividad local. Ha sido el hermanamiento cluster y competitiveness lo que ha llamado la atencin de los policy-makers britnicos en los aos 90, luchando contra una amenazante Globalizacin, un sector manufacturero en total exfoliacin y un sector servicios -sobre todo aqullos de mayor valor aadido- nicamente concentrado en Londres. A diferencia de las lneas que ha seguido el debate britnico, el presente anlisis quiere ser un intento de verificar el fenmeno de los distritos industriales marshallianos en Gran Bretaa, aplicando una ya bien testada metodologa que hace referencia a los trabajos de Sforzi en Italia y Boix en Espaa. Este anlisis para Gran Bretaa permite una comparacin internacional ms amplia sobre los distritos, y esto tiene valor sobre todo en contraposicin a la ms reciente tendencia a explorar el tejido econmico ingls, y verificar con instrumentos ms cualitativos si hay todava realidades territoriales que tienen un espesor socio-econmico.
ABSTRACT
The paper presents some considerations on the role and extent of industrial districts in Great Britain. One hundred years after the seminal work by Marshall on industrial districts and localised industries, these were reawakened and brought to renewed attention by Becattini (1987) in his work on industrial districts in Tuscany. Such contribution paved the way for a new approach to analyse the dynamics, functioning and trends of places, firms and socio-economic communities. The breath of the academic literature that followed has had resounding effects also in the Anglo-Saxon debate, which, however, has tended to be pegged to Porter's concept of clusters (1990) and to consider clusters as factors of local competitiveness. The twinning between clusters and competitiveness has crucially caught the attention of British policy-makers, especially since the 1990s. Increasingly aware of the challenges and threats of production globalisation, manufacturing decline especially across the English regions and Wales, and the rise of the high value added service sectors mostly in London, British policy-makers have turned to clusters as possible objects and vehicles policy actions. Unlike the current line of investigation in the UK, this paper presents the findings of an analysis of the industrial district phenomenon across England, Scotland and Wales, drawing on similar studies in Italy by Sforzi and in Spain by Boix. This paper contributes to a possible international comparison of the phenomenon; whilst at the same time providing a picture of local production system across regions.
Artculo publicado en el nm. 13 de la Coleccin Mediterrneo Econmico: "Los distritos industriales" Coordinador: Vicent Soler - ISBN: 978-84-95531-40-7 - Depsito Legal: AL-728-2008 Edita: CAJAMAR Caja Rural, Sociedad Cooperativa de Crdito - Producido por: Fundacin Cajamar
203
1. Introduccin
Este ensayo presenta una reflexin sobre la presencia y el papel de los distritos industriales en Gran Bretaa. Como sabemos, el concepto de distrito industrial fue introducido por Marshall para describir las industrias localizadas de los Midlands, la cuna de la Revolucin Industrial. Cien aos despus de los escritos de Marshall, los distritos industriales fueron redescubiertos por Becattini (1987), y se han convertido en el objeto de una dramtica revuelta en el estudio sobre las dinmicas de los lugares, de las empresas y de las comunidades. Los ros de contribuciones que les han seguido han contribuido conceptualmente a expandir y detallar tal modelo, mientras que los estudios casusticos han establecido que tal forma de sistema local est presente en todo el mundo, incluso aunque no sean siempre llamados como tales.
* Universidad de Birmingham.
LOS DISTRITOS INDUSTRIALES
De hecho, y sobre todo en los pases anglosajones, a menudo se encuentran realidades similares o diversas a los distritos industriales, pero en todos los casos muy groseramente reagrupadas bajo el trmino porteriano de clusters. Frente al debate acadmico entre naciones que ha desmenuzado las varias formas de sistema local y ha analizado cada matiz: -vanse los distritos industriales (Pyke et alii, 1990; Becattini et alii, 2003); los medios innovadores (Camagni, 1995) y los sistemas regionales de innovacin (Cooke, 2001; Braczyk et alii 1998), corrindose tambin el riesgo de crear caticas repeticiones y confusiones (Martn y Sunley, 2003)-; los anlisis que han resultado en Gran Bretaa han participado ms de la literatura americana y filo-porteriana, que ha introducido el concepto de cluster y lo ha propuesto como el motor vital de la competitividad local. Fue el hermanamiento clusters y competitiveness lo que atrajo la atencin de los policy-makers britnicos en los aos 90, luchando contra una amenazante Globalizacin, un sector manufacturero en total exfoliacin y un sector servicios -sobre todo aqullos de mayor valor aadido- nicamente concentrado en Londres. El debate sobre los clusters en el Reino Unido se ha desarrollado con estudios casusticos y poco con contribuciones sustancialmente conceptuales. Los policy-makers ingleses han absorbido la idea de que los sectores estn geogrficamente concentrados y que esto tiene razones, potencialidades y a veces lmites, slo desde 2001. Siguiendo un documento publicado por el Department of Trade and Industry, donde se hace una cartografa de los clusters en Gran Bretaa (Inglaterra, Gales y Escocia), las Agencias de Desarrollo Regional (Regional Development Agencies) empezaron a ver sus economas regionales con distintos ojos; esto es, como cogulos de sectores en lugares particulares, y a focalizar las acciones de poltica sobre stos. No hay duda de que para los observadores expertos el descubrimiento en Gran Bretaa del papel de los sistemas locales y de su uso como objeto de intervencin para acciones para el desarrollo regional, se ha llevado a cabo sin una verdadera y profunda comprensin del fenmeno, y con la idea de que fuese la ltima tendencia en los trminos de policy-making, similar a un tren en movimiento al que debemos saltar o dejar atrs. A diferencia de las lneas que ha seguido el debate britnico, el presente anlisis quiere ser un intento de verificar el fenmeno de los distritos industriales marshallianos en Gran Bretaa, y de encuadrar el fenmeno en una discusin crtica de su papel en las polticas regionales.
204
2. El debate actual
Las razones de este relativismo son dos: por una parte, en los pases anglosajones la literatura sobre el desarrollo local y sobre los sistemas locales ha sido, por as decirlo, porterizada, por lo que, a partir de una moda, el uso del trmino cluster se ha multiplicado, dejando a interlocutores o lectores a menudo no muy seguros de lo que verdaderamente se est hablando.
LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS
La porterizacin del debate sobre sistemas locales ha sido tambin avalada por la desconexin entre sociedad y territorio que caracteriza la realidad econmica britnica. El concepto de cluster propuesto por Porter no parte del territorio, de las localizaciones de tradiciones productivas, sino de la estructura y organizacin de la produccin; esto es cierto, en lnea con la literatura americana de la flexible specialisation, que siempre ha considerado los sistemas locales como el resultado de la desmembracin de la gran empresa en el ocaso del sistema fordista. En otras palabras, los sistemas locales siempre han sido considerados sobre la base de su funcionalidad productiva. En Gran Bretaa, la erradicacin de las actividades productivas del territorio puede explicarse por varios factores: en primer lugar, el hecho de que Inglaterra se haya industrializado mucho antes que el resto de Europa hizo que empezase a terciarizarse tambin antes; en segundo lugar, el capitalismo ingls es quiz ms parecido al americano que al europeo, empresas de dimensiones ms grandes, un desarrollado sistema de venture capital y una economa muy abierta a las inversiones extranjeras. Precisamente este ltimo fenmeno, con muchas de las mayores empresas inglesas ahora de propiedad extranjera, ha acelerado el susodicho proceso de desconexin, por lo que decisiones que impactan sobre una localidad vienen tomadas en head offices a miles de kilmetros de distancia (Bailey y Driffield, 2007). Este fenmeno se ha llamado wimbledonizacin, en referencia al hecho de que Wimbledon, uno de los eventos tensticos ms importantes de la temporada, lleva desde hace aos sin campeones ingleses. Traducido en trminos econmicos, esto quiere decir que la presencia de propiedad extranjera en empresas y sectores punteros en Gran Bretaa, ha producido una separacin entre territorio, sociedad y economa que se ha reflejado en el debate sobre el desarrollo local y los sistemas locales. La segunda razn es que el debate britnico sobre los sistemas locales est fragmentado y desunido, en cuanto que las varias disciplinas que podran concurrir a desarrollar el concepto de manera multiforme no se hablan, siendo stas, por ejemplo, la Economa, la Geografa y la Sociologa. La fragmentacin disciplinar y la desconexin territorio/economa lleva a considerar el distrito industrial marshalliano (Becattini, 1987, 1994, 2000 y 2001) como una realidad muy compleja y casi un ideal-tipo al cual aspirar, sin ser alcanzado jams.
205
3. Los distritos industriales en la poca de Marshall

No creo que sea posible escribir un ensayo sobre los distritos industriales en Gran Bretaa sin comenzar por Marshall, y en cierto modo por la Revolucin Industrial en los Midlands. El propio Marshall, en Industry and Trade y en los Principles, hace referencia explcitamente a algunas realidades concretas, como las industrias metalrgicas en Staffordshire,
Shropshire y, ms vagamente, en Gales y Escocia; el potteries en Staffordshire; chair-making en Buckinghamshire; el straw plaiting en Bedfordshire; la cuchillera en Sheffield; los cotton mills de Lancashire; y el textil de los Midlands. No era el inters de Marshall hacer un inventario de todas las industrias localizadas. De hecho, otros interesantes distritos emergen deshojando los trabajos de historia econmica. Como se ha dicho anteriormente, la investigacin de los historiadores econmicos ingleses no ha sabido apreciar el modelo distritual y crear un filn de estudios sobre esto; por tanto, cuando he intentado desentraar el fenmeno a caballo del novecientos, me he encontrado juntando numerosos estudios que describen varias industrias y localidades sin un acople conceptual a la manera de las industrias localizadas de Marshall. Esta estimulante caza del tesoro ha sacado a la luz una multitud de distritos en el ochocientos y a caballo del novecientos, exactamente en los tiempos de Marshall. stos incluyen, por ejemplo: La industria de los guantes en Worchester y Taunton (Coopey, 2003). La industria del ribbon-making machinery en Coventry Popp (2003) sostiene que despus se ha convertido en la industria de las bicicletas, primero, y de la mecnica de automviles despus; y el textil en Manchester (Lancashire) (Wilson y Singleton, 2003). La elaboracin de la lana en Est Anglia, del lino en Norkfolk y de la seda en Essex; la camisera en Leicester; la elaboracin del metal (metal bashing) en los Midlands y en Yorkshire; la cuchillera en Sheffield y en Yorkshire (Hudson, 2004). La elaboracin de la lana en Yorkshire; la produccin de toys and button en Birmingham (Berg, 1994). El iron district en Dudley; las destileras de ginebra en Londres y de spirits en Escocia; llaves y candados en Wolverhampton; la guarnicionera en Walsall1; la produccin de encajes (bone-pillow lace industry) en Buckinghamshire; la produccin de botones para camisas en Shaftesbury y Blandford (Clapham, 1930). La produccin de zapatos y botas para hombre en Northampton, y de zapatos de mujer en Norwich y Leicester; el mecano-textil en Lancashire; la produccin de bicicletas en Coventry y Birmingham (Aldcroft, 1968). El jewellery quarter en Birmingham (Wise, 1950; De Propris y Lazzeretti, 2007). El gun quarter en Birmingham (Wise, 1950).
206
El distrito de la guarnicionera de Walsall est en rpido crecimiento por el boom de la equitacin a la inglesa, sobre todo en los Estados Unidos.
Mapa 1. Algunos distritos industriales en Inglaterra y Gales (1892)
207
4. Sistemas productivos locales y distritos en Gran Bretaa

En Gran Bretaa ha habido pocos estudios que hayan dado lugar a un mapa de los clusters: un informe del Ministerio de Comercio e Industria (Department of Trade and Industry, 2001), titulado Business clusters in the UK, presentaba una cartografa general de los clusters en Gran Bretaa para el sector manufacturero y terciario; un ensayo de Crouch y Farell (2001), donde se usan los cocientes de localizacin para diferenciar cogulos ocupacionales; y, finalmente, mi contribucin (De Propris, 2005), donde aplico los cuatro criterios de Sforzi (1990) para localizar varias formas de sistema local y los distritos industriales en particular. El informe del DTI (2001) ha provisto de algn modo una primera descripcin, si bien muy aproximativa, del fenmeno de los clusters en las distintas regiones britnicas. La contribucin ms importante de este informe ha sido la de dar a conocer el fenmeno de los sistemas locales a los policy-makers, a los que las resonancias del debate acadmico no les haban llegado. Para cada regin el informe seala una serie de sectores que se configuran como aglomerados, usando un cociene de localizacin que de hecho asevera la presencia de concentraciones sectoriales sobre la base de densidad de ocupaciones a nivel regional (con un LQ inferior a 1). En particular, los high points de esta polvareda de clusters son aqullos cuyo LQ es mayor que 1,25, y la ocupacin del sector corresponde al 2% de las ocupaciones regionales. A esta debilidad metodolgica, se une el hecho de haber usado informaciones cualitativas y ad hoc para clasificar tales clusters segn el nivel de desarrollo (embrionarios, en crecimiento o maduros); la profundidad, entendida como extensin y complejidad de las relaciones entre empresas (profundos, superficiales o desconocida); el crecimiento (en crecimiento, estables o en declive), y la relevancia a nivel regional, nacional o internacional. En De Propris (2005) se hace un esfuerzo por proceder a un anlisis-diagnstico ms completo de los sistemas locales, proponiendo una metodologa que combina un anlisis espacial a nivel regional o nacional (sobre la base de Sforzi, 1990 y Brusco y Paba, 1997), con estudios casusticos cualitativos sobre la organizacin de la produccin y sobre el aumento de importancia de las instituciones.
208
4.1. Metodologa para el anlisis espacial

Para disear un mapa de sistemas locales hay que considerar cuatro criterios: a) intensidad manufacturera; b) dimensin de las empresas; c) especializacin industrial; y d) conjuntamente la especializacin industrial y la dimensin de las empresas (Sforzi, 1990). La aplicacin de estos criterios requiere una apropiada definicin ya sea de la clasificacin sectorial, ya de la unidad geogrfica de referencia. Sforzi (1990) considera sectores de dos cifras y sistemas locales del trabajo. Los sistemas locales de produccin que surgen de la aplicacin de estos criterios pueden ser mltiples, incluidos los distritos industriales.
Mapa 2. Mapa de los clusters en el Reino Unido elaborado por el DTI (2001)
209
En Gran Bretaa, la Oficina Nacional de Estadstica (Office of National Statistics) facilita datos sobre la ocupacin clasificados por sectores (dos, tres o cuatro cifras) y por sistemas locales de trabajo (297 travel-to-work-area2). Esto ha hecho posible aplicar los cuatro criterios y disear un mapa de los sistemas locales en general, y de los distritos industriales en particular en Gran Bretaa.
A. Intensidad manufacturera La intensidad manufacturera mide las economas de aglomeracin, en tanto que asume que la proximidad de mltiples sectores manufactureros genera externalidades positivas para las empresas en trminos de intercambio intersectorial, transferencia de conocimiento y tecnologa. Desgraciadamente, el sector manufacturero en Gran Bretaa se est retirando de manera visible, aunque no siempre para dar paso a un emergente terciario; no obstante, los sistemas locales de trabajo de alta intensidad manufacturera son ahora ms numerosos que los no manufactureros: 165 sobre un total de 297.
E (sll , manuf .) E (nat., manuf .) E (nat.) E (sll )
(1)
Notas del traductor: Las abreviaturas utilizadas son las originales en italiano.
210 B. Dimensin de las empresas El segundo criterio sirve para aseverar la composicin de la poblacin de las empresas en un cierto sistema local; en otras palabras, est caracterizado por pequeas, medianas o grandes empresas. Las clases de dimensin pueden variar segn la disponibilidad de los datos estadsticos. En Gran Bretaa, la ONS clasifica las empresas con menos de 99 trabajadores (que podremos llamar pequeas empresas) en 152; empresas con 100-299 trabajadores (medianas empresas), 186; y empresas con ms de 300 trabajadores (grandes empresas), 113.
E (sll , manuf ., occupati ) E (nat., manuf ., occupati ) E (sll , manuf .) E (nat., manuf .)
(2)
En De Propris (2005) se usa la definicin de sistema local del trabajo de 1998, sobre la base del censo de poblacin de 1991.
C. Especializacin industrial El tercer criterio sirve para identificar la especializacin sectorial del sistema local. Esto mide las economas de localizacin del sistema, esto es, las externalidades que surgen de la acumulacin de conocimientos y competencias especficas por un cierto sector en un cierto lugar. Las economas de localizacin hacen a un lugar distinto de otro en cuanto contenedor de conocimientos no transferibles sino radicados. El anlisis del ndice de especializaciones industriales puede suministrar importantes informaciones no slo sobre el sector dominante de un sistema local, sino tambin indicar sectores secundarios que, sin embargo, forman parte de la misma filire de produccin. En el caso britnico hemos aplicado el ndice de especializacin industrial al sector de dos cifras por un total de 23 sectores.
E (sll , settore ) E (nat., settore)
E (sll , settore)
E (nat., manuf .)
(3)
D. Especializacin industrial y dimensin de las empresas En definitiva, si combinamos el ndice de especializacin industrial con la dimensin de las empresas, estamos en condiciones de verificar si un cierto sistema local del trabajo especializado en un sector particular tiene un sistema local de pequeas, medianas o grandes empresas. Este ndice es muy importante porque, a partir de los datos sobre la ocupacin, nos permite individualizar aglomeraciones de empresas y no de trabajadores. Adems, estudios sobre la governance (De Propris, 2001; Markusen 1997) han sugerido que la dimensin de las empresas influye sobre la capacidad de decisin y de negociacin de las mismas, lo cual a su vez define la governance del sistema local.
211
E (sll , settore, occupati ) E (nat ., settore , occupati ) E (sll , settore ) E (nat ., settore )
(4)
4.2. Resultados
En resumen, la aplicacin de la metodologa para el anlisis espacial de los sistemas productivos locales en Gran Bretaa se ha llevado a cabo con datos sobre la ocupacin de 1997, estratificados por 23 sectores manufactureros3 y por 297 sistemas locales del trabajo, desde Andover (SO) a Wick (Scotland). La dimensin de las empresas haban permitido distinguir entre pequeas (menos de 99 trabajadores), medianas (entre 100 y 299 trabajadores) y grandes (ms de 300 trabajadores). A diferencia de Sforzi (1990), primero se ha distinguido y luego clasificado los sistemas locales sobre la base de tres coordenadas: intensidad manufacturera (manufacturero o no manufacturero); especializacin sectorial (sector primario o sector secundario); y, por ltimo, la dimensin de la empresa (pyme o gran empresa). Dada la avanzada terciarizacin del sistema econmico ingls, tras la progresiva contraccin del sector manufacturero que de 1984 al 2004 ha perdido ms del 30% de la ocupacin, al inicio de 2000 ste supone slo el 15% de la misma. Se decidi entonces considerar tambin sistemas en contextos de baja intensidad manufacturera. En segundo lugar, se decidi tener en cuenta tambin sistemas locales cuya especializacin sectorial en pequeas y medianas empresas, o bien grandes empresas, correspondiese a sectores secundarios y no primarios, es decir, con un LQ segundo en el ranking. Finalmente, se han distinguido sectores caracterizados por pequeas y medianas empresas de aqullos dominados por grandes empresas. Haciendo esto se han obtenido ocho categoras de sistemas locales: (1) proto-distrito4; (2) proto-distrito no especializado (porque el sector en el que se encuentra una aglomeracin de pequeas y medianas empresas no es aqul con el LQ ms alto); (3) sistema local no manufacturero / especializado de pequeas y medianas empresas; (4) sistema local no manufacturero / no especializado de pequeas y medianas empresas; (5) sistema local manufacturero / especializado de grandes empresas; (6) sistema local manufacturero / no especializado de grandes empresas; (7) sistema local no manufacturero / especializado de grandes empresas; y (8) sistema local no manufacturero / no especializado de grandes empresas (ver Tablas 1 y 2). Respecto al informe del DTI (2002), este ejercicio ha permitido detectar de manera exacta y rigurosa diversos tipos de sistemas locales (vale la pena considerar que esta clasificacin inicial se basa puramente en datos estadsticos agregados; si se efectuase tambin un extenso anlisis cualitativo surgira una ms detallada heterogeneidad); y entre stos, formas distrituales.
212
Los 23 sectores manufactureros son: alimentos y bebidas; metales bsicos; productos del tabaco; productos fabricados de metal; textiles; maquinaria y equipo; vestir ropa/pieles; equipos de oficina y ordenadores; cuero; maquinaria y aparatos elctricos; madera y corcho; radio, TV, equipo de comunicaciones; pasta papelera, papel y productos de papel; instrumental mdico de precisin; publicaciones, impresin, soportes grficos; vehculos de motor, remolques; coque, productos de petrleo refinado; otros equipos de transporte; productos qumicos; mobiliario; de caucho y plstico; mercancas; reciclaje; otros productos no metlicos. Brusco y Paba (1997).
De hecho, la aplicacin de los cuatro criterios ha sealado 47 proto-distritos, contradiciendo la famosa afirmacin de Zeitlin (1995) de que no habra distritos industriales en el Reino Unido. Ms genricamente, se han encontrado 165 sistemas locales en reas de alta intensidad manufacturera y 132 en reas de baja intensidad manufacturera; 105 sistemas locales de grandes empresas y 192 de pequea y mediana empresa. Los sistemas locales de grandes empresas se concentran mayoritariamente en los Midlands, el Noroeste, el Sur de Gales y las Tierras Altas del Sur de Ecocia. Mientras los sistemas locales de pequeas y medianas empresas estn espolvoreados por todas partes. En particular, se detectan ms distritos industriales en los Midlands, el Este de Gales y el Norte (ver Mapas 3 y 4). En particular, a los sistemas locales especializados correspondera el 21% de la ocupacin manufacturera; pero si se consideran tambin las especializaciones secundarias (que a menudo forman parte de una filire productiva o bien son el resultado de economas de urbanizacin), entonces la relevancia de los sistemas locales alcanza cuotas del 50% (vase el caso de Coventry). Tabla 1. Tipos de sistemas locales (1997)
Proto-distrito Proto-distrito no especializado SL de PMI no manuf. y especializado SL de PMI no manuf. y no especializado SL de LI manuf. y SL de LI manuf. y especializado no especializado SL de LI no manuf. y especializado SL de LI no manuf. y no especializado Otros
1) Intensidad manufacturera 2) empleados <100 empleados <300 empleados >300
Sector primario Sector secundario
3) Especializacin sectorial Sector primario 4) Espec.+ PI Espec.+ PMI Sector primario
Sector secundario
Sector primario
213
Sector secundario Sector secundario Sector secundario
Sector secundario
Sector primario
Sector secundario Sector secundario 5 Sector primario 28 Sector secundario 5 6
Espec.+LI Sector primario No SLT 47 31 80 22 73
Fuente: ONS. Elaboracin propia.
Tabla 2. Tipos de sistemas locales (1997)

LS manufactureros SL especializados SL no especializados
SL no manufactureros SL de PMI SL de PMI
PROTODISTRITOS PROTO-DISTRITOS NO-ESPECIALIZADOS
Mapa 3. Mapa de los sistemas locales de gran empresa en Gran Bretaa (1997)
214
Fuente: ONS (2003). Elaboracin propia.
Mapa 4. Mapa de los sistemas locales de pequea y mediana empresa en Gran Bretaa (1997)
215
Fuente: ONS (2003). Elaboracin propia.
5. Los distritos industriales

En el anlisis presentado en De Propris (2005), se haban localizado 47 proto-distritos industriales. stos estaban distribuidos por el territorio britnico, concentrndose sobre todo en el ms manufacturero Centro y Norte, si bien dejando an fuera centros importantes como Birmingham, donde se ha registrado un alto impacto de grandes empresas; Manchester, donde se ha destacado un sistema local pero de un sector secundario; o finalmente Liverpool, donde el sector manufacturero ha alcanzado dimensiones modestas. Los sectores en los que se han encontrado formas distrituales van desde aqullos ms tradicionales, como alimentos y bebidas en Escocia (Whiskey Valley), o el textil en Leicester, hasta aqullos con un contenido ms alto de tecnologa, como el sector mecnico, la composicin de automviles, el plstico y la goma (ver Tabla 3). En 1998 hubo una reclasificacin de los confines de los sistemas locales de trabajo, y se necesit entonces verificar el impacto del fenmeno con las nuevas fronteras. Siempre considerando 297 sistemas locales de trabajo y 23 sectores manufactureros, recientemente se ha actualizado este trabajo con datos de 2002. Se ha reconsiderado el fenmeno de los distritos industriales en Gran Bretaa, siempre aplicando la metodologa de Sforzi (1990), pero distinguiendo los distritos de pequeas empresas de los distritos de mediana empresa (ver Tabla 4). Uno de los primeros resultados de relieve y que, reteniendo la informacin sobre los sistemas de pequea empresa (sin amalgamar juntos sistemas de pequea y mediana empresa, como en De Propris, 2005), surgen realidades interesantes donde existen proto-distritos industriales slo de pequea empresa. En algunos casos, hay proto-distritos de pequea y mediana empresa y en otros casos slo proto-distritos de mediana empresa. Conjuntamente se han localizado 77 distritos industriales que cubren un poco todas las regiones, desde el sector textil al mdico. En particular, el estudio ha individualizado 18 distritos de pequea empresa, 9 de pequea y mediana empresa y 54 distritos de mediana empresa. Algunos de estos distritos son histricos distritos marshallianos, como el metalrgico en Dudley/Sandwell y Worchester en los Black Countries; el de Sheffield (cuchillera) y el textil (ropa, lencera, peletera) de Leicester. El famoso distrito cermico de las potteries de Stoke-onTrent ha sido detectado en nuestros estudios como un sistema local, pero de grandes empresas y, por lo tanto, no ya como un distrito industrial. En general, por sectores como el metalrgico, mecnico, elctrico y de los transportes, que estn presentes en los Midlands, es fcil ver la relacin indirecta con las industrias localizadas a las que Marshall haca referencia en sus obras. Otros sectores son tpicamente tradicionales, como el alimentario, de la elaboracin de la madera y el textil. Finalmente, otros distritos estn asociados a sectores modernos, como el de la goma y el plstico, el mdico, el de los ordenadores/office machinery, el de las comunicaciones y el de las ciencias ambientales (reciclaje).
216
Tabla 3. Los distritos industriales en Gran Bretaa (1997)

SPL Campbeltown Fishguard y St Davids Knighton y Radnor Haltwhistle Kelso y Jedburgh Poole Welshpool Dudley and Sandwell Gainsborough Matlock Diss Fakenham Kings Lynn Workington Banff Girvan Keith y Buckie Peterhead Camelford Devizes Launceston Shaftesbury Wadebridge y Bodmin Llangefni y Amlwch Goole y Selby Retford Andover Bedford Worcester Calderdale Rhymney y Abergavenny Falmouth Harlow Horncastle Malvern Leominster Wellingsborough Leicester Nottingham Blackburn Rochdale East Ayrshire Galashiels y Peebles Leek Huddersfield Keighley y Skipton Huntly REGIN Escocia Gales Gales Noreste Escocia Suroeste Gales Medio Oeste Medio Este Medio Este Este Este Este Noreste Escocia Escocia Escocia Escocia Suroeste Suroeste Suroeste Suroeste Suroeste Gales Yorkshire y Humber Medio Este Sureste Sureste Medio Oeste Yorkshire y Humber Gales Suroeste Sureste Medio Este Suroeste Medio Oeste Sureste Medio Este Medio Este Noroeste Noroeste Escocia Escocia Medio Este Yorkshire y Humber Yorkshire y Humber Escocia SECTOR Ropa Ropa Metales bsicos Qumica Metal Metal Metal Metal Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Alimentos y bebidas Maquinaria y equipo Maquinaria y equipo Maquinaria y equipo Maquinaria y equipo Maquinaria y equipo Vehculos a motor Otros transportes Publicidad y edicin Caucho y plstico Caucho y plstico Caucho y plstico Curtidos / cuero Textiles Textiles Textiles Textiles Textiles Textiles Textiles Textiles Textiles Madera
217
Tabla 4. Distritos industriales en Gran Bretaa (2002)

Sistemas productivos locales (clasificacin de 1998) ESTE ESTE ESTE ESTE ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE MEDIO ESTE NORESTE NORESTE NORESTE NORESTE NORESTE NOROESTE NOROESTE NOROESTE NOROESTE ESCOCIA ESCOCIA ESCOCIA ESCOCIA ESCOCIA ESCOCIA ESCOCIA ESCOCIA ESCOCIA ESCOCIA ESCOCIA ESCOCIA SURESTE SURESTE SURESTE SUROESTE SUROESTE SUROESTE SUROESTE SUROESTE SUROESTE SUROESTE SUROESTE SUROESTE Kings Lynn Peterborough Wisbech Huntingdon Mildenhall Gainsborough Leicester Retford Chesterfield Matlock Stamford Worksop Louth Horncastle Boston Skegness y Mablethorpe Kettering y Corby Bishop Auckland Sunderland y Durham Haltwhistle Berwick-upon-Tweed Hartlepool Wigan y St Helens Blackburn Nelson y Colne Rochdale Keith y Buckie Newton Stewart North Ayrshire Peterhead Dingwall Huntly Fraserburgh Girvan Kelso y Jedburgh East Ayrshire Forfar Hawick Bedford Southend Wellingborough Evesham Holsworthy Okehampton Camelford Launceston Stroud Falmouth Gloucester Poole Otros equipos de transporte Reciclaje Instrumental mdico de precisin Alimentos y bebidas Metales bsicos Alimentos y bebidas Instrumental mdico de precisin Otros equipos de transporte Otros equipos de transporte Otros equipos de transporte Maquinaria y equipamiento Coque y petrleo Curtido de pieles y vestido Alimentos y bebidas Alimentos y bebidas Maquinaria de oficina y ordenadores Alimentos y bebidas Metal Metal Alimentos y bebidas Papel Radio, TV, comunicaciones Textiles Textiles Textiles Madera Prendas de vestir y pieles Reciclaje Metales bsicos Metales bsicos Productos no metlicos Productos no metlicos Papel Prendas de vestir y pieles Madera Madera Curtido de pieles y vestido Maquinaria y equipo Vehculos a motor Caucho y plstico Madera Madera Productos no metlicos Textiles Textiles Textiles Distritos industriales de pequea empresa (0-99) Distritos industriales de mediana empresa (100-299) Alimentos y bebidas Maquinaria y equipo Papel Caucho y plstico Madera
218
Continuacin Tabla 4. Distritos industriales en Gran Bretaa (2002)

Sistemas locales de trabajo (clasificacin de 1998) SUROESTE SUROESTE SUROESTE SUROESTE GALES GALES GALES GALES GALES GALES MEDIO OESTE MEDIO OESTE MEDIO OESTE MEDIO OESTE MEDIO OESTE MEDIO OESTE MEDIO OESTE MEDIO OESTE MEDIO OESTE Y&H Y&H Y&H Y&H Y&H Y&H Y&H Y&H Y&H Bridgwater Malvern Chard Wells Welshpool Knighton y Radnor Rhymney y Abergavenny Pontypridd y Aberdare Merthyr Newtown Leominster Ludlow Wolverhampton y Walsall Dudley y Sandwell Worcester Stafford Oswestry Kidderminster Leek Calderdale Keighley y Skipton Sheffield y Rotherham Scarborough Bridlington y Driffield Barnsley Pickering Wakefield Huddersfield Otros equipos de transporte Alimentos y bebidas Textiles Textiles Metales bsicos Maquinaria elctrica Alimentos y bebidas Productos no metlicos Otros equipos de transporte Prendas de vestir y pieles Textiles Madera Caucho y plstico Curtido de pieles y vestido Metales bsicos Metales bsicos Metales bsicos Maquinaria elctrica Madera Textiles Textiles Curtido de pieles y vestido Madera Metales bsicos Maquinaria elctrica Mobiliario Maquinaria y equipo Textiles Caucho y plstico Distritos industriales de pequea empresa (0-99) Distritos industriales de mediana empresa (100-299) Caucho y plstico Caucho y plstico Curtido de pieles y vestido Curtido de pieles y vestido
219
6. Conclusiones
Las reflexiones presentadas en este ensayo constituyen un primer intento de describir y analizar el fenmeno de los sistemas locales, y en particular de los distritos industriales en Gran Bretaa. Los resultados de la cartografa aportan importantes indicaciones a tres niveles: (1) muestran la relevancia espacial de los sistemas locales; (2) ayudan a individualizar tipologas de sistemas locales; y finalmente, (3) suministran una exacta descripcin espacial de los distritos industriales (proto-distritos). No hay duda de que esto es slo el primer paso hacia un anlisis ms detallado de las especficas realidades territoriales de los distritos industriales, para el que es necesario un diverso y ms complejo set de datos e indicadores. Tal anlisis tiene un gran valor, dada la relevancia que los sistemas productivos locales han asumido en el debate poltico en el Reino Unido; de hecho, el Gobierno central ha delegado a las agencias locales de desarrollo (regional developemt agencies) la tarea de identificar prioridades econmicas sobre las que focalizar fondos y energa. En este sentido, al viejo enfoque de policy que vea los sectores como objetivo de acciones y decisiones, a partir de 2001 tales agencias han revisado sus estrategias de poltica, sealando priority clusters como pilares del desarrollo econmico regional. La individualizacin de tales sistemas locales es, por lo tanto, no slo relevante, sino sobre todo necesaria.
220
7. Bibliografia
ALDCROFT, D. H. (1968): The Development of British Industry and Foreign Competition. Londres, George Allen and Unwin. BAILEY, D. y DRIFFIELD, N. (2007): Industrial Policy, FDI and Employment: Still a Missing Link; en Journal of Industry Competition and Trade (7); pp. 189-211. BECATTINI, G. (1987) : Mercato e Forze Locali: Il Distretto Industriale. Bolonia, Il Mulino. BECATTINI, G. (1994): The Development of Light industry in Tuscany: An Interpretation; en LEONARDI, R. y NANETTI, R. Y., eds.: Regional Development in a Modern European Economy. Londres, Pinter; pp. 69-85. BECATTINI, G. (2000): Il Bruco E La Farfalla: Prato Nel Mondo Che Cambia. Florencia, Le Monnier. BECATTINI, G. (2001): Il Caledoscopio dello sviluppo locale: trasformazioni economiche nellItalia che Cambia. Turn, Rosemberg & Sellier.
BECATTINI, G.; BELLANDI, M.; DEI OTTATI, G. y SFORZI, F. (2003): From Industrial Districts to Local Development. Cheltenham, Edward Elgar. BERG, M. (1994): Factories, workshops and industrial organisation; en FLOUD, R. y McCloskey, D., eds.: The Economic History of Britain since 1700. Cambridge, Cambridge University Press; vol. 1. BRACZYK, H.; COOKE, P. y HEIDENREICH, M., eds. (1998): Regional Innovation Systems. Londres, UCL Press. BRUSCO, S. y PABA, S. (1997): Per una storia dei distretti italiani dal secondo dopoguerra agli anni novanta; en BARCA, F., ed.: Storia del capitalismo italiano. Roma, Donzelli; pp. 265-333. CAMAGNI, R. P. (1995): The Concept of Innovative Milieu and its Relevance for Public Policies in European Lagging Regions; en Papers in Regional Science (74, 4). CLAPHAM, J. H. (1930): An Economic History of Modern Britain. Cambridge, Cambridge University Press. COOKE, P. (2001): Regional Innovation Systems, Clusters and the Knowledge Economy; en Industrial and Corporate Change (10, 4); pp. 945-974. COOPEY, R. (2003): The British Glove Industry, 1750-1979: The Advanategs and Vulnerability of a Regional Industry; en WILSON, J. F. y POPP, A., eds.: Industrial Clusters and Regional Business Networks in England, 1750-1970. Aldershot, Ashgate. CROUCH, C. y FARREL, H. (2001): Great Britain: Falling through the Holes in the Network Concept; en CROUCH, C.; LE GALS, P.; TRIGILIA, C. y VOELZKOW, H., eds.: Local Production Systems in Europe. Rise or Demise? Oxford, Oxford University Press. DEPARTAMENT OF INDUSTRY AND TRADE (2001): Business Clusters in the UK. A First Assessment. Londres. DE PROPIS, L. (2001): Systemic Flexibility, Production Fragmentation and Local Industrial System Governance; en European Planning Studies (9, 6); pp. 739-753. DE PROPIS, L. (2005): Mapping Local Production Systems in the UK: Methodology and Application; (39, 2). Enseguida DE PROPIS, L. y LAZZERETTI, L. (2007): The Birmingham Jewellery Quarter: A Marshallian Industrial District; en European Planning Studies (15, 10); pp. 1.295-1.325.
221
HUDSON, P. (2004): Industrial Organisation and Structure; en FLOUD, R. y JOHNSON, P., eds.: The Cambridge Economic History of Modern Britain 1700-1860. Cambridge, Cambridge University Press; vol. 1. MARKUSEN, A. (1996): Sticky Places in Slippery Space: A Typology of Industrial Districts; en Econ. Geogr. (72, 3); pp. 293-313.
MARSHALL, A. (1920): Priciples of Economics. Londres, MacMillan (8 edicin). MARTIN, R. y SUNLEY, P. (2003): Deconstructing clusters: chaotic concept or policy panacea?; en Journal of Economic Geography (3); pp. 5-35. PORTER, M. E. (1998): Clusters and the new economics of competition; en Harvard Business Review; pp. 77-90. PYKE, F.; BECATTINI, G. y SENGENBERGER, W. (1990): Industrial districts and inter-firm cooperation in Italy. Gnova, ILLS. SFORZI, F. (1989): The Geography of Industrial Districts in Italy; en GOODMAN, E. y BAMFORD, J., eds.: Small Firms and Industrial Districts in Italy. Londres, Routledge; pp.153173. SFORZI, F. (1990): The Quantitative Importance of Marshallian Industrial Districts in the Italian Economy; en PYKE, F.; BECATTINI, G. y SENGENBERGER, W., eds.: Industrial Districts and Inter-firm Cooperation in Italy. Gnova, IILO; pp. 75-107. SFORZI, F. (1996): Italy: Local Production Systems of Small and Medium-sized Firms and Industrial Changes. OCDE, Networks of Enterprises and Local Development. SFROZI, F. (1999): Economic Change; en BONAVERO, P.; DEMATTEIS, G. y SFORZI, F., eds.: The Italian Urban System. Aldershot, Ashgate. WILSON, J. F. y SINGLETON, J. (2003): The Manchester Industrial District, 1750-1939: Clustering, Networking and Performance; en WILSON, J. F. y POPP, A., eds.: Industrial Clusters and Regional Business Networks in England, 1750-1970. Aldershot, Ashgate. WISE, M. J. (1950): On the Evolution of the Jewellery and Gun Quarters in Birmingham; en Institute of British Geographers (15). ZEITLIN, J. (1995): Why are there no Industrial Districts in the United Kingdom?; en BAGNASCO, A. y SABEL, C. F., eds.: Small and Medium Enterprises. Derby, Pinter.
222
udies. Vol. 39.2. pp. l')7-211. Apnl 2t)()5
O Routledge
Mapping Local Production Systems in the UK: Methodology and Application

LISA DE PROPRIS
Institute for Industrial Deuelopmetit Policy, Birmingham Business School, University of Bimiitigham, Edgbaston, Birmingham B15 2'rr, UK. Email: l.de_propris@bham.ac.uk
(Received June 20{)2: in revised form December 2003) Dr PnoPHis, L, (2005) Mapping local production systems in the UK: mcthodolot!;y and application. Regional Studies 39. 197211. The paper outhnes a possible methodology to map and study local production systems. Tht- three-level diagnostic methodology enables researchers to map, classify and analyse in depth firms' agglomerations in regions or countries where there is little intorniation about the presence and location oflocal production systems. The spatial diagnostic procedure is applied to ilie UK to ni.ip local production systems. Location Industrial districts Agglomerations
is, L. (2(105) L'elaboration des systemes de production iocaux au Royaume-Uni; une methodologie et son application. Regional Studies 39, 197-211. Cet article cherche a esquisser une methodologie eventuelle qui sert a elaborer et a etudier les systemes de production Iocaux. Cette methodologie diagnostique a trois niveaux permet aux chercheurs d'elabiirer, de classer et d'approfondir les agglomerations des entreprises dans les regions ou les pays ou il y a peu de renseignements a propos ou de la presence, ou de la localisation, des systemes de production Iocaux. La demarche diagnostique geographique se voit appliquer au Royaume-Uni afin d'elaborer les systemes de production Iocaux. Localisation Districts industriels Agglomerations
D E PROPRIS, L. (2005) Kartierurig iirtlichcr Produktionssyteme im Vereinigten Konigreich: Methodik und Anwcndnng, Regiotiiil Studies 39, 197-21 1. Dieser Autsatz unireilJt eine in Betracht zu ziehende Methodik der Kartierung und Untersuchung ortlicher Produktionssysteme. Die diagnostische Methodik auf drei Ebenen gestattet Forschern, Firnienballungen in Regionen oder Landern. wo es an Information liber Vorhandensein und Standort ortlicher Produktioiissysteme nuiigclt. zu kartieren, zu klassifizieren und eingehend zu analysicren. Das raumlich-diagnostische Vertahren wird zur Kartierung ortlicher Produktionssysteme angewandt. Standort Industriegebiet Dallungen
Dr. FROPKIS, L. (2005) El mapeo de los sistemas de produccion locales en el Reino Unido: planteaniiento metodologico y aplicacion, Regional Studies 39, 1972] ], El articuio resume un posible planteaniiento metodologico para niapear y estudiar los sistemas de produccion locales. La metodologia de diagnostico de tres niveles propuesta permite a los mvestigadores que la utilicen mapear, clasificar y analizar en profundidad aglonieraciones de cmpresas en aquellas regiones o paises donde existe escasa intormacion sobre la presencia y la localizacion de sistemas locales do produccion. El procedimiento de diagnostico espacial se aplica al Rcino Unido para mapear sistemas locales de produccion. Localizaci6n Distritos industriales Aglomeraciones
JEL classification: R12
INTRODUCTION The flourishing debate on local production systems (LPSs). e.g. industrial districts, clusters and networks, has mostly involved in-depth analyses of existing phenoniena in Europe and the USA in the attempt to understand their nature and dynamics. These studies aimed at identifying case-specific elements, as well as
drawing from them generalizable observations. One conimon element across all these studies was their confirmation that LPSs have been crucially important for industrial development, regional development and employment. Over the last decade, studies on industrial districts in Italy (LEONABDI and N A N E T T I . 1994; COSSENTINO et ai, 1996; BECATTINI, 2001), on clusters in Portugal (PORTER, 1998), the USA
DOI: t0.1U80/U03434O()5200059983
01)34-3404 pnnt/13WI-()5<91 online/05/020147-15 2005 Regional Studies Association li ttp://www.rfgioiial-studies-assoc.ac.uk
198
(PORTER,
Lisa De Propris (1998) distinguish between evolutionary, reversible and steady-state networks. The set of tools outlined in this paper is a methodology that allows one to carry out the following diagnostic study of production systems: (1) to localize agglomerations of small, medium and large firms in an area producing in the same sector; (2) to map such production systems across a territory; (3) to provide a taxonomy' of LPSs according to basic categories (e.g. sector, firm size) in order to identity patterns of regional specialization; (4) to understand the structure of interfirm relationships; and (5) to assess the role of the institutional framework. The diagnostic methodology has three levels of analysis: spatial diagnostic, diagnostic analysis of interfirm relationships and analysis of the institutional framework. It relies both on statistical and primary data collected via questionnaires. Qualitative and quantitative tools tend to be complementary, allowing for a rounded research approach that moves from a general mapping of LPSs to the in-depth study of each individual case or a selection of them. The mix of data and case studies suggested in this methodology is coherent with the types of information it brings together. One is not distinguishing between measurable and nonmeasurable variables, since case studies can also produce quantifiable information sometimes as dummy variables (e.g. belonging to a civic club or association) or discrete variables (e.g. the output share of a firm's largest buyer). Rather, what matters is the depth of analysis. The general description of the phenomenon (levels 1 and 2) must rely on genenilized indexes and formulas that require data. However, LPSs are characterized by more than employment concentration and inter-sectoral input-output flows: there are intangible factors and system-specific elements that can be only analysed properly with a case study. To some extent, the databased approach enables general conclusions to be drawn on the phenomenon, whereas the case study approach provides precise information that can be used to show the variants and shades in which the phenomenon presents itself in reality. The possibility of identifying LPSs since their origin and studying them throughout their development would enable research to move away from a typical expost static approach (e.g. PORTER'S (1998) work on the wine valley in California) that explores successful systems and tries to explain them by looking back at the causes of their success. Rather, researchers could adopt a more dynamic approach where the patterns of production systems' evolution can be evaluated as they unfold. Note too that the analysis of 'success stories' only provides a partial description of the more general phenomenon of local industrial systems because it neglects to study 'failure stories', i.e. production systems that, although displaying industrial specialization and geographical proximity, have not emerged or continued successfully but rather have lingered before dechning.
ARCANGELI
2000) and Norway (BJOKG and ISAKSEN, 1997), and innovative milieu in France ( L O N G H I , 1999) and developing countries (RABELLOTTI, 1997; GuERRiERi ct al. 2001) have all strengthened the argument that agglomerations of small- and mediumsized enterprises (i.e. firms) (SMEs) can catalyse regional industrial competitiveness. Most LPSs, however, tend to be studied once they become so successful as to catch the attention of geographers or economists. At that point, they are analysed to trace their evolution, to understand their functioning and the reasons for their success and, finally, to puU out elements ot replicability. In other words, researchers' analysis of production systems tends to start only after the phenomenon has become visible to the eye. One shortcoming of this 'approach" is that research tends to omit two important categories of LPSs: emerging production systems in industrialized countries and those in embryonic form in developing countries. In fact, in industrialized countries, there are sectors, subsectors or even product niches that tend to grow quickly and show a localized concentration. If, as happens in most cases, localization contributes to agglomeration and extemal economies, which in turn boost competitiveness, then such phenomena could be singled out before they become common knowledge. The thrust of this paper is, therefore, to present a set of tools to map and diagnose LPSs in a region or country where there is httle or no knowledge about the existence and location of such systems. LPSs are referred to generally as the geographical agglomeration of firms specialized in one or a few complementary sectors. Such producdon systems are characterized by an external division of labour, a more or less developed social capital and a more or less engaged institutional framework. This definition was kept as general as possible since any further categorization would imply the need for more detailed and accurate information about individual clusters. In the current literature many typologies of LPSs have been proposed to the point
where, as argued by MARTIN and SUNLEY (2003), the
concept of a cluster has been given so many labels and has been defined in so many ways that it has become a chaotic and ambiguous one. After the first conceptualization ofindustrial districts (BECATTINI, 1987, 2000, 2001; PYKE et ai, 1990), further definitions include hub-and-spoke districts (MARKUSEN, 1996), clusters (PORTER, 1998, 2000; C O O K E , 2002) and, if the emphasis is on inter-firm innovation and technology processes, innovative milieu ( C A M A G N I , 1995). Similarly, G O R D O N and M C C A N N (2000) distinguish between three forms of clustering; the classic model of pure agglomeration, the industrial complex model and the social network model. Meanwhile, GARCJFOLI'S (1991) typology of LPSs includes areas of specialized production, local productive systems and system areas. SIMMIE and SENNETT (1999) add cities to MARKUSEN'S (1996) typology; while BELUSSI and
Mappinji Local Production Systems in the UK

Yet, LPSs have also proved to be engines ofindustrial development (BELLANDI, 2001) alternative to large firms, and for [his reason they are deserving special attention in developing countries. More recently, BELLANDI and SFORZI (2001) recognize that a multiplicity of hybrid forms of local systems can be found by combining four elements: small firms, large firms, and rural and urban environments. In particular, they identify five possible categories of LFSs: industrial poles, rural local systems, big dynamic cities, buyerdependent local systems and industrial districts. Each form has the potential to trigger and catalyse regional development. The usefulness of this methodology with respect to the clustering phenomenon is related not only to its exploratory value, but also to the fact that it can provide information for the design and definition of objectives for regional policy. Once LPSs have been mapped and analysed, regions have a better understanding ot their strengths and weaknesses. This awareness can assist policy-makers in formulating a set of policy measures to create the conditions for LPSs to grow and consolidate. These measures could include training, business-tobusiness marketing, research and development, exporting, finance and specialized service provision. In the UK, the publication of the Department of Trade and Industry's (UTI) report on clusters in the UK
(DEPARTMENT OF T R A D E AND INDUSTRY, 2001) has
199
data from various databases." In contrast, in the spatial diagnostic procedure, a case is made for using travelto-work areas (TTWAs) data supplied by the UK Office for National Statistics. The reason for using these data is that as they correspond to self-contained working and living areas, they are more appropriate to identify LPSs. As P A N I C C I A (2002, p. 48) notes, TTWAs 'encapsulate the features of a locally bounded community, given that it identifies a restricted area in which interactions between firms and populations are very dense'. Second, the DTI methodology relies on a data-based analysis and qualitative information gathered from local institutions and organizations about the identified clusters. The objective of the data-based analysis is to measure employment density via a location quotient and data about industries' regional employment. The report highlights "high points' as those industries tor which regional employment accounts f~or at least 0.2% of the total regional workforce and for which the location quotient shows that the regional industry is 25% more concentrated than at the national level. For sector /, region r and country n, the local quotient (LQ) is defined as follows:
>1.25
ln other words, the data-based analysis of the DTI only measures industrial concentration as mirrored by employment density. Qualitative analysis is aimed at ranking and classifying the identified clusters under four headings: stage of development (embryonic, established, mature), deptb (deep, shallow, unknown), employment dynamics between 1991 and 1998 (growing, declining, stable) and significance (regionally, nationally, internationally). The spatial diagnostic procedure presented in this paper is more complete in that in addition to measuring industrial concentration (step 3), it considers firms' size and the use of data on units of production by sector and by TTWAs. Both elements provide a much more informative picture of firm clustering as opposed simply to mapping employment density. Furthermore, the analysis of case studies by means of the quantitative input-output analysis and the qualitative analysis of the institutional framework would provide a richer picture of the phenomenon. For instance, in the input-output analysis, one could provide imponant information on the flow of inter-industry exchanges and on inter- and intra-LPS sectoral networks. The paper is structured as follows. The second to fourth sections will discuss the three levels of the diagnostic methodology. The fifth section wall present the results of the application of the spatial methodology to map LPSs in the UK. Finiilly, the sixth section will present some concluding remarks.
been used by regional development agencies across the country to incorporate the targeting and support of clusters in their economic strategy. The author feels this methodology would be even more valuable to inform policy-makers in transition and developing countries where the need to trigger the combined regional and industrial development often lies initially in catalysing the further development of existing competencies and specializations. Little work has been done so far to diagnose LPSs.
MARTIN and SUNLEY (2003) divide the empirical
work done to identify and map clusters in two broad approaches: top-down mapping exercises and bottomup qualitative studies. Both approaches seem to have fundamental shortcomings. Pt)RTER's (199H) mapping of the wine cluster in California and the footwear cluster in Portugal possibly falls under the latter category since the study of these clusters was based on knowledge and information gathered on the ground. In the UK, there have been two main quantitative mapping studies; the DTI report Busimss Clusters in the UK (2001) and work of C R O U C H and FABREH- (2001). The DTI report, in particular, has opened the debate on what are the possible and appropriate methodologies to diagnose LPSs, which can then be applied to map known and unknown realities. The main differences between the D T I methodology and that presented herein are manifold. First, different geographical units of analysis are referred to for statistical data. The DTI uses both regional and local authority
200
SPATIAL DIAGNOSTICS
Lisa De Propris of economic activities. Local labour markets rarely correspond to standard geographical classifications in that they often cut across regions or counties and group together towns and villages that would i>therwise not have anything in common. C R O U C H and FARHELL (2001) use T T W A data for employment and units of production figures to identify clusters in the UK. In particular, they identified four types of firm agglomeration: industrial districts, concentrated clusters, weak clusters and simple clusters. Before going any further, note that there is no reason for restricting analysis only to the manufîcturing sector. However, the author feels that the mapping of service local systems might require a slightly difference set of criteria. For instance, firm size might not be so relevant for the service sector. The location quotient a.s defined in the third criterion could, nevertheless, provide useful information about the geographical concentration of service industries (for tlie mapping of Italian urban systems of services, see SFORZI, 1999).'' For the spatial diagnostic analysis described in this paper, four criteria need to be considered: the share of employment in the manufacturing sector; the size of firms; industrial specialization; and industrial specialization and the size of firms.
Share of employment in the nuiiiufacturing sector
The obvious first level of the diagnostic study of LPSs in a country or region is to pinpoint on a map where they are and to attempt an initial understanding ofwhat tliey are. For this purpose, it is important to have a consistent statistical data set that contains information about firms' production activities according to three coordinates: sector, size and location. A nuip of firms" agglomerations provides the basic information about production systems across a country or region: in some cases, it might confirm the existence of already known phenomena, while in some other cases, it can flag up new infant systems. A methodology already tested and proved successful is that used by SFORZI (1990) to map Italian industrial districts and other types of LPSs. Until then, industrial districts were only perceived as regional experiences and almost unique cases attached to local traditional expertise. The application of this methodology highlighted the presence of LPSs across Italy in the 19H()s and gave an initial picture of the magnitude of the phenomenon." Furthermore, it showed that in addition to industrial districts, there were other types of production systems that played a diflerent, and crucial, role for regional industrial development such as 'nonmanutacturing local production systems' or 'local production systems of large firms' (BHUSCO and PABA, 1997). Sforzi's methodology 'does not identify districts, but local systems whose production structure is compatible with that of districts" (13RUSCO and PABA, 1997, p. 278; author's translation) and this represents its strength, especially when applied with an open perspective with respect to what types of LPSs could be found. In fact, even before the case-based analysis, it enables an initial classification of LPSs according to sector, firm size and urban/non-urban area. The methodology relies on data on employment shares in firms (considered as units of production) according to size, sector and location. Small firms are seen as having fewer than 100 employees, SMEs have fewer than 250 employees and large firms have more than 2S() employees."' Manufacturing sectors are defined according to the national classifications, which can be more or less aggregated. Moving from a two- to a three-digit sector classification might alter the mapping ot local systems, but it tends to be a very interesting exercise if one wants to single out extremely dynamic product niches.^ Finally, location is defined by the statistical geographical units that partition the national or regional territory. These can correspond to regions, counties, provinces, cities or any meaningful combination of them. SFORZI (1990) argued that the most appropriate geographical units to identify LPSs tend to overlap with local labour markets. As self-contained TTWAs, they reflect the overlapping of the community of people living in a certain area and the population
This criterion assesses the proportion of employment in the manufacturing sector (manuf.) in each T T W A out of the total non-agriculture employment (nat.), which is benchmarked against the national employment share of the manufacturing sector out of the national non-agricultural employment: E(TTWA, manuf.) H(nat., manuf~. E(nat.)
(1)
E(TTWA)
If the ratio of manufacturing to total employment in the T T W A is greater than the national average, it means that manufacturing employment dominates the local labour market in question. In contrast, if the ratio is below the national average, it indicates that service sectors are dominant. SFORZI (1990) and 13RUSCO and PABA (1997) stress that one of the key features of industrial districts is that of being embedded in a manutacturing environment. In a further application of Sforzi's methodology, SOLINAS and B A R O N I (1998) distinguish between manufacturing local systems and non-manufacturing local systems to highlight the increasing importance of local systems (not necessarily industrial districts) in de-industrialized or still agricultural areas where manufacturing sectors account for a limited share of the local economy.
Size of firms
The second criterion concerns firms' size and assesses the composition of firms' population in the local labour
Mapping Local Production Systems in the UK market according to whether firms are mostly small, medium or large sized. In particular, it considers the proportion of employment in firms with fewer than 100, fewer than 250 and more than 250 employees out of the total TTWA's manufacturing employment. Again, the benchmark is the share of the national manufacturing employment by firms' size out of total manufacturing employment: H(TTWA, manuf., size) E(TTWA. manuf.) .E(nat., manuf, size) H(nat., manuf) (2) This criterion allows researchers to ascertain the dominance of small, medium or large firms in the local labour market. In the case ofindustrial districts, for instance, SFORZI (1990) argued that firms had to have fewer than 250 employees. In contrast, if medium or large firms dominate, other t>'pes of local systems can be found. Finns" size is an important element when starting analysis of LPSs. It can be argued that inter-firm dynamics and the governance of LPSs cliange greatly according to whether the system is dominated by small or large firms (DE PROPRIS, 2001). Besides, firms' size is important in order to uncover geographical divides in the organization of production activities.
Industrial specialization
201
In other words, the proportion of employment in a sector out oi the total TTWA's manufacturing employment has to be greater than the share of the sector at the national level. Industrial concentration mirrors, therefore, the accumulation of sector employment across all firms' size. In so doing, it can be said that an area is specialized in one sector. As already mentioned, the same area can be specialized in a few sectors; sometimes they tend to be related in a type of sector chain (Jiiicre), or can be completely unrelated. In the latter case, one possible explanation is that the area has been or is still manufacturing-intensive so that a variety of sectors have developed and are still present. Another explanation could be the presence of urban areas that tend to be centres of accumulation of many industrial activities. Sforzi argued that industrial districts tend to be specialized in only one sector.
Industrial specialization and firms' size
The third criterion to assess LPSs is to look at their industrial specialization. In fact, whichever types of agglomeration of firms one aims to find, they tend to be characterized by production specialization in one sector or, in the case of a filicre, in a few related and complementary sectors. Manufacturing sectors can be classified by relying on different levels of aggregation: a two-digit sector classification usually allows researchers to identify the main production sectors (e.g. textiles, chemicals), while a three-digit sector classification allows a breakdown of each sector into sub-sectors. Within the present methodology, the use of either classification has advantages and disadvantages. On the one hand, the use of a sub-sector classification means the possibility of identifying production systems on the basis of a precise producdon specialization such as ceramic tiles (a subsector of the non-metallic mineral product sector) or hosiery (a sub-sector of the textiles sector). However, it means having to handle a large data set given that sub-sector data are in turn broken down by TTWAs. On the other hand, using the two-digit classification makes the handling of the methodology more straightfbi-Vk'ard, although some information might be lost. SFORZI (1990) relied on the two-digit sector classification. Industrial specialization is given by the local quotient: H(TTWA, sector)/E(TTWA, manuf) H(nat., sector)/E(nat., manuf) (3)
Finally, the last criterion brings together information about firms' size and industrial specialization so as to clarify whether the sector(s) in which a certain area is specialized is (are) characterized by small, medium or large firms. This means looking at whether the proportion of employment in sector / by firms' size out of the total TTWA's sector employment is smaller or greater than the share of the national employment by sector and firms' size out of each sector's national employment: (TTWA, sector,size) E(TTWA, sector) '(nat., sector,size) E{nat,, sector) (4)
This last criterion enables one to understand whether a certain agglomeration of firms is formed by small or large firms, or whether some sectors are more likely to be characterized by small or large firms. In industrial districts, industrial specialization must be characterized by SMEs, whilst other types of production systems can be completely different in their make-up. For instance, hub-and-spoke districts are formed by one or a few large firms and many small firms respectively producing in related industries.
Pros aud cons of the spatial diaj^noslic analysis
The spatial diagnostic analysis just described produced the first systematic assessment of Italian industrial districts and identified a wide range of LPSs different from industrial districts. In particular, to identity industnal districts, SFORZI (1990) suggested that local labour markets: (1) had to be dominated by manufacturing sectors; (2) had to be characterized by firms with fewer than 25(1 employees; (3) had to be specialized in one sector; and (4) the specialized sector had to be characterized by firms with fewer than 250 employees. The result was the well-known maps of Italian districts
202
Lisa De Propris
Table 1. Local production systems (LPS) iisin;^ Sforzi's methodology
LPS of specialized small LPS ot medium firms and large firms sectors outside the industrial district Noniract LPS
Pre-district {]) Manufacturing dominance {2a) Dominance of small firms {fewer than 25(1 employees) (2b) Dominance of large firms {more than 250 employees) (3) liidiiscrial specialization (4a) Size and specialization (fewer than I(H) employees) (4b) Size and specialization {fewer than 251) employees) {4c) Size and specializarion (more than 250 empioyecs!
LI'S of large firms
Soimes: BBUSCO and PAHA (1'W7), SOLINAS and B A R O N I (1998), SFOBZI (199U, 1996).
in Sforzi. More precisely, what was found were 'predistricts' (j)roto-disrretti). namely, LPSs with the spatial characteristics of districts (small firms, sector specialization, agglomeration), but that could be labelled as such only after a deeper analysis of the structure of interfirm relationships and of the institutional framework. Overall, its main advantage is to provide an automatic procedure to carry out a spatial diagnostic analysis of LPSs. It relies on statistical data available at the regional or national levels. It must be stressed that the four criteria do not Just enable the identification of only industrial districts, but all sorts of production systems. SFORZI (199(3) also mapped 'northern and central urban systems', 'southern urban systems' and 'northern manufacturing local systems'. In the same way, BRUSCO and PABA (1997) mentioned 'specialized sectors of small firms', 'non-specialized sectors of small firms', 'small- and medium-sized sectors' and 'large firm sectors' (Table 1). The breadth of the classification that emerged from the use of Sforzi's methodology showed that it can be a powerful tool to carry out spatial diagnostics when one keeps an open perspective on what could be found. This is especially true if one begins to pinpoint LPSs on a blank geographical map. The main drawback of the procedure is that it only captures localized industrial specialization according to three dimensions: firms' size, sector and location. It is silent, however, on firm interactions. Only a deeper analysis can help in terms of quahfying the nature and governance of the LPS. This explains why Sforzi's methodology constitutes only a first step. The second step is to analyse inter-firm relationships to understand the nature of the production system. The fifth section presents the findings of an application of the spatial diagnostic procedure to map LPSs in the UK ( D E PROPKIS, 2003).
DIAGNOSTICS OF INTER-FIRM RELATIONSHIPS The second level in the assessment of LPSs is to analyse the structure and nature of inter-firm relationships. Before attempting this, two questions need to be raised: Do firms in LPSs necessarily interact?; and How do they interact? It cannot be taken for granted that in local systems firms engage in some form of relationship, but there is much evidence showing that the localized agglomeration of production activities is tightly related to the presence of firms' interdependencies (traded and untraded). Firms locate close to other firms to interact with them as much as they are encouraged to interact with other firms because of proximity. C O O K E and MORGAN (1998) emphasize the associational nature of localities; PioRE and SABEL (1984) refer to regional conglomerations of firms; M A R K U S E N (1996) accentuates the stickiness of some places; whilst STORPER (1996) looks at the locality as a nexus of untraded interdependencies. It is also well established that proximity and firm cooperation contribute to agglomeration and external economies." There are many ways in which firms can interact. The most common is through the exchange of outputs and inputs along the production chain. LPSs are almost by definition complex networks of buyers and suppliers, together contributing to the production of the final goods. Subcontracting relationships can often be accompanied by other forms of interaction, such as cooperation over innovation,'^ joint ventures, joint purchase of inputs and joint training.'" However, one needs to distinguish between quantitative and qualitative aspects of inter-firm relationships. The existence of production relationships between buyers and suppliers can be detected and measured with
Mapping Local Production Systems in the UK an input-output analysis, whereas the nature of such relationships and the dimensions of inter-firm cooperation can be only detected with case-by-case qualitative studies that rely on questionnaires or interviews. Therefore, the need to handle a multifaceted methodology that involves both quantitative and qualitative approaches is again stressed. Input-output analysis has been traditionally used to measure inter-industry trade at the macro level, interindustry multipliers and inter-industry linkages
{RASMUSSEN, 1956; CHENERY and WATANABE,
203
1958; IsARD el ai, 1998). In reality, it can also be an extremely useful instrument for the study of LPSs. Input-output tables enable researchers to quantify interindustry exchanges and localize the origin and destination of such exchanges (i.e. interregional flows), inputoutput tables quantify the monetary value of sales (or purchases) from (to) sector i to (from) sector j . It is possible to construct a map of the network of interindustry exchanges that mirrors the network of the external division of labour (inter-firm production linkages). This can be a national or regional map depending on the data available froni countries' statistical offices. In the USA, such data are supplied at the state level; in the UK, there are data only at the national level and for Scotland. EUROSTAT provides input and output tables for the 15 European Union Member States. An example of input-output analysis is in FESER and IiER(;MAN (2000), who map the 'vehicles manufacturing cluster' in the USA, namely the complex network of inter-industrial linkages. Such linkages are assumed to be underpinned by inter-firm linkages, so that the result is a map of the buyer-supplier network around the automotive sector. FESER and BERGMAN (2000) map the automotive cluster across the USA providing no information about the geographical dimension of the cluster. However, when regional data are available, it is possible to map an inter-industry network at the local level. Input-output analysis can therefore be a useful tool: (1) to identify the set of industries involved in an LPS; (2) to see the direction of the exchange flows {i.e. who buys from whom); and (3) to evaluate the relative strength of the direct and indirect inter-industry linkages. Fig. 1 shows in stylized form the map that can be derived from an input-output analysis, where the direction of the exchange between buying and supplying industries is described by the arrows between industries (e.g. F buys from Y and sells to C). Some industries, like C, are the hubs of the systems because many inputs are channelled in that direction where they are assembled into a final or semi-final output that in turn is sold to a downstream industry before approaching the fmal market. The thickness of the arrows corresponds to the intensity of inter-industry flows. The main limitations of the input-output analysis
Fig. 1. huer-indusfry linkages
are twofold. One is that it needs a predefined geographical unit of analysis; the other is that there is no information about firm size. In fact, if the input-output analysis is carried out at the national level, it will map the network of inter-industry linkages across the entire country with no information about localization or agglomeration. Because of these limitations, inputoutput analysis is best seen as complementary to Sforzi's spatial diagnostic analysis in that it adds key information about the dynamics of inter-firm linkages between subsectors. A further step to understand the nature of such relationships would involve a qualitative assessment via questionnaires and case studies. Given that a qualitative methodology is very labour intensive and time consuming, it should be adopted only to carry out an in-depth analysis ot one particular LPS, and in any case, it cannot be the dominant methodology for the study of LPSs. A qualitative assessment can clarify in how many types of linkages firms tend to engage in addition to production inputoutput exchanges; u^hether firms share or exchange information and knowledge, cooperate over innovation, training, marketing and exports, or embark on different sorts of joint ventures. Moreover, a qualitative assessment could also capture
204
Lisa De Propris As far as the first two points are concerned, it is well known that the growth of SMEs is often constrained by their limited access to finance and management skills, and that for this reason they might struggle to carry out lu-house innovation projects, training schemes or marketing/export strategies. External support is needed to bypass these problems. Business support organizations need to tailor their support to satisfy precisely those needs offering either individual or, wherever possible, collective help. In LPSs, collective initiatives can work particularly well because they can be activated in an environment where there already exists a high degree of inter-firm networking due to subcontracting linkages. The brokerage role of business support organizations is aimed at encouraging or strengthening linkages between firms and between firms and organizations.'"' This is particularly important where the competitiveness of the LPS is hindered by "rusty' or malfunctioning inter-firm relationships caused by changes in the competitive scenario, sector decline, internal restructuring, the disrupting entry of new competitors, etc. All these factors can temporarily alter the equilibrium of the network and stall the fluidity of the dynamics of interfirm linkages. Tbe system can take time and eflort to adjust to the new context. In the best-case scenario, it will redefine its functioning possibly with a different set of firms. This adjustment process can be eased by an external intervention that can recreate or facilitate the re-organization of the production network. In the worst-case scenario, where the system is incapable of altering its structure to accommodate the new context and where there is no external support to help, the system might never recover from the exogenous shock and decline until it disappears. Needless to say, some LPSs do not have a set of business support organizations on which to rely. The lack of an institutional framework leaves firms on their own to solve both individual and collective problems, exposing the entire system to decline. Both inter-firm networking and business support organizations allow firms to overcome their internal constraints so that they can improve their innovative capability, grow and be competitive. The assessment of the institutional framework necessarily involves a qualitative analysis based on primary information gathered through questionnaires from both firms and business support organizations, as well as on secondary information. The objective of this assessment is twotold: (1) to list and map all organizations and institutions of the region or locality that have contacts with firms of the LPS; and (2) to appraise the role and involvement of such organizations and institutions in the production system's life. For the first objective, secondary information can be used together with direct questions to firms. The outcome should be a map of the various organizations and institutions that specifies their objectives, activities and types of firms they aim
elements that cannot be measured such as trust, cooperation, the degree of firms' einbeddedness and the role of social capital. These intangible elements are crucial for the functioning of LPSs. However, they can only be identified by means of primary information gathered directly from agents (e.g. firms and institutions) that operate within the local system. Because of their daily involvement in the lite ot the local system, local agents can be unique sources of information about the degree of trust between firms, the degree of cooperation between firms and insritutions, the role of the civic society behind the functioning of the production system and, finally, the trade-ort~ between einbeddedness and openness for the sustainability of the LPS. To summarize, once a map of LPSs has been drawn by means of Sforzi's methodology, the analysis of interfirm relationships enables a better understanding of the nature and dynamics of the linkages between firms. Here an in-depth analysis should rely on both inputoutput analysis and a qualitative assessment of the relationships.
INSTITUTIONAL FRAMEWORK Tlie development, and often survival, of LPSs can be linked to the presence and role of the local and regional institutional framework. This comprises a network of public and private institutions related in many different ways and to different degrees to hrms in an LPS. Analysing this institutional framework is the final stage ill the diagnostic analysis of the LPSs. The geographical dimension of the institutional framework is crucial to define its interaction with firms. LPSs are often bounded within a territory that is narrower than the administrative regional borders and. therefore, need local business support organizations and bodies. These are meant to reflect the specific interests and needs of firms in the LPS, for instance in relation to their sector specialization. Local and regional bodies can include universities, laboratories, science parks, trade unions, trade associations, entrepreneur associations, training and other service centres. Besides these, there could be public institutions such as city, provincial and regional councils, and government offices. Examples from Italian industrial districts and German clusters have illustrated how crucial the role of business support organizations can be in addressing situations of market failure that damage firms." In particular, there are three main functions that business support organizations undertake:'"^ To provide advice and support to individual firms (e.g. training, fmance, innovation, export). To provide collective goods (e.g. training, computeraided design, testing centres, joint marketing or branding, joint export). To act as brokers to facilitate and promote businessto-business and business-to-institution networking.
Mapping Local Production Systems in the UK

C O target. For the second objective, questionnaires have to be designed to address specific issues directed at firms and organizations. This enables an evaluation of business support organizations' initiatives for firms and localities. To conclude, analysis of the institutional framework is the hnal stage of the diagnostic analysis of an LPS. It should not be surprising that the rise and growth of an LPS often tends to be related to the effectiveness of the institutional setting, which can intervene in situations of market failure (i.e. fmance and innovation) and trigger virtuous circles of cooperation, competition, innovation and competitiveness. The study of effective institutional frameworks can also be used as a reference point to enhance and strengthen the institutional capability of localities and/or systems experiencing challenges and changes. LOCAL PRODUCTION SYSTEMS IN THE UK To illustrate how such an approach can be used, the application of the spatial diagnostic analysis to the U K is now presented. The data set constructed for mapping UK LPSs consists of 1997 industry employment data supplied by the Office for National Statistics, originally broken down into five size classes: 1-24, 25-49, 5()-99, 100-299 and more than or equal to 300 employees. Such data are aggregated in three main
205
bands: small firms with fewer than 100 employees, SMEs with fewer than 3(K) employees and large firms (LFs) with more than or equal to 300 employees. There are 23 manufacturing sectors (two-digit level)'"' and 297 TTWAs from Andover to Wick. The application of the four criteria of the spatial diagnostic procedure identified eight categories of LPSs according to their manufacturing intensity, degree of specialization and firm size. First, LPSs were distinguished according to whether the economy of the TTWAwas dominated by manufacturing sectors or not {in line with criterion 1). Second, LPSs' sector specializations were identified and such specializations were ranked according to the location quotient of criterion 3. In so doing, we distinguished between primary and secondary specializations. The former corresponds to the sector specialization with the highest location quotient, while the latter corresponds to the second or third ranked specializations. Finally, LPS were distinguished between those dominated by SMEs and those dominated by large firms (according to criteria 2 and 4). Based on these divides, a matrix of types of LPS in the UK was constructed (Table 2 and Fig. 2). The eight categories of LPSs were identified as follows: (1) pre-district; (2) non-specialized pre-district; (3) non-manufacturing and specialized LPS of SMEs; (4) non-manufacturing and non-specialized LPS of SMEs; (5) manufacturing and specialized LPS of LFs;
Jiihlc 2. Types of ]oai\ production systems (LPSs)

NonNonNonmaiiufacManufacNonman utactnanufaf- CuHng and M;inufac- turing and manufac- turiiig and Curiiig and nonturing and iionturing and nonNonspecialized specialized specialized specialized specialized specialized Pre- specialized LPS of LPS of LPS of LPS of LPS of LPS of district prc-district SMEs SMEs large firms larue firms larije firms laree firms (1) Manufacturing specialization (2) Size < ! 0 0 Size <3(X) Size >300 (3) Specialization primary secondary sector sector (4) Specialization plus size <100 Specialization plus size <300 primary secondary sector sector Specialization plus size > 300 primary sector secondary sector y primary sector 47 31 73 / secondary sector / primary sector / secondary sector primary sector secondary sector primary sector secondary sector primarysector secondary sector
Other
Number of travel-rn-work areas
Source: Liata from the Office of National Statistics (ONS), author's elaboration. Note: SMEs, small- to medium-sized firms.
206
Lisa De Propris
Manu^cturing LPSs Non-maniifecturing LPSs
Specialized LPSs
NS Pre-district of SMEs
Non-speciailzed LPSs
Fig. 2. Types of heal production systems (LPSs) in the UK

S-NM, specialized non-inanufacturing; S-M, specialized manufacturing; NS, non-speciaiized: N S - N M , non-specialized non-manufacturing; NS-M, non-spccializcd manufacturing; LF. large firm; SMEs. small- and medium-sized enterprises {i.e. firms). Source: Data are from che Office of National Statistics, author's elaboration
(6) manufacniring and non-specialized LPS of LFs; (7) non-manufacturing and specialized LPS of LFs; and (8) non-manufacturing and non-specialized LPS of LFs.'^ A few points are worth mentioning. First, with this methodology, we were able to identify diverse types of LPSs in the UK directly from the statistical data before any qualitative assessment was made. This is a very useful step to provide a broad {e.g. country-wide) picture of the LPSs' phenomenon and, at the same time, single out diversity across such systems by means of a quantitative analysis. This is in contrast witb the DTI (2001) report where only clusters were identified and differences across them were purely based on qualitative information. Second, 47 industrial districts were identified that strictly satisfy criteria 1-4. This finding contrasts with the statement by Z E I T L I N (1995) that there are no industrial districts in the UK. Third, more than half the LPSs were characterized by SMEs: 170 LPSs were dominated by SMEs against 111 being dominated by large firms. The geographical distribution of LPSs in the UK is shown in Figs 3 and 4. LPSs of LFs are distributed across England, with a concentration
in the Midlands and the North East, and in South Wales and the Southern Upperlands of Scotland. On the other hand, LPSs of SMEs are scattered across England and cover most of Wales and Scotland. In particular, industrial districts are scattered across the entire country with a significant presence in the Midlands and the North of England. Manufacturing LPSs are concentrated in Scodand, Wales and the South of England. Fourth, there seems to be more nonmanufacturing LPSs (185 were counted) than manufacturing LPSs (156). This finding is consistent with the rapid and extensive expansion of the service sector across the UK. Finally, those LPSs that were strictly identified according to criteria 1 4 accounted for 21% of the total UK liianufactiiring employment. This result is not insignificant given that the rigid application of the criteria identifies only the main localized industry for each T T W A . For instance, for the Coventry T T W A , only its main LPS in the motor vehicles sector were considered as satisfying criteria 1 4 and it accounted for 22% of the sector national employment. However, in Coventry there are also two other sectors for which the location quotient was greater than tbe
207
Fig. 3. Local producfioii systems of large firms in the UK hy trauel-to-u>ork area. For ahbreuiations, see Fig. 2
Soiirtr. Figs 3 and 4 are the author's elaboration with data from the Office of National Statistics (ONS). O N S , 1998; map produced fay O N S , 2003
208
Lisa De Propris
NM-NS-LPS NM-5-LPS
N5-ID
ID
Fig. 4. Local production systems ofsmall and medium-sizefrms
in the UK by travcl-to-work area. For abbreviations, sec Fig. 2
Mapping Local Production Systems in the UK national average: the fabricated metal product sector, and machinery and equipment. These account for a considerable share of national employnienc: the metal prodLict sector accounted for 13% of the sector national employment, and the machinery and equipment for 12%. Therefore, should one consider such secondary LPSs for each T T W A , it would certainly be found that their total employment share of the U K manufacturing employment is ahnost 50%. An evaluation of these initial fmdings already reveals how a better Linderstanding of the location, type and (if one were also to apply the second and third steps of the methodology) kinctioningot UK LPSs is extremely important for regional industrial development catalysing around localized industries and, associated with this, the formulation of appropriate industrial development policies.
209
To conclude, this methodology discusses the desirability of criteria and measures to study LPSs; the author feels that the debate on the methodology to map LPSs shoLild drive rather than be constrained by the availability of statistical data. In other words, che academic debate should challenge the 'goodness' of the present data provision, should this be insufficient to provide a clear picture of a particular phenomenon. For instance, the absence of regional or sub-regional inputoutput tables is a serious limit to the study of LPSs in the UK. Acknowledgements - The author thanks four anonymous referees aTid Marco BcUandi, Christos Pitelis and David Bailey for useflil comments and suggestions. The usual disclaimers apply.
NOTES
CONCLUSIONS
The paper has presented a methodology to map and analyse LPSs in regions and countries where there is little or no knowledge about whether LPSs exist, where these are and what form they take. The diagnostic niethodology presents three stages: the spatial diagnostic, the diagnostic of inter-firm relationships and the assessment of the institutional framework, hi particular, the spatial diagnostic procedure is aimed at equipping us with an autoniatic procedure that relies on statistical data and that can provide basic but fundamental information about LPSs. The three-tiered methodology involves both quantitative and qualitative analyses, and it proceeds from the big to the small, from the general to tbe specific. In fact, spatial diagnostic analysis reveals a map and a classification of LPSs; after this first stage, tbe further two stages apply to in-depth case studies on specific production systems. The present paper fills a gap in the debate on localized industries and firm agglomeration in [hat it suggests a methodology to identify, map, classify and analyse indepth LPSs. The methodology' has the potential to flag up any agglomeration of firms: inflint systems, successful systems, systems in decline. This means that it enables research to explore the factors driving the rise, development and success of production systems, as well as the causes of their decline and possibly disappearance. The second half of the paper presented an application of the spatial diagnostic analysis to the UK (excluding Northern Ireland). It identified eight types of LPS, inchiding industrial districts, according to sector specialization, firm size and manufacturing intensity. The three-step methodology has also been applied to map LPSs in Central and Eastern Europe. It has provided a very useful insight into the presence of LPSs in Lithuania, Bulgaria, Romania and Bosnia, where they can be sparks for regional industrial development {D E PROPRIS and P I T E L I S , 2000).
1. For a very interesting discussion on the conceptual difference between typology and taxonomy, see GRANDORI (1990). A typology is defined as an ex-antc classification that identifies analytical categories to predict variables not included in it. By contrast, a taxonomy is a classification generated ex-post from empirical observation. 2. The main data sources for the DTI cluster analysis are; the Inter-departmental Business Register, the Dun and Bnidstreet database, the National Online Manpower Information System, ;ind local aucbority- data. 3. The niain contributions in Italian industrial districts
include: LEONARDI and N A N E T T I (1994) on Tuscany; B R U S C O (1982) and LAZERSON (1990) on Emilia
4.
5.
6.
7.
Romagna, and RABI;1.I.OTTI (1997) on the footwear district in che Marche and Veneto. Thf Eumpean Union considers small firms as having fewer than 50 employees and SMEs as having fewer than 250. Note that SIC-s are a device designed to a.ssist the otfice of Customs and Excise and the collection of statistical data. However, they do not always provide a realistic grouping of production activities. For this reason, one needs to consider sector classifications with caution especially if used to map industrial agglomerations. Here the degree of specialization of production activities makes it very difficult to match them with standard SIC codes. Most studies have looked at LPSs in the manufacturing sectors. However, there has been some work on clusteK in the service sector, such as the entertainment industry in Hollywood, CA (PORTER, 1998), and the financial service cluster in Hong Kong (ENBIGHT, 2000). On the case of tbe 'three Italies', see KlNc; (1985) and
SFOBZI (1999).
8. According to MARSHALL (1920), agglomeration economies are related to firms' external economies of scale and a rise from information spillovers, local non-traded inputs and a pool of local skilled labour ( M C C A N N . 2001). In the case of LPSs, external economies are associated with the organization of production and firms' specialization. Production fragmentation and specialization enable firnis to reap economies that are not internal
210
Lisa De Propris
Wuttemberg and Emilia Romagna, see COOKE and
MORGAN (1994).
to the firm (although producing increasing return to scale) but external to the firm and internal to the production system of which the firm is a part. Seminal contributions on agglomeration economies include MARSHALL (1920), PER RO U X (1950) and, more recently. PORTER (1998) and STCJRPER (1996). For empirical work, see BRESNAHAN ct ai (2001), R O S E N THAL and STRANGE (2001) and PARR (2002). 9. For a case study on the link between production networking and cooperation over innovation, see D E
PROPRIS (2000).
10. There is a fundamental difference between inter-firm relationships in the industrial district model and the postFordist model, i.e. the structure of governance mirrored by input output exchanges. In post-Fordist systems, where the main buyer outsources its non-core activities, governance tends to be characterized by external vertical control. In contrast, in industrial districts, subcontracting tends to be associated with a more dispersed and horizontal system of governance. For fijrther information, seeBECATTiNi (1994). 11. For a description of the institutional frameworks in Baden
12. This analysis draws on D E PROPRIS and PITELIS (2000). 13. Oil Real Service Centres in Emilia Romagna, see L E O N ARDi and NANETTI (1990). 14. The 23 manufacturing sectors are as follows; Food products and beverages; Basic metals; Tobacco products; Fabricated metal products; Textiles; Machinery and equipment; Apparel dressing/dying fur; Office machinery and computers; Tanning/dressing of leather; Electrical machinery/apparatus; Wood/products/cork; Radio, television, communications equipment; Pulp, paper and paper products; Medical precision instruments; Publishing, printing, recorded media; Motor vehicles, trailers; Coke, refined petroleum products; Other transport equipment; Chemicals and chemical products; Furniture: Rubber and plastic goods; Recycling; and Other non-metallic products. 15. Six T T W A cannot be classified because criterion 4 is not respected: Barnsley, Bolton, Burnley, Dundee, Havi'ick, and Sheffield and Rotherhani.
REFERENCES
BECATTINI G. (1987) Mcrcam e Forze Locali: II DimeUo liiditslriale. II Mulino, Bologna. BECATTINI G. (1994) The development oflight industry in Tuscany: an interpretation, in LtoNAKDl R. and NANKTTI R . Y. (Eds) Regional Devehypment in a Modern European Economy, pp. 69-85. Pinter, London. BECATTINI G. (2000) H Bmco e la Ftirfalla: Prato ml Mondo cite Canihia. Le Monnier, Florence. BECATTINI G. (2001) [I Calcdoscopio dello Sviluppo Locale: Trastoriua.zioi)i Eamomichc m-IVltalia che Gambia. Rosenberg & Sellier, Turin. BELLANDI M . (2001) Local development and embedded large firms. Entreprencurship and Regional Development 13, 189-211. BELLANDI M . and SFORZI E (2001) La molteplicita' dei sentieri di sviluppo locale, in BECATTINI G. et al. (Eds) // calcdoscopio dello sviluppo locale, pp. 4tf>6. Rosenberg &: Sellier, Turin. BELUSSI F and ARCANGELI R (1998) A typology of networks: flexible and evolutionary tirnis. Research Policy 27. 4 1 5 ^ 2 8 . BjORc; A. and ISAKSHN A. (1997) Location, a^lomeration and innovation; towards regional innovation systems in Norway, European Planning Studies 5, 299-331. URESNAHAN T , GAMBARDELLA A . and SAXENIAN A. (2001) Old economy inputs for new economy outcomes: cluster formation in the New Silicon Valley, Industrial and Corporate C/JIJ/JSJC 10, 853860. BRUSCO S. (1982) The Emihan model: productive decentralisation and social interaction, Cambridge Journal of' Ecouomia 6, 167-184. BRUSCO S. and PABA S. (1997) Per una stoHa dei distretti italiani dal secondo dopoguerra agli anni novanta. in BARCA F (Ed.) Storia del capitalismo ilaliaiio, pp. 265-333. Donzelli, Rome. CAMAGNI R . P. (1995) The concept of innovative milieu. Its relevance for public policies in European lagging regions. Papers in Regional Scien^ê 74, 317-340. CHENERY H . B . and WATANABE T (1958) International comparison of the structure of production, Econometrica 26, 487-521. COOKE P. (2002) Knowledge Economy: Clusters, Learning and Cooperative Advantage. Routledge. London. COOKE P. and MORGAN K. (1994) Growth regions under duress: renewal strategies in Baden Wiitteniberg and Emilia Romagna, in AMTN A. and THRIFT N . (Eds) Globalisation, Institutions and Regional Development in Europe, pp. 91-117. Oxford University Press, Oxford. COOKE P. and MORGAN K. (1998) Tlte Associational Economy. Oxford University Press, Oxford. CosspNTiNo F, PYKK F and SEGENBERGER W (1996) Local and Regional Responses to Glohal Pressure: The Case of Italy and its Industrial Districts. International Institute for Labour Studies, Geneva. C R O U C H C . and FARRELL H . (2001) Great Britain; falling through the holes in the network concept, in C R O U C H C , LE GALfes P., TRIGILIA C . and VOELZKOW H . (Eds) Local Production Systems in Europe. Rise or Demise? Oxford University Press, Oxford. DEPARTMENT OF TRADE AND INDUSTRY (2001) Business Clusters in the UK- A First Assessment. D T I , London. D E PROPRIS L. (2000) Innovation and inter-tirin cooperation: the case of the West Midlands, Economic Innovation and New Technology 9, 4 2 1 ^ 4 6 . D E PROPRIS L. (2001) Systemic flexibility, production fragincntation and cluster governance, European Planning Studie.^ 9, 739-753. D E PROPRIS L. (2003) Mapping Local Production Systems iti Britaiu. Working Paper No. 14. Birmingham Business School. D E PROPRIS L. and PITELIS C . (2000) Glusier Diagnostics. Phare A C Project Report.
211
ENRICH r M. J. (2000) Regional clusters and multinational enterprises, Intemaliotial Studies in Mariaêmetit and Organization 30, 114-138. FESEB E. J. and BERGMAN E. M . (2000) National industry cluster templates: a framework for applied regional cluster analysis. Regional Studies 34, 1-19. GABOFOLI G. (1991) Local networks, innovation and policy in Italian industrial districts, in Br;Rc;MAN E., MAIER G. and ToDTLiNG E (Eds) Regions Reconsidered, pp. 119-140. Mansell, London, GORDON I. R. and M C C A N N P. (2000) Industrial clusters: complexes, agglomeration and/or social networks, Urban Studies 37, 513-532. GRANDORI A. (1990) Epistemologia ed economia aziendale: note per un dibattito, L'indmtria 11, 76-103. Gui'.RRiERi P., IAMMARINO S. and PL[^TROBBI.I.I L. (2001) The Global Change and Industrial Districts. Edward Elgar, Cheltenham. IsARP W.. Azis I. j . , DRBNNAN M . P.. Miii-ER R. E., SALT/^MAN S, and THORBEKE E. (1998) Methods of Interregional and Regional Analysis. Ashgate, Aldershot. KING R . (1985) Tlie Industrial Geography of Italy Groom Helm, London.
LAZERSON M . H . (1990) Subcontracting in the Modena knitwear industry, in PYKE E , BECATTTNI G. and SENGENBERGER
W. (Eds) (1990) Industrial Districts and Inter-jirni Cooperation in Italy. International Institute for Labour Studies, Geneva. LEONARDI R . and NANETTI R . (1990) The Region and European Integration: The Case of Emilia Romagna. Pinter, London. LEONARDI R . and NANETTI R . (1994) Regional Development in a Modern European Economy: Tlie Case of Tuscany. Pinter, London. LoNGHi C. (1999) Networks, collective learning and technology development in innovative high technology regions: the case of Sophia Antipolis, Regional Studies 33, 333-342. MARKUSEN A. (1996) Sticky places in slippery space: a typology of industrial districts. Economic Geography 72, 293313. MARSHALL A. (1920) Principles of Economics, 8th Edn. Macmillan, London. MARTIN R . and SUNLEY P. (2003) Deconstructing clusters: chaotic concept or policy panacea?, Jowmd/ of Economic Geography 3, 5-35. M C C A N N G . (2001) Urhan and Regional Economics. Oxford University Press, Oxford. PANICCIA I. (2002) Industrial Districts. Edward Elgar, Cheltenham. PARR.J. B. (2002) Missing elements in the analysis of agglomeration economies. International Regional Science Reuiew 25, 151-170. PERROUX H (1950) Economic space: theory and application. Quarterly Journal of Economics 64, 89104. PioKE M. and SABEL C . {19K4) Tlie Second Industrial Divide. Basic, New York. PORTER M . P. (1998) Clusters and the new economics competition, Harrard Busitiess Rei'iews November-December, 77-90. PORTER M . P. (2000) Location, competition and economic development: local clusters in a global economy. Economic Deuelopment Quarterly 14, 15-24. PYKE H, BECATiiNt G. and SENGENBERGER W. (Eds) (1990) Industrial Districts and Inter-firm Cooperation in Italy. International Institute for Labour Studies, Geneva. RABELLOTTI R . (1997) External Economies and Cooperation in Industrial Di.trnWi. Macmillan, Basingstoke. RASMUSSEN P. N. (1956) Studies in Inter-sectoral Relations. North Holland, Amsterdam. RoSENTHAi., S. S. and STRANGE W. C . (2001) The determinants of agglomeration, yoMrii/ ofUthan Economics 50, 191-220. SFORZI E (1989) The geography of industrial districts in Italy, in GOODMAN E . and BAMFORD J, (Eds) Smalt Eirnis and Industrial Districts in Italy, pp. 153-173. Routledge, London. SFORZI E (1990) The quantitative importance of Marshallian industrial districts in the Italian economy, in PYKE F . , BECATTINI G. and SENGENBERGER W. (Eds) Industrial Districts and Inter-firm Cooperation in Italy, pp. 75-107. International Institute for Labour Studies, Geneva. Sr-ORZi E (1996) Italy: Local Production Systems of Small and Medium-sized Eirms and Industrial Changes. O E C D Networks of Enterprises and Local Development, Paris. SFORZI E (1999) Economic change, in BONAVERO P., DEMATTEIS G . and SFORZI E (Eds) Tlte Italian Urban System. A.shgate, Aldershot. SiMMiE J. and SENNETT J. (1999) Innovative clusters: global and local linkages?. National Institute of Economic Reviews 170, 87-98. SOLINAS G. and BARONI D. (1998) / sislemi locali manufatturieri in Italia: 1991-1996. Mimeo, Dipartimento di economia politica, Universita' di Modena e Reggio Emilia. STORPER M . (1996) The resurgence of regional economies, ten years later: the region as a nexus of untraded interdependencies, European Urban and Regional Studies 2. 191-221. ZKITLIN J. (1995) Why are there no industrial districts in the United Kingdom?, in BAGNASCO A. and SABEL C . E (Eds) Stnall and Medium Enterprises. Pinter, Derby.
Urban Studies http://usj.sagepub.com/
Making Location Quotients More Relevant as a Policy Aid in Regional Spatial Analysis
Andrew Crawley, Malcolm Beynon and Max Munday Urban Stud published online 20 December 2012 DOI: 10.1177/0042098012466601 The online version of this article can be found at: http://usj.sagepub.com/content/early/2012/12/20/0042098012466601
Published by:
http://www.sagepublications.com
On behalf of:
Urban Studies Journal Foundation
Additional services and information for Urban Studies can be found at: Email Alerts: http://usj.sagepub.com/cgi/alerts Subscriptions: http://usj.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Dec 20, 2012 What is This?
Downloaded from usj.sagepub.com at UNIV OF TEXAS ARLINGTON on April 7, 2013
116, 2012
Making Location Quotients More Relevant as a Policy Aid in Regional Spatial Analysis
Andrew Crawley, Malcolm Beynon and Max Munday
[Paper first received, October 2011; in final form, June 2012]
Abstract Location Quotients (LQs) remain an important tool for geographical analysis, particularly in terms of assessing industrial specialisation and clustering. LQs as decision aids are typically understood through the use of arbitrarily set cut-off values. However, LQs are rarely accompanied by an associated level of variance that can be connected with the estimated data used to calculate them. This paper reveals the importance of understanding this variance and shows how confidence intervals can be estimated for employment-based LQs. A systematic process is introduced, through which the arbitrariness of cut-off-value choice can be mitigated and borderline industry cases in terms of their LQ values and the considered cut-off value, identified. A case from a UK region is used to illustrate the issues covered in the paper.
1. Introduction
Since its conception with the advent of the economic base model (Haig, 1926), the location quotient (LQ) has provided one means of assessing the relative specialisation of a particular characteristic within a population.1 Its popularity has not diminished and its use is still prevalent (see recent work by Bishop et al., 2003; Tonts and Taylor, 2010). The LQ has been used in different types of analysis to accomplish numerous analytical tasks. One factor encouraging the analytical use of LQs, according to Isserman (1977), is that relatively few data are required for their computation and this may also have encouraged its use as a decision aid. Indeed, the notion of the LQ has been consistently used in geographical analysis since the 1940s (Gibson et al., 1991). Its application has increased recently with its adoption in a number of studies investigating industry specialisation and clustering. For example, Porter (2000) and the UK cluster mapping study (DTI, 2001), used LQs to identify clusters of industries
Andrew Crawley, Malcolm Beynon and Max Munday are in the Business School, Cardiff University, Colum Drive, Cardiff, CF10 3EU, UK. E-mail: CrawleyAJ@cardiff.ac.uk, BeynonMJ@Cardiff.ac.uk and mundaymc@cf.ac.uk
0042-0980 Print/1360-063X Online 2012 Urban Studies Journal Limited DOI: 10.1177/0042098012466601
ANDREW CRAWLEY ET AL.
at sub-national scale (see also for recent studies, Boix and Galletto, 2009; Feser and Isserman, 2009; Chiang, 2009; Carroll et al., 2008). The LQ is also used by the European Cluster Observatory (ECO) programme, a policy-making group of the European Union (Crawley and Pickernell, 2012). Many researchers are interested in the practical application of LQs. For example, epidemiologists examine the spatial distribution of diseases (see Clayton and Hills, 1993), whilst criminologists often want to understand differences in reported crimes in different neighbourhoods (see de Frances and Titus, 1993). Climatologists have used the technique to measure certain meteorological effects (see Wong and Chen, 2009). The LQ has also been widely used in psychology as a way of establishing patterns of psychological disorders in populations (see Metraux et al., 2007). Returning to the traditional use of LQs in industry analysis, a common problem is the use of arbitrary cut-off values to designate sets of industries based on calculated LQ values. Moreover, while LQ-based analysis leaves the analyst with a set of included and excluded industries there are inevitably borderline industries with LQs close to the considered cut-off value. The systematic process developed in this paper elucidates the interest potential of identified borderline industries, in terms of the uncertainty of their membership to either of the two designation sets (inclusion or exclusion). In particular, this paper demonstrates that such analysis should take careful account of the underlying variance in calculated LQs. To illustrate the points made in our paper, we examine LQs calculated for Welsh manufacturing industries. The structure of the rest of the paper is as follows. The second section reviews how LQs are typically used to aid decisionmaking within industrial specialisation analysis, and reviews the usefulness of the
measure. This section focuses on the problems associated with practically using LQs without taking into account associated variances. The third section discusses one method that can be used to analyse variance associated with LQs. This demonstrates how employment-based LQs,2 calculated for the purpose of industrial specialisation analysis, might be understood when the confidence intervals around them are considered. The fourth section introduces a systematic process by which the arbitrariness of cut-off value choice can be mitigated and how the interest potential for borderline industries, with LQs near a cutoff value, can be identified. The fifth section offers conclusions and directions for future research.
2. The Usefulness of Location Quotients: A Review

LQs were originally used as a means of determining the levels of manufacturing versus service industries within a given spatial area (see Haig, 1926). They have since been used to quantify and identify numerous spatial phenomena. The method quickly became one means of determining the presence of industrial complexes (Czamanski, 1974). Its use to describe industrial specialisation within national and regional economies has increased in the past 20 years following the academic work by, among others, Porter (1990, 1999, 2000), Bergman and Feser (2000) and Carroll et al. (2008). Gibson et al. (1991) describe the LQ as demonstrating how strongly an industry is represented in a region. Studies have used the technique to identify industrial clusters which might become the focus of government support (see DTI, 2001; see also World Economic Forum, 2007). LQs have further applications. For example, the estimation of LQs forms one basis for
MAKING LOCATION QUOTIENTS MORE RELEVANT
regionalising national inputoutput tables in the absence of extensive regional survey data (see Isard et al., 1968; Miernyk, 1976; Brand et al., 1998). In non-survey input output table development, LQs are used to adjust national industry coefficients to take into account regional differences in industry structure (for a full discussion on this, see Miller and Blair, 2009). The use of LQs in these applications is not without problems. For example, in the case of developing non-survey tables, there are a series of assumptions typically made (i.e. with respect to spatial differences in labour productivity, cross-hauling, etc.), constraining the use of LQs (see Schaffer, 1999). Moreover, in terms of the methodology itself, it has been argued both for cluster analysis (Malmberg and Maskell, 2002) and inputoutput table construction (West, 1986), that LQs lack reliability and consistency in estimation. Linked to these issues, practical analysis using LQs often means the employment of arbitrary cut-off values to determine designation sets of industries (see Bishop et al., 2003). The cut-off value is simply a value against which the industry LQs are compared, whether an industry LQ value is above or below it, so forming two different designation sets. For example, in one of the largest UK studies utilising LQs, undertaken by the DTI (2001), a cut-off value of 1.25 was employed. This allowed the inclusion and exclusion of industries for further study. Table 1 summarises some of the other common cut-off values used in prior research. An issue arising from the variation in the previously employed cut-off values shown in Table 1 is that the key to understanding which value is the most appropriate (as a cut-off value) might not be the most important issue given the applications for which the LQ is typically used. The exercise should be to establish whether or not a particular LQ value is noticeably greater than some arbitrarily developed cut-off value. It is also
important to recognise when calculating LQs with government employment data in industries, that the data themselves may be estimated. The accuracy of small-area employment data has been called into question before and this has been shown to have impact on LQ estimates (see Silcocks, 1994; Thrall et al., 1995; Beyene and Moineddin, 2005). For example, in treating with these problems ODonoghue and Gleave (2004) suggest a standardised LQ approach which reveals agglomerations as being made up of locations with statistically significant (rather than arbitrarily defined) LQ values for industries being considered. The problem of the LQ method generating single point estimates is also examined by Beyene and Moineddin (2005). Basically, there is a need to consider the measurement of statistical error, or the degree of accuracy of a particular LQ value, based upon the overall dataset being employed. A simple technique that could be applied is the calculation of a t-test comparing the LQ for a particular industry in a region, with that of other LQs calculated from other regions. Thrall et al. (1995) adopt a similar approach when looking at the calculation of LQs for the number of mortgages given by different financial institutions across the US. They used a t-test to determine how similar an individual LQ for one institution was compared with an aggregated LQ determined by looking at all the institutions. This, however, still had the drawback of assuming that the calculated point estimates were absolute rather than estimates. Beyene and Moineddin (2005) chose to utilise a more sophisticated approach, based upon the linear approximation of the variance, allowing the establishment of confidence intervals around LQs. This calculation is of benefit as it is constructed to take into account the inherent uncertainty contained within the estimation procedure involved in the initial calculation of the LQs.
Table 1.
Arbitrary cut-off values used previously in economic studies References Malmberg and Maskell (2002) Held (1996); Bishop et al. (2003); Tonts and Taylor (2010) Bergman and Feser (2000) lvell et al. (2003) So Kumral and Deger (2006)
LQ cut-off value 3 1 1.25 2 1.25 to 5
This paper in the next section applies the Beyene and Moineddin (2005) method to industrial employment data.3 However, we then use the resulting LQ confidence intervals in a novel way to offer an informed expectation to the use of arbitrary cut-off values, in terms of designation of industries. By analysing the relative numbers of industries designated as included or excluded, based on different cut-off values, this paper provides a unique graphical-based systematic process to allow decision-makers a consistent statistically based alternative to arbitrary cutoff-value choice.
3. Estimating Confidence Intervals for LQs

In this section, and the following, we use an example to illustrate the intended systematic process. The case used is the estimation of LQs for manufacturing industries in Wales in the UK. To test the effectiveness of the method, a region needs to be selected that allows a manageable and yet representative sample of a nation as a whole. Wales has around 5 per cent of the UK population and has a representative number of industries that also feature at the national UK level. To explore industry specialisation, we use the most disaggregated data possible. The UK Office of National Statistics holds estimated data on firms and industry size in terms of employment. These data are classified by the Standard Industrial Classification (SIC),4
within which the level of detail is denominated by the number of digits used to express the data. The other potential level of disaggregation is spatial, with SIC employment data being available at national, regional, travel-to-work area and local authority levels. When calculating LQs, the spatial domain is important (see McCann and Dewhurst, 1998). Mulligan and Schmidt (2005) also found that LQs are sensitive to changes in the reference geographical scale. The issue known as the moveable area unit problem (Openshaw and Taylor, 1979) means that if one uses different spatial domains it changes the findings dramatically. To this effect, for the purposes of this paper, a large region case is used to give a fair representation of what the method can allow. In summary, to illustrate the confidence interval approach, we use LQs calculated for SIC 5-digit manufacturing industries within Wales for 2010. For illustrative purposes, we also initially adopt an arbitrary cut-off value for the LQ of 1.25 to highlight the issues we are seeking to address later. For Wales, there were 218 different manufacturing industries (5-digit SIC) where data were available.5 In order to explore the issue of cut-off values, this work uses the LQ formulation given by de Propris (2005), for measuring manufacturing specialisations (for industry i)6 xi LQi = ni x n
where, xi and ni represent the levels of manufacturing employment in industry i at the regional and national levels (in our case Wales and the UK); and x and n are the total levels of manufacturing employment at the regional and national levels. In what follows, the word regional is taken to represent some geographical scale below the national. It follows, for industry i, that if the LQi is greater than one, then the region has a greater share in that industry than the national average and thus it could be inferred that the regional area is more specialised in the given industry. To estimate the variance associated with an LQi value, the Delta method is adopted (see Oehlert, 1992), which utilises the Taylor series expansion. The Taylor series is the representation of any given function by an infinite sum of terms. These are calculated from the values of its derivatives at a single point (see Thomas and Finney, 1996). For full details on the utilisation of this method with respect to LQs and their associated variances, see Moineddin et al. (2003) and Beyene and Moineddin (2005). In summary, xi x , and r = n given the two expressions ri = n i ri (so LQi = r ), denoting respectively the regional share of national employment in manufacturing industry i, and then the regional share of total national manufacturing employment, it can be shown the confidence interval limits at the 1001 a per cent confidence level, with associated za=2 z-score, are given by ri 6za=2 rs 2r2 ri (1 ri ) r2 i (1 ri ) i (1 ri ) + nr 3 nr 3 ni r 2 It follows that the relative size of industry employment and the regional area in which it operates, both have a role in determining the relative variance associated with the
estimated LQ. For example, if the industry employment is relatively large and the regional area is small, the LQ will appear large and the relative variance will appear small. Conversely, if the industry employment is relatively small and the regional area employment is larger, then the LQ will be small and the associated variance will be much higher. With the use of regions it is possible to make the assumption that the reference area is partitioned into k non-overlapping regions. It is assumed that the distribution of xi is binomial with the parameters of total population ni , and the true incidence rate ri . Following Moineddin et al. (2003) and Beyene and Moineddin (2005), we can make the assumption that x/n is fixed not random (the incidence rate is the same in all areas for an industry, ignoring any spatial variation). It is acknowledged that this is a restrictive assumption and means that a greater component of the measure is said to be non-random. This approximation is used as a substitute for individual incidence rates as, with large sample sizes, the relative change in the true incidence rate is negligible (Moineddin et al., 2003). Assumptions can then be made regarding the covariance if spatial autocorrelation is considered negligiblethat is, if the presence of an industry in one regional area has no impact on whether that industry is represented in areas close by. With these data, the lack of continuitythat is to say, not all regions have a presence from every industrymeans that autocorrelation becomes less of an issue. Due to the large number of industries and the resulting observations for Wales used in this study, it can be assumed that the statistic has a Gaussian distribution, subject to a z-score (here considered over a range of the 1001 a per cent confidence interval). The size of the confidence interval then depends upon just the three variables, ri , r and n. Hence the earlier
assertion that the magnitude of the variance varies with the levels of spatial and industrial disaggregation. In Table 2, we present the constituent LQ values and 95 per cent (a = 0.05 in 1001 a per cent) confidence intervals for a sample of the manufacturing industries. These were not randomly chosen from the 218 manufacturing Welsh industries considered. Instead, they were identified subject to a chosen LQ cut-off value (and confidence level). As noted earlier for illustrative purposes, an LQ cut-off value of 1.25 is considered, this being a common value in previous studies (see Table 1).7 The choice of 1.25 is without loss of generality to the consideration of any LQ cut-off value. These industries are then of interest since each of their respective 95 per cent confidence intervals surrounds the considered 1.25 LQ cut-off value (whereas for the other 203 industries at the 95 per cent confidence interval, their confidence intervals are either fully below or above the 1.25 cut-off value). In this situation, due to the consideration of the variances of industry LQs, there is the possibility that, if the point LQ is above or below the considered LQ cut-off value, the actual LQ could be below or above this respectively. Even within the reported sample of industries shown in Table 2, the variation in the LQ point values and associated variances is demonstrated. The subsequent impact on the constructed confidence intervals is also displayed in Table 2. Rather than only considering the usual single confidence interval for a specific 1001 a per cent confidence level (see Moineddin et al., 2003; Beyene and Moineddin, 2005) as shown in Table 2, the analysis here is extended in Figure 1 to show the relative change in the confidence intervals as the respective 1001 a per cent confidence level changes from 0 per cent (point values) to 95 per cent (often
employed), for the 15 industries highlighted in Table 2. In Figure 1, in the main part of the graph, each defined industry, from Table 2, is respresented by a two-dimensional trumpet shape. At the back of the graph, on the confidence level axis labelled Point(0%) and 95 % Confidence Interval is the point LQ value and 95 per cent confidence interval representation (see Moineddin et al., 2003; Beyene and Moineddin, 2005), for all 218 industries including the 15 industries whose values are reported in Table 2. The trumpet shape shows the change in the associated confidence interval of the industry LQ value as the 1001 a per cent confidence level changes. Clearly, as the 1001 a per cent confidence level increases, so the confidence interval range increases with the limits represented by the respective top and bottom points along the trumpet (at the relative confidence level). The important point to note here is the variability of the confidence interval ranges across the different industries considered, a consequence of the different variances associated with each LQ value. Figure A1 in the Appendix summarily shows the trumpet shapes associated with all 218 manufacturing industries. The results in Figure 1 show the industries, in terms of their LQ trumpet representations, which, subject to a 95 per cent confidence interval, could be considered in one of two following incidences (the dashed line shows the 1.25 cut-off value across the whole figure) (1) Their LQ value is above the 1.25 cut-off value, but their 95 per cent confidence interval includes values below 1.25 (so could actually be below it). (2) Their LQ value is below the 1.25 cut-off value, but their 95 per cent confidence interval includes values above 1.25 (so could actually be above it).
Table 2. Manufacturing 1.352 1.305 1.197 1.155 1.055 1.232 1.317 0.918 1.206 1.155 1.271 1.195 1.234 1.227 1.281 0.013732 0.010778 0.003116 0.002964 0.003234 0.006518 0.017445 0.000494 0.007599 0.057111 0.006838 0.003395 0.012700 0.000992 0.003462 LQ Var (s2)
LQ confidence intervals for 15 Welsh manufacturing industries (original point LQ, variance and 95 per cent confidence interval) 95 per cent confidence interval (LQ 6 1.960s) (1.2431, 1.4619) (1.1981, 1.4115) (1.0859, 1.3088) (0.9967, 1.3131) (0.7964, 1.3141) (1.1886, 1.2758) (1.1464, 1.4881) (0.4500, 1.3867) (1.0442, 1.3684) (1.0404, 1.2688) (1.0503, 1.4921) (1.1331, 1.2566) (1.1188, 1.3495) (0.9976, 1.4569) (1.0771, 1.4841)
SIC no.
10910 16100 16290 17211 18202 22230 23140 23490 24520 25210 25300 27120 28230
28940 30920
Prepared feeds for farm animals Sawmilling and planing of wood Other products of wood Corrugated paper and paperboard; manufacture of sacks and bags of paper Reproduction of video recording Builders-ware of plastic Glass fibres Other ceramic products Casting of steel Central heating radiators and boilers Steam generators, except central heating hot water boilers Electricity distribution and control apparatus Office machinery and equipment (except computers and peripheral equipment) Machinery for textile, apparel and leather production Bicycles and invalid carriages
Figure 1. LQ confidence intervals for manufacturing industries in Wales (with varying confidence levels of 100(1 a) per cent), for industries whose LQ confidence interval at the 95 per cent confidence level is inclusive of 1.25 cut-off value.
A count of these industries shows 15 (6.9 per cent) out of the 218 considered industries fall into one of these two incidences. Two example industries are highlighted in Figure 1 with their trumpet and SIC codes in bolder lines, for illustration purposes, Other Ceramic Products is SIC 23490 (and so labelled in Figure 1) and Electricity Distribution and Control Apparatus (hereafter Electricity Distribution etc.) is SIC 27120. These two industries illustrate the variation in the levels of variance associated with their respective LQ values. Other Ceramic Products has a large variance (wide trumpet) as the confidence level 1001 a per cent increases, whereas for Electricity Distribution etc. there is a smaller variance (thin trumpet). These results illustrate the earlier assumption that employment size and variance are interconnected. Other Ceramic Products employ just 14 employees whereas Electricity Distribution etc. has relatively large employment (1328) compared with the UK average (470). Inspection of these two industries in Figure 1 shows that they have LQ values below the cut-off value
1.25 and, as the confidence level increases, they each subsequently have confidence intervals inclusive of this cut-off value (both fall into incidence (2) described previously). The implication is that industries whose derived LQ confidence intervals surround the chosen cut-off value might be flagged up for further consideration, such that there may be contradictions in their initial designations.
4. Implications
The method described in the previous section shows a way of determining the variances associated with the calculation of LQs. A consequence of the consideration of variances is the potential uncertainty in the industries designations based on a chosen LQ cut-off value. As demonstrated in Table 1, numerous different reports and academic papers have chosen to adopt arbitrary cutoff values, with little statistical rationale. Understanding this uncertainty issue next allows us to consider whether this method,
using the associated LQ variance values, can play a role in improving one of the longstanding issues of the utilisation of LQsnamely, the arbitrariness of the choice of cut-off values used (ODonoghue and Gleave, 2004). This paper proposes a systematic process whereby utilising the confidence interval method one is able to develop a statistically based interpretive rationale to the inclusion or exclusion of industries for further analysis. The process is generated on the principle that statistical variance should be considered when choosing industries for inclusion, not simply how close they are to an arbitrary cut-off value based on their LQ values. Essentially, this process allows the assessment of the potential of borderline industries to be included or excluded from further analysis. Figure 1 illustrated that the ranges of the confidence intervals associated with the industry LQs varied markedly (over the same 1001 a per cent confidence level). It follows that, for those industries whose LQ values are close to the considered cut-off value, the associated confidence intervals, beyond some 1001 a per cent confidence level value, may include the cut-off value. This means that parts of the associated confidence intervals are on the opposite side of the cut-off value from their respective point LQsthe implication being that an industrys LQ confidence interval suggests a possible contradiction in what the LQ value implies, relative to the considered cut-off value. Figure 2 further illustrates the issue, examining the industries which have a potential for contradiction in their initial designation based on their LQ confidence interval (now considered up to the 99 per cent confidence interval).8 In the industry potential graph reported in Figure 2, each point represents an industry, which has a point LQ value below or above the cut-off value 1.25, but which has a concomitant
confidence interval, beyond a value of 1001 a per cent, which goes above or below the considered cut-off value respectively. The implication, following on from this, is that the industry could have been inaccurately included or excluded from further analysis (since there is a part of the confidence interval suggesting the opposite of what the industrys LQ value was implying). The vertical axis in the industry potential graph is the LQ scale and, for each industry point shown, their value against the vertical axis is their point LQ value. Along the horizontal axis, the level of potential is shown (least 1001 a per cent confidence level), here from 0 per cent to 99 per cent, signifying the confidence level at which the considered cut-off value is included in an industrys LQ confidence interval. Here, the closer to 99 per cent the less potential (left to right in decreasing potential). For example, Other Ceramic Products (SIC code 23490) has a LQ value of 0.917 (on the vertical axis) and a confidence level potential value of 91.8 per cent (on the horizontal axis) indicating it would require the consideration of the LQs confidence interval at the 91.8 per cent confidence level or above for it to suggest potential for contradiction of its initial designation. There are two points to note from Figure 2. First, industries occurring above or below the line need to be considered separately: above linethese industries have the potential to be excluded even though they were originally included based on the considered LQ cut-off value; below linebecause these are industries which have the potential to be included even though they were originally excluded based on the considered LQ cut-off value. Secondly, the scattered nature of the points, either side of the cut-off line, is due to the variations in confidence interval ranges as shown in Figure 1 (different-sized trumpets). This is a consequence of the different levels
10
Figure 2. Industry potential graph, with example LQ cut-off value of 1.25.
of LQ variance found amongst the considered industries (see, for example, Table 2). In this analysis, considering the 218 Welsh manufacturing industries and up to a 99 per cent confidence interval, out of the 50 of these whose LQ values are above the cutoff value 1.25, six have the potential to fall below based on their associated confidence intervals (for different levels of 1001 a per cent), and out of the 168 whose LQ values are below 1.25, 12 have the potential to go above the cut-off value (examples of these industries are shown in Figure 1 labelled with their SIC codes). Clearly, these 18 industries might be further considered. The industry potential graph provides decision-makers with a tool to assess the potential of industries to change their initial designation. To utilise the graph in Figure 2, decision-makers read, from left to right, the order of the points they encounter, which denotes the decreasing potential for the associated industries to be considered, in terms of the possible contradiction of their initial designation. There is an important point here: the industry order of potential shown in Figure 2 is not simply based on the order of the industries LQ values moving away, either down or up, from the considered cut-off value. As well as the nearness of an industrys LQ value to the considered cut-off
value (on the vertical axis), their potential also needs to be considered (on the horizontal axis). To illustrate this point, take the two industries highlighted earlier. Other Ceramic Products, labelled 23490, has an LQ value further away from the cutoff value 1.25 than the other, Electricity Distribution etc., labelled 27120. It may not be the case that an industry with an LQ value further away from the cut-off value should be thought of any less than an industry with an LQ value closer; it is the assessment of its associated potential that is also a factor (based on 11001 a per cent confidence level). For these two industries, the larger variance associated with Other Ceramic Products means that its 1001 a per cent confidence-levelbased confidence interval goes over the cut-off value 1.25 for a smaller 1001 a per cent confidence level than that of the industry Electricity Distribution etc. with an LQ value nearer the cut-off value. The example results and inference on industries potentials for inclusion or exclusion from further analysis so far described here are referenced on the specific cut-off value of 1.25. Moreover, the industry potential graph shown in Figure 2 is specific to the cut-off value 1.25; it follows that, if a different cut-off value were to be considered, then a new associated industry potential
11
Figure 3. Industry potential graphs, with example cut-off values of 1.75 (above) and 0.75 (below).
graph, like that shown in Figure 2, would have to be considered. Figure 3 displays the industry potential graphs for two further theoretical cut-off valuesnamely, 1.75 (Held, 1996) and 0.75 (Bergman and Feser, 1999). In Figure 3, the two industry potential graphs shown reveal the potential for industries whose LQ associated variances imply that they are possibly worthy of further consideration, based on the cut-off values 1.75 and 0.75. The results show different numbers of industries, above and below the respective cut-off values, from those reported in Figure 2. Instead of choosing an arbitrary or extant-literature-based cut-off value, this analysis suggests that consideration of a range of cut-off values, and also the concomitant variances of the established industry
LQ values, be taken into account. This is an important pointnamely, that the prescribed systematic process will naturally consider a continuous range of cut-off values rather than the typical single cut-offvalue approach. Figure 4 graphically displays the relationship over a range of cut-off values and the subsequent number of industries included or excluded in designation terms based on different cut-off values in this range. Using this, decision-makers can make an initial, more subjective, choice of cut-off value. This means choosing a cut-off value which offsets the overinclusion or underinclusion or exclusion of industries. In Figure 4, the two curves represent, for a specific cut-off value shown on the horizontal axis, the number of industries which would be included or excluded, based on whether their LQ values are above or below
12
Figure 4. Industry inclusionexclusion graph.
Figure 5. Numbers of potential industries identified in inclusion or exclusion groups, depending on considered cut-off value.
the respective cut-off value. With 218 industries considered here, the inclusion and exclusion lines monotonically decrease and increase from 218 and 0 respectively, as the considered cut-off value increases. It is of note that the range of considered cutoff values is from zero to around 16; the upper bound of near 16 was necessary due to one industry having an LQ value of near 16. This graphical information can be taken further over the same range of cut-off values as considered in Figure 4. Figure 5 shows the relationship between a considered cut-off value and the number of industries which could be incorrectly designated for inclusion
or exclusion, termed here (see Figure 2) as having the potential to change (contradict) their initial designation. In Figure 5, the two graphs report the numbers of industries whose LQ values are above (top graph) and below (bottom graph) the considered cut-off value indicated on the horizontal axis, but have the potential to be on the other side of the cutoff value. When combined with the information in graphs such as Figures 2 and 3, and the choice of a cut-off value, this graphical information can give valuable insight into industry designation and potential. It follows, using the information contained, such as that shown in Figures 2, 3, 4 and 5,
13
that a systematic process can be constructed elucidating the impact of cut-off-value choice and industry designation. To summarise, the goal of this overall process is to provide an understanding of cut-off-value choice and the resulting impact on the numbers of industries considered for inclusion and exclusion. The process also highlights which industries are borderline cases in which case their potential may need to be reconsidered, as follows. First, Figure 4 offers the decision-maker two possible options to begin the process. They can ascertain the chosen cut-off value by selecting the number of industries they wish to include or exclude from further analysis. Alternatively, they can initially select a cut-off value and establish how many industries will be included or excluded from further investigation. Secondly, using the selected cut-off value from Figure 4, Figure 5 offers the ability to establish how many industries within the two designations (inclusion or exclusion) have the potential to have their designation changed. Finally, after the consultation of Figures 4 and 5, and after the selection of a cut-off value, an industry potential graph can be constructed, similar to those shown in Figures 2 and 3, reporting the possible order in which the industries have the potential to change their designation (based on inclusion of cut-off value in their associated confidence interval around their LQ value). As described previously, this ordering is based on moving left to right across the graph. In summary, the systematic process outlined allows a consistent robust technique for the consideration and interpretation of LQ values. By incorporating a diagnostic element into the interpretation of LQs, some of the problems of arbitrariness associated with more conventional analyses are avoided.
5. Conclusions
The context of this paper was the continued use of location quotients (LQs) for decisionmaking purposes, particularly with respect to the establishment of designation sets in industry cluster analysis. Of practical concern was the use of arbitrary cut-off values to establish industries into two designation setsnamely, inclusion and exclusion from further analysis. This issue has been a long standing concern in the literature (see ODonaghue and Gleave, 2004). While there is a literature that questions the wisdom in selecting key clusters of industries, there are parallel concerns on the methods to accomplish this selection of industries on which to target policy resources. This paper suggests that where such policies are being adopted, real care needs to be taken in selecting designation sets of industries based on arbitrary cut-off values. However, the paper aimed to go one step further and establish a test of significance for employment-based LQs. This test obviates against the use of arbitrary cut-off values for LQ-based industry designation. Moreover, we argue that undertaking this more rigorous analysis permits greater precision in the definition and identification of spatial specialisations. Simply using the traditional point estimates for LQs provides little indication of the significance attatched to such estimates. Furthermore, this contribution provides a systematic process through which the analyst can develop a statistically based rationale for a designation set of industries adopted within a cluster or key-sector policy. Importantly, the process described permits the analyst to consider the potential for borderline industries to be included or excluded from designation sets and provides a ready visualisation of the technique for clearer interpretation purposes. The method derives answers that are simple in calculation and ease of use, and
14
yet which demonstrate the variance that is overlooked when using the standard formulation of the LQ. In short, we believe that future studies using LQs need to be a little more transparent in the limits of the approach and more ready to acknowledge the ways in which uncertainty around estimates and resulting decision choices can be better accounted for. Future research might include an explicit analysis of the determinants of the range of the confidence interval associated with an LQ. Furthermore, there is a need to consider how far the better understanding of uncertainty around LQ estimates can be incorporated into techniques that are commonly used to regionalise inputoutput tables for economic and geographical analysis.
Notes
1. Using Google scholar, it is possible to identify over 700 articles using the LQ methodology published in 2008 and 2009 alone. 2. The paper here utilises the employmentbased LQi.e. for industry i
5. This work utilised 5-digit industry data from the 2010 Annual Business Inquiry, Office of National Statistics, UK, that the authors feel provides a good coverage of the UK and Wales manufacturing base. This list is not exhaustive, but was used to illustrate the usefulness of the technique. 6. For the purposes of this illustration, the de Propris (2005) form of the LQ has been used, but other forms would yield similar results. 7. The 1.25 cut-off value is used as an example because this was adopted in DTI (2001), but later we show pictorially the consequences of changing cut-off values. 8. The reason for 18 industries shown in Figure 2 and only 15 shown in Figure 1 is that in Figure 2 the upper limit on the considered confidence level is 99 per cent in contrast to 95 per cent in Figure 1. The differences in confidence levels employed are only to allow the clarity of the associated graphs to be maintained.
Funding Statement
This research received no specific grant from any funding agency in the public, commercial or notfor-profit sectors.
xi LQi = ni
x n
References
Bergman, E. M. and Feser, E. (1999) Industrial and Regional Clusters: Concept and Comparative Applications. Web book, Regional Research Institute, West Virginia University (http://www.rri.wvu.edu/WebBook/Bergman Feser/contents.htm). Bergman, E. M. and Feser, E. J. (2000) National industry cluster templates: a framework for applied regional cluster analysis, Regional Studies, 34(1), pp. 119. Beyene, J. and Moineddin, R. (2005) Methods for confidence interval estimation of a ratio parameter with application to location quotients, BMC Medical Research Methodology, 5(32), pp. 17. Bishop, P., Gripaios, P. and Bristow, G. (2003) Determinants of call centre location: some evidence for UK urban areas, Urban Studies, 40(13), pp. 27512768.
where, the numerator term is employment in the industry divided by the national employment in the industry; and the denominator is the reference area employment over national employment. 3. Although employment data are being using in this illustration, this method could also be applied to any estimated economic data used to calculate location quotients such as, unit or GVA data. 4. Traditional industrial specialisation analysis has utilised sectors. Some have argued that commodities are a more accurate way of identifying these agglomerations nder, 2011). UK official surveys do (Voigtla not collect employment data by commodity groups, but rather industry groups.
15
Boix, R. and Galletto, V. (2009) Innovation and industrial districts: a first approach to the measurement and determinants of the I-district effect, Regional Studies, 43(9), pp. 11171133. Brand, S., Hill, S. and Roberts, A. (1998) Welsh InputOutput Tables for 1995. Cardiff: University of Wales Press. Carroll, M., Reid, N. and Smith, B. (2008) Location quotients versus spatial autocorrelation in identifying potential cluster regions, Annals of Regional Science, 42(3), pp. 449463. Chiang, S. (2009) Location quotient and trade, Annals of Regional Science, 43(2), pp. 399414. Clayton, D. and Hills, M. (1993) Statistical Models in Epidemiology. Oxford: Oxford University Press. Crawley, A. and Pickernell, D. (2012) An appraisal of the European cluster observatory, European Urban and Regional Studies, 19(2), pp. 207211. Czamanski, S. (1974) Study of clustering of industries: Halifax, Nova Scotia. Institute of Public Affairs, Dalhousie University, Canada. DTI (Department of Trade and Industry) (2001) Business clusters in the UK: a first assessment. DTI, London. Feser, E. and Isserman, M. (2009) The rural role in national value chains, Regional Studies, 43(1), pp. 89109. Frances, C. J. de and Titus, R. M. (1993) Urban planning and residential burglary outcomes, Landscape and Urban Planning, 26, pp. 179191. Gibson, J., Miller, M. and Wright, G. (1991) Location quotient: a basic tool for economic development analysis, Economic Development Review, 9(2), pp. 6568. Haig, R. M. (1926) Toward an understanding of the metropolis, Quarterly Journal of Economics, 40, pp. 421433. Held, J. R. (1996) Clusters as an economic development tool: beyond the pitfalls, Economic Development Quarterly, 10(3), pp. 249261. Isard, W. H., Langford, T. W. and Romanoff, E. (1968) Philadelphia region inputoutput study. Working Papers, Regional Science Research Institute, Philadelphia, PA. Isserman, A. (1977) The location quotient approach to estimating regional economic
impacts, Journal of the American Institute of Planners, 43(1), pp. 3341. Kumral, N. and Deger, C. (2006) An industrial cluster study: as a basis for the Aegean regions development policy. Working Paper No. 06/01, Department of Economics, Ege University. Malmberg, A. and Maskell, P. (2002) The elusive concept of localization economies: towards a knowledge-based theory of spatial clustering, Environment and Planning A, 34(3), pp. 429449. McCann, P. and Dewhurst, J. (1998) Regional size, industrial location and inputoutput expenditure coefficients, Regional Studies, 32(5), pp. 435444. Metraux, S., Caplan, J., Klugman, D. and Hadley, T. (2007) Assessing residential segregation among Medicaid recipients with psychiatric disability in Philadelphia, Journal of Community Psychology, 35(2), pp. 239255. Miernyk, W. H. (1976) Comments on recent developments in regional inputoutput analysis, International Regional Science Review, 1, pp. 4755. Miller, R. E. and Blair, P. (2009) InputOutput Analysis: Foundations and Extensions, 2nd edn. New York: Cambridge University Press. Moineddin, R., Beyene, J. and Boyle, E. (2003) On the location quotient confidence interval, Geographical Analysis, 35(3), pp. 249256. Mulligan, G. F. and Schmidt, C. (2005) A note on localization and specialization, Growth and Change, 36(4), pp. 565575. ODonoghue, D. and Gleave, B. (2004) A note on methods for measuring industrial agglomeration, Regional Studies, 38(4), pp. 419427. Oehlert, G. (1992) Note on the Delta method, American Statistician, 46, pp. 2729. Openshaw, S. and Taylor, P. J. (1979) A million or so correlation coefficients: three experiments on the modifiable areal unit problem, in: N. Wrigley (Ed.) Statistical Applications in the Spatial Sciences, pp. 127144. London: Pion. Porter, M. E. (1990) The Competitive Advantage of Nations. Basingstoke: Macmillan. Porter, M. E. (1999) Clusters and the new economics of competition, Harvard Business Review, November/December, pp. 7790. Porter, M. E. (2000) Location, competition and economic development: local clusters in a
16
global economy, Economic Development Quarterly, 14, pp. 1524. Propris, L. de (2005) Mapping local production systems in the UK, Regional Studies, 39(2), pp. 197211. Schaffer, W. (1999) Regional Impact Models. Web book, Regional Research Institute, West Virginia University (http://www.rri.wvu.edu/ regscweb.htm). Silcocks, P. (1994) Estimating confidence limits on a standardised mortality ratio when the expected number is not error free, Journal of Epidemiology and Community Health, 48, pp. 313317. , Lindqvist, G. and Ketels, C. (2003) lvell, O So The Cluster Initiative Greenbook. Stockholm: Ivory Tower AB. Thomas, G. B. Jr and Finney, R. L. (1996) Calculus and Analytic Geometry, 9th edn. Reading, MA: Addison Wesley. Thrall, G., Fandrich, J. and Elshaw-Thrall, S. (1995) Location quotient: descriptive geography for the Community Reinvestment Act, Geo Info Systems, 5(6), pp. 1822.
Tonts, M. and Taylor, M. (2010) Corporate location, concentration and performance: large company headquarters in the Australian urban system, Urban Studies, 47(12), pp. 26412664. nder, N. (2011) Skill bias magnified: interVoigtla sectoral linkages and white-collar labor demand in U.S. manufacturing. Economic Working Paper Series, University of California, Los Angeles. West, G. R. (1986) A stochastic analysis of an input output model, Econometrica, 54(2), pp. 363374. Wong, N. H. and Chen, Y. (2009) Tropical Urban Heat Islands: Climate Building and Greenery. New York: Taylor and Francis. World Economic Forum (2007) Global competitiveness report 20072008 (http://www.gcr. weforum.org/).
Appendix
This appendix reports the trumpet shapes for all 218 defined industries considered (see Figure A1).
Figure A1. LQ confidence intervals for SIC manufacturing industries in Wales, with varying confidence levels of 100(1 a) per cent, between 0 per cent and 95 per cent.

Mapping Local Production Systems in The UK

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Mapping Local Production Systems in The UK

Diunggah oleh

Hak Cipta:

Format Tersedia

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN.

LOS DISTRITOS INDUSTRIALES

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

3. Los distritos industriales en la poca de Marshall

LOS DISTRITOS INDUSTRIALES

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

Mapa 1. Algunos distritos industriales en Inglaterra y Gales (1892)

LOS DISTRITOS INDUSTRIALES

4. Sistemas productivos locales y distritos en Gran Bretaa

4.1. Metodologa para el anlisis espacial

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

LOS DISTRITOS INDUSTRIALES

E (sll , manuf .) E (nat., manuf .) E (nat.) E (sll )

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

LOS DISTRITOS INDUSTRIALES

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

1) Intensidad manufacturera 2) empleados <100 empleados <300 empleados >300

Sector primario Sector secundario

3) Especializacin sectorial Sector primario 4) Espec.+ PI Espec.+ PMI Sector primario

Sector secundario Sector secundario Sector secundario

Sector secundario Sector secundario 5 Sector primario 28 Sector secundario 5 6

Espec.+LI Sector primario No SLT 47 31 80 22 73

Fuente: ONS. Elaboracin propia.

Tabla 2. Tipos de sistemas locales (1997)

SL no manufactureros SL de PMI SL de PMI

PROTODISTRITOS PROTO-DISTRITOS NO-ESPECIALIZADOS

LOS DISTRITOS INDUSTRIALES

Fuente: ONS (2003). Elaboracin propia.

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

Fuente: ONS (2003). Elaboracin propia.

LOS DISTRITOS INDUSTRIALES

5. Los distritos industriales

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

Tabla 3. Los distritos industriales en Gran Bretaa (1997)

LOS DISTRITOS INDUSTRIALES

Tabla 4. Distritos industriales en Gran Bretaa (2002)

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

Continuacin Tabla 4. Distritos industriales en Gran Bretaa (2002)

LOS DISTRITOS INDUSTRIALES

LOS DISTRITOS INDUSTRIALES EN EL MUNDO ANGLOSAJN. EL CASO BRITNICO / LISA DE PROPRIS

LOS DISTRITOS INDUSTRIALES

udies. Vol. 39.2. pp. l')7-211. Apnl 2t)()5

Mapping Local Production Systems in the UK: Methodology and Application

JEL classification: R12

01)34-3404 pnnt/13WI-()5<91 online/05/020147-15 2005 Regional Studies Association li ttp://www.rfgioiial-studies-assoc.ac.uk

Mappinji Local Production Systems in the UK

Industrial specialization and firms' size

Pros aud cons of the spatial diaj^noslic analysis

LI'S of large firms

Fig. 1. huer-indusfry linkages

Mapping Local Production Systems in the UK

Jiihlc 2. Types of ]oai\ production systems (LPSs)

Number of travel-rn-work areas

Fig. 2. Types of heal production systems (LPSs) in the UK

Mapping Local Production Systems in the UK

Fig. 4. Local production systems ofsmall and medium-sizefrms

in the UK by travcl-to-work area. For abbreviations, sec Fig. 2

Mapping Local Production Systems in the UK

Urban Studies http://usj.sagepub.com/

>> OnlineFirst Version of Record - Dec 20, 2012 What is This?

Downloaded from usj.sagepub.com at UNIV OF TEXAS ARLINGTON on April 7, 2013

ANDREW CRAWLEY ET AL.

2. The Usefulness of Location Quotients: A Review