Geographical Agglomeration in Australian Manufacturing

Abstract

Leahy A., Palangkaraya A. and Yong J. Geographical agglomeration in Australian manufacturing, Regional Studies. This paper investigates geographical agglomeration of Australian manufacturing industries from 1993–1994 to 1996–1997. It is found that although agglomeration increased substantially during the period, it was still less than in other developed economies such as the United States, the UK, France, and Ireland. Using a common industry classification and comparable spatial units, little correlation is found between Australian industries' agglomeration and that of Belgium, Ireland, and Portugal. However, for industries at the top and bottom ends of the agglomeration scale, a common pattern reflecting common drivers such as proximity to raw materials and production facilities is observed.

Leahy A., Palangkaraya A. et Yong J. Le taux d'agglomération géographique des industries australiennes, Regional Studies. Cet article cherche à étudier le taux d'agglomération géographique des industries australiennes de 1993–1994 à 1996–1997. Il s'avère que le taux d'agglomération reste inférieure à ce qu'elle ne l'est dans d'autres économies développées, telles les Etats-Unis, le Royaume-Uni, la France et l'Irlande, malgré qu'elle s'intensifie sensiblement au cours de la période en question. A partir d'un classement industriel commun et des délimitations géographiques comparables, il n'est qu'une corrélation faible entre le taux d'agglomération des industries australiennes et ceux de la Belgique, de l'Irlande, et du Portugal. Cependant, pour les industries en haut de et en bas de l'échelle des taux d'agglomération, on note une distribution commune qui reflète des forces motrices, telles la proximité des matières premières et des installations de production.

Taux d'agglomération Concentration géographique Industries australiennes

Leahy A., Palangkaraya A. und Yong J. Geografische Agglomeration im australischen Produktionswesen, Regional Studies. In diesem Beitrag untersuchen wir die geografische Agglomeration der produzierenden Industrien Australiens im Zeitraum von 1993–1994 bis 1996–1997. Wir stellen fest, dass die Agglomeration in diesem Zeitraum zwar erheblich zugenommen hat, doch immer noch niedriger ausfiel als in weniger entwickelten Wirtschaften, wie z. B. in den USA, Großbritannien, Frankreich oder Irland. Mit Hilfe einer gemeinsamen Branchenklassifikation und vergleichbarer räumlicher Einheiten stellen wir fest, dass zwischen der Agglomeration der Industrien von Australien und der von Belgien, Irland und Portugal wenig Korrelation besteht. Allerdings ist für Industrien am oberen und unteren Ende der Agglomerationsskala ein gemeinsames Muster zu beobachten, in dem sich gemeinsame Einflussfaktoren wie z. B. eine Nähe zu Rohstoffen und Produktionsanlagen widerspiegeln.

Agglomeration Geografische Konzentration Australisches Produktionswesen

Leahy A., Palangkaraya A. y Yong J. Aglomeración geográfica en el sector manufacturero australiano, Regional Studies. En este ensayo investigamos la aglomeración geográfica de las industrias manufacturas australianas de 1993–1994 a 1996–1997. Observamos que si bien la aglomeración aumentó sustancialmente durante el periodo, todavía era menor que en otras economías desarrolladas tales como las de los Estados Unidos, el Reino Unido, Francia e Irlanda. Usando una clasificación industrial común y unidades espaciales comparables, hallamos poca correlación entre la aglomeración de las industrias australianas y la de Bélgica, Irlanda y Portugal. Sin embargo, para las industrias en los extremos superiores e inferiores de la escala de aglomeración, observamos un modelo común que refleja los desencadenantes comunes tales como la proximidad de materias primas e instalaciones de producción.

Aglomeración Concentración geográfica Sector manufacturero australiano

Keywords:

• Agglomeration
• Geographic concentration
• Australian manufacturing

JEL classifications:

• R11
• R12

Acknowledgements

This paper was funded by an ARC Linkage Grant (Grant Number LP0348635). The authors are grateful to their Linkage Partners the Australian Bureau of Statistics, IBISWorld, Productivity Commission, Austrade, and the Victorian Department of Treasury and Finance for their support. The authors thank John Creedy and two anonymous referees for very helpful comments and suggestions on an earlier draft of this paper.

Notes

Other measures exist (for example, Duranton and Overman, 2005).

Both measures have been shown to provide relatively the same pattern of agglomeration in manufacturing industries (for examples, see Maurel and Sèdillot, 1999; and Maré, 2005). Thus, different results might indicate data quality or other unknown problems.

The authors thank an anonymous referee for pointing this out.

For examples and more detailed discussions on the difference between the two indices, see Maurel and Sèdillot (1999) and Maré (2005).

It is possible for an establishment to consist of a group of plants in different locations, but these multiple locations are always within the same State or Territory. In practice, the majority of establishments operate at only one location (ABS, 1997a). A location is defined at the Statistical Location Area level, which is explained below in this section.

The raw census data imply annual average entry and exit rates of 22% and 16%, respectively. For comparison, Disney et al. (2003) reported annual entry and exit rates of around 18% and 21%, respectively, for manufacturing establishments in the UK during the period 1986–1991. It is worth noting that the Australian data were collected during a period that Australian manufacturing was growing, while the UK data were collected when UK manufacturing was in decline.

SLAs are further divided into Census Districts.

For more details, see ABS (1996) or more recent ASGC documents produced by the ABS.

ABS (1994, 1996) are used as references.

There were some changes in terms of names, codes, and, in certain cases, delimited areas of the SLAs and the higher geographical units (ABS, 1996). The authors do not think the changes are significant enough to cause problems for the findings, but note that some SLAs were reduced in size, while others were enlarged. Unfortunately, the documentation is too vague to allow a concordance to be constructed.

This table is produced based on a series of ABS publications (ABS, 1997b, 1997c, 1997d, 1997e, 1997f, 1997g, 1998a, 1998b). Unfortunately, the size of each location/area is only reported for some States.

The authors thank an anonymous referee for pointing this out.

This increase in agglomeration likely reflects the significant consolidation of the industry which, as in many other developed countries, was facing a steep decline (a 30% per capita reduction) in tobacco consumption (Winstanley et al., 1995).

Unfortunately, due to confidentiality restrictions, the authors cannot provide the actual numbers of establishments.

The authors thank an anonymous referee for pointing this out and for suggesting they study the sensitivity of the results with respect to the spatial unit.

Note that the number of establishments in 1996–1997 is different from 1993–1994. For brevity, Table 8 is constructed based on the number of establishments that existed in 1996–1997. Similar results were obtained when Table 8 was constructed using the number of establishments that existed in 1993–1994.

Though the numbers based on the SD measures seem to suggest a slight tendency towards less agglomeration, in relative terms it can probably be stated that the SD measures exhibit a lack of movement.

For example, there are only two and three SD areas for the Australian Capital Territory and Northern Territory, respectively.

Note that Australia's results for 1993–1994 were even lower.

See note 22.

For example, ANZSIC 2411 is described as ‘Paper stationery manufacturing’, while 2412 is described as ‘Printing’. The concordance indicates that ANZSIC 2411 is equivalent to the combination of parts of ISIC 2109 (Manufacturing of other articles of paper and paperboards) and parts of ISIC 2221 (Printing), while ANZSIC 2412 is equivalent to parts of ISIC 2221. In this case, ISIC 2109 is assigned to ANZSIC 2411 and ISIC 2221 is assigned to ANZSIC 2412.

Barrios et al. (2008) indicate that their common classification is ISIC Rev. 1.0. However, after comparing their tables and the classification at the United Nations Statistical Division's Classifications Registry (http://unstats.un.org/unsd/cr/registry/), the present authors believe that their classification corresponds to ISIC Rev. 2.0.