Geographical Agglomeration: the Case of Belgian Manufacturing Industry [ Working Paper 14-02 - 30/12/2002]

Information
Classification

The aim of this working paper is to carry out a descriptive analysis of the geographical concentration of the manufacturing sector in Belgium, from which geographical agglomeration of sectors of activity is analyzed at a fine industrial level, i.e. NACEBEL 4-digit industries.

PDF & Download

Publication (en - 787 Kb)

Authors

Jehan Decrop (A)

A : Author, C : Contributor

Publication type

Working Papers

The Working Paper presents a study or analysis conducted by the Federal Planning Bureau on its own initiative.

To assess the degree of concentration, individual plant data on wage and salary earners is used, with a relatively fine geographical breakdown (districts and townships). The short period of investigation (1997-2000) is informative from a descriptive point of view, but one must remain very cautious when it comes to addressing policy issues, which is not at all the primary aim of this paper.

We further analyze a number of specific problems, which arise when studying concentration patterns at specific geographical and industrial disaggregated levels. First, we try to assess whether broad industries (NACEBEL 2-digit) are concentrated because of the concentration within or across their sub-industries (NACEBEL 4-digit). For most industries, concentration within sub-industries is prevalent, which is an argument in favour of the presence of localization economies. Four industries, however, do highlight significant concentration across sub-industries: textile, clothing, printing and publishing and precision instruments.

We also study the relation between agglomeration and plant scale, which appears to be positive: “hard-core” manufacturing, which generally takes place in large plants, seems to be more geographically concentrated than smaller plants within the same industry.

Finally we study the issue of spatial autocorrelation on a fine geographical scale, namely townships. Autocorrelation tries to assess the degree of value similarity between neighbouring geographical units.

All in all, this paper provides some descriptive features with respect to agglomeration in the manufacturing industry. Although the study gives some hints about potential explanatory factors, the intention is not to build an explanatory framework or to suggest policy measures. Further research into these fields is to be expected.