With respect to type C companies, the location quotient appears to be significant, but negatively
correlated to the municipal IVA. Type C companies are small size business that are spatially scattered
and are not exporters. Hence these companies should be expected to have a limited influence on the
scale of municipal IVA. In fact, this is what has been observed: higher municipal IVA figures are
associated to a smaller concentration of type C companies (negative QLC coefficient).
Such “exclusion” of type C companies from large industrial agglomerations may be related to
the difficulties experienced by type C companies in sharing economic spaces with leading industrial
companies (type A companies and, secondarily, type B companies). The high costs associated with
urban agglomerations can only be supported by those companies that do add more value to their
products (through product and/or process innovation) and this is not, by definition, the case of type C
companies. However, in order to remain active, such companies tend to be located in smaller, more
scattered industrial centers where costs are lower than in urban areas. To have access to major markets,
these companies (or their customers) must bear the costs of transportation. Exceptionally, type C
companies are found present in major agglomerations, occupying interstices of the metropolitan space
and offering products of low unit price and high transportation cost, including some standardized food
products.12
4.3. International Insertion - Export and Import
International trade of major industrial agglomerations is highlighted in Tables 7 and 8. To
analyze this aspect, we considered the total exports and imports of each municipality as a synthetic
measure of competitiveness of the industrial agglomerations.
In Brazil, imports and exports diverge in terms of technological content, sectoral structure and
competitiveness. The same tends to apply to how industrial space is organized (Haddad & Azzoni,
1999). The estimated model in Table 7 looks into some of these features. In this case, the dependent
variable is the total industrial export per municipality.
12 Lemos et al (2005-b) present a more detailed studies on the determinants of the location of firms A, B and C and their
spatial interactions. Lemos et al (2005-c) presents another and similar study that stresses the locational differences between
domestic and foreigner firms.
20