The theory of regional economic stability differentials : analysis, reformulation, and empirical evidence

The objectives of this study were to analyze the four theories of regional economic stability differentials--the industrial mix, growth differentials, city size, and industrial diversification hypotheses-to arrive at a more precise formulation of an hypothesis; to develop measures of regional economic instability and industrial diversification which overcome the conceptual and empirical shortcomings of previous measures; to test the reformulated hypothesis that industrial diversification, combined implicitly with city size and industry mix elements, is a factor in explaining regional economic stability differentials; and to compare these empirical results with those of previous studies.

The traditional measures of industrial diversification and regional economic instability were shown to have a number of shortcomings. It was argued that the entropy measure of industrial diversification and the coefficient of variation measure of economic instability overcome these shortcomings.

These two measures were calculated for a sample of l06 metropolitan areas in the U.S. and for three data sets--nonagricultural, manufacturing, and basic employment. Previous diversification measures were also constructed for comparative purposes. The index of economic instability used quarterly total nonagricultural employment data covering the period 1967 through 1976, and the indexes of industrial diversification used 1972 annual average two-digit nonagricultural employment data. These data were obtained from the U.S. Department of Labor, Bureau of Labor Statistics.

Several major conclusions based on a number of empirical tests conducted in this study were apparent. First, the econometric problems of multicollinearity and heteroscedasticity were found to plague regional economic stability models. Second, chi square tests of independence indicated dependence between industrial diversification and regional economic stability. Third, the entropy measure of industrial diversification was shown to perform better than previous measures as an explanatory variable of regional economic stability differentials, especially when adjusted by city size to correct for heteroscedastic error variances. Fourth, industry mix descriptions proved useful supplements to the industrial diversification explanation for cases of small cities specialized in traditionally stable economic activity such as education or public administration.

Quite obviously, this study could not attempt to answer all questions about regional stability differentials. Still to be explored is the question of to what extent does industrial diversification of regions reduce the economic instability of the nation. Second, what are the differences in the timing and duration of regional instability, and what are their causes? Third, to what extent will the availability of data disaggregated at the three- and four-digit levels improve the explanatory power of the industrial diversification variable? Finally, what are the potential uses of the decomposition property of entropy in explaining, either in an empirical or descriptive sense, the differences in regional economic stability?

The results of this study did indicate, however, that a region which desires to reduce the degree of its economic instability, may use industrial diversification as at least a partial means to accomplish this goal.