Modeling regional endogenous growth: an application to the non-metropolitan regions of Australia

Research paper by Robert J. Stimson, Alistair Robson, Tung-Kai Shyy

Indexed on: 16 Apr '08Published on: 16 Apr '08Published in: The Annals of regional science


An exploratory modeling approach to investigate spatial variation in the levels of regional endogenous employment growth and decline over the decade 1991–2001 is developed and applied to an analysis of the non–metropolitan regions (Local Government Areas) in each of the five mainland States of Australia. For the dependent variable, the summation of the regional shift component for change in total employment in major industry sectors1 over the decade 1991–2001, standardized by the size of the labor force at the beginning of the period, is used as a proxy measure of regional endogenous growth. A general OLS model incorporating a set of 27 independent variables (measuring aspects of industry structure, unemployment, occupational structure, population size and growth, human capital, income distribution, and proximity to the coast and the state metropolitan region) is run, followed by a backward iterative statistical procedure to reduce the complexity of the general model by eliminating statistically insignificant variables to arrive at a specific model for each State.