Theories of Local Economic Growth (Part 2): Model Specification and Empirical Validation

Abstract

This study is an attempt to produce a theoretically informed econometric analysis of dynamic regional economic performance. The first paper in this two-part study highlighted the problems and possibilities of translating the propositions contained in six ‘soft’ theories of local economic growth into measurable dimensions and sets of proxy variables that are hypothesized to determine local economic growth. This second paper focuses on issues of econometric-model design and empirical validation. Given the practical limitations imposed by data availability, the ‘soft’ theories of local economic growth are modeled by a simple conditional ‘gap’ convergence model in which prevailing regional unemployment relativities are determined by a common trend in unemployment and a set of regionally specific variables. The empirical validity of the competing ‘soft’ theories of local economic growth is evaluated by applying test restrictions imposed by the ‘soft’ theories to a general model specification containing eight regionally specific variables that have been identified as potential drivers of local economic growth. The econometric modeling of Australian data suggests that the processes of regional economic performance might be somewhat different from those specified in individual theoretical models. Significantly, the role of local ‘enterprise culture’ is revealed—built on specialization, technological leadership, human resources, and the local integration of firms—though significant caveats are attached to the roles of access to information, institutional support, and interregional trade as promoters of local economic growth.

Labor Market Dynamics and Technology Adoption: Theory and Empirical Evidence

“…A similar sectoral selection is proposed in several other studies (e.g. Plummer and Taylor, 2001a;2001b;Andonelli, 1990;Alderman, 2004;Alderman and Fischer, 1992).…”

Modelling Local and Regional Economic Development in Turkey: A “Curate's Egg”

“…where εit,t‐1 is assumed to be normally distributed with an expected value of zero, errors independent of the set of explanatory variables, uniform variance, uncorrelated, and with fixed regressors in repeated samples. To establish whether this general model captures any specific information that is not embodied in the seven theoretical models, a variance encompassing procedure is employed to test the validity of the restrictions that are imposed on this general model by the seven models (see Plummer and Taylor ). The encompassing model is defined as the model in which variance dominates the set of alternative model specifications in the sense that the other models contain no information capable of improving the model (Hendry and Mizon ; McAleer ).…”

“…where ro is the residual sum of squares of the restricted model and r is the residual sum of squares in the unrestricted model (Maddala ; Plummer and Taylor ). According to the probability ratios of the theoretical models, for both the full and reduced sample cases, there is evidence to suggest that the assumption “the set of linear restrictions imposed on the general model by all the models are correct” can be rejected.…”

Should Economic Geographers Count?

“…Plummer and Taylor (2001a,b) demonstrate effectively the utility of statistical analysis for choosing among various ‘soft’ theories about local economic growth. Their theoretical models were informed by qualitative and quantitative research on industrial districts, learning regions, and networks.…”