A Choice-Theoretical Labour-Market Model: Empirical Tests at the Mesolevel

Abstract

This paper contains an analysis of a multiregional labour-supply model for Austria. In the approach suggested here attempts are made to combine the advantages of random-utility-based discrete choice theory and partial reduced-form estimation. Two recursive submodels, the labour-force participation submodel, and the commuting and employment submodel, are developed. Three different types of model specifications at the mesolevel are used to analyse the consequences of choosing a spatial framework of overlapping regional labour markets upon the model parameters, to investigate the usefulness of the model approach in relation to different regionalisation variants, and to carry out a sensitivity analysis with respect to the effects of the model parameters used to delineate the labour-market regions.

Spatial Interaction Models and the Role of Geographic Information Systems

“…The decision on the selection and design of an appropriate spatial framework is a crucial issue in the model building process which affects both the interpretation and acceptability of the models (Openshaw 1977, Baumann et al 1988. GIS offers spatial (dis)aggregation capabilities as well as interactive graphics display possibilities which combined with rudimentary exploratory spatial data analysis provides a considerable power to design zoning systems meeting specific requirements (see Batty and Xie 1994a).…”