A number of integration strategies that embed input-output relations within dynamic econometric models of regional employment determination are examined. By viewing the alternative approaches in the literature as specific cases of restricted estimation a number of important methodological issues associated with model misspecification are identified and examined analytically. Additional insight as to the importance of these issues is provided by means of a series of Monte Carlo simulations. The results suggest that the relative performances of the various approaches towards integration are sensitive to errors associated with the form of the intersectoral linkages, labor-productivity coefficients, and regional purchase coefficients. The previous finding that the embedding strategies provide indicators of the strength of regional intersectoral linkages is shown to be potentially misleading.
ASJC Scopus subject areas
- Geography, Planning and Development
- Environmental Science (miscellaneous)