Applying an Industrial Diversification Decision Model to Small Regions

Diversification models based on the analysis of regional industrial portfolios were first developed in the 1970s and applied to the problem of minimizing employment instability in U.S. metropolitan areas.Since then proponents of the portfolio approach have asserted that it can be a useful tool in devising both long and short term strategies to guide the growth of a region. However, none of these authors has applied it to small regions such as a county located in a non-metropolitan area. The purpose of this paper is to assess the applicability of the industrial portfolio analysis approach in four non-metropolitan North Alabama counties. The authors contend that the lack of detailed time-series data seriously degrades the utility of this technique, not only in non-metropolitan regions but also in previously studied larger regions. When faced with these constraints, researchers in the past either have employed highly aggregated local data in their models or have substituted national data for missing local data. This leads to a misspecification of the variance-covariance matrix used to optimize the industrial mix of the region and consequently to erroneous policy recommendations.