The shift-share regression

Abstract

"The so-called Shift-Share-Regression is used to analyse the development of employment. This does not imply a deterministic decomposition such as in classical Shift-Share-Analysis (Dunn 1960; Loveridge and Selting 1998). Instead, Shift-Share-Regression is a powerful and flexible econometric tool, which is especially suitable for testing theory-based hypotheses. In a basic version it was introduced by Patterson (1991) as a method for analysing and testing regional industrial developments. In contrary to the deterministic Shift-Share-Analysis employment development was examined in a linear model. In Patterson's analysis the industrial sector structure was used as the sole determining factor alongside the location effects and the national trend, parallel to those of the deterministic analysis. Möller and Tassinopoulos (2000) extended Patterson's approach by an additional variable for regional concentration. Further theory-based influences were then integrated in various IAB analyses (Blien and Wolf 2002). Some results of these studies are presented below, following an overview of the method." (Author's abstract, IAB-Doku) ((en))

Cite article

Blien, U., Eigenhüller, L., Promberger, M. & Schanne, N. (2014): The shift-share regression. An application to regional employment development in Bavaria. In: K. Kourtit, P. Nijkamp & R. Stimson (Hrsg.) (2014): Applied regional growth and innovation models, p. 109-137. DOI:10.1007/978-3-642-37819-5_6