The Emergence of a More Positive View of Workplace Codetermination: Evidence from Some Larger Datasets

Abstract

Studies in the next phase of econometric research were able to draw on large-scale data sets, principally the Hanover Firm Panel (Hannoveraner Firmenpanel) and the NIFA-(Neue Informationstechnologien und Flexible Arbeitssysteme/New Information Technologies and Flexible Work Systems) Panel. The population of the first data set is all manufacturing establishments with at least five employees in the state of Lower Saxony. The actual sample of plants is stratified according to firm size and industry. It comprises 1,025 establishments in 1994, declining to 709 establishments by the time of the fourth and final wave in 1997 because of sample attrition (for a description of this data set, see Brand, Carstensen, Gerlach, and Klodt, 1996; Gerlach, Hübler, and Meyer, 2003). The second survey covers all establishments with 20 or more employees in the German machine tool industry. This panel has eight waves, 1991–1998. The sample base is approximately 6,000 companies, and the realized sample ranges from a maximum of 1,642 in the first year to a minimum of 1,038 in the last year (see Schmidt and Widmaier, 1992; Widmaier, 2001). One advantage of the NIFA-Panel is that it allows for differentiation between “types” of works councils on the basis management assessment of their “attitude” (see below).¹ Unfortunately, this question was only asked in the 1996 wave.