Abstract
Traditionally, scholars of comparative capitalisms expect a solid link between institutions and outcomes, such as exporting and innovation. For Germany seminal approaches rely on an ideal-typical firm-level profile with strong innovative capabilities and high traditional institutional embeddedness. Current literature on firm-level diversity in Germany exposes that the empirical links between institutions, de facto firm-level profiles, and their outcomes are not well understood. An analysis of 988 German firms reveals no direct link between institutional variables and outcomes. A latent class analysis identifies five innovative capability profiles, which display diverse institutional embeddedness patterns and outcome levels. The prevailing profile conforms to several traditional expectations about the ideal-typical German firm-level profile. However, a second profile achieves similarly high export and innovation rates without traditional institutional embeddedness. The analysis shows the importance of diverse innovative capability profiles for the de facto links between institutions and firm-level outcomes within Germany's economy.
Acknowledgement
The research presented in this article is based on a joint project of the University of Hamburg and the Institute for Employment Research (IAB). The Author would like to thank Jürgen Beyer, Lutz Bellmann and the whole MINO-project team as well as the anonymous reviewers for their valuable feedback.
Notes
1. The term establishment differs from firm or company, because a firm or company can consist of several establishments (plants, subsidiaries, and headquarters). In many cases, the terms establishment, firm, and company coincide – if a firm only has one establishment. For the following analysis, establishments are treated as the primary firm level where practices take place and strategies are conceived.
2. An alternative operationalization of the profiles as independent variables produces similar results. The regression models are not a manifestation of an implied causal direction. The analyses therefore describe the association between institutional variables, the profiles, and their outcomes.
3. Wolfang Streeck recently made a similar argument for the institutional level (see Crouch et al. Citation2005).