Does Collective Wage Bargaining Restore Efficiency in a Search Model with Large Firms?


  • We are especially grateful to two anonymous referees and the editor, Steve Pischke, whose excellent and detailed feedback helped us to greatly improve the manuscript. We thank Daron Acemoglu, David Andolfatto, Rüdiger Bachmann, Tobias Böhm, Matthias Doepke, Renato Faccini, Andreas Haufler, Christian Holzner, Gerhard Illing, Leo Kaas, Philipp Kircher, Jochen Michaelis, Espen Moen, Joao Montez, Sven Rady, Martin Schneider, Daniel Schultz, Robert Solow, Michèle Tertilt, Mark Trede, Achim Voß and seminar participants at the Universities of Munich, Münster, and Regensburg, the 3rd EALE – SOLE joint meeting, and the 2010 annual congresses of the EEA, EARIE and the German Economic Association for valuable comments and suggestions. Moreover, we thank Brian Bloch for editing and proofreading the article. Bauer gratefully acknowledges financial support from the German Research Foundation (DFG) through GRK 801 and Grant No. HA 3195/8.


Existing search and bargaining models show that firms hire an inefficiently large number of workers. We ask whether decentralised collective wage bargaining may result in a second-best allocation. Collective bargaining restores efficiency when the bargained wage is independent of employment; conditions that we characterise. Firms then behave as if collective bargaining was over both wages and employment, thus linking the large-firm search and bargaining environment to the efficient bargaining model (McDonald and Solow, 1981). Under more realistic conditions, workers can bargain for a share of output, so that the wage is then a function of employment. In equilibrium, firms are too large and firm entry is inefficient.