Conflict, arbitration, and dispute resolution in the German workplace

Abstract

PurposeThis paper investigates the role of arbitration panels in German employment relations. It is the main aim of the analysis to improve our knowledge of the ways through which employment related conflict is formalized within a transparent procedure.Design/methodology/approachHypotheses are generated from a review of relevant publications on German arbitration panels. To test the hypotheses with evidence, the paper uses data from the 2006 WSI works council survey and applies binary logistical regression analysis.FindingsThe statistical analysis reveals that works councils oppose management hostility by way of making more frequent use of arbitration panels. While in a multivariate model the use of arbitration panels could be positively associated with certain attitudes of works councils and management, results for a number of institutional variables were other than expected. While foreign ownership of a particular establishment did not have a significant impact on the use of arbitration panels, firm‐level agreements, as negotiated by unions without the direct involvement of works councils, are associated with a more frequent use of such panels.Research limitations/implicationsThe model applied in the article could not verify establishment age because data on this subject was not available. Future surveys should fill this void.Practical implicationsThe findings of this paper suggest that arbitration panels are a valuable tool in the process of conflict resolution. Because there is little evidence for an excessive use of such panels, future debates on the reform of the Works Constitution Act should consider strengthening this instrument rather than restricting it.Originality/valueWhile data on the use of arbitration panels had already been documented about 20 years ago, the author presents a unique set of data, which for the first time allows the analysis of the use of arbitration panels in all industries within the private sector and is also significant in making it possible to statistically test relevant hypotheses on the usage of such panels.