When No Bad Deed Goes Punished: A Relational Contracting Experiment in Ghana

Abstract

This paper uses experimental methods to study the impact of limited enforcement and reputation on employer-worker relations in labour markets in Ghana. Participants, students recruited from universities in Accra, Ghana are designated as either employers or workers and play a gift-exchange game on a tablet computer. In this game, employers make wage offers to workers, who can then choose to accept or reject and, after accepting, what effort level to exert. Five treatments were used to assess the impact of limited enforcement, competition between employers and reputation. Each participants plays four games, consisting of five trading periods. We find different results from earlier experiments in developed countries: while these experiments have found strong evidence for relational contracting and conditional reciprocity, we do not find evidence for this. We find that a subgroup of workers exerts very low effort levels, but that this low effort of the workers is not punished by employers, who are not responsive in their wage offers to what the workers did previously. As a result, on average, the workers capture most of the profits. Introducing competition or a multilateral reputation mechanism does not significantly improve this.

Citation

Davies, E.; Fafchamps, M. When No Bad Deed Goes Punished: A Relational Contracting Experiment in Ghana. CSAE Economics Department, University of Oxford, Oxford, UK (2015) 32 pp. [CSAE Working Paper WPS/2015-08]

When No Bad Deed Goes Punished: A Relational Contracting Experiment in Ghana

Published 1 January 2015