Working papers

We study relational contracting patterns in Ghanaian labour markets by conducting a gift-exchange game lab experiment, in which subjects interact in a principal-agent setting as workers or employers. In this game, employers make wage offers to workers, who can then choose to accept or reject this offer and, after accepting and being paid, what effort to exert. The employers and workers interact repeatedly over several periods. While in earlier experiments in developed countries relational contracting, in which cooperation is sustained by a threat of punishment of non-cooperative behaviour, emerged naturally (e.g. Brown et al., 2004, 2012), we do not find evidence for this. In particular we do not find conditional reciprocity on behalf of the employers: employers in our experiment do not punish low effort provision. As a result, employers fail to discipline a subgroup of “selfish” workers, resulting in a low average effort and low and often negative employers’ earnings. Set identification of Fehr-Schmidt preferences of the workers shows that the share of “selfish” workers in our experiment is not substantially different from earlier experiments. Introducing competition for workers or a reputation mechanism does not significantly improve workers’ effort.

Firm surveys have shown that labour management in developing countries is often problematic. Earlier experimental research (Davies & Fafchamps, 2017) has shown that managers in Ghana are reluctant to use monetary incentives to motivate workers. This paper presents the results from a gift-exchange game experiment in Ghana in which the worker can make a promise to the employer before a contract is offered (ex ante communication) and in which the employer can send negative or positive feedback to the worker after the worker has chosen effort (ex post communication). The results indicate that feedback can help sustain cooperate behaviour (high effort provision), but only if the wage offered is high enough. Feedback reinforces reciprocity concerns on the behalf of the worker. In particular positive messages (praising) leads to higher effort provision, no significant relation between negative feedback and effort can be found. Promises are related to higher effort, but do not necessarily lead to higher wages.


This paper explores the determinants of firm survival in Ghanaian manufacturing and the contributions of growth and selection to the evolution of the firm size distribution. For this analysis a survey of 1000 firms in Ghana was conducted in 2013, the sample being randomly selected from the firms interviewed in the 2003 Ghanaian National Industrial Census. This survey thus created a two wave panel spanning 10 years and allowed us to track the exit and growth or decline of firms between 2003 and 2013. We find strong differences in exit patterns by region and firm size. We show that both growth and selection played only a small in the evolution of the firm size distribution, contradicting earlier work on Ghana, and suggesting that this evolution is driven by changing patterns of entry. Overall, the picture we paint of manufacturing in Ghana is not a positive one: total employment by firms operating before 2003 decreased from 134 863 in 2003 to 74 319 in 2013, although we cannot explore to what extent this was compensated by new employment by firms that entered after 2003, who were not surveyed.

  • Who Gains From Competition? The Ultimatum Game in a Labour Market Setting in Ghana (with Marcel Fafchamps). Book chapter in CESifo Seminar Series on Social Economics (MIT Press Volume). Latest version here.
  • Who benefits from introducing competition in the setting of an ultimatum game? We introduce a multiplayer version of the ultimatum game to subjects in Accra, Ghana, framed in a labour market setting. In this version three Proposers (employers) can make offers to three Responders (workers) at the same time. Subjects also participate in a treatment without competition. In this treatment one Proposer faces one Responder, just as in the classical ultimatum game. Even though in the competition treatment the number of Responders and Proposers is equal, we find some evidence that the amounts proposed increase in the treatment with competition. A potential explanation for this are bidding effects, where Proposers bid offensively for the Responders with lower reservation payoffs, to increase their chances of having this Responder accept their offer. This bidding increases the amounts that Proposers propose to give to the Responders. This is in particular beneficial to the Responders, who now capture a larger share of the surplus.

  • Between Conquest and Independence: Real Wages and Demographic Change in Spanish America, 1530-1820 (with Leticia Arroyo Abad and Jan Luiten van Zanden). Explorations in Economic History 49 (2), pp. 149-166.
    Available as: CGEH Working Paper.