Deferred Compensation in Multiperiod Labor Contracts: An Experimental Test of Lazear's Model

Abstract

This paper provides the first experimental test of Edward Lazear's (1979) model of deferred compensation. We examine the relationship between firms' wage offers and workers' effort supply in a multi-period environment. If firms can ex ante commit to a wage schedule with deferred compensation, workers should respond by supplying sufficient effort to avoid dismissal. We contrast this full-commitment case to controls with no commitment and computer-generated wages in order to examine the roles of monetary incentives, social preferences, and reciprocity. Finally, we examine a setup without formal commitment, but where firms can build a reputation for paying deferred wages. (JEL D86, J22, J31, J33, J41)

Repeated moral hazard and contracts with memory: A laboratory experiment

“…Taken together, in our view the recent findings of Huck et al (2011) and the results reported in the present paper suggest that more experimental work on the incentive effects of long-term contracts guided by contract-theoretic research might be very promising.…”

“…To our knowledge, Huck et al (2011) is the only other experiment exploring the intertemporal allocation of wages. They test Lazear's (1979) theory of deferred compensation (i.e., wage profiles that are increasing over a worker's lifetime; cf.…”

“…Akerlof and Katz, 1989;Lazear, 1981). While the models motivating the experiments are quite different, 18 Huck et al (2011) also find that agents respond strongly to monetary incentives and provide support for 18 In the deferred compensation model, the principal can always deduce effort from output (so that with full commitment the first-best solution could be implemented with a Maskin mechanism; see Maskin, 1999;Maskin and Sjöström, 2002). Moreover, the wage in a given period does not depend on the output of that period, but the agent is dismissed with a certain probability after a period in which he exerted low effort.…”

“…Following the approach taken by Huck et al (2011), we can characterize which decisions are Fehr-Schmidt rationalizable and which decisions cannot be explained by any values of   and   (see Appendix A for details).…”

Evaluating replicability of laboratory experiments in economics

“…(33)(34)(35)(36)(37)(38)(39)(40)(41)(42)(43)(44)(45)(46)(47)(48)(49)(50) The deadline for inclusion in the study was that the paper should be published or posted as accepted/in press at the website of the journal at August 1, 2014. There are a number of different possible experimental designs. The most "classical" design is the randomized controlled trial (RCT) design, where participants are randomly allocated to two or more treatments and the outcome is compared between treatments.…”