Reciprocity in Labor Relationships

Numerous economic experiments suggest that a substantial part of individuals exhibit reciprocal preferences. It is also well documented in the economic literature that reciprocity plays an important role in employment relationships. As a source of gift exchange between employer and employee, positive reciprocity offers a prominent explanation for non-minimal wage offers and effort choices beyond the selfishly rational minimum effort. On the other hand, negative reciprocity can result in shirking and sabotage activitiesThe focus of this thesis lies on the significance of reciprocal preferences for labor market outcomes and employment conditions. It comprises three independent research papers and is organized as follows: Section 1 theoretically investigates the labor market success of workers with heterogeneous reciprocal preferences in a competitive search equilibrium. Section 2 proposes a theoretical model to explain hidden costs of control based on eterogeneous reciprocal preferences of workers. The model presented in Section 3 explores the matching between heterogeneous reciprocal employers and workers in a Principal-Agent setting and the corresponding implications for the profitability of different matches. Concluding remarks and perspectives can be found in Section 4.SECTION 1. The model presented in Section 1 examines identical profit-maximizing firms that consist of one job and compete for workers with heterogeneous preferences for gift exchange in a labor market with search frictions. If a firm is successful in hiring a worker, the latter can exert observable but non-verifiable effort to produce output. To motivate the worker, the firm offers a linear incentive contract. With perfect information about workers’ social preferences, e.g., due to screening, the firm can identify the worker it wants to hire. As a consequence, the considered labor market divides into several submarkets, where each sub-market is characterized by the type of worker and the expected wage he is offered by the firm. If firms are identical, they prefer to hire the same type of worker, which results in waiting queues of firms for the best types. I show that in the described labor market reciprocal workers with higher reciprocity concerns are approached by more firms than workers with lower reciprocity concerns. Thus, they find a job more quickly, get higher expected wages, and exert higher efforts compared to low reciprocity types. These results are in line with empirical studies showing that positively reciprocal workers are associated with higher wages, efforts, and a higher probability to be employed compared to selfish workers. The presented model further suggests that labor market regulations in form of binding minimum wages result in lower profits for firms in the corresponding labor market. Consequently, fewer firms engage in the labor market which implies a higher unemployment rate and more long-term unemployed workers. To the best of my knowledge, this model is the first attempt to introduce social preferences as well as linear incentive contracts into a competitive search model. In the presented model workers’ heterogeneous reciprocity concerns affect the provision of incentives in ex ante homogeneous firms and thus represents a different approach to explain output differentials compared to traditional labor search models. In those models productivity differentials are either captured by random firm- or match-specific outputs or heterogeneous worker types by simply assuming that higher types produce higher output or have lower opportunity costs of labor captured by different unemployment benefits. In contrast, the model presented here is based on incentives provided by optimal contracts and workers’ heterogeneous reactions to these incentives.SECTION 2. In Section 2, I consider a moral hazard setting where agents with heterogeneous reciprocal preferences based on Self-Determination Theory as proposed by psychological literature can choose whether to shirk or to work. The agent’s choice is restricted to a binary variable to capture the idea that some jobs leave only little scope for agents to choose their effort. Moreover, depending on the production process, agents might not be able to increase their productivity by exerting more effort.The model results suggest that with observable agent types, selfish agents are always monitored while with reciprocal agents monitoring is not always necessary to ensure effort provision. The reason is that without monitoring, the principal’s contract offer is perceived as more friendly by a reciprocal agent which provides incentives for the agent to reciprocate by choosing high effort. These additional incentives from gift exchange can be high enough to balance shirking incentives such that the trust strategy dominates the control strategy. Whether monitoring and pay are substitutes or complements depends on the agent’s reciprocal preferences. Thus, heterogeneous reciprocity concerns can serveas an explanation for mixed empirical results on the relationship between monitoring and wages. In addition, reciprocity and self-determination have important implications for the complementarity of optimal firm policies. For example, The presented model can explain why white collar jobs tend to offer more discretion than blue collar jobs. The reason is that white collar jobs are associated with higher monitoring expenses compared to blue collar jobs. Higher monitoring costs imply higher distrust if monitoring is introduced. Consequently, an introduction of monitoring affects the principal’s friendliness stronger in white collar jobs compared to blue collar jobs which results in more discretion for the former and less for the latter in equilibrium. Moreover, the interdependence of monitoring and pay results in a complementarity between a firm’s recruiting policy and the discretion it offers to its workers which might explain why firms undertake considerable recruiting efforts that are designed not only to screen for ability but also the willingness to perform well. Even with unobservable agent types the principal can benefit from lower employment costs under full discretion contracts by screening for applicant’s type with separating contracts. Monitoring contracts then always offer higher wages than full discretion contracts because wages under non-monitoring contracts can be reduced to an extent which ensure both, effort provision and self-selection, if agents care enough about not being monitored. This result also holds when introducing a competitive labor market and alternative screening devices.SECTION 3. The model presented in Section 3 is based on a moral hazard setting with one manager (firm) and one worker. Both, manager and worker, can either be selfish or reciprocal. The manager offers a contract that lets the worker participate in output. Depending on matching, the offered contracts can differ among firms, making some firms more profitable than others although the production technology remains unchanged.I show that a purely reciprocal match is characterized by a strictly higher output share for the employee and a strictly higher effort, as compared to a mixed match led by a selfish manager if the worker is positively reciprocal. In contrast, both manager types offer the same low share to a selfish worker because they anticipate that selfish workers will not provide costly gifts. Both manager types can increase their utilities by employing highly positively reciprocal workers, while the workers’ preferences for managers are only based on the size of the offered share of output but not on the type of employer. In the competition subsection, I focus on managers’ preference for the favored reciprocal workers to investigate its implications for the labor market. I introduce a labor market without frictions where the total number of workers available exceeds the total number of vacancies in the market but the share of highly positively reciprocal workers is not sufficient to fill all vacancies. In this setting, competing managers might offer higher shares than without competition to attract highly positively reciprocal workers. The resulting competitive matching allocates those preferred reciprocal workers to reciprocal managers. Consequently, all scarce preferred reciprocal workers will be employed, while selfish workers, followed by unfavored reciprocal (i.e., less positively and negatively reciprocal) workers are only hired if there are still vacant jobs in the market. This result is in line with empirical studies that find that, compared to selfish workers, positively reciprocal workers are associated with higher wages, higher efforts, and a higher probability to be employed, while negatively reciprocal workers are associated with lower efforts, and a lower probability to be employed