We present a new methodology to estimate dynamic discrete choice models with aggregate data; the estimation allows for a multi-dimensional state space, but still retains signicant computational benets. We specically build upon the literature pertaining to the dynamic single-agent models with conditional choice probabilities by including both observed and unobserved population state variables in estimation. We demonstrate that the approach performs well in accurately recovering the estimated parameters via Monte Carlo simulations, and that it compares favorably with the current state-ofthe-art methods. We illustrate with an empirical application to assess the impact of dynamics in the digital camera market.

This paper seeks to understand the relational factors that may affect the decisions of both third-party raters and service providers in a setting where service providers compete with one another. We employ laboratory economics experiments to examine how removing anonymity and allowing for repeated interactions between the rater and the service provider impact both the ratings assigned by the rater and the quality levels expended by the service provider. Our methodology enables us to observe the true quality level chosen by a service provider, which allows us to accurately detect any bias in the assessment of the third-party rater. The experimental results show that the decisions of both the rater and the service provider are very sensitive to the relational factors that govern their interaction. When the rater and the service provider know each other’s identities, we observe a high proportion of overrating even though raters earn less monetary rewards for doing so, and the propensity to overrate is even stronger with repeated interactions. Furthermore, the service provider chooses low quality levels. We develop and estimate a model that captures the rater’s psychological trade-off between remaining objective and helping the service provider compete in the marketplace, and the evolution of the service provider’s beliefs about the rater’s preferences. Data are available at https://doi.org/10.1287/mnsc.2018.3082 . This paper was accepted by Juanjuan Zhang, marketing.

We model an expert review system where two producers competing for market share each are evaluated by two raters. Employing economics experiments, the paper examines how the rater’s incentive to provide objective feedback can be distorted in the presence of social ties and different penalty structures for assigning unobjective ratings. The results reject the self-interested model. We find that raters assign more biased ratings to help the producer they know compete, and this distortion is exacerbated when the reputation cost for rating unobjectively is lowered. Counterintuitively, when both of the raters know the same producer, the likelihood of biased ratings drops significantly. To explain the empirical regularities, we develop a behavioral economics model and show that the rater’s utility function should account not only for social preferences toward the producer, but also the rater’s psychological aversion toward favoring a producer more than the other rater. Our findings demonstrate that it is critical for policymakers to be cognizant of the nonpecuniary factors that can influence behavior in expert review systems. This paper was accepted by John List, behavioral economics.

When using group-based commission plans to motivate their sales force, should firms always compensate salespeople based on the average of team members’ sales outcomes? The theory suggests that when team members are heterogeneous in sales abilities, the proposed Maximum contract (where the team output is set by the largest individual sales output) dominates the Average contract (where the team output is determined by the average output of team members) in terms of overall team effort. This is because the stronger team member will exert higher effort under the maximum contract, compared to the average contract, and this increase exceeds the decrease in the weaker team member’s effort. The authors validate the theoretical predictions by employing two laboratory experiments to provide a causal test of the theory, and two randomized field experiments to deliver additional corroborating evidence. The experimental results are overall consistent and broadly confirm the theoretical predictions, pointing to the substantial gains from implementing the maximum contract when team members are heterogeneous in abilities. Interestingly, the weaker team members exert similar effort across the maximum and average contracts, although the theory predicts higher effort under the latter.

No abstract