We analyze information acquisition in a strategic trading model with multidimensional uncertainties. Our main result is that the equilibrium can exhibit multiplicity because of the integer nature of the number of informed traders and the strategic interactions across different types of informed traders. We show that the number of equilibria can be arbitrarily large. Moreover, asymmetric equilibria can arise even in symmetric settings. As information costs rise, fewer traders acquire information, yet prices can become more informative. These results shed light on the instability of financial markets and yield testable empirical predictions. This paper was accepted by Lukas Schmid, finance. Funding: This work was supported by the National Natural Science Foundation of China [Grants 72472171 and 72372011]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.08463 .
Online labor markets have become increasingly popular among companies seeking global access to workers. Some platforms have adopted a contest-based scheme, allowing employers to review multiple submissions before selecting the best one. This practice raises a crucial question: Given a predefined budget, should employers choose a contest or contract scheme for tasks? This study examines the comparative effectiveness of contests versus contracts in achieving better job outcomes, focusing on design tasks characterized by high uncertainty. Using a comprehensive data set from a leading online labor market, we employ multiple empirical strategies, combined with machine learning and text mining techniques, to examine the impact of compensation schemes on job quality. Our results suggest that contract-based jobs yield better performance than contests in the context of ideation tasks. Further analysis, such as examining image similarity among pairwise solutions within a task, identifies “copying” behaviors as a key mechanism undermining the effectiveness of contests. In the public setting of contests, workers can observe others’ submissions and revise their own, leading to imitation rather than innovation. As competition intensity increases, workers have weaker incentives to exert effort, reducing solution quality. Consequently, the negative incentive effects in contests outweigh both the potential positive incentive and parallel effects. Moreover, although information availability can enhance work quality, contests inherently offer limited informational advantages compared with contracts. Taken together, our findings suggest that contract-based schemes are more effective than contests for design tasks, offering important implications for both employers and platform designers in online labor markets. This paper was accepted by Karan Girotra, operations management. Funding: N. Zhang acknowledges the support from the National Natural Science Foundation of China [Grant 72302053]. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2024.07082 .
Competitive environments often leave room for “dirty” practices such as sabotage, retaliation, or dishonesty. We use an online experiment to document aggregate levels and individual differences in the willingness to engage in dirty competition and in the willingness to enter competitions where the opponent may play dirty. We then use the experimental data to validate a set of survey questions that capture willingness to engage in dirty competition above general willingness to compete. We elicit these questions in a representative survey panel and show that willingness to engage in dirty competition is a strong predictor of holding a management or supervisory position and of working in the private—versus the public—sector, but also of worse self-esteem, worse social relationships, and increased feelings of guilt and shame. Men, younger people, and lower-educated people are on average more willing to engage in dirty competition. This paper was accepted by Dorothea Kübler, behavioral economics and decision analysis. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2025.01218 .
Left-digit bias in price perception has been widely documented. We study how it affects firms' joint pricing and quality choices. The bias increases quality at interior prices but decreases quality at 99-ending prices. At 99-ending price points, the bias discourages price adjustment, leading firms to absorb cost shocks through quality changes and thereby, contributing to shrinkflation (quality reductions with nominal prices held fixed). In channels, the bias can either exacerbate double marginalization via a prisoner's dilemma or-because retailers are rigid at 99-ending prices-induce wholesale price adjustments that may strengthen coordination. When manufacturers choose quality, for empirically relevant bias magnitudes, the bias can generate 99-ending retail pricing and raise total channel profits; its coordination effects depend on whether the induced move to 99-ending prices is upward or downward. In decentralized channels, consumer surplus can sometimes increase alongside total channel profit.
Experimentation is increasingly used across domains such as healthcare and online platforms to inform decision making, with a central goal often being the estimation of the average treatment effect (ATE). However, in many real-world settings, treatment effects, or even the treatments themselves, evolve over time, making classical experimental designs, which assume stationarity, less effective or even misleading. This paper studies nonstationary experimental design under structured trends, addressing two key objectives: (i) accurately estimating the dynamic treatment effect and (ii) minimizing regret (e.g., revenue or welfare loss) during the experiment. We propose a flexible design framework that can be tailored to achieve Pareto-optimal tradeoff between the two objectives. We further analyze how time-varying noise levels affect this tradeoff. Additionally, we establish asymptotic normality of the estimators and show that different trend orders yield different convergence rates, enabling efficient statistical tests for the existence of high-order trends.
