The Breakdown and Recovery of Cooperation in Large Groups: Exploring the Role of Formal Structure Using a Field Experiment

Abstract : Cooperation among employees is crucial for the success of organizations but can unravel with size. We study cooperation in the context of a workplace safety methodology: workers are voluntarily enrolled and are trained to provide advice to co-workers on safe behavior. Using archival data, we show that while cooperation is useful in reducing accidents, it breaks down as the number of enrolled workers increase. We show that this is due to decreasing marginal rewards for the additional enrolled worker. This supports recent research that highlights the crucial role of marginal benefits in shaping the relationship between cooperation and size. Then, we manipulate the safety methodology in a field experiment by randomly structuring workers in groups. This produces a recovery of cooperative effort and a reduction in risky behavior and accidents. We show that the likely mechanism behind this recovery are repeated interactions among advisors and workers, rather than group identity or social control such as peer pressure or reputation. This result suggests that the core function of structure is not only grouping workers to favor the division of specialized labor –as emphasized in prior research–, but also to promote cooperation at scale.

---

Choice Architecture and Gender Differences in Propensity to Compete and Productivity: an Experimental Study

Abstract : Research on competitive behavior typically demonstrates that women are less likely to participate in competitions than men are. We show that this difference can be eliminated by changing the framing of a competitive task from a default where applicants must actively choose to compete to a default where applicants are automatically enrolled in competition but can choose to opt out. We provide experimental evidence that this choice architecture reduces gender disparities by affecting the perception of prevailing social norms about competition without negatively affecting performance or well-being, or decreasing the likelihood that a woman is selected for a job.

---

The Value of Delegation in Hiring

Abstract : Delegation in hiring (DIH) is becoming a mainstream management practice among business chains worldwide, yet theoretical and empirical evidence of its effects remains scarce. DIH refers to the use of store managers to do their own recruiting, rather than relying on the headquarters human resource (HR) department. To estimate the impacts of DIH on employee recruitment, productivity, and firm performance, we partnered with a large Chinese firm with 111 retail stores on a 12-month field experiment. Results show that DIH increases both the mean and the variance of individual productivity, with an overall positive impact on store-level productivity by 7.1%. The growth in store-level performance is driven by both direct effects of recruiting higher-performing workers and indirect effects of positive spillovers from new to existing employees. Consistent with our theory, DIH achieves better outcomes in stores that have better incentives, larger shares of repeated customers, and that are less busy. To test the validity of our experiment, we find similar results from observational data on this firm and a second large multinational firm in the spa industry. Overall, this study illustrates the value of information held by local managers in hiring productive employees.

---