Perceptions at Work
Policy Context
Good managers matter. Variation in the quality of individual managers contributes significantly to differences in productivity both within firms and across firms. But how can firms find good managers? The fact that there is such large dispersion in observed managerial quality attests to the difficulty of solving this problem. This issue is also particularly important in developing countries with relatively low human capital, where a “missing middle” in management is often cited as a key barrier to firm growth. In practice, many firms rely on the discretion of existing managers to select workers for promotion. Whether this is optimal for firms is an open question. On the one hand, supervisors work closely with workers and may thus have the best knowledge of which workers would perform well in new roles. However, supervisors may also have bias, favoritism, or personal preferences when selecting workers for promotion that are not aligned with the firm’s interests.
Study Design
In partnership with one of the largest garment manufacturing firms in Tanzania, researchers implemented a series of field experiments to examine supervisor discretion in the context of selecting workers for promotion to managerial roles. In the status quo, the partner firm typically relies on supervisors to select workers for promotion. To study this discretionary selection, Ho first conducted a field experiment where all supervisors are given the opportunity to refer up to two workers for promotion. During the referral process, Ho randomized whether supervisors face financial incentives, worth roughly 10% of their monthly base wage, based on the quality of their referrals. In a complementary experiment with workers, researchers also studied how different selection methods affect workers’ decisions to apply for promotion. Specifically, they randomized whether supervisor selection or selection by objective performance metrics are emphasized on the application forms that workers receive.
Results and Policy Lessons
Randomly assigned financial incentives lead supervisors to refer workers with 13% higher measured managerial quality relative to non-incentivized supervisors. Through the lens of a simple conceptual framework, these results indicate simultaneously that supervisors have private information about worker quality and that there are trade-offs between supervisors’ personal preferences and the firm’s objectives in the status quo. Supervisors also show preferences consistent with gender bias and favoritism, and the referral bonus does not significantly change these patterns.
There are also sorting effects from discretion, as randomly emphasizing supervisor referrals in the selection process significantly reduces the number of workers who apply for promotion, a 6.5 percentage point (12%) decrease relative to the control group. As a direct result, discretionary selection reduces the number of high quality candidates the firm receives. These application results are consistent with workers’ stated preferences over different selection methods.
Finally, Ho compared supervisor selection to more objective selection methods, which are both more transparent and preferred by a majority of workers. Compared to promotion by seniority or observed performance metrics, supervisors select workers with 21-24% higher managerial quality. Thus, while discretion does have costs, it also leads to the selection of workers with significantly higher managerial quality relative to more objective selection methods.
