Intervention (Hidden)
We will implement two interventions.
In the first intervention, we propose an experimental design that allows us to test whether giving workers the right to vote on a compensation policy improves performance. In the treatment group, we will give workers in some teams the right to vote over options that define their performance pay. Currently, the workers’ bonus pay is based on production metrics. This will remain an option that workers can choose but we will introduce an alternative option to the calculation of their bonus. Instead of reflecting only individual-level production metrics, the bonus component of their pay can be determined in part by anonymous evaluations by the peers in the team. The treatment teams will be allowed to vote over their compensation scheme and the most popular outcome will be implemented. Whatever is the popular choice within the team will be implemented at the team level. In the control teams, the workers will not be able to vote on their preference and we will randomly assign teams to the two compensation options in a ratio that corresponds with the number of teams that select each policy among the treatment teams.
In the second intervention, our design focuses on manager compensation and the effects of allowing workers to evaluate their managers. In the treatment teams, workers will be allowed to provide anonymous feedback on the several dimensions of the performance of their managers each month and these evaluations will be given some weight in determining the monthly bonus of managers. In control teams, the workers will not provide monthly feedback on their managers and the managers bonus pay each month will depend only on the productivity metrics; this is the status quo.