Observability and peer effects: Theory and evidence from a field experiment

This paper designs a field experiment in a real job environment, which incorporates differential observability in a job with multiple tasks (quality and quantity) to study the effect of peer pressure on outputs. The treatment group was informed of each individual's output quality information, whereas the control group only knew the group mean. The treatment group produced better output quality but lower output quantity, implying that workers adjusted their efforts between tasks. After switching off the treatment, results from the follow-up experiment show that the output quality produced by the treatment group rolled back to the baseline low level, whereas their output quantity continued to decrease. These results suggest that peer pressure, as a tool for promoting workers' productivity, should be adopted with caution.