See citation below for complete author information.
With the rise of online labor markets, remote work, and other forms of computer-mediated production, firms have an increasing ability to tailor the information environment of workers, including their engagement with virtual peers. Using a series of field experiments conducted in an online labor market, we find strong evidence that even brief exposure to peers strongly affects output. Workers exposed to high-output peers subsequently raise their own output. However, there is no “free lunch”; exposure to high-output also causes some workers to exit. Even having one worker evaluate another worker—a common quality control pattern—affects subsequent output. We show that this peer effect is not solely due to social learning about employer expectations. When we remove all uncertainty about an employer’s expectations, workers still increase output, even beyond the employer’s expectations when exposed to other workers doing the same.
Horton, John J., and Richard J. Zeckhauser. "Wielding Peer Effects in Online Production: Evidence from a Series of Field Experiments." October 2018.