The idea that worker utility is affected by co-worker wages has potentially broad labour market implications. In this paper, published in the Quarterly Journal of Economics, Breza, Kaur and Shamdasani (2018) report on a month-long experiment with Indian manufacturing workers, where they randomize whether co-workers within production units receive the same flat daily wage or different wages (according to baseline productivity rank). For a given absolute wage, pay inequality reduces output and attendance by 0.24 standard deviations and 12%, respectively. These effects strengthen in later weeks. Pay disparity also lowers co-workers’ ability to cooperate in their self-interest. However, when workers can clearly observe productivity differences, pay inequality has no discernible effect on output, attendance, or group cohesion.