One reason to be concerned about income inequality is the idea that people not only care about their own absolute income, but also their income relative to various reference groups (e.g. co-workers, friends, neighbors, relatives, etc.). We use Canadian linked employer-employee data to estimate the casual effect of co-worker pay on a worker's reported job and pay satisfaction. Since worker satisfaction can affect the worker's productivity, organizational commitment, turnover, creativity and innovation, as well as the firm's productivity and profitability, this is an issue that requires more attention and careful examination. In models that control for a rich set of workplace characteristics, we find that coworker pay has a large positive and significant effect on both pay and job satisfaction. In our preferred models with establishment-level fixed effects, the effect of coworker pay on pay satisfaction is half as large, and the effect on job satisfaction completely disappears, suggesting that part (all) of what previous studies attribute to the effect of coworker pay on worker pay (job) satisfaction is driven by unobserved heterogeneity across firms or establishments. Our results also suggest that the effect of coworker pay on worker satisfaction is much stronger for workers who leave their job during the following year. Finally, we find that while coworker pay has a positive effect on pay satisfaction among Canadian-born whites, it has a negative effect among immigrants and Canadian-born visible minorities.