This paper studies how managers affect workers through feedback. Feedback conveys both information about work and a tone that can motivate effort. I construct data from software development teams on GitHub, covering over 200 million code review feedback messages across 1.7 million teams. Reviewers perform managerial functions by providing feedback and deciding whether code is accepted. Using large language models, I classify feedback into types (e.g., toxic, positive, and constructive) and estimate their causal effects on productivity. My instrumental variables design exploits the fact that some teams adopt random reviewer assignments. I find that feedback tone matters: toxic feedback reduces future code quantity and quality, whereas respectful criticism (negative but non-toxic) has no such detrimental effects. Positive feedback improves future code quality and raises retention, as measured by LinkedIn job histories. Finally, I measure reviewer quality by their contribution to developer productivity growth and show that feedback explains much of the variation in reviewer quality.

Does the division of labor increase team productivity? This paper provides new evidence challenging the conventional view that specialization increases productivity. I create a panel dataset from GitHub, covering 35 million task allocations across 64,400 software development teams from 2017 to 2023. My result shows a negative relationship between team specialization and various productivity metrics, including output quality, quantity, and user issue resolution time. To identify causal effects, I exploit GitHub’s introduction of an automatic task assignment feature, which evenly distributes tasks across team members. Using a matched difference-in-differences design, I find that adoption of this feature reduces specialization and leads to significant gains in productivity: output quality rises by 4%, output quantity by 21%. Team communication also increases by 13%, suggesting that improved interaction and knowledge exchange are a key mechanism behind these productivity gains. These findings highlight a trade-off in non-routine production: while specialization increases task-specific human capital, it impedes cross-task knowledge spillovers that are essential for innovation.

We examine how political preferences shape migration decisions of college-educated workers in the United States. Using county-level voting patterns and migration flows, we show that workers are more likely to leave politically misaligned areas and move to places aligned with their political views. This effect is stronger for younger workers and those in occupations with high geographic mobility. Our findings highlight the role of political polarization in shaping labor mobility and regional economic dynamics.

Using high-frequency monthly precipitation data, we investigate the demographic consequences of extreme rainfall variability, comprising extreme droughts and heavy rainfall, in rural China. We find that both forms of extreme weather increased fertility rates, particularly for male births, between 1990 and 1999. More than a decade later, individuals who experienced extreme weather conditions, especially heavy rainfall, in utero showed lower educational attainment compared to their unexposed counterparts. Notably, we identified the second trimester of pregnancy as the critical period, with exposure during this period exerting the most significant negative impact on adult educational outcomes. Analyzing the mechanism, we find that extreme weather events reduce household incomes and increase child mortality. These findings highlight the enduring demographic impacts of environmental disasters.