This study examines how winter weather conditions affect remote work patterns across U.S. metropolitan statistical areas. Combining individual-level telework data from the Current Population Survey with daily weather records from the National Oceanic and Atmospheric Administration, I employ a two-way fixed effects model and a continuous treatment difference in differences framework to estimate the causal impact of winter weather on labor supply decisions and trace the dynamic dose- response relationship between snow depth and telework hours. The empirical re- sults demonstrate that adverse winter conditions, specifically lower temperatures and greater snow depth, significantly increase teleworking hours, with this effect markedly stronger in occupations characterized by high telework feasibility. Furthermore, while snowpack sharply reduces on-site work, telework partially compensates for the lost hours, mitigating the overall decline in total labor supply. The estimated dose-response curve confirms a robust, positive causal relationship between snow depth intensity and remote work substitution. These findings suggest that spatial reallocation of workplaces serves as an effective adaptation mechanism against weather-induced la- bor disruptions, offering important implications for climate adaptation policies and infrastructure planning.

This study investigates the impact of heat waves on the working hours of construction laborers in U.S. metropolitan statistical areas (MSAs). Utilizing daily temperature data and working hours data, the research explores how extreme heat affects labor output in short-term heat shocks, contrasting with prior studies that often focus on long-term climate adaptation. Using a two-way fixed effects model, this paper investigates the impact of increasing heat waves on construction laborers, accounting for both individual and temporal heterogeneity. The results suggest that in samples restricted to weeks with low precipitation, an additional day per week with temperatures exceeding 90°F reduces working hours by about 0.22 hours, and the occurrence of heat waves at least once a week decreases working hours by about 0.75 hours. Despite the statistical significance of some results, implications must be drawn carefully as the magnitudes are not large.

Research on climate as a local amenity has focused on the level and dispersion of temperatures, but the order in which daily temperatures arrive may also affect where people choose to live. This paper asks whether consecutive-day anomaly changes in temperature anomalies, holding their dispersion fixed, are associated with county-level internal migration in the United States. I measure the mean absolute change between consecutive daily anomalies, so that the remaining variation concentrates on how quickly anomalies reverse. Linking NOAA GHCN-Daily station records to IRS migration flows for 859 counties over eight filing-year pairs from 2011 to 2019, I relate migration to the three-year average of absolute day-to-day anomaly changes, controlling for within-month dispersion, the distribution of daily temperatures, precipitation, and seasonal amplitude, and including county and state-by-year fixed effects. A 1°C increase in consecutive-day anomaly changes is associated with a 0.39 percentage point rise in the out-migration rate, though the 95% confidence interval includes zero. The results show that the temporal ordering of temperatures can be distinguished empirically from dispersion, while evidence of a stable migration response remains limited.