Collaborative Research: Using Weather Forecasts to Identify Economic Behavior

This award funds an innovative research on information economics. It examines how acting on weather forecasts enables agents to improve outcomes that are sensitive to weather. The first part of this project examines how improvements in weather forecast accuracy benefit society by helping people avoid mortality and why the benefits from these improvements differ across groups. The second part of this project tests whether people use forecasts differently when they are aware that forecasts have become more accurate. These types of reactions could be important for valuing persistent changes in forecast quality. The third part of this project uses routine weather forecasts to trace out the geographical footprints of firms’ economic activities, a fundamental but heretofore elusive feature of the economy. More generally, this research will further the state of science surrounding the value and utilization of information. It will also inform public policies on investment in information that can improve human health and economic prosperity, including to disadvantaged communities, domestically and across the world. The results will also help to establish the US as a global leader in the use of weather forecast to improve human wellbeing. This research will use three projects to study how ubiquitous information interventions—routine weather forecasts—impact economic behavior. The first project develops a theory-driven method for estimating the benefits of improvements in weather forecast accuracy using mortality events and National Weather Service temperature forecasts in the United States. Preliminary evidence shows that forecasts are greatly beneficial overall, but on average, has differential effects across different groups. The first project seeks to understand the sources of this disparity. The second project tests a crucial assumption underlying many recent information studies: that agents’ responses to messages do not depend on their beliefs about the persistent accuracy of information with quasi-experimental methods. The third project takes advantage of the granular nature of weather forecast data to identify the location of firm-level economic activity. This research contributes to the theoretical and empirical understanding of the value of information, including for persistent increases in accuracy and the utility of public information interventions. It will also contribute to understanding how anticipatory behavior may mitigate costly weather extremes that plague society. The results will also inform public policies on investment in information that can improve human health and economic prosperity, domestically and across the world. The results will also help to establish the US as a global leader in the use of weather forecast to improve human wellbeing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.