The Hutchins Roundup brings the latest thinking in fiscal and monetary policy to your inbox. Have something you'd like us to include in the next Roundup? Email us and we'll take a look.
This edition was written by Elijah Asdourian, Sam Boocker, Georgia Nabors, Comfort Oshagbemi, and David Wessel.
Using data on hurricane forecasts, damages, and protective and recovery spending for 18 out of 29 Atlantic hurricanes from 2005-2020, Renato Molina of the University of Miami and Ivan Rudik of Cornell show that improvements in hurricane forecasting from 2007 to 2020 reduced total costs by $5 billion per hurricane, or 19%. This exceeds the annual budget for all federal weather forecasting. In particular, the authors show that a one-standard-deviation improvement in forecasting reduces total costs (including preventive) by $30 million per county, on average. Weather forecast errors are also consequential: underestimating the wind speed of a forecast by 1 meter per second increases costs by $2,000 per person or $75 million per county.
Non-competitive labor markets can generate worker rents—a wage that exceeds what is necessary to retain a worker in their current job. Using U.S. data from 1980 to 2016, Daron Acemoglu of MIT and Pascual Restrepo of Boston University find that automation results in larger wage declines among workers in high-rent tasks and reduces variation in wages across workers. Slightly more than 50% of the rise in wage inequality between workers from different demographic groups since 1980 is attributable to automation, the authors estimate. Of this share, 42 percentage points arise from the direct displacement effects of automation. Rent dissipation—the disproportionate displacement of workers in high-rent tasks—accounts for the remaining 10 percentage points. The authors find that rent dissipation offsets 60% to 90% of productivity gains since 1980 because tasks targeted for automation are high-rent tasks rather than the most costly, producing allocative inefficiency.
Though increased sleep has been shown to improve productivity, little is known about the effectiveness of interventions designed to increase sleep. In a field experiment with over 1,100 university students, Osea Giuntella of the University of Pittsburgh, Silvia Saccardo of Carnegie Mellon, and Sally Sadoff of the University of California, San Diego, find that interventions can increase both sleep and academic performance. Students in the treated group were sent bedtime reminders each night and received $4.75 each weekday morning that their Fitbit showed they got at least seven hours of sleep. On average, students got an extra 19 minutes of sleep while the incentives were in place, and extra eight minutes of sleep in the weeks even after the incentives were taken away. Average academic performance improved by a tenth of a standard deviation, suggesting a cost-effective method for universities to improve educational outcomes.
"First, they are a new type of systemic risk. Unlike rare tail events known as “black swans”, climate change represents a break from the classical Gaussian probabilistic universe because it is doomed to happen if not addressed by appropriate policies… The second factor is the irreversibility of their impacts. If we do not reverse the course of climate change and environmental degradation, we will reach points of no return… Third, the impacts of climate and nature-related risks will spread throughout the economy, affecting central banks’ tasks… Finally, these risks are also unique in that, unlike for instance geopolitical risks, central banks can contribute towards mitigating them. There is broad political consensus on the need to tackle the causes of climate risks. And while governments are in the lead and have the most powerful tools, we can also play our part, within our mandate, in preparing the economy for the future.”
