Working Paper: NBER ID: w27027
Authors: Sumedha Gupta; Thuy D. Nguyen; Felipe Lozano Rojas; Shyam Raman; Byungkyu Lee; Ana Bento; Kosali I. Simon; Coady Wing
Abstract: This paper examines the determinants of social distancing during the COVID-19 epidemic. We classify state and local government actions, and we study multiple proxies for social distancing based on data from smart devices. Mobility fell substantially in all states, even ones that have not adopted major distancing mandates. There is little evidence, for example, that stay-at-home mandates induced distancing. In contrast, early and information-focused actions have had bigger effects. Event studies show that first case announcements, emergency declarations, and school closures reduced mobility by 1-5% after 5 days and 7-45% after 20 days. Between March 1 and April 11, average time spent at home grew from 9.1 hours to 13.9 hours. We find, for example, that without state emergency declarations, event study estimates imply that hours at home would have been 11.3 hours in April, suggesting that 55% of the growth comes from emergency declarations and 45% comes from secular (non-policy) trends. State and local government actions induced changes in mobility on top of a large response across all states to the prevailing knowledge of public health risks. Early state policies conveyed information about the epidemic, suggesting that even the policy response mainly operates through a voluntary channel.
Keywords: COVID-19; social distancing; government policy; mobility; epidemic response
JEL Codes: I0
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Emergency declarations (H84) | Time spent at home (D13) |
| Announcement of the first confirmed case (H12) | Social mixing (Z13) |
| School closures (I21) | Social mixing (Z13) |
| Stay-at-home orders (C69) | Social mixing (Z13) |
| Emergency declarations (H84) | Increase in time spent at home (J29) |