COVID-19 Effects on the U.S. Unemployment: Nationwide and State-Level Evidence from Google Trends
74 Pages Posted: 24 Mar 2020 Last revised: 26 Jun 2020
Date Written: June 26, 2020
This paper investigates the relationship between nationwide Google search queries of "covid-19" and "unemployment" using daily data from the United States. The investigation is achieved by employing a structural vector autoregression model, where nationwide queries of "interest rate" and "inflation" are also included. Historical decomposition analysis suggests that the spike in the nationwide search interest of "unemployment" is mostly explained by that of "covid-19" in mid-March 2020. One unit of a positive nationwide "covid-19" shock results in about 8.1 units of a significant cumulative increase in nationwide "unemployment" after two months, ranging between about 2.5 and about 9.3 across U.S. states when the additional state-level search query of "unemployment" is included, suggesting unequal "unemployment" effects of "covid-19" across states. Regarding monetary policy, one unit of a negative nationwide "interest rate" shock results in 0.4 units of a significant cumulative reduction in nationwide "unemployment" after two months, although the corresponding effects range as between 1.3 units of an increase and 0.5 units of a reduction in "unemployment" across U.S. states, suggesting evidence for redistributive effects of nationwide monetary policy.
Keywords: COVID-19, Coronavirus, Unemployment, Interest Rate, Inflation
JEL Classification: J63, F66, I10
Suggested Citation: Suggested Citation