Geography of Unemployment in the Midwest During the COVID-19 Pandemic: A Spatial Analysis

Abstract

Unemployment surged across the U.S. following the onset of the COVID-19 pandemic. We analyze the spatial pattern of county-level unemployment in the Midwest during the second quarter (Q2) of 2020 when it peaked, followed by the fourth quarter (Q4) of 2020 when local economies began to recover. We find clustering, with most counties in the Eastern section of the Midwest showing unemployment in the 75^th quantile of the distribution (hardest-hit counties). In contrast, counties in the Western section mostly showed unemployment in the 25^th quantile (least-hit counties). Our spatial lag regression analysis of the unemployment change between Q2 and pre-COVID (Q1) finds a higher increase in unemployment in counties with a high COVID-19 Community Vulnerability Index (CCVI). In addition, counties that relied on Leisure & Hospitality, and Manufacturing, and had a mask mandate, showed a larger increase in unemployment in Q2 and persisted through Q4 albeit to a lesser degree.