Does paid sick leave encourage staying at home? Evidence from the United States during a pandemic

Abstract

We study the impact of a temporary U.S. paid sick leave mandate that became effective April 1st, 2020 on self-quarantining, proxied by physical mobility behaviors gleaned from cellular devices. We study this policy using generalized difference-in-differences methods, leveraging pre-policy county-level heterogeneity in the share of workers likely eligible for paid sick leave benefits. We find that the policy leads to increased self-quarantining as proxied by staying home. We also find that COVID-19 confirmed cases decline post-policy.

Keywords: infectious disease spread; paid sick leave; pandemic disease; physical mobility.

Figure 1.. Correlates of county-level nonessential workers

Notes: N (counties) = 2706. See text for data sources. Data are weighted by the county population. The unit of observation is a county, there is observation per county. All regressions are estimated with least squares. Omitted categories for demographics are female and white race. All variables converted to standard deviations prior to estimation. Standard errors are clustered at the county-level and are reported in parentheses. ***;**;* = statistically different from zero at the 0.1%, 1%, 5% level.

Figure 2.. Effect of FFCRA on physical mobility outcomes using an event study design

Panel A. Hours away from home Panel B. Percent away from home >8 hours per day Notes: Data source is SafeGraph Social Distancing Metrics files March 13^th, 2020 through April 30^th, 2020; weekends are omitted. Data are weighted by the county population. The unit of observation is a county in a day. All regressions are estimated with least squares and control for social distancing policies, weather covariates, county fixed-effects, and date fixed-effects. Coefficient estimates are reported with black circles. The omitted category is March 13^th, 2020 to March 24^th, 2020. 95% confidence intervals account for within-county clustering and are reported with vertical lines. The vertical dashed line indicates April 1^st, 2020 which is the first day that FFCRA became effective.

Figure 3.. Effect of FFCRA on physical mobility outcomes using a difference-in-differences style model: Allow for separate industry code interaction with nonessential worker industry codes

Notes: Data source is SafeGraph Social Distancing Metrics files March 13^th, 2020 through April 30^th, 2020; weekends are omitted. The unit of observation is a county in a day. Data are weighted by the county population. All regressions are estimated with least squares and control for weather covariates, county fixed-effects, and date fixed-effects. Coefficient estimates on the interaction between industry shares and the post-FFCRA indicator are reported in black circles. 95% confidence intervals account for within-county clustering and are reported with horizontal lines.

Figure 4A.. Effect of FFCRA on average hours away from home using a difference-in-differences style model: Falsification testing

Panel A: Scatter plot (nonessential worker establishments left; nonessential workers right) Panel B: Histogram (nonessential worker establishments left; nonessential workers right) Notes: Data source is SafeGraph Social Distancing Metrics files March 13^th, 2020 through April 30^th, 2020; weekends are omitted. The unit of observation is a county in a day. All regressions are estimated with least squares and control for social distancing policies, weather covariates, county fixed-effects, and date fixed-effects. Panel A: black diamond is the coefficient estimate from our preferred specification and small white circles capture coefficient estimates generated in equation (1) after randomly re-shuffling the treatment variable (Post FFCRA*% nonessential worker establishments) across counties and dates.

Figure 4B.. Effect of FFCRA on share away from home for more than eight hours using a difference-in-differences style model: Falsification testing

Panel A: Scatter plot (nonessential worker establishments left; nonessential workers right) Panel B: Histogram (nonessential worker establishments left; nonessential workers right) Notes: Data source is SafeGraph Social Distancing Metrics files March 13^th, 2020 through April 30^th, 2020; weekends are omitted. The unit of observation is a county in a day. All regressions are estimated with least squares and control for social distancing policies, weather covariates, county fixed-effects, and date fixed-effects. Panel A: black diamond is the coefficient estimate from our preferred specification and small white circles capture coefficient estimates generated in equation (1) after randomly re-shuffling the treatment variable (Post FFCRA*% nonessential worker establishments) across counties and dates.

Appendix Figure 1A.. Geographic distribution of nonessential worker establishments and workers across U.S. counties

Panel A: Nonessential worker establishments Panel B: Nonessential workers Notes: Data source is Quarterly Census of Employment and Wages 2019. We include the following NAICS codes: 111, 112, 114, 115, 22, 23, 311, 3121, 3221, 32222, 32223, 32229, 3251, 3253, 3254, 3256, 3259, 33311, 3341, 3342, 3343, 3345, 3344, 3346, 3352, 3391, 4233, 4234, 4237, 4241, 4242, 4244, 4245, 4246, 4247, 4248, 4249, 4441, 44511, 44512, 4452, 4453, 4523, 454110, 44611, 447, 481, 482, 484, 4851, 4852, 4853, 4854, 4855, 4859, 491, 492, 493, 51111, 515, 517, 5182, 51913, 521, 52211, 52219, 52212, 52213, 5222, 5223, 523, 5241, 5412, 5415, 5416, 5417, 54194, 5525, 5617, 56173, 562, 616, 6211, 6212, 6213, 6214, 6215, 6216, 6219, 6221, 6222, 6223, 6231, 6232, 6233, 6239, 6241, 6242, 6244, 7211, 722, 8111, 8112, 8113, 8122, 8123, 92111, 92112, 92113, 92114, 92115, 92119, 922, 923, 924, 925, 926, 927, and 928. We exclude the following NAICS codes: 311811, 42491, 44413, 517311, 56173, 62131, 62132, 7224, and 811192.

Appendix Figure 1B.. Frequency distribution of nonessential worker establishments and workers across U.S. counties

Panel A: Nonessential worker establishments Panel B:Nonessential workers Notes: Data source is Quarterly Census of Employment and Wages 2019 quarter 1. We include the following NAICS codes: 111, 112, 114, 115, 22, 23, 311, 3121, 3221, 32222, 32223, 32229, 3251, 3253, 3254, 3256, 3259, 33311, 3341, 3342, 3343, 3345, 3344, 3346, 3352, 3391, 4233, 4234, 4237, 4241, 4242, 4244, 4245, 4246, 4247, 4248, 4249, 4441, 44511, 44512, 4452, 4453, 4523, 454110, 44611, 447, 481, 482, 484, 4851, 4852, 4853, 4854, 4855, 4859, 491, 492, 493, 51111, 515, 517, 5182, 51913, 521, 52211, 52219, 52212, 52213, 5222, 5223, 523, 5241, 5412, 5415, 5416, 5417, 54194, 5525, 5617, 56173, 562, 616, 6211, 6212, 6213, 6214, 6215, 6216, 6219, 6221, 6222, 6223, 6231, 6232, 6233, 6239, 6241, 6242, 6244, 7211, 722, 8111, 8112, 8113, 8122, 8123, 92111, 92112, 92113, 92114, 92115, 92119, 922, 923, 924, 925, 926, 927, and 928. We exclude the following NAICS codes: 311811, 42491, 44413, 517311, 56173, 62131, 62132, 7224, and 811192.

Appendix Figure 2A.. F-statistics for structural breaks for average hours away from home

Notes: Data source is SafeGraph Social Distancing Metrics files March 13^th, 2020 through April 30^th, 2020; weekends are omitted. Lines are the F-statistic from a test for a level- and slope-shift in a linear time trend at the indicated date in models controlling for county fixed-effects, weather, and state policies. F-statistic based on covariance matrix that is clustered on county.

Appendix Figure 2B.. F-statistics for structural breaks for fraction away from home more than eight hours

Notes: Data source is SafeGraph Social Distancing Metrics files March 13^th, 2020 through April 30^th, 2020; weekends are omitted. Lines are the F-statistic from a test for a level- and slope-shift in a linear time trend at the indicated date in models controlling for county fixed-effects, weather, and state policies. F-statistic based on covariance matrix that is clustered on county.