Table 2.

Effects of the suspension requests on sales change from March to April.

	(1)	(2)	(3)	(4)
VARIABLES	Sales change	Sales change	Sales change	Sales change
Suspension request	$-$9.84***	$-$10.06***	$-$9.99***	$-$9.70***
	(1.87)	(1.94)	(1.93)	(2.04)
Infection risk	0.59	0.58	0.63	$-$1.22
	(1.00)	(1.02)	(1.01)	(1.01)
Sales change Feb-to-Mar			0.04*	0.04
			(0.02)	(0.02)
Education $\times$ Suspension request				9.36
				(5.83)
Travel & Hotel $\times$ Suspension request				4.24
				(24.24)
Education				0.99
				(4.82)
Travel & Hotel				$-$15.45
				(22.93)
New infection change Mar-to-Apr	$-$2.91
	(2.33)
Observations	3,230	3,230	3,230	3,230
Pref FE	NO	YES	YES	YES
Mean (Suspension request)	$-$17.79	$-$17.79	$-$17.79	$-$17.79
Mean (No suspension request)	$-$9.082	$-$9.082	$-$9.082	$-$9.082

Notes: This table uses the sample of firms in B-to-C service industries. ”Sales growth change in March to April” is the change in monthly sales growth (relative to the same month in the last year) from March to April, measured in percentage points. “Suspension request” takes 1 if business suspension was requested of the firm’s industry in April. The infection risk measure is from Benzell et al. (2020). “New infection change Mar-to-Apr” is the change in the number of infection cases per 10,000 population in the prefecture from March to April. The second column controls for the prefecture fixed effects. The third column additionally controls for the sales change from February to March, when the infection cases were increasing but business suspension was not requested, to take into account the firm’s sensitivity to the pandemic. “Education” is a dummy for education-support industry, which reportedly used online teaching to avoid business suspension. “Travel & Hotel” is a dummy variable for travel and hotel industries. Aside from education, travel, and hotel industries, business suspension was requested to restaurants (including bars) and entertainment industries in some prefectures. Standard errors clustered at the prefecture level are in parentheses. The observations are weighted to match the number of firms in the Economic Census. ${}^{***}$ for $p<0.01$, ${}^{**}$ for $p<0.05$, and ${}^{*}$ for $p<0.1$.