Table 2.

Transmission equation (IV) conditioned on different urban covariates.

	(1)	(2)	(3)	(4)	(5)	(6)
	Full sample			First month
${\Delta Infectious}_{ot}$	1.24***	−6.18	−7.00	0.81***	6.63	8.52
	(0.32)	(8.71)	(9.36)	(0.29)	(7.04)	(7.09)
Interacted with..
log density	−0.01	−0.04		0.07**	0.12***
	(0.06)	(0.07)		(0.03)	(0.01)
Public transport share		0.25***	0.22***		−0.01	0.07***
		(0.09)	(0.08)		(0.03)	(0.02)
Health insurance coverage		−0.46	−0.05		2.15**	2.03**
		(1.24)	(1.11)		(0.88)	(0.92)
log average wage		0.97*	0.92*		−0.57***	−0.62***
		(0.50)	(0.47)		(0.11)	(0.12)
Unemployment rate		−3.12	−2.96		−2.70*	−1.92***
		(5.09)	(4.96)		(1.54)	(0.56)
Proximity index of jobs		−0.51	−0.39		−0.87	−1.11
		(2.01)	(2.12)		(1.01)	(1.01)
Work-from-home		−6.24*	−6.44*		0.46	0.86
Share of jobs		(3.40)	(3.44)		(1.27)	(1.18)
Share elderly ($>$70)		8.35***	8.98***		−0.49	0.24
		(2.81)	(3.10)		(1.19)	(1.35)
Share children ($<$18)		5.65***	6.75***		3.93**	4.42***
		(1.28)	(1.92)		(1.62)	(1.56)
Observations	87,460	86,620	86,620	12,234	12,234	12,234
State-day fixed effects	Yes	Yes	Yes	Yes	Yes	Yes
Interaction arguments	Yes	Yes	Yes	Yes	Yes	Yes

Notes. ***$p$ $<$ 0.01, **$p$ $<$ 0.05, *$p$ $<$ 0.1. The dependent variable is the county-day level of cases. Estimated in first differences. Twoway (county and day) clustered standard errors in parentheses. $\Delta Infectiou{s}_{ot}$is the 6-day lagged 5-day average number of infected, weighted for each destination county with the origin-normalized commuting flows, multiplied with the origin population. Public transport share (of all commuters), health insurance coverage, unemployment rate, work-from-home share of jobs (of all jobs), share elderly and share children are shares of population or subpart of population where indicated — ranging from 0 to 1, see descriptive statistics for details on covariates.