Table A3.

NYC cases by borough.

	(1)	(2)	(3)	(4)	(5)	(6)
	ln($Case{s}_i$) NYC	ln($Case{s}_i$) The Bronx	ln($Case{s}_i$) Brooklyn	ln($Case{s}_i$) Manhattan	ln($Case{s}_i$) Queens	ln($Case{s}_i$) Staten Island
	Panel A: OLS
$\%\Delta Trip{s}_i$	0.010${}^{***}$	0.001	0.015${}^{*}$	0.007	0.004	0.008
	(0.003)	(0.005)	(0.008)	(0.010)	(0.006)	(0.004)
R-Sq.	0.437	0.645	0.544	0.624	0.262	0.892
Obs.	159	20	34	32	48	10
	Panel B: Reduced form IV
$ShareTel{e}_i$	7.542${}^{***}$	0.711	2.125	3.075	7.132${}^{***}$	4.462
	(0.883)	(1.968)	(2.212)	(2.054)	(2.640)	(5.153)
$ShareEs{s}_i$	7.660${}^{***}$	1.024	9.305${}^{***}$	3.478${}^{*}$	5.345${}^{*}$	3.188
	(0.859)	(2.813)	(1.806)	(2.021)	(2.945)	(5.446)
R-Sq.	0.629	0.704	0.770	0.673	0.370	0.826
Obs.	159	20	34	32	48	10
	Panel C: IV
$\widehat{\%\Delta Trip{s}_i}$	0.068${}^{***}$	0.007	0.058${}^{**}$	0.135	0.081	0.006
	(0.020)	(0.008)	(0.025)	(0.156)	(0.062)	(0.011)
Root MSE	0.487	0.079	0.288	0.577	0.551	0.088
Obs.	159	20	34	32	48	10
F-Stat.	22.199	10.058	4.028	2.613	1.607	1.754
	Controls
$\%AfA{m}_i$	X	X	X	X	X	X
$ln\left(Ag{e}_i\right)$	X	X	X	X	X	X
$ln\left(In{c}_i\right)$	X	X	X	X	X	X

Notes: This table is analogous to Table 2 in the main text, but compares cases across boroughs in NYC instead of different cities. All columns control for log of healthcare employment. Panels A shows results from Eq. (1). Panel B shows the reduced form IV regression results from $ln\left(TotalCase{s}_i\right)=\alpha +{\beta }_{1}ShareTel{e}_i+{\beta }_{2}ShareEs{s}_i+\Gamma X_i+{\epsilon }_i$. Panel C shows results from Eq. (1.2), adding additional demographic controls, $X_i$: $ln\left(TotalCase{s}_i\right)=\alpha +\beta \widehat{\%\Delta Trip{s}_i}+\Gamma X_i+{\epsilon }_i$. Robust standard errors in parentheses. Significance: ${}^{*}p<0.10{,}^{**}p<0.05{,}^{***}p<0.01.$