Skip to main content
. 2022 Jan 19;143:104003. doi: 10.1016/j.euroecorev.2021.104003

Table 3.

Covid-19 cases and school opening.

Specification Dependent variable: Change in ln of Covid-19 weekly cases
Full sample date<December 2020 De-biased ABC IPW Spatial correlated errors Neighbors effect
(1) (2) (3) (4) (5) (6)
School opening (2nd lag) 0.026*** 0.027*** 0.023*** 0.025*** 0.012*** 0.037***
(0.001) (0.001) (0.001) (0.001) (0.003) (0.003)
School opening neighboring sections 0.017***
(0.003)
Covid-19 (ln, 2nd lag) −0.787*** −0.759*** −0.745*** −0.786*** −0.786*** −0.558***
(0.004) (0.005) (0.006) (0.008) (0.008) (0.011)
Covid-19 (ln, 3rd lag) 0.010*** 0.036*** 0.000 0.013*** 0.013*** 0.086***
(0.003) (0.004) (0.003) (0.005) (0.005) (0.006)
Covid-19 (ln, 4th lag) −0.020*** 0.035*** −0.027*** −0.015*** −0.015*** 0.046***
(0.003) (0.005) (0.004) (0.005) (0.005) (0.008)
Covid-19 (change ln lag) −0.864*** −0.855*** −0.806*** −0.864*** −0.864*** −0.733***
(0.003) (0.004) (0.002) (0.006) (0.006) (0.008)

Time dummies Y Y Y Y Y Y
Census area FE Y Y Y Y Y Y
Spatial autocorrelation N N N N N Y
Propensity scores weighting N N N Y Y Y
Number of Census areas 33,184 33,184 33,184 33,184 33,184 33,184
Observations 530,944 464,576 530,944 530,944 530,944 790,768
R2 within 0.43 0.41 0.43 0.43 0.35

***, **, and * denote significance at 1%, 5%, and 10%, respectively. Standard errors are clustered at census area level in columns 1, 2, 4 and 6.