4hijt
t are wave xed eects interacted with health shocks, which allow for a
exible time trend in
the eects of health changes (e.g., from country-wide changes in health care). 4hijt
j are county
dummies interacted with health shocks, which control for persistent dierences among counties in
8
the severity of health shocks or the quality of health care. To capture any common shock to the
outcome variable across households within the county and over time, standard errors are clustered
at the county level.9