#StackBounty: #econometrics #subsampling What we should do when the result is only significant to one country but the whole sample?

Bounty: 100

In a staggered Difference-in-Differences setting, Dasgupta, 2019 has a formula for such a static setting is

$Y_{it}$ = $alpha$ + $beta$ $pt_{kt}$ + $delta$$X_{ikt}$ + $theta$$_t$ + $gamma$$_i$ +$epsilon$$_{it}$ (1)

where $i$, $k$, and $t$ index firms, countries, and years respectively. $X_{ikt}$ is a vector of the different firm, country, and industry control, while $gamma$ and $theta$ are firm and year fixed effects.

When replicating table 3 in his paper, I focus on columns (2) and (6). In these two columns, I have the same control variables sets, fixed effects sets, but only difference in sample. While columns (2) controlled for the whole sample but column (6) controlled for non-US sample.

enter image description here

As we can see from the Table, the coefficients of $pt_{kt}$ tell us the effect of laws stronger in US countries but weaker in non-US countries. So, I need to address this issue

What I tried is that I add US_dummy equalling to 1 to specification (2) if this firm is in the US to the equation. But I faced the omitted issue because the the information in US_dummy is carried by the year fixed effect already.

That said, I simply cannot solve the weakened results in non-US sample problem by using the US_dummy in all regressions.
Can you suggest me any other way to deal with this issue?

Get this bounty!!!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.