Khoa

My quickest QJE rejection was when I thought about submitting a paper there then got a rejection notice.

So novel 🥳🥳🥳

You are not a development economists if you haven't read this classic.

I was giving up on bluesky but Klaus brought me back with this.

Hahahahahahahahahahahahahaha

First year PhD student be like "I can't find interesting research ideas"

The research ideas:

Society of Labor Economists (SoLE) Annual Conference will be May 1-2, 2026 in Denver, Colorado. Submission portal is now open! Deadline for submissions is October 31--don't forget to submit! It will be a great conference! (organized by me and @jrothst.bsky.social)

mailchi.mp/sole-jole/so...

Synthetic Control Method be like

You can make your treatment variable "as-if random" by sorting the variable randomly in your dataset. Follow for more econometric tips.

Just revisited a previous draft of the paper that I'm working on. July Khoa was a terrible writer.

Does renaming my independent variable as "def_exogeneous_var" make my regressions causal?

Corinne Low! She's on here.

It's pretty clear that people are only gay if the marginal benefit is less than the marginal cost of dating men.

Essentially you take a bunch of variables (sometimes also over time) and then ask "can I somehow cluster observations according to the variables."
Step 2: ???
Step 3: Profit, or at least a publication.

God forbid a man trying to keep his job for more than a month.

"But did this BLS Commissioner Nominee even have an econ PhD from a top econ program" is the most expected dunk from economists.

At this point I am not even sure what neoclassical economics even means.

25%

Can finally write about the US would have a better economy if it had a Western institution.

Data-driven synthetic control method be like

My optimal setup as an applied microeconomist.

PhD student: "I think I will wait another year before going on the job ma..."

Adviser:

Tariffs *have* shown up in the data (just not at total CPI level) #econsky
econbrowser.com/archives/202...

When you are caught starting a new project instead of finishing all of the current ones.

When the hardest proof of the paper "is left for readers as an exercise".

Using time series graphs to make causal claims be like

Seeing too many diff-in-diff reviews and and summary articles but not sure where to start? We propose a new way to synthesize this growing literature using a Review in Reviews approach.

I actually forgot that he did that lol