Clint Harris

New paper with @instrumenthull.bsky.social and Michal Kolesár on leniency/judge IV designs.

This article is targeted for the Journal of Economic Perspectives, and so we wrote it with the goal of being accessible to a wide range of users, including advanced undergrads!🧵

arxiv.org/abs/2511.03572

Well you look great!

Literally crushing his L4-L5. I must be explaining the joke.

What evidence could even convince us
That no one is actually addicted to porn
and/or
That no one can be,
or whatever on earth the actual claim is
?

Rich men whose wealth does not enrich them:

Proud towers,
to swift destruction doomed.

Surely the value of knowing the effects of multilingualism is that people could choose to learn another language to gain benefits?

It seems an RCT assigning people to language training would get a more relevant treatment effect than the whole effect (whatever it is) of a lifetime as multilingual.

I think even Manski bounds are constrained by the support of the variable when it's observed.

I think these bounds are out of (Manski) bounds.

Maybe if you click yes they throw out the person's whole application.

Only the people with the most conscientious connections need apply.

If you are engaged in unlawful activity by my definition (e.g. occupying a city) you should not be legally protected from injuries you receive from someone engaged in self-defense that is proportional to the illegality of your actions.

Not endorsing violence! But I'd acquit if it was a brick.

My belated take on sandwich guy.

1. I think it is assault for someone to throw a sandwich at me.

2. I think it is not assault for a lumberjack to acidentally bop a coworker with some bark while chopping down a tree.

3. I think ICE/etc being injured by someone while working is case 2.

Register for IPUMs International Online Session Please use this form to register for the IPUMs International Session hosted by the Historical Economic Demography Group at LSE. The session will be on Zoom from 15:15-16:30 UK Time on 12 November 202...

Curious about using census microdata in your research? 📊

Join us for a webinar on IPUMS International, the world’s leading repository of harmonized census data.

🗓️ 12 Nov 2025 | 🕒 15:15–16:30 UK | 💻 Zoom
Register: forms.gle/oqTDNU4Zpn2s...

Hosted by the LSE Historical Economic Demography Group.

I think there is gold in it, which is kind of a stone?

Econ of Ed (and other) friends: what are the best studies linking in-college interventions to labor market outcomes? Especially interested in those that improve outcomes via different mechanism than completion. E.g. if a college wants to improve earnings of grads, what should they do?

Hard to imagine a real researcher having an opinion.

I have a mapping from assumptions to conclusions...

but I'll die before I tell you whether I believe my assumptions!

If California has a true, justified belief that something causes cancer, then it must actually cause cancer.

If it does not, saying that it is known to the State of California to cause cancer is a Lie and should be treated as false advertising.

I'm not sure about that.

My favorite thing about being an academic is realizing how obvious things are after wracking my brain about them for several years.

Econ job market candidate intros.

Candidates, communicate:
1) your name, program, & fields,
2) image of JMP title & abstract + Alt Text,
3) #EconJMP, &
4) a link to your website.

Employers, browse & interact respectfully.

Everyone else, shhhh.
📉📈

I use a more conservative approach in which I recall the max and min of garlic cloves I have observed, add those together, and double it.

New levels of making up people to be mad at.

There's the feeling I was looking for when I logged onto Bluesky. I can log off now.

Seems like we could just say white people aren't responsible for the historical accident of the circumstances that allowed their domination of other groups. I.e. their power isn't because they are smarter or harder working.

Then we can say whatever true thing we want about specific atrocities.

Worst case bounds uninformative again.

The Rubin Causal Sermon on the Mount says thus.

For binary sin indicator D and circumstances Z, a person is guilty of D if D(z)=1 for any z in the support of Z.

Blessed are the never-takers.

Saying you learn something about decision-makers (all people) from data in which nobody made any real decisions is misleading even if not strictly lying.

But I only briefly looked over the study, so I won't cast the accusation aside from stating it generally.

the president has destroyed the east wing of the white house without so much as a word of input from anyone else to build a royal ballroom paid for with bribes

They only had D&D BCE back then.

The comments on this are just like audience reviews of The Woman in the House Across the Street from the Girl in the Window.

Impossibly, they are even funnier than the show.

tinyurl.com/3w46k6u7

Very straightforwardly detained by secret police for speech. I honestly don't know what establishment figures are telling themselves any more to avoid cognitive dissonance over this shit.

Exclusion is just one weird guy in the SUTVA party trying to follow his cool friends into the club.