Can LLMs help humans to write books? TLDR: sort of. @imkereimers.bsky.social and I have a new paper with a longer answer. 1/n www.nber.org/papers/w34777

Great post, and an excellent example of the Bitter Lesson of AI: generalized models are always over time going to beat specialized models. substack.com/home/post/p-...

very cautiously optimistic about her as Dot

Applications for the Economic History Association Graduate Dissertation Fellowships are due in two days (January 14) eh.net/graduate-dis...

Outing myself as a Python notebook enjoyer but now that so much work happens via Claude Code and Git PR review I've found the format to be basically unworkable.

How do people write numerical tests for an estimation routine? I have three layers: 1) unit tests: simple numerical examples that I compute _by hand_ (literally on paper with a calculator), 2) synthetic datasets coded up separately, and 3) checking that some properties of the estimator hold... [1/2]

Rest in peace, Kate Ho.

sites.google.com/view/memoriesofkateho My advisor Kate Ho passed away two weeks ago. Her students have put together a memorial for her here. With permission from her family, we are sharing the site with the academic community. We'd love to hear from everyone who knew and loved Kate... [1/2]

Devastating news. Kate was a great economist and a generous person.

At Princeton's holiday party in 2023, I recorded a version of "Sheena is a Punk Rocker" with the lyrics changed to "Kate Ho is the Best Advisor." I still feel that way. I couldn't have asked for a better advisor. I can't believe she's gone.

The brilliant Kate Ho has passed away. She was an amazing economist and a genuinely kind human being. Her work shaped how we think about healthcare markets and her generosity touched everyone lucky enough to know her. The profession and the world are poorer without her.

Every IO economist's favorite weekend of the year!

man this is so exciting

Pitch to Shohei you cowards

Trey Yesavage’s career single-game highs entering tonight: 6 IP (May 1, Single-A Dunedin) 12 K (May 13, Dunedin) 94 pitches (Sept. 27, Toronto) 19 whiffs (Sept. 15, Toronto) Tonight: 7 IP, 12 K, 104 pitches, 23 whiffs It is Game 5 of the World Series against the defending champions.

this is too much baseball

i've seen enough, this is rigged against Toronto

hm blue jays looking suspiciously good

if toronto scores first i will simply turn the tv off and never watch baseball again

The end of “visuals as proof” means that people will more than ever need sources that can help them tell apart real from fake. That might be the key to the survival of journalism. www.nytimes.com/2025/10/09/t...

torontooooo

Collusion doesn't just happen in smoke-filled rooms. It can be out in the open in earnings calls. New research uses NLP on 300k+ transcripts to spot the language that firms use to coordinate publicly. Regulators can use this tool to detect anticompetitive behavior. papers.ssrn.com/sol3/papers....

Demand for human-generated, trust-certified internet content is skyrocketing. There is a major business opportunity, in *every* country, for media firms to meet this rising demand with innovation. Fascinating new experiment by @filipecampante.bsky.social et al.—> filipecampante.org/wp-content/u...

Debunking #immigration myths with data egc.yale.edu/research/lea... @yaleeconomics.bsky.social #econsky

super excited to join and then promptly forget about my team by GW2 five years in a row

tired: comprehensive and clear replication package wired: hiding Easter eggs all over the place for future grad student replicators

football is a business where you invest hundreds of millions into important assets and also tens of millions into a guy to operate those assets and then you sell the most productive assets that the guy happens to not like in order to keep the guy happy

This is also where I ended up. My god we were terrible but I’ve never been happier in 20 ish years as a Tottenham fan. Will I remember how little possession we had, in ten years? I doubt it.

I can’t stress enough how silly it is that I’m about to say what I’m about to say but this was one of the best days of my life I love Tottenham so much

ICE COLD

Well a nice thing happened to my soccer team

come on you spurs

Super cool: SFJAZZ has an archive of about 750 jazz concerts they hosted from 1983 to 2012, and they just posted the recordings online over at the Internet Archive. Jazz fans, check them out here: archive.org/details/sfja...

Has anyone successfully streamed the AEA continuing education lectures on iOS? The webcast app is amazing on the computer but it keeps freezing up on my iPad. www.aeaweb.org/conference/w...

this clasico is reinforcing my strongly held belief that every high level soccer defender is a policy failure

El Clasico ref is too scared to send the only defender on the pitch off

seeing some joke comments about the subject matter of this paper so it seems important to clarify that studying competition between platforms and the effect of regulations in a setting with potentials for truly massive externalities is just a very high value added activity for economists to do

🚨 New working paper 🚨 Demand for LLMs: Descriptive Evidence on Substitution, Market Expansion, and Multi-Homing A key question for the business of AI is the extent to which LLMs are differentiated from each other. I use data from OpenRouter to take a first look. andreyfradkin.com/assets/deman...

Step 1: Don’t doom on the timeline. Step 2: Don’t doom at all. Step 3: Do things. Small things count. Step 4: Repeat forever. This is both exhausting and liberating.

IO motto

happy new rm essay day for all who celebrates

It's here. @shanegreenstein.bsky.social 's annual review of the digital economy. digitopoly.org/2024/12/26/d...

Confident in my ability to properly Stata, I post a question on Statalist. I did not read Nick Cox's forum policies. He could still be nice to me, I think. He answers my question. His comment telling me to type in `help regress` passes cleanly through my skull, killing me instantly

#econsky What is, in your opinion, the biggest question IO has credibly settled? Seems most of our contributions are either (a) fancy methods; (b) negative results (e.g. don't regress price on HHI); (c) narrow in scope (yogurts, etc); or (d) contested (have markups actually risen?)

in order for me to fully buy in to the bluesky thing i'm gonna need a Tottenham Hotspur starter pack