@DanielYiXu Love the paper, though as a French person living in the US, do not love the conclusion *at all* :D
English
Adrien Matray
1.2K posts
Adrien Matray
@AdrienMatray
Economist studying trade policy, finance, and macroeconomic outcomes. Research on tariffs, industrial policy, and global trade.
Atlanta, GA Katılım Ağustos 2020
112 Takip Edilen3.3K Takipçiler
@SamuelKaplan13 @JonSteinsson @BusinessInsider right but with this reasoning you can never prove markets are wrong. They just do not have the "right" information at time t...
English
@AdrienMatray @JonSteinsson @BusinessInsider But Adrien, markets are forward looking with the available information set as of t. If new information arrives at t+1, then expectations are updated.
English
IBM stock fell 13% on a new Anthropic product that makes finding issues in Cobol easier.
What I find most interesting about this is that the market didn't see this coming. Isn't the market supposed to be forward looking?
businessinsider.com/ibm-stock-pric… via @businessinsider
English
A big big congrats to @JulianeBegenau, that is so richly deserved
Stanford Institute for Economic Policy Research@SIEPR
👏 Congratulations to SIEPR's @JulianeBegenau and @shoshievass, also @StanfordGSB associate profs, on their 2026 Sloan Research Fellowships! The award recognizes early-career scientists whose work marks them as next-gen leaders. Story 👇 #EconTwitter #AcademicResearch
English
As anyone tried a comparison between Codex and Claude code? Codex seems to be able to operate stata and I have noticed for more sophisticated task chatgpt pro tends to be better than claudeai @ahall_research
@paulnovosad
@cblatts
#econtwitter @alexolegimas
English
@paulnovosad is that 9 out of 10 times it will produce something very very wrong and it is not trivial to identify the problem.
English
@paulnovosad I am curious: how well did you know the data before? Did you have already existing code that AI could rely on? My experience so far is that in particular if the data is not super plain vanilla in its structure and I do not know them well (so I cannot tell AI to be ultra careful)
English
I wrote a decent paper with AI.
It took me about 3 hours from start to finish, including an interactive choose-your-own-border-RD-adventure, it’s a what are we even doing here kind of day.
Paul Novosad@paulnovosad
I was writing about land reform in West Bengal last night and was curious if it had persistent effects on the ownership distribution. So I did what anyone would do, I* wrote an academic paper on it Turns out — yes! 1/
English
what an incredible public goo @ricardo_hausman has provided to the community!
Ricardo Hausmann@ricardo_hausman
Our new version of the Atlas of Economic Complexity is out. New data (2024), new classification of products, new ways of cleaning the data, new method to calculate complexity variables. The 🧵below highlights the main findings. Read on!!
English
@AdrienMatray That bias is baked in. I'll try to add a journal-specific option in the future (e.g., if you say you're submitting to Journal of Finance, it'd bias the recommendations in favor of finance profs.)
English
🚨 Are you about to submit a paper to an Economics journal?
Curious who the potential referees may be?
Upload a PDF of your paper to this free tool and find out...
#EconTwitter
🧵👇
English
@pereztruglia I would to see if it can adapted to more finance journals :)
English
@pereztruglia thanks so much for providing this this is super useful! I have tried on some of my papers and my impression is that the list of potential referees misses obvious people in finance and tilts toward more "econ" people. Would there be a way to access the raw instructions?
English
@cblatts @startupecon @ahall_research @paulnovosad ok super cool thanks a lot Chris! and thanks Michael, Claude kept telling me that claude code was not able to do online research but apparently there is a way to go around it. Really useful!
English
So I'm not sure if I believe Claude, but it told me that ChatGPT is better at search and more contemporaneous updates, especially when it comes to Claude code, which is relatively recent and advice is coming out daily. So I actually programmed all my Claude code prompts and code to tell me when I should use ChatGPT instead, and it tells me to do that pretty often. But I'm not sure if I over-programmed it or not
English
Hello productivity boosters with claude code! What are your favorite ways of using claude code for doing preliminary lit review in econ and finance? @ahall_research @paulnovosad @cblatts #econtwitter
English
@startupecon @ahall_research @paulnovosad @cblatts Thanks! So one big issue i have with claude code is that it cannot do online search so you miss all the last two or three years of publications on the "local model", but yes i love otherwise claude code
English
@AdrienMatray @ahall_research @paulnovosad @cblatts Open Claude Code CLI: “I need a system to perform a preliminary lit review in topic X. Ask me a lot of questions about what I want then build it” that’s the secret!
English
@DiegoRestuccia is one of the greatest. I know what I will be doing on February 12th!
Christian Volpe Martincus@christianvolpem
On February 12, @DiegoRestuccia from @UofT will present his paper "On Productivity and Distortions across Countries" at @the_IDB's Trade Policy Research Seminar Series. To participate online, please register at: bit.ly/46b5vnF. #Trade #Productivity
English
I'm very excited to share that I joined @AtlantaFed as a Research Economist and Assistant Advisor! Grateful for my incredible colleagues at @federalreserve and looking forward to this next chapter in Atlanta!
English
Could not have said it better! @HannoLustig is really providing an incredible service!
Luis Garicano 🇪🇺🇺🇦@lugaricano
Clear writing and good analysis on the pitfalls of joint debt without joint taxation. A Hamilton moment involves both. When we put together the RRF, we thought both would be coming. The failure is to expand joint borrowing when we know joint fiscal capacity will not happen.
English
Adrien Matray retweetledi
Productivity, firm size, and why distortions hurt developing economies
Diego Restuccia @UofT
ow.ly/aOrO50Y9aoo
English
Adrien Matray retweetledi
A new substack by @HannoLustig and @RWacziarg, i expect it to be interesting. I suggest you follow it.
Hanno Lustig@HannoLustig
In this blog post, @RWacziarg and I explain why the EU's 90bn Euro Ukraine deal in December 2025 was really a bad deal for Europe by moving the EU closer to a half-baked fiscal union. open.substack.com/pub/thetwocent…
English