Amazing session on one of my favorite topics at GRAPE Summer Workshop on Macro and Finance: determinacy and uniqueness. Lukasz Rachel and Tobias Kawalec (who will be on the job market).

Bottom line: if you want your structural estimates to be more robust to model misspecification, be sure to use recentered instruments! arxiv.org/abs/2504.04056

When knowledge spillovers are local, firms may under-invest in geographic expansion. This is studied using a spatial growth model, and detailed data on US firms, from Craig A. Chikis, Benny Kleinman, and Marta Prato nber.org/papers/w34010

Very grateful to Noah Smith 🐇 for sharing my London talk and using my slide deck to highlight key demographic challenges for the coming decades. I’m a strong believer in open science: academics should share slides, papers, and materials openly to foster honest, rational debate.

🎧 The latest episode of Value Investing with Legends featuring Clifford Asness (co-founder of AQR Capital Management) is truly exceptional; one of the best podcast episodes I’ve ever listened to: podcasts.apple.com/us/podcast/cli… Let me first share why I always listen to this podcast

🎉 Thrilled our paper "The Sectoral Origins of Heterogeneous Spending Multipliers" is now accepted at Journal of Public Economics! We explore why the output response to government spending crucially depends on which industries receive the funds. Details in thread below! 👇📈 #EconTwitter

I'm so excited to see this paper coming out in RESTUD. Fun facts about this paper: 1) Bruno and I had never met in person until after we completed the first draft. 2) It's my first and only paper that was never rejected by a journal. 3) I can now claim to understand macro.

📦 Can industrial policy work? Yes—the East Asian experience shows it can (at least partially). But its success rests on a key condition: labor control. 🇯🇵🇰🇷🇹🇼 Japan, South Korea, and Taiwan industrialized rapidly under authoritarian or semi-authoritarian regimes. Wages and

It’s tiresome to debate labels, so I observe: China’s economic development has been inclusive. Almost everyone has benefited enormously. But clearly, capitalists (the people who own the capital stock & mostly get to keep the returns) are still the primary economic beneficiaries

We offer a winter school to bring you (from almost nothing) to the frontier in HANK modeling including estimation, higher order solutions, and ambiguity-averse households. For PhD, Postdocs, and Professionals!

🎉 Thrilled to share that my JMP "Uncertainty and Unemployment Revisited: The Consequences of Financial and Labor Contracting Frictions" is accepted at JPE Macro! In short: 𝐔𝐧𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲 𝐢𝐬 𝐜𝐫𝐮𝐜𝐢𝐚𝐥 𝐟𝐨𝐫 𝐮𝐧𝐞𝐦𝐩𝐥𝐨𝐲𝐦𝐞𝐧𝐭. 🔗 yajiewang.info/research

I'll post the paper in a little bit, but here's a preview of some results on the dynamic propgation of stochastic shocks and their worst case amplfication -- which also covers model uncertainty -- over networks. I focus on production networks and information propagation.

More teasers: Title page plus an interesting result Most models of production networks are static, while I analyze dynamic flows on networks. But for worst-case/H∞ performance & robustness, only frequency zero/steady state matters. (maximum singular value of Leontief inverse)

Introducing **China Modern Gazetteer Full-Text Database** — a data project w/ Leo Yang 杨阳 & Ting Chen ct. Gazetteers cover major county-level events & stats in early PRC, Republican Era, and late Qing. Visit: cngazetteer.com youtube.com/watch?v=Coh6tw…

Interested in Micro-to-Macro Development research? Save the date of this workshop on December 9-10, 2025, in Montevideo And send us your best papers on this topic Submission Deadline: September 13th Submission Link: ridge.org.uy

Tomorrow, Wed. 20, Matteo Maggiori and I will review the state of Geoeconomics at the Lunchtime Session of the 2025 World Congress of the Econometric Society 👉 eswc2025.org I’ve had a sneak preview of Matteo Maggiori’s slides, and they are fantastic. This promises to be a

📢📢 Call for Papers 📢📢 We're organizing a *macroeconomics workshop* in Amsterdam 15-16 December 👌 ✅ Any topic in macro broadly defined ✅ Two nights in a'dam + EU travel covered ✅ Keynote by virgiliu midrigan Midrigran Submit by October 5. Call here: tinbergen.nl/event/2025/12/…

Almost a decade ago I was the Harvard economist that said that bitcoin was more likely to be worth $100 than 100k. What did I miss? I was far too optimistic about the US coming to its senses about sensible cryptocurrency regulation; why would policymakers want to facilitate tax

Programming with AI is insanely fun. Process is: 1. generate code 2. read & understand code that was generated 3. make small changes "manually" (still with great autocomplete) 4. test & debug 5. make big changes with new prompt 6. go back to step 1 Pure vibe coding skips step 2

Totally ahead of its time. Still learning from that paper. RIP to a giant.