Sign in

Eduardo Levy Yeyati

@eduardoyeyati.bsky.social
257 followers 43 following 25 posts

Economista & escritor | DiTella & @Brookings.edu & AdCap | Finanzas + IA & Futuro del Trabajo + Literatura de género | Últimos libros: #Automatizados (Planeta) & #ElJuegodelaMancha (Random House)

PostsRepliesMedia
Reposted by Eduardo Levy Yeyati
VoxEU @ CEPR @voxeu.org · 27/08/2025
Digital currencies pegged to fiat money face a built-in tension between credibility and competition. @eduardoyeyati.bsky.social‬ & Sebastian Katz analyse competing frameworks in the #US and #China to address this #stablecoin paradox. cepr.org/voxeu/column... #EconSky
Graph of stablecoin market growth vs US M2 and Treasury holdings, 2020-2025.

Digital currencies pegged to fiat money face a built-in tension between credibility and competition, creating a ‘stablecoin paradox’. This column analyses the competing frameworks of the US and China to address the paradox. The US (under the GENIUS Act) combines strict reserve rules for privately issued, fully backed dollar stablecoins with market incentives that encourage expansion. Meanwhile, China’s approach centres on a central bank digital currency deployed domestically and extended across borders. Further expansion of the digital currency system will require backstop access, global coordination on regulation, limits on intermediation, and integration with payment systems.
045
Reposted by Eduardo Levy Yeyati
Centre for Economic Policy Research @cepr.org · 23/09/2025
On 22 September, the US treasury secretary said that the US "stands ready to do what is needed" to stabilise the Argentinian economy. Revisit this 2024 VoxEU column by S Katz & @eduardoyeyati.bsky.social to see how #Argentina diverged from other developed economies. cepr.org/voxeu/column... #EconSky
cepr.org
When did Argentina lose its mojo?
There are recurrent debates about the ongoing underperformance of Argentina’s economy. This column proposes a metric to characterise and date Argentina’s divergence from a benchmark group of countries with comparable initial incomes per capita. The authors find Argentina’s divergence to be longer than usually conjectured, with marked tranches in the first half of the 20th century and the post-WWII period. They identify specific dates for the inflection points, discuss the context in each case, offer an explanation for the enduring divergence, and describe Argentina’s volatile chapter since the 1990s.
022
Reposted by Eduardo Levy Yeyati
VoxEU @ CEPR @voxeu.org · 24/11/2025
AI fundamentally alters the balance between formal credentials and verified experience in labour markets. @eduardoyeyati.bsky.social‬ reviews employment record systems from 8 countries to establish design principles and draw lessons from implementation failures. cepr.org/voxeu/column... #EconSky
The rise of AI fundamentally alters the balance between formal credentials and verified experience in labour markets, but without verified employment records, experience remains invisible. This column reviews employment record systems across eight advanced and emerging economies, as well as EU-level initiatives, to establish design principles and draw lessons from implementation failures. The analysis reveals that technical infrastructure alone does not guarantee success. Effective systems require aligned incentives, continuous stakeholder engagement, and careful sequencing.
042
Reposted by Eduardo Levy Yeyati
VoxEU @ CEPR @voxeu.org · 02/02/2026
@eduardoyeyati.bsky.social shows that non-degree credentials can deliver substantial wage gains if they are job-relevant, especially for workers without college degrees and those early in their careers. cepr.org/voxeu/column... #EconSky
Credentials aren’t distributed evenly across the labour market. The figure shows that they cluster heavily in knowledge-intensive occupations: 23.4% of computer and mathematical workers hold non-degree credentials, along with 16.9% in business and finance, and 16.4% in architecture and engineering – all more than triple the 7.2% overall average.

Non-degree credentials are rapidly multiplying as alternatives to traditional higher education, but in a largely unregulated market. This column draws on nearly 40 million US worker resumes to analyse the returns to non-degree credentials. The findings suggest that they can deliver substantial wage gains if they are job-relevant, especially for workers without college degrees and those early in their careers. However, these are also the workers who are least likely to hold such credentials. Changing this requires not just more credentials, but better information, lower barriers, and quality assurance that protects workers from low-value programmes.
143
Reposted by Eduardo Levy Yeyati
VoxEU @ CEPR @voxeu.org · 19/03/2026
@eduardoyeyati.bsky.social argues that automation is the right place to start the debate about AI and jobs but not the whole problem. The relevant race in the AI transition is now more to do with adoption speed versus retraining capacity. cepr.org/voxeu/column... #EconSky
Figure: The parameter κ denotes the speed of adoption: higher κ means the same long-run automation occurs more quickly. Panel (a): labour force participation — permanently lower under fast adoption. Panel (b): non-routine employment — paths cross as early gains reverse later. Panel (c): non-routine wages — mirror image of panel (b). Panel (d): displacement flow — the same total displacement compressed into a much shorter interval under fast adoption.

Most debate about AI and jobs still starts with the automation frontier: how many tasks machines can eventually perform. This column argues that is the right place to start, but not the whole problem. The next big labour shock may come less from new technological breakthroughs than from the race between retraining capacity and the – still incipient – adoption of tools that are already out there.
042
Reposted by Eduardo Levy Yeyati
VoxEU @ CEPR @voxeu.org · 15/06/2026
Foreign exchange reserves are generally associated with lower sovereign spreads. @eduardoyeyati.bsky.social shows that only privately financed reserves are robustly associated with tighter spreads and lower crisis risk. Debt-financed reserves show a weak connection. cepr.org/voxeu/column... #EconSky
Figure: Effect on the sovereign spread, by financing source. A one-percentage-point increase in privately financed reserves as a share of GDP is associated with spreads roughly 1.3% lower, while the public-borrowing component carries a smaller coefficient significant only at the 10% level, and IMF lending is insignificant. Within private flows, current account surpluses do most of the work: a one-point increase is associated with spreads about 1.8% lower, against roughly 1.0% for financial inflows. Translated into levels at the sample-average spread of roughly 350 basis points, a one-point-of-GDP leaning-against-the-wind accumulation tightens spreads by about 4.4 basis points, rising to about 6.4 points from the current account and falling to about 3.6 points for financial inflows.

Emerging market economies hold foreign exchange reserves to insure against currency crises and to cushion exchange rate movements. Reserves are generally associated with lower sovereign spreads. This column argues that how those reserves are built up is also important. Using a balance-of-payments classification that separates reserves accumulated through private inflows from those financed by public external borrowing, it finds that only privately financed reserves are robustly associated with tighter spreads and lower crisis risk, while debt-financed reserves show a weak or insignificant connection. For inflation-targeting floaters, the findings leave room for measured, macroprudential interventions rather than habitual currency management.
032
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 19/07/2025
Latin LLMs: Buy, Build or Fine-Tune? LatamGPT aims at a regional language model trained in regional data at Chile’s solar-powered supercomputing center. Switzerland’s own in-house LLM shows that sovereign AI is possible—but is it the best path for Latin America? www.linkedin.com/pulse/latin-...
linkedin.com
Latin America's search for AI sovereignty
From the foothills of the Andes, a group of Chilean engineers is building an ambitious AI experiment: LatamGPT, a regional language model trained not in California, but at Chile’s solar-powered superc...
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 30/06/2025
"Latin America faces a choice that will determine its economic future: develop AI governance on its own terms, or become a regulatory colony of Silicon Valley and Brussels." Escribí sobre cómo pensar la regulación de la IA en América Latina. americasquarterly.org/article/regu...
americasquarterly.org
Regulating AI on Latin America’s Terms
The region has a narrow window to develop its own approach, taking inspiration from international frameworks while adapting them to local needs.
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 16/06/2025
"Credentials say what you know, employment records show what you’ve done. If we can’t track experience, we can’t recognize it and reward it. And experience will soon be our last chance to complement the algorithm." My new piece for @brookings.edu. www.brookings.edu/articles/emp...
brookings.edu
Employment records: The missing piece in the US labor market?
Eduardo Levy Yeyati discusses employment records systems and explains how they could potentially transform the U.S. labor market.
001
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 13/06/2025
“We have public education, health, and transportation policies, but we don’t have public digital connectivity policies. That is absolutely inexplicable today.” Chile-backed model Latam-GPT plans to make its public debut in September. foreignpolicy.com/2025/06/13/l...
foreignpolicy.com
Where Does Latin America Stand in the Global AI Race?
Chile-backed model Latam-GPT plans to make its public debut in September.
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 02/06/2025
On Argentina's stabilization and growth, and automation & the future of work (in Spanish). www.ambito.com/economia/edu...
ambito.com
Eduardo Levy Yeyati: “Una fuente de riesgo es que en 2017 parecíamos haber salido de la crisis, pero duró poco“”
El economista analiza las condiciones actuales para atraer inversión productiva. Resalta la estabilidad del dólar y el resultado electoral como requisitos clave para invertir. También advierte por la ...
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 22/05/2025
With fewer legacy systems, growing tech ecosystems, and a strong tradition of social rights-based regulation, Latin America is a natural sandbox for forward-looking AI governance. In a recent piece, I lay out the key ingredients and essential steps for smart AI regulation in LAC.
110
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 21/04/2025
www.infobae.com/educacion/20...
infobae.com
Eduardo Levy Yeyati: “La inteligencia artificial pone en crisis la centralidad del trabajo en nuestra sociedad”
En un encuentro organizado por Ticmas y el Grupo de Fundaciones y Empresas (GDFE), el economista conversó sobre el impacto de la inteligencia artificial en el trabajo, la educación y la estructura soc...
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 14/04/2025
We wrote with Juan Carlos Hallak for Americas Quarterly about Latin America's failed experiment with import substitution industrialization (ISI), and the lessons that can be drawn for the U.S. tariff war. americasquarterly.org/article/fail...
americasquarterly.org
Failed Protectionism: What Latin America Can Teach Us
The region’s failed experiment with import substitution industrialization (ISI) offers a stark warning: embracing it may breed inefficiency, corruption, and stagnation.
010
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 24/03/2025
Una lección de las corridas de los 80-90 fue que una flotación sucia es preferible (más resiliente) a una banda porque no ofrece al especulador un blanco (el techo de la banda) que atacar. De ahí la prevalencia de los regímenes flexibles en el nuevo milenio. Fuente: ideas.repec.org/p/sad/wpaper...
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 23/03/2025
Gen AI offers an opportunity for education—not to replace teachers, but to support them. In this study we show how AI can act as a scaffold—not a shortcut—for learning, and why LAC is well-positioned to lead the integration of AI in the classroom. Link (English & español): utdt.edu/ver_contenid...
131
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 21/03/2025
"Digital advertising excels at predicting cultural fit. [...] If these factors can be used to sell cars, it is hardly a leap to imagine they could also help predict job fit." My piece on alternative job scoring for @brookings.edu. www.brookings.edu/articles/dig...
brookings.edu
Digital footprints and job matching: The new frontier of AI-driven hiring
Eduardo Levy Yeyati and Ian Seyal discuss how AI can be responsibly leveraged to assess the more qualitative attributes important to hiring, like skills and cultural fit.
010
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 19/03/2025
Our paper on What Works for Active Labor Market Policies is already available online. Ya está disponible online nuestro artículo sobre qué funciona en politicas pro empleo, donde comparamos evaluacion de experiencias recientes en el mundo. Link: economia.lse.ac.uk/articles/10....
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 18/03/2025
History teaches that harder than containing inflation with an exchange rate anchor is exiting the anchor in time. When the moment comes, nobody wants to change a winning strategy; once the moment´s passed, there is no easy way out. "You can check out any time you like, but you can never leave."
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 17/03/2025
La confusión entre democracia y liberalismo nubla el debate sobre la nueva derecha, que es a la vez democrática (popular, incluso populista) y antiliberal (aversa a la libertad de expresión y a la separación de poderes). Su oposición natural no sería la izquierda iliberal sino un liberalismo amplio.
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 16/03/2025
Escritores britanicos sobre el cuento corto hecho por el modelo de escritura creativa de Open AI: “un elegante vacío” (Nick Harkaway); “genuinamente inteligente” (David Baddiel); “hermoso y emotivo” (Jeanette Winterson). www.theguardian.com/books/2025/m...
theguardian.com
‘OpenAI’s metafictional short story about grief is beautiful and moving’ | Jeanette Winterson
I think of AI as alternative intelligence – and its capacity to be ‘other’ is just what the human race needs
110
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 19/02/2025
"DeepSeek suggests the future may be defined not by a single dominant technology but by a range of competing standards. Regardless, LatAm must prevent swapping one form of dependence for another." My piece on the implication of DeepSeek for LatAm's AI agenda: americasquarterly.org/article/deep...
americasquarterly.org
DeepSeek Reveals Latin America’s AI Crossroads
China’s new large language model, LLM, offers the region a chance to seize AI on its own terms but also brings risks of dependency.
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 17/02/2025
[ARCHIVO] Hace un par de años escribimos sobre el mundo cripto con un querido amigo y distinguido "econochanta". #libra #milei www.brookings.edu/articles/cry...
brookings.edu
Cryptomania: The good, the bad, and the ugly
Eduardo Levy Yeyati and Sebastian Katz critique the crypto revolution and the advent of central bank digital currencies.
101
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 21/12/2024
Se la va a extrañar. QEPD. www.youtube.com/watch?v=UeGE...
youtube.com
Beatriz Sarlo, con Levy Yeyati: “A los políticos los juzgo viendo cómo les fue en su territorio”
YouTube video by Todo Noticias
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 21/12/2024
Mi nota sobre el rol del conocimiento práctico en la complementación entre humanos y algoritmos para el blog de at @brookings.edu. My piece on the role of experiential knowledge in strengthening the complementarity between human workers and algorithms. www.brookings.edu/articles/why...
brookings.edu
Why gen AI can’t fully replace us (for now)
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 29/11/2024
Cómo regular el cambio tecnológico en su fase exponencial sin trabarlo o mirarlo por el retrovisor? Para orientar el debate escribimos con Angeles Cortesi esta guía introductoria sobre regulación de IA en LAC (versión en español pronto en este mismo link): utdt.edu/ver_contenid... Translate post
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 25/11/2024
Hay un meme circulando donde alguien pregunta a ChatGPT qué puede decirle de sí mismo y el algo le da una descripción "profunda" de su personalidad. No te dejes engañar pensando que hay algo profundo ahí: es sólo un algoritmo actuando como un horóscopo. simonwillison.net/2024/Oct/15/...
simonwillison.net
ChatGPT will happily write you a thinly disguised horoscope
There’s a meme floating around at the moment where you ask ChatGPT the following and it appears to offer deep insight into your personality: From all of our interactions what …
000
Eduardo Levy Yeyati @eduardoyeyati.bsky.social · 24/11/2024
Con Luciana Vazquez hablamos de cómo la IA trae oportunidades o riesgos para el país según nos preparemos, por qué Elon & Sam promueven el ingreso universal, y cómo creo que el #hechoxhumanos será la marca identitaria que defina la frontera del trabajo. youtu.be/MWGj918EO-I?...
youtu.be
Milei y la utopía argentina de la IA: ¿por qué no alcanza con macro ordenada y desregulación total?
YouTube video by LA NACION
000