Agus Capdevila

Economist & Msc in Finance.
Continuously leveraging knowledge in Finance, Crypto, Blockchain, and Economics. Always.

[Agus Capdevila @agusscapdevila]

Everyone keeps saying 2025 was about US equities and AI.

I disagree.

The real story of the year was the value of money and how misleading it has become to measure everything in dollars.

In 2025, the USD weakened materially:
-0.3% vs JPY
-4% vs CNY
-12% vs EUR
-13% vs CHF
-39% vs gold

Gold, the only major non-fiat monetary asset and arguably the second global reserve currency, returned +65% in USD terms.

The S&P 500 returned +18% in USD.
Measured in gold terms? -28%.

Same asset. Same year. Completely different reality depending on the unit of account.

This is the part most investors still miss:
when your benchmark currency weakens, asset returns look better than they really are.

A US investor saw +18%:
A euro-based investor saw ~+4%.
A CHF-based investor ~+3%.
A gold-based investor lost almost a third of their purchasing power.

That’s not a rounding error, that’s a regime shift.

The same logic applies to bonds.
10Y USTs returned ~+9% in USD, but -34% in gold terms. Cash did even worse. No surprise foreign demand for USD bonds keeps deteriorating unless heavily hedged.

At the same time, capital quietly rotated away from the US:
– European equities outperformed US stocks by ~23%
– China by ~21%
– UK by ~19%
– Japan by ~10%
– EM equities returned ~34%

This wasn’t a global risk-on into the US.
It was diversification away from it.

Yes, US earnings grew (+12%), margins expanded, and multiples went up. But valuations are now stretched, credit spreads are tight, and expected equity returns sit around ~4.7%, barely above bond yields (~4.9%). The equity risk premium is thin.
In other words: a lot of future returns have already been pulled forward.

The Fed will likely keep easing to manage debt dynamics and avoid funding stress, which supports asset prices nominally, buuuut.... it also continues to erode the real value of money. The adjustment isn’t default. It’s dilution.

This is why looking at markets exclusively through a USD lens has become increasingly misleading.

For years now, the dollar has been a convenient benchmark, not a neutral one.

If you measure performance in a weakening unit of account, you confuse nominal gains with real wealth creation.

That’s the macro backdrop we’re in. Not a single bubble waiting to pop... but a slow monetary repricing where the benchmark itself is the problem.

[Agus Capdevila @agusscapdevila]

For portfolio managers focused exclusively on crypto:

Beyond directional bets (which can work, depending on the asset), anyone running strategies anchored to USD-pegged stablecoins in a declining rate environment should rethink the benchmark.

At these levels, USD stablecoin yield will likely underperform emerging markets, FX carry, and global macro opportunities.
The opportunity cost is real.

And it highlights what crypto still lacks today: a proper on-chain FX market.

Who’s actually building it?

[Agus Capdevila @agusscapdevila]

@bourbonni @DamianCatanzaro Tienen misma altura, simetría, y estructura de balcón, es muy probable.

[Agus Capdevila @agusscapdevila]

@ErnestoCettour 100% de acuerdo. No con el 100% del capital que tenés, pero si es una pata y no tenes que volver a reconstruir capital desde 0, es el camino correcto, no financiero, pero si a nivel calidad de vida.

[Agus Capdevila @agusscapdevila]

@JuanSamitier Si llegan a necesitar un moderador en alguna sesión yo estoy, tengo varios años de mafia con amigos.

[Agus Capdevila @agusscapdevila]

@0xKoller Espectacular el mafia.

[Benjamin Cowen @benjamincowen]

In a few months profits will temporarily flow out of TradFi (cool off phase after IPOs).

People will want to know where to put their money because everything already went up so much.

And there Bitcoin will be, ready to begin the next four year cycle.

[Agus Capdevila @agusscapdevila]

@JuanSamitier El pt de usdat o stusdat? El pt de usdat si no me equivoco va contra la gente que está farmeando los points de Saturn no?

[Agus Capdevila @agusscapdevila]

Envejecio bien este twitt. Por muy lindo que sonara el APY y el tokenomics detras, ni en pedo lockeaba a 90 / 120 dias en una estrategia que compraba STRC y me "repartia" los dividiendos con un premium. Habia un monton de supuestos de como STRC va a manejar sus financials en base a cuando bajo pueda ir BTC de aca a fin de año.

Probablemente todo termine bien y ojala sea el caso, pero porque suscribir ese riesgo?

[Agus Capdevila @agusscapdevila]

Cuántas más estrategias en pendle van a meter para tener exposición a strc?

[Agus Capdevila @agusscapdevila]

Empiezo a dudar de lo Hawkish de Warsh en esta entrevista

@LeanZicca

[Milk Road @milkroaddaily]

Kevin Warsh just made an argument that, if he's right, sends interest rates significantly lower than the market expects (save this).

His claim is that we're at the front end of a deflationary wave driven by AI making the cost of production fall across nearly every industry.

His deeper concern is that the Fed is still running on economic models from 1978, with no institutional framework for recognizing what happens when a technology-driven productivity boom permanently changes the relationship between growth and prices.

He pointed to Alan Greenspan's famous bet in the 1990s...

Greenspan held rates steady and let the economy run hot because he believed the internet productivity boom was real and would show up in the data.

He was right. Output per hour grew 2.7% annually while inflation dropped to 1.9%, one of the most beneficial macro environments in modern history.

Warsh thinks that bet is on the table again, and that a Fed anchored to old models risks tightening into a productivity boom and strangling growth that would have otherwise been non-inflationary.

The data backing the thesis goes like this:

The cost of running frontier AI has collapsed from $30 per million input tokens in early 2023 to pennies today, and by the end of 2026, today's frontier capabilities will likely be available near-free.

When the price of intelligence approaches zero, the cost structure of every industry that uses intelligence starts falling with it.

But there's a flip side worth taking seriously.

AI-related price pressures have already added roughly 0.3 percentage points to core PCE through hardware prices, software hikes disguised as AI upgrades, and electricity costs from data center power consumption.

Microsoft raised M365 by 30%. Adobe raised Creative Cloud Photography by 50%. Intuit raised QuickBooks by 45%. All justified with AI features, all showing up as price increases in the inflation data.

Almost everyone agrees AI will eventually be deflationary if it delivers on the productivity promise.

The main fight is over timing. Is the payoff 2 years away or 10? And should the Fed act on that expectation now or wait for the data to confirm it first?

Nearly 60% of economists surveyed by the University of Chicago's Center for Markets said Warsh's AI thesis would have minimal impact on inflation or rates over the next two years.

If he's right, we're entering an environment where growth stops being a rate hike trigger, deflation comes through the supply side, and the companies investing most aggressively in AI now pull further ahead.

If he's wrong and the old models still apply, rates stay higher for longer and the productivity payoff gets pushed further out than the market has priced in.

The market is betting he's at least partially right.

If you want to track how this plays out, what it means for rates, AI infrastructure investment, and the risk assets (like crypto) positioned to benefit from a Fed that finally sees that the productivity boom is real...

Join Milk Road PRO, link bio: @MilkRoad

[Agus Capdevila @agusscapdevila]

Have been reading this carefully and want to share a few honest questions. Saying this with respect, from someone who's been holding ETH for years, and as honest thinking out loud rather than a critique.

If the EF is as confident as this suggests, and I share that confidence in the network itself, there's still something I can't quite reconcile. A lot of people who built on Ethereum or have held ETH for years no longer have clarity on the role of ETH as an asset going forward. People who viscerally don't want to sell, but honestly can't find solid reasons to keep holding instead of rotating. So either ETH is winning more clearly than it feels from where most of us sit, or there's a real gap between how the EF sees the future and how that gets translated into a thesis for the token.

Ethereum the network is still by far the best: most decentralized, best community, deep work on privacy and scalability, never down once. The question today is not whether Ethereum the network wins. I think most of us believe it does. The question is whether ETH the asset captures that win, and that's where it gets less obvious.

A few things on that.

1) Post-Dencun, value capture shifted. The roadmap moved activity to L2s, which is great for users, but L1 fee accrual got thinner. Base generates billions in revenue for Coinbase; the equivalent flow back to ETH holders is much smaller. The "ultrasound money" framing quietly inverted, with staking issuance outpacing burn most weeks now. None of that means Ethereum is losing. It means the mechanism by which network success becomes ETH price has to be re-explained.

2) Crypto's killer app turned out to be stablecoins, and the value goes to the issuers, not to the L1 they run on. Tether and Circle take the float. And now Tempo (Stripe + Paradigm) and Arc (Circle) are coming online, both designed so that flow doesn't have to pass through Ethereum. That deserves an answer.

3) At the institutional level, the story is harder to carry than it should be. BTC has one clean line: digital gold, SoV. ETH has to be a productive asset, plus ultrasound money, plus global computer, plus restaking yield, plus economic bandwidth, all at the same time. When institutions try to model it, they don't really know whether to price it as a bond, an equity, or a commodity. Tom Lee, BMNR and Etherealize are doing real work pushing the story in Wall Street rooms and they deserve credit for it. But it isn't landing yet at the scale that would move the asset, and that is not on them. It's a signal that the message itself needs sharper framing further upstream.

4) On RWA, I want to push back on the usual reassurance, the "when RWA breaks out, ETH benefits by default" line. RWA is not a future event, it's already happening. BUIDL, Ondo, Centrifuge, BlackRock's tokenized funds, most of it is on Ethereum or Ethereum L2s and Ethereum is winning the share. The issue is not whether RWA shows up. The issue is that TVL on its own doesn't move the price of ETH. Narrative does. And the narrative hasn't compounded into asset demand yet, even with the real adoption underneath. That gap is the whole problem.

5) There's also been some unbundling of the "global computer". Perps moved to Hyperliquid, NFTs largely to Solana, payments to dedicated chains. What's living on Ethereum mainnet today is RWA, blue-chip DeFi (Aave, Sky, Morpho), and settlement. A great business, just a more focused one than what was sold a few years ago.

6) And, said gently, maybe the sharpest piece: time horizon mismatch. The EF thinks in post-quantum, credible neutrality, 50-year uptime. That is exactly right for Ethereum the network. But holders are pricing 2 to 5 year cycles. "Hold for 30 years until we're the global economic hub" is a beautiful belief, but it is not an investable thesis. The investable thesis is what's missing.

None of this is a sell call to anyone else. But to be honest about where I personally landed: I rotated part of my ETH into BTC on december. I didn't want to. Over a 2 year window I see less uncertainty in BTC than I see beta in ETH, and absent a clearer asset-level thesis I couldn't justify holding the same size.

The point is humbler than it sounds. If Ethereum's roadmap and dominance are as strong as the EF believes, and I think they are, then what's missing is not engineering and it's not adoption. It's a clear, repeatable, digestible line of communication that translates all of that network value back into ETH as an asset, in a way the community and investors can actually hold on to.

The moment we get that line, a lot of the noise goes quiet. And honestly, the moment we get that line, I rotate back. But somebody really has to draw it.

[⟠ @ryanberckmans]

Not sure if this may clear things up or muddy it further - I just know we are winning and it's time to get fully back to building and remember that our patience will be rewarded:

imo it's a common misconception that the EF and/or Vitalik don't care about the price of ETH.

They do care, very much, because they want Ethereum to be globally ubiquitous for a thousand years, and they know that this audacious goal requires lots of resources and economic security which can only come from a terrific ETH valuation.

The reason the EF has often over the years appeared not to care about the price of ETH is two-fold:

One, the EF is overall insanely confident in Ethereum and in ETH. As they should be. They've earned it and earn it every day. So when we are bearish or scared about the spot price, it's just effectively noise from their perspective of strong conviction and focus.

Two, the EF cares about price in long term structural ways that are incomprehensible to many of us.

We want to know why Didn't Number Go Up in Q4 Or Yesterday.

Whereas they want to know, "How will Ethereum remain dominant after quantum computers?" and, "How will Ethereum be the world's economic hub for trillions in assets and thousands of L2s across a hundred countries?" These are inherently bullish questions. And their programs/answers in response are gigabullish.

The EF departures are not because the people departing feel differently about Ethereum and our trajectory vs. the people staying at EF or vs. community folks like me.

The EF departures are because -

Even benevolent special smart wonderful people naturally have internal politics and differences of opinion over substrategies, policies, etc.

Vitalik and his leadership team feel the EF should be run in a certain way. Some folks disagreed. Some tiny number were asked to leave for Reasons. Some few others left immediately due to Reasonable Net Feelings. Some more are leaving because the Wheel is Turning and they feel that while we all love Ethereum and are extremely excited for our roadmap and proud of past wins, the time for new blood is here.

New blood means genzeth and also young up and comers who are ready to take the reins of their teams and departments.

What's important is that the EF's determination is as strong as ever and its strategy and focus are better than ever before. Credible neutrality. Decentralization. 100% Uptime. Postquantum. Privacy. Scaling L1+L2. Unifying/improving UX for L1+L2. The EF is on it.

What's also important is that the EF is now complemented and balanced by a growing cohort of deeply invested elite eth orgs across the stack/verticals/technologies/go-to-markets, including top L2s like Base, Arb, and zkSync, DATs like BMNR and SBET, and enterprise groups like Etherealize and Consensys, and too many more orgs and kinds of orgs to list here.

We will miss the great EFers leaving this season.

The EF is not only going to be fine, they are going to be amazing. Let the wheel turn. We're ready.

This bear market- secularly in crypto and in terms of global issues- is unfortunate, but that's the industry.

ETH will hit multi trillion in due course. Strap in, be patient. Help out. Get involved. I've been here for 8 years now full-time and it's never more felt like I'm just getting started.

Ethereum.

[Agus Capdevila @agusscapdevila]

Yo entiendo lo que dice Ari, no se si lo dijo particularmente para tus primero 100k. Pero si creo que si logras acumular un capital tal que la compra de la casa/depto sea < 40-50% de tu capital total, por mas que no sea lo optimo financieramente, lo banco bastante. Es una realización material importante de todo el esfuerzo acumulado de años y te da otra tranquilidad de cara a la vida. El resto de los 50/60% los seguis invirtiendo y componiendo, esta claro que con la compra de una propiedad no te podes quedar en 0 en la comitente.

[Agus Capdevila @agusscapdevila]

You’re oversimplifying things.

Ethereum in 2020–2021 was in the middle of DeFi Summer, with extremely high gas costs. That meant that as network activity exploded (DeFi growing, NFTs booming, and overall on-chain demand increasing) fee burn was massive, which supported a much stronger ETH valuation model.

After Dencun and the broader L2 scaling push, Ethereum intentionally prioritized scalability so L2s could grow and businesses on top of Ethereum could operate more efficiently. That made sense strategically, but the value accrual to ETH itself became significantly weaker than it was in 2020–2021.

That’s a big part of the price action we’re seeing today.

And this is coming from someone who still believes Ethereum is the best blockchain network and wants it to succeed. But the reality is that the Ethereum story in 2020 is very different from the Ethereum story today.

[Agus Capdevila @agusscapdevila]

Si quieren entender a fondo el delirio de lo que acaba de reportar NVIDIA 👇

[Matias Scalbi @MatiasScalbi]

Reportó $NVDA y te explico todo sencillo

- Revenue USD 81,6B (+85% YoY)
- Data Center USD 75,2B (+92% YoY)
- Gross margin 75%
- FCF USD 48,6B en el trimestre
- Nuevo buyback de USD 80B + dividendo x25

Las 4 megacaps gastan USD 725B en CapEx. NVIDIA los está cobrando 👇

[Agus Capdevila @agusscapdevila]

Excelente Mati, te agrego un dato de cara a futuro por si no lo tenías en mente:

https://t.co/ds65FywBCC .

Si esto efectivamente se lleva a cabo, los costos de financiamiento de los prestamos para comprar gpus se reducen bastante porque están hedgeados con contratos futuros, potencialmente aumentando aún más la demanda que existe hoy .

[Ornn @OrnnExchange]

Every trader can now take a position on the most important commodity of the AI infrastructure buildout: compute.

@OrnnExchange x @ICE_Markets are partnering to list the first transaction-based compute index on a regulated exchange today.

Compute is a commodity.

https://t.co/hP0vyJU08G

[Agus Capdevila @agusscapdevila]

Estamos en un momento raro, donde los precios en cripto estan todavia relataivamente deprimidos, y a su vez, la tokenizacion esta tomando vuelo a niveles muy superiores a lo que pense y con una curva acelerando mas la pendiente aún.

No paro de charlar con empresas y entender mas a lleno el alcance que esta teniendo y me vuelve loco, es lo que crei que iba a pasar hace varios años y nunca terminaba de verlo y ahora efectivamente es imparable. Me pone muy contento y me reafirma la pasión.

Una vez se despeje el ruido de Irán (que debería ser pronto por una cuestion de como se esta comportando la curva de bonos del tesoro, la inflación estacional y como eso afectar al view de path de tasas + midterms de trump) creo que vamos a ver despegar los precios rapido.

Sinceramente le perdi un poco la fe a ETH y rote mayoritariamente a BTC. Aunque ETH pueda tener mas beta, el value accrual y la cantidad de competencia me dejan mucha duda de como puede performar a futuro.
Mi bet era que la EF tenia un plan para meterle mas fuego a la L1 y cambiar el business model de las L2 + tracción de los ETF con staking como una especie de onchain yield para institucionales con exposición direccional en un mercado evidentemente creciente.

Pero hoy ya no lo tengo tan claro, prefiero apostar mas safe, por ende la rotación a BTC.

Esperemos el balance de NVIDIA y buenas noticias sobre Iran.

[Agus Capdevila @agusscapdevila]

@belisarioAT Coincido, me junte con varias alycs en las últimas semanas y, obvio que hay gente que sabe una banda, pero muchos de los FA no tienen real conocimiento del mercado, entonces venden empaquetados, sin gestión de riesgos y sin evaluar perfil o temporalidad en serio.

[Agus Capdevila @agusscapdevila]

@fibobeta Que flojo el flujo de eth la verdad. Cada vez estoy más cerca de soltarle la mano del todo.

[Agus Capdevila @agusscapdevila]

@BTC_for_Freedom They buy OTC, different market.

[Agus Capdevila @agusscapdevila]

@aetchebarne @JMilei No vale la pena resignar la posibilidad de tener política monetaria por un miedo futuro, sobre todo si tiene 4 años más para terminar de encaminar la macro y con superávit comercial como se proyecta. Cortenla con esto.

[Agus Capdevila @agusscapdevila]

Unpopular opinion: pero armar un agente de AI, si bien requiere agency, es mas facil que saber usar excel (sin la API de claude).

[gaston parisier @maghico22]

Vamo’ $GLOB

[Agus Capdevila @agusscapdevila]

Ok, esto está muuuuy bien.

[Centrifuge @centrifuge]

Instant redemption for Centrifuge's tokenized assets, in partnership with @grovedotfinance.

Through Grove Basin, investors can access up to $1B in committed near-instant daily liquidity.

Among the first supported assets: $JTRSY, Janus Henderson Anemoy Treasury Fund.

A turning point for tokenized Treasuries to become institutional-grade cash management instruments, with the liquidity profile to match.

Powered by Centrifuge.