What this covers
Vitalik Buterin discusses cryptoeconomics—economics applied under the hard constraints of code, anonymity, and cryptographic verification—as the conceptual foundation for blockchain design. The conversation moves across technology, economics, finance, and philosophy, examining why blockchains matter and what they actually solve. Rather than defending decentralization as a categorical good, Buterin argues that blockchains' real value lies in [credible commitment](concept:Credible Commitment) against monopolistic control, a function served through logical centralization alongside architectural and political distribution. He contends that the persistence of financial rent—from bid-ask spreads to high-frequency-trading profits to cryptocurrency exchange fees—reflects fundamental economic frictions and signaling behavior, not technological inevitability.
The discussion ranges over several domains where conventional thinking misses economic structures beneath the surface. On finance, Buterin explains why adverse selection and information asymmetry make bid-ask spreads necessary, and why high-frequency-trading profits remain constant even as latency has collapsed—the profits depend on an arms race of speed relative to competitors, not absolute speed. On blockchain design, he clarifies types of decentralization, addresses the limitations of smart contracts when external data is required, and argues that proof of stake would reduce institutional trust requirements relative to proof-of-work mining. On broader culture, he explores cost disease in healthcare and education as a function of approval structures, economies of scale, and costly signaling rather than technological regression, and examines how geography, regulation, and valuation models shape cryptocurrency adoption.
Buterin argues that cryptoeconomics is real economics applied under unique constraints (full code specification, anonymity, inability to push utility below zero), and that blockchains' core value is credible commitment and logical centralization rather than full decentralization, while many perceived market and social 'constants' (financial fees, cost disease, HFT profits) reflect rent extraction and signaling rather than technology.
- Crypto mechanisms must be specified to programmer-level exactness and cannot punish anonymous actors beyond their staked capital
- Blockchains are logically centralized and that is good; their unique advantage is credible commitment against monopolistic behavior
- Persistent fee/profit constants in finance and HFT reflect rent extraction independent of technological speed
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
Bid-ask spreads must exist for fundamental reasons: a zero-fee market maker effectively earns the geometric mean rather than the arithmetic mean of two assets and thus loses money, and adverse selection means that anyone willing to take your offer is evidence the offer may be mispriced, so traders rationally avoid pricing bids too close to others' asks—making spreads a mix of technological and fundamental inefficiencies.
“instead of earning the average of the earnings of two assets, you’re earning the geometric mean of those two assets, which is lower than the arithmetic mean, and so you lose a bit of money.”
Although the time for price disequilibria to snap back has fallen from about 1 second in 2000 to about 50 milliseconds, high-frequency-trading profits have stayed roughly constant, because the rate of rent extraction depends on the arms race between canceling stale orders and snapping up mispriced ones—not on absolute technological speed—so the rate of profit and social loss is independent of speed.
“even though that time has decreased from something like 1 second to something like 50 milliseconds, the rate of profit that these HFT firms have been making has stayed pretty much constant.”
A blockchain is one of the few tools that allows a platform to credibly commit to not turning into a monopolistic gatekeeper, making it an alternative to platform monopolies.
“the general idea that a blockchain is one of the few tools that allows you to credibly commit to not turning into a monopolistic jerk is actually a really interesting insight”
Quantum computing will break much modern cryptography—digital signatures, public key encryption, and SNARK zero-knowledge proofs—but replacements were developed over a decade ago (hash-based signatures, STARKs, elliptic curve isogeny-based encryption), so it forces a transition rather than catastrophe; the cryptocurrencies that suffer most will be those whose governance is too stalled to adapt in time.
“quantum computing will make a lot of cryptography that’s used in modern times just not work any more. Digital signature algorithms won’t work any more.”
Cost disease in healthcare and education arises from several factors: societies lacking it have a 'just get things done' culture while afflicted ones require many approvals (including intra-institutional conformism); getting things done cheaply requires getting in people's way, which is harder in established societies with more expectations; wealthier societies do more signaling expenditure (e.g., impressive campuses) without admitting it; and some efficiencies require economies of scale absent in low-density contexts where decentralized solutions make more sense.
“places that seem to not have cost disease have more of a “just get things done” culture. Whereas places that do seem to have more of this culture of “oh, you need five kinds of approvals””
Smart contracts are frictionless only when everything stays within the system; for purely on-chain transfers or financial derivatives they are frictionless, but for events requiring external real-world data the Oracle problem raises costs—publicly verifiable data (like a price feed in MakerDAO) is manageable, while transaction-specific data (like verifying a house fire for insurance) is expensive, so smart contracts alone won't solve it and built-in arbitrator mechanisms are needed.
“moving ether from one account to another is absolutely frictionless. Converting ether into fiat currency is absolutely not frictionless.”
Contrary to common perception, moving from proof of work to proof of stake would on net require less institutional trust, because proof of work is already highly centralized—Jihan Wu controls mining pools totaling 42 percent of Bitcoin's network, and 42 percent is enough for selfish mining that drives out competitors and enables a 51 percent attack, so users already trust large miners not to collude.
“In Bitcoin, there is one person, Jihan Wu, who controls the largest two mining pools in bitcoin that add out to 42 percent of the network.”
Physically attending conferences is an indispensable way to learn both what people are thinking about and how people think in a space, knowledge that cannot easily be found on the internet.
“It’s always kind of an indispensable opportunity to physically go to places and actually talk to people, because that way you learn things about both what people are thinking about and how people are thinking in the space that you can’t easily find from the internet.”
Decentralization has three distinct dimensions: architectural (not running on one piece of hardware, fault tolerant), political (not run by one person or small cabal), and logical (whether the system is one canonical thing); blockchains pursue architectural and political decentralization but are logically centralized, like the English language which is architecturally and politically decentralized yet logically one thing.
“I wrote this blog post a couple of years ago where I talk about different kinds of decentralization. One I talk about is architectural decentralization”
A blockchain functions like a world computer with a hard drive storing accounts, smart contract code and memory, accepting signed transactions and processing them by rules; on top of ordinary computation it adds trust guarantees that the computer will run as expected and cannot be subverted by a few people going out of business, getting hacked, or turning evil.
“a blockchain, as a whole, functions like a computer. It has a hard drive, and on that hard drive, it stores what all the accounts are.”
Even if cryptocurrencies do nothing intrinsically, if there is an established equilibrium where they grow at roughly the same rate as other assets but in a way at least somewhat uncorrelated, then pure portfolio theory justifies holding some of your wealth in them—consistent with the gold portfolio-competition framing.
“as long as they grow in a way that’s at least somewhat uncorrelated from other things, then it makes sense from a pure portfolio theory to have some of your money in them.”
Prediction markets are thin partly due to regulatory barriers and the friction of traditional financial systems, and crypto's higher liquidity and frictionlessness could help blockchain prediction markets succeed; but it is also possible that this kind of betting simply isn't interesting enough for amateurs to provide volume for professionals—still, with only a few years and a couple of approaches tried, it is too early to call prediction markets a failed experiment.
“I don’t really think it’s fair to say that prediction markets as a category can be really called a failed experiment yet, given that there have only been a few years of trying to make them work”
Because rents in Silicon Valley are too high and California keeps blocking new housing supply, basic economics dictates most developers will live outside major centers; as remote-communication technology improves, the premium for being inside a center will shrink, so capital-light, lone-wolf-friendly industries (especially crypto) will become more geographically distributed.
“If rents continue being too damn high, then it’s a matter of basic economics and math that the majority of developers are going to live in places other than the big centers.”
The cryptocurrency ecosystem has already gotten people to accept micropayments—specifically paying transaction fees—even though many thought that was impossible.
“the cryptocurrency ecosystem has, first of all, managed to get people to accept micropayments — specifically paying for transaction fees — when a lot of people thought that that’s impossible.”
The theory that you can go back in time but fate ensures your actions still produce the present is computation-theoretically implausible since it would require a conscious god pushing consequences to nullify them; if backward time travel is possible at all, it means traveling diagonally into a different universe rather than altering one's own timeline.
“I definitely don’t believe in the theory that says you can go back in time but somehow the forces of fate ensure that whatever you cause to happen ends up turning into whatever is the modern world. That just seems computation theoretically incredibly implausible.”
Issuing a cryptocurrency is costless, but issuing one that people value is not; the equilibrium is a recognized 'club' of valuable cryptocurrencies, and joining it requires costly signaling expenditure (burning capital or consuming something unique) difficult enough to prevent costless proliferation that would otherwise hyperinflate the entire space.
“issuing a cryptocurrency by itself is costless, but issuing a cryptocurrency that people care about is not costless. The equilibrium is that there exists this club of cryptocurrencies that people recognize as having value.”
Cryptoeconomics is genuine economics specialized to particular constraints: mechanisms must be fully specified to the exactness of computer code, actors are anonymous so you cannot drag their utility below zero (you can only take the capital they stake), combined with insights from cryptography, information theory, and distributed systems to design mechanisms achieving desired properties.
“Cryptoeconomics is basically taking economics with those particular constraints and then adding together insights from fields that are fairly close by to the cryptocurrency space — particularly cryptography, information theory, math, and distributed systems”
One way to value a cryptocurrency is to treat it like a corporation collecting transaction-fee revenue (especially when fees are burned or redistributed to proof-of-stake validators) and compute the net present value of those fees; for Ethereum, ~$500,000/day in fees (~$180 million/year) implies a 'P/E ratio' in the low 200s—high for a company but not absurd—while alternative models include MV=PT medium-of-exchange and store-of-value framings.
“Ethereum’s transaction fees tend to be about $500,000 a day recently, which is about $180 million a year. If you tried to value the ether market cap as some kind of corporation, then the “P/E ratio” is only somewhere in the low 200s”
Ethereum's biggest technical bottleneck is scalability: current capacity is about 15 transactions per second, while putting Uber rides on-chain needs ~12/sec, PayPal needs hundreds, and more complex applications need thousands; solutions include sharding and base-layer upgrades plus Layer 2 technologies like state channels and plasma, alongside major needed improvements in user experience and security.
“Ethereum blockchain’s capacity right now is about 15 transactions a second. If you even consider something like putting all of the Uber rides on the blockchain, that’s 12 transactions per second already.”
Regulatory changes would not help much on the technology-development side because the US already has code-is-free-speech precedents that most other countries follow in principle; the harder regulatory question concerns wider deployment, where easing large money into crypto exchanges (e.g., into low-financial-literacy markets full of Ponzi schemes) can be counterproductive, while fundraising innovations like DAICOs and airdrops require an experimentation-friendly environment such as Asian regulatory sandboxes.
“the US already has all of these code-is-free-speech precedents, and most other countries, even those without significant free speech protections, are still willing to follow along in principle.”
Libertarian tax havens and havens in general (online and offline) suffer from attracting both freedom-lovers and rich people who aren't particularly interesting, which—along with small populations, inaccessibility, and a finance-heavy crowd—makes places like Zug overrated as places to live, a risk now also appearing in crypto-valley projects in Puerto Rico.
“libertarian tax havens and libertarian havens in general, both online and offline, is that they attract both people who like freedom and rich people who are not particularly interesting.”
Modern epic myths mislead by portraying the world as good guys fighting unified evil (orcs, Sauron, the Empire), whereas reality and better fiction (e.g., Star Wars Episode VIII) show that conflicts often have the protagonists losing for much of the time and enemies with their own internal conflicts; Buterin wants social science fiction exploring how political and economic systems work and fail without anyone being literally Hitler.
“one of the ways in which all of these modern epic myths misled me when I was much younger is, they do make it look like the world is a bunch of good guys fighting against the evil orcs, Sauron, the Empire”
Because all actors in a crypto mechanism are anonymous and can switch identities, you cannot throw them in jail or socially ostracize them; the only way to motivate them is by having them lock up capital that the mechanism can put at risk.
“You cannot throw me in jail. You cannot socially ostracize me so I can’t earn any money again because I can always just switch identities. But to the extent that I’m willing to lock that ether up and make it vulnerable to a mechanism, then you have the ability to motivate me to that extent.”
Tourist sightseeing yields little beyond what Wikipedia provides, but travel that is incidental to doing something—interacting with people and observing how a culture thinks and operates—is valuable precisely because it reveals practices better or worse than your own that you could not learn by reading about those places online.
“the kind of traveling that is valuable is the kind where it’s not even set up as explicit travel for the sake of travel. You’re traveling there in order to do something”
Traditional financial systems carry inefficiencies beyond regulation—moving money into an application often requires bank wires and varied systems with friction—whereas crypto is a grand experiment in a fully frictionless financial system where moving ether into a smart contract is as simple as logging into a website.
“I see cryptocurrency in part as this grand experiment of, what happens if you create a financial system that really is fully frictionless”
Blockchains are logically centralized because there is one shared ledger and canonical history that you are either part of or not, and Buterin argues this logical centralization is actually a good property.
“I do think that blockchains are logically centralized, basically because there is one blockchain. It is a shared ledger. It is a canonical history... I argue that this logical centralization is actually good.”
Crypto protocols cannot prohibit bribery because there is no simple way to define a bribe in code, and parties can arrange a bribe outside the protocol where it has no visibility; rules must be expressible as exact program code, a tighter constraint than law-writing.
“you can’t say in cryptoeconomics, “It’s illegal to bribe people,” because there’s really no simple way to define what a bribe is. If someone really wants to bribe someone else, he can just go and do that outside of the protocol”
The innovation in the crypto space is not just technological or political but largely cultural, with the community performing a social coordination function and pioneering new ideas about how projects organize, what leadership and team membership mean, and how ideas spread.
“the innovation in the crypto space isn’t just technological innovation or political innovation. It’s also to a large degree cultural innovation”
Ethereum is not best described as a company; there are many overlapping ways to be part of it (Foundation employee, grant recipient, groupie, application builder, lone wolf, or contributor inside a traditional corporation), and some of the biggest contributions come from people outside the formal Foundation, which is a source of its successful innovation.
“There’s definitely people who work outside of the Ethereum Foundation as a formal organization that have made much bigger contributions than even some people inside of the foundation.”
A belief in progress is a social delusion that should be more prevalent in modern times, even though progress is only true in complicated ways and parts of the belief are definitely false in many situations.
“In modern times, possibly a belief in progress.”
Buterin learned Chinese not via a three-month mobile app but by going through podcasts and Learn Chinese podcasts, using flashcard apps for the most popular 500–1,000 characters and words, reaching an intermediate level enabling WeChat and in-person conversation, then improving by immersion and eventually watching movies on airplanes.
“I also started going through flash-card apps that taught me the most popular — maybe 500–1,000 — characters and words. Once you get to an intermediate level, you’re high enough that you can participate in WeChat conversations”