The sovereignty premium

Self-custody used to be a slogan. It is now, increasingly, a price — a measurable spread between executing a trade while still holding your own keys and executing it after handing them to someone else for the afternoon. The crypto industry has spent fifteen years pretending that spread is zero. It is not zero, and the people who price it best are going to eat the people who price it worst.

What you are paying when you send to an exchange

The instant your BTC lands in a centralized exchange's deposit address, you have extended that exchange a short-duration unsecured loan. The credit terms are favorable to the borrower: no collateral, no covenants, no maturity date you control, and a withdrawal queue the exchange can throttle whenever it finds itself short. In a more polite corner of finance this instrument has a name. It is called a deposit at a non-bank financial intermediary, and the spread on such deposits is non-trivial and observable. Crypto users pay it without noticing because the line item is not on the trade ticket.

Mt. Gox was the first time the industry was forced to mark that spread to market. FTX was the second. Celsius and Voyager and Genesis and BlockFi each repeated the lesson in a slightly different key. The cumulative tuition bill is somewhere north of twenty billion dollars and the curriculum is the same every time: the people you trusted to hold the thing for an afternoon also held it overnight, and overnight is when the institution failed.

Wrapped Bitcoin is a credit product

Most people think of WBTC as a token. It is more accurately described as a depositary receipt: a claim on a real Bitcoin held by a custodian, redeemable on a redemption ladder that depends entirely on the custodian's honesty, solvency, and continued willingness to deal with the merchant making the request. Every DeFi position collateralized by WBTC is, underneath, a leveraged bet on the custodian's operational integrity. That this bet has paid out for years does not change what the bet is. It changes only how confident the participants have become that the bet pays.

This is fine, in the way that most credit products are fine most of the time. The problem is the user almost never understands that they are holding a credit product. They think they hold Bitcoin. Bitcoin is non-defaulting by construction. Their WBTC is not.

What intent settlement actually changes

The interesting feature of intent settlement is not that it is faster or cheaper, although it often is both. The interesting feature is that the user's custody is continuous through the trade rather than bracketed by it. A solver holds inventory; the user holds their own coins; the protocol guarantees that the two transfers happen in a way that cannot leave the user holding nothing. Nobody, at any point in the trade, takes possession of the user's Bitcoin for a duration that exceeds the settlement window.

This is not a marketing distinction. It is a structural one. The sovereignty premium — the cost of being the kind of user who refuses to extend an unsecured loan to an exchange in order to rotate into stablecoins — was, until recently, large enough that most people gave up and paid it. The premium is now small. It will get smaller. Flip is a bet that the premium should approach zero, and that when it does, the marginal user will not bother to use anything else.

The thesis in one sentence

The market for moving Bitcoin around without giving it up has been served, for a decade, by products that quietly assumed nobody really wanted to keep custody. That assumption was always wrong. It has just taken until now for the infrastructure to catch up to the preference.

The architectural background to all of this lives in intent settlement vs bridges. This essay is the financial one.