Value accrual

Three steps, and the two that get skipped

Fees are what users pay. Revenue is the portion the protocol keeps after paying whoever provides the service, which in an exchange means the liquidity providers. Value accrual is the third step: a built, running mechanism that moves the retained portion toward the token. Decks routinely present the first step and call it the third.

The gap matters because each step has its own failure mode. A protocol can have enormous fee volume and keep none of it. It can keep a large share and spend all of it on operations. It can retain a treasury for years and never decide what to do with it. None of those is an accident; each is a governance choice that somebody either made or avoided making.

So the question to ask about any token is not whether value accrues but where the route stops. In our experience it usually stops at step two, and the deck describes step one.

A fee switch is not a distribution decision

Uniswap's governance forum has the cleanest statement of this anywhere, and it predates the implementation by years. The fee switch design-space thread notes explicitly that turning on fee retention does not create any expectation that the retained tokens will be paid out to UNI token holders.²

Read that as a template. Retention and distribution are separable votes with separate legal, tax and operational consequences, and a governance process can pass the first while never scheduling the second. When a founder tells us their token accrues value because a fee switch exists in the contract, this is the distinction we open with.

What UNIfication actually built

In late 2025 Uniswap governance approved the UNIfication proposal, activating the protocol fee switch. Coin Metrics describes the resulting architecture as a pipes model: trades on v2, v3 and Unichain generate fees, a portion goes to the protocol with the rest to liquidity providers, all protocol fees flow into a single vault contract called TokenJar on each chain, and value can only leave TokenJar if UNI is burned through a contract called Firepit.¹

That is an unusually strict design. There is no discretionary treasury spend at the end of the pipe and no cash distribution to holders. The only exit is destruction of supply, which means the entire value-accrual claim rests on the market pricing a shrinking float.

The numbers, and what a 207x multiple is telling you

Coin Metrics published the first read on 13 January 2026, twelve days into the new configuration. Cumulative protocol-level fees reached roughly 0.8 million USD over that window, which the note converts into an illustrative annualised run rate of about 26 to 27 million USD if conditions persisted. Total UNI burned reached about 100.17 million UNI, worth roughly 557 million USD, or about 10.1% of the original one billion UNI supply, most of it from a retroactive one-off burn.¹

Ongoing burns, excluding that one-off, were running at roughly 4 to 5 million UNI per year. Against a UNI valuation of about 5.4 billion USD, the note puts the implied revenue multiple at around 207 times and reads that as high growth expectations already embedded in the price.¹ Every figure here is twelve days of data extrapolated, and the authors say so; treat it as a first reading, not a run rate.

The arithmetic is the useful part regardless of where the numbers land later. A 207x multiple means the current price is not being supported by the current burn. It is being supported by an expectation about a much larger burn later. Whether that expectation is met is a business question about trading volume, not a tokenomics question, and no mechanism can answer it.

Burn, distribute, or redeploy

There are three ends to the pipe and they are not equivalent. Burning destroys the retained value as supply. Distributing pays it out, in the fee asset or the native token, to holders who meet some condition. Redeploying keeps it as a working asset inside the protocol.

Joel Monegro of Placeholder argued against the first in 2020, and the argument has aged well enough to be worth stating here. For an asset whose value comes from participation in a pool of resources rather than from being spent, burning does not create new value; it redistributes existing value among a smaller group. He adds that whenever the token price does not immediately grow at the same rate as the burn, which he says is most of the time, the burn reduces the network's overall market capitalisation.⁴

Direct distribution is the other end of the range. Course material on token economics cites GMX's model, in which a documented share of trading fees is paid to stakers, as the cash-flow version of the same idea.³ It is operationally and legally heavier than a burn, which is a large part of why burns are more popular than the economics alone would explain.

What we ask before signing off a value-accrual claim

Four questions, in order. What do users pay, in what asset, per period. What share does the protocol retain after paying the people who provide the service. Which contract holds the retained share, and who can move it. What happens to it, by what rule, with no human deciding at the moment it happens.

If any of those four cannot be answered with a contract address or a passed proposal, the claim is a plan. Plans are fine. Presenting a plan as an accrual mechanism is not, because it invites holders to price something that does not exist yet.

And the thing under all four: the protocol has to earn something. Value accrual is plumbing. It moves what the business produces to where the token is, and it cannot move what the business does not produce.