How the Uniswap Fee Switch Rewired Its Revenue

The Uniswap fee switch is the governance setting that routes part of every swap fee on Uniswap to the protocol instead of sending all of it to liquidity providers. It funds an ongoing burn of UNI, the protocol's governance token on Ethereum. Traders search it because it changed what holding UNI actually pays for.

The mechanism is stranger than a simple revenue share, and the part that trips most people up is the condition attached to the money.

The Uniswap Fee Switch at a Glance

What Is the Uniswap Fee Switch?

Uniswap has always charged traders a swap fee. For most of the protocol's life every basis point of that fee went to the people supplying the tokens in each liquidity pool, and the protocol itself earned nothing from the trades it settled. The fee switch is the setting that changes the split, and it sat unused in the contracts for years while governance argued about tax exposure, legal structure and where liquidity would go if Uniswap started taking a cut.

Turning it on took two separate votes eight months apart, and conflating them is the most common mistake in coverage of this topic. The first vote did the hard part, and the second simply widened the pipe.

On a standard 0.30% v2 pool the switch moves the split to 0.25% for liquidity providers and 0.05% for the protocol. Uniswap's own protocol fee documentation describes the same idea on v3 as a fraction of the LP fee rather than a flat number, one quarter on the cheapest tiers and one sixth on the 0.30% and 1% tiers. The trader's all-in cost barely moves. The money simply stops being entirely an LP payment.

What makes this different from a normal protocol treasury is where the money goes next, and that is the section worth reading twice.

Why Did the Fee Switch Become a Story in 2026?

Because the 0.3% figure that nearly every Uniswap explainer on the internet quotes stopped being the whole picture, and that includes ours. Our Uniswap explainer carries an update stamp of 16 December 2020 and still tells readers the platform charges 0.3% per trade. That was accurate for five years, and it is the version of Uniswap most people still hold in their heads.

It is also the reason this article exists. When a protocol changes the destination of its fees rather than the size of them, the headline number stays recognisable while the economics underneath it turn into something else, and older reference pages across the whole of DeFi quietly fall out of date without ever looking wrong.

The second reason is that the numbers finally got large enough to argue about. The Defiant and Crypto Briefing both reported protocol revenue running near $114,000 a day earlier in July 2026 and around $325,000 a day once v4 pools came into scope. Two independent outlets landing on the same pair of figures is about as good as attribution gets for a metric this young.

And UNI itself started moving. The token closed Tuesday 1 September 2026 at $5.844, up 11.78% on the session and 36.96% across the seven sessions from Tuesday 25 August, with its 50-day average sitting 13.64% above its 200-day. None of that proves the fee switch caused the move, and anyone telling you the two are cleanly linked is selling a story rather than reading one.

How Do TokenJar and Firepit Actually Work?

This is the part that makes Uniswap's design unusual, and it is worth slowing down for.

Protocol fees do not go to a multisig. They accumulate in an immutable contract called TokenJar, one per chain, which collects whatever tokens the pools happen to produce. TokenJar has no spending logic of its own. It has a single authorised withdrawer, and that withdrawer is a contract called Firepit.

Firepit will release funds on one condition, which is that somebody burns UNI first.

The caller burns, the caller collects. Anyone can call the release function, burning a governance-set threshold amount of UNI, and in exchange the contract hands them the top assets sitting in the jar, valued at slightly more than what they just destroyed. Uniswap's documentation works the example at a 100 UNI threshold with the released basket worth marginally above it.

That "slightly more" is the whole engine. It turns the burn into an arbitrage that strangers compete to perform, rather than a treasury operation somebody at Uniswap has to remember to run. The protocol never has to sell the fee tokens, never has to hold them, and never has to decide when to burn. Fee income and supply destruction become the same event, executed by whoever spots the spread first.

Think of it as a vending machine that accepts one currency only and destroys every coin you feed it. The goods inside are worth a little more than the coin, so the machine never stays full for long, and every purchase permanently removes supply.

The design has an unglamorous virtue too. Because the release is permissionless and the incentive is mechanical, the burn keeps happening through the stretches when nobody at the DAO is paying attention.

What Proposal 100 Changed and What It Did Not

Proposal 100, executed Monday 27 July 2026, did not turn protocol fees on. That had already happened. The proposal extended the existing fee machinery to v4 pools and carried it across seven networks at once, covering Ethereum, Arbitrum, Base, BNB Chain, Polygon, OP Mainnet and Robinhood Chain. It passed with 46.6 million UNI in favour against 1.27 million opposed.

The distinction matters for anyone reading older commentary. UNIfication was posted on 10 November 2025, covered at the time by CoinDesk, and approved in December 2025 with near unanimous support. That vote is what activated the switch and authorised the retroactive burn of 100 million UNI from the treasury, an amount Uniswap Labs framed as roughly what would have been burned had fees been live since launch.

UNIfication's own rollout plan said the first phase would cover v2 and selected v3 pools representing most of Ethereum mainnet LP fees, with Ethereum Layer 2 networks and v4 to follow through later governance. Proposal 100 is that follow-up arriving. It is an expansion of scope rather than a change of policy, and the roughly threefold jump in daily burn revenue is what a scope expansion looks like when the new scope is where the volume already lives.

UNI Under the Fee Switch vs Bitcoin

Traders reach for a Bitcoin comparison because Bitcoin is the reference asset for supply-driven value. The comparison earns its place precisely where it breaks down.

The sharpest line in that table is the last one. Bitcoin's supply schedule cannot be voted away. UNI's burn can, because on-chain governance put it there and the same process can take it back. A revenue mechanism that exists at the pleasure of a token vote is a different animal from one enforced by consensus rules, and pricing the two the same way is a mistake that costs money.

What Can Move the UNI Price?

Swap Volume Across the Seven Networks

Protocol revenue is a fixed percentage of trading activity, so the burn rate tracks volume almost mechanically. A quiet quarter in DeFi shrinks the burn without any governance decision being taken at all. This is the cleanest input to watch and the one most people skip in favour of headlines.

Further Governance Votes on Fee Scope

Proposal 100 widened the fee base, and another proposal could widen it again to remaining v3 tiers, additional chains or new pool types. Each expansion changes the revenue base rather than the rate. Governance can also narrow it, which markets tend to price a good deal faster than they price expansions.

The Balance Between Burns and the Growth Budget

UNIfication granted Uniswap Labs an annual growth budget of 20 million UNI, vesting quarterly from 1 January 2026. Net supply change is burns minus that budget, and a reader tracking only the burn number is looking at half of the equation.

Liquidity Provider Behaviour

The 0.05% taken from a v2 pool comes out of what LPs used to earn, and if providers migrate capital to venues paying the full fee, depth thins, routing worsens and volume follows. Uniswap's founder has publicly argued the effect on LP yields is minimal. That claim is testable over time rather than settled at the moment of activation.

Broader Conditions on Ethereum

Gas costs, stablecoin flows and general risk appetite set the ceiling on how much swapping happens in the first place. Uniswap's revenue is a derivative of Ethereum activity more than it is a derivative of anything Uniswap governance decides.

Risks of Buying or Trading UNI

The Burn Can Be Switched Off

Everything described in this article was created by a vote and can be undone by one. Treat the fee switch as a policy rather than as a property of the token, because policies get revisited when circumstances change.

Net Supply May Still Rise

Burns run against a 20 million UNI annual growth budget, and depending on volume the arithmetic can land on either side of zero. Deflation is a claim to verify against the actual numbers rather than something to accept from a headline.

Revenue Is Concentrated, Not Evenly Spread

The seven networks do not contribute equally to protocol revenue, and a large share arrives from a small number of them. Concentration means one chain's decline or one integration changing terms can move the total far more than the seven-chain framing suggests.

Fee Competition From Other Venues

Zero-fee and rebate-driven trading venues exist, and a protocol that starts taking a cut hands competitors a number to undercut. The counterargument is that Uniswap's routing and depth are hard to replicate, which has held so far and is not guaranteed to keep holding.

Wrong-Contract and Copycat Risk

UNI has a single canonical ERC-20 contract on Ethereum, and tokens using the same name and ticker appear regularly on other chains and in new pools. A pool being live with plausible liquidity proves nothing about which token you are actually buying. Verify the contract address against Uniswap's own documentation or a major data aggregator's coin page before touching anything called UNI outside a venue you already trust.

Governance Concentration

Votes clearing quorum with 46.6 million UNI, in a supply approaching a billion, tell you participation is thin. Thin participation means a small number of large holders decide fee policy, and their interests are not automatically aligned with yours.

How to Research the Fee Switch Safely

The research trap on this topic is that most secondary coverage collapses two proposals into one event. Reading a summary of a summary is how you end up believing Proposal 100 turned fees on.

Read the executed proposal, not the recap. The UNIfication governance thread and the on-chain proposal record show what was voted on and what was implemented, including the phased rollout that made Proposal 100 necessary in the first place.

Separate activation from expansion. Any claim about protocol revenue needs a date attached to it, because the same protocol produced very different numbers before and after 27 July 2026.

Cross-check revenue against two feeds. Fee dashboards disagree, sometimes materially, on what counts as protocol revenue versus what counts as LP income. Print the spread rather than quietly picking the friendlier number.

Check the burn against the emission. Look up both figures over the same period before drawing any conclusion about net supply. A burn number quoted with no mention of the growth budget is an incomplete picture rather than a wrong one, and incomplete pictures are harder to spot.

Confirm the contract before you trade. Pull the canonical address from official documentation and compare it against the pair you are actually trading, particularly on chains where anyone can deploy a token called UNI in an afternoon.

Is UNI a Good Investment?

The fee switch gives UNI something it did not have before, which is a mechanical link between protocol usage and token supply. That is a genuine change in what the asset is, and it moves UNI out of the category of tokens whose only function is voting on things.

It does not make UNI cheap, and it does not make the burn large relative to supply. At around $325,000 a day the annualised figure is meaningful for a protocol and modest against a token with a billion-unit cap, and the growth budget offsets part of it. The fair description of UNI after the fee switch is a governance token with a real and measurable revenue link, sized smaller than the enthusiasm around it usually implies.

Anyone giving you a target price for UNI based on the burn is running a discounted cash flow on a number that a vote can change. The useful question is not where UNI goes. It is what happens to daily protocol revenue over the next two quarters, because that figure is observable, it updates constantly, and it is the one input in the whole story that governance cannot fake.

Final Thoughts

The series to track from this point forward is daily protocol revenue on a per-chain basis, because it answers the volume question, the concentration question and the burn-versus-emission question at once. Watch the governance forum for any proposal touching remaining v3 tiers or additional networks, since scope expansion is what produced the move from roughly $114,000 to about $325,000 a day and the same lever is still sitting there. If the DAO ever proposes narrowing the fee base, that is the signal worth acting on rather than the burn totals that get quoted after the fact.

The broader lesson has nothing to do with UNI. A protocol can change what its fees do without changing what they cost, and every reference page in the industry keeps quoting the old sentence anyway.

Frequently Asked Questions

Does the Uniswap fee switch make trading on Uniswap more expensive?

Barely. On a 0.30% v2 pool the total swap fee stays where it was and only the split changes, with 0.25% going to liquidity providers and 0.05% to the protocol. The cost the trader pays is close to unchanged, while the destination of part of it is not.

Do UNI holders receive the protocol fees?

No, and this is the most common misreading of the design. Fees accumulate in TokenJar and can only be released by burning UNI through Firepit, so holders benefit through supply reduction rather than through a distribution. A passive holder receives nothing directly and never has to claim anything.

Is UNI deflationary after the fee switch?

Only in periods when burns exceed the 20 million UNI annual growth budget granted to Uniswap Labs. Treat deflation as a condition to verify against live burn data rather than a permanent property of the token.

What is the difference between UNIfication and Proposal 100?

UNIfication, approved in December 2025, activated protocol fees and authorised the 100 million UNI retroactive burn. Proposal 100, executed on 27 July 2026, extended those same fees to v4 pools across seven networks and roughly tripled daily burn revenue without changing any of the underlying rules.

This article is for informational purposes only and does not constitute financial or investment advice. Cryptocurrency trading involves substantial risk. Always conduct your own research before making trading decisions.