Six of the ten largest DeFi protocols by total value locked have activated or expanded token buyback and fee-switch mechanisms since December 2025, directing an estimated $1.8 billion in annualized protocol revenue toward token holders. The shift represents the single largest structural change in...
"Buybacks are forever, the rate is just the one the DAO has voted for the next 6 months to start slow and at a conservative rate." — Marc Zeller, Founder, Aave Chan Initiative
Six of the ten largest DeFi protocols by total value locked have activated or expanded token buyback and fee-switch mechanisms since December 2025, directing an estimated $1.8 billion in annualized protocol revenue toward token holders. The shift represents the single largest structural change in DeFi tokenomics since the yield-farming era of 2020-2021, and it is moving in the opposite direction: from emission-funded subsidies to revenue-funded deflation.
Uniswap activated its fee switch in December 2025 and has since generated approximately $23 million in protocol revenue routed to UNI burns. Hyperliquid has spent over $1 billion buying back HYPE tokens on the open market. Aave launched Aavenomics 3.0 with automated buybacks running at $30 million annually. Jito approved JIP-38 on July 14, 2026, committing 100% of its JTX revenue share to JTO buybacks and burns. Lido proposed a $20 million stETH-funded LDO buyback. Yield Basis, a Curve-adjacent protocol, activated its fee switch after attracting $130 million in Bitcoin deposits. The common thread: protocols are converting usage into direct economic value for token holders, not governance rights alone.
The term "fee switch" refers to a protocol's decision to redirect a portion of transaction fees — previously retained entirely by liquidity providers or the protocol treasury — toward token holders via buybacks, burns, or direct distributions. For years, DeFi governance tokens traded as pure voting instruments with no claim on protocol cash flows. That structure is now being dismantled at scale.
The trigger was Uniswap's "UNIfication" proposal, passed on December 25, 2025, with overwhelming governance support. The largest decentralized exchange began routing 17% of swap fees on enabled pools to a smart contract called TokenJar, from which value can only exit if UNI tokens are burned through the Firepit contract. The rollout started with Ethereum in late December 2025 and expanded to additional networks through March and June 2026.
Within six months, the pattern replicated across the sector. According to data aggregator DefiLlama, protocols that have activated holder revenue mechanisms now collectively generate more than $2 billion in annualized fees, with a growing share flowing to token holders rather than to treasuries or liquidity providers alone.
Eight crypto projects have recorded buybacks outpacing growth in circulating supply since January 2026, according to MEXC research, with Meteora leading at 71% of its circulating supply repurchased.
Uniswap's fee switch applies a 0.05% protocol fee on enabled pools (liquidity providers retain 0.25%). All protocol fees flow into TokenJar contracts on each deployed chain. Withdrawal is possible only through UNI burning via the Firepit contract.
Key metrics as of mid-July 2026:
The one-time 100 million UNI burn represented approximately 10% of total supply and was framed as retroactive compensation for years of missed value capture.
Hyperliquid operates the most aggressive buyback program in DeFi. The protocol's Assistance Fund directs 97% of protocol fees into continuous, automated market purchases of HYPE tokens.
Key metrics:
Quarterly buyback volumes have declined from $316.76 million in Q3 2025 to $255.05 million in Q4 2025 and $192.25 million in Q1 2026, reflecting both lower trading volumes and the mechanical reality that fewer tokens remain available at higher prices.
Aave replaced its discretionary buyback committee with an automated, non-discretionary mechanism under Aavenomics 3.0. All Aave Protocol and GHO revenue now routes to AAVE holders without committee sign-off.
Key metrics:
The budget reduction from $50 million to $30 million, passed in March 2026, reflects a pragmatic adjustment to declining borrow-fee income. The mechanism itself, however, was made permanent.
Jito's governance approved JIP-38, committing 100% of the Jito DAO's revenue share from its forthcoming JTX trading platform to programmatic JTO buybacks and burns. The policy applies to 80% of JTX platform fees; the remaining 20% funds platform development.
Key terms:
The proposal formally designates JTO as the primary value-capture token across Jito's product suite, which includes JitoSOL, the Block Engine, and the upcoming JTX derivatives platform.
Lido proposed spending up to 10,000 stETH (~$20 million) to buy back LDO tokens, routing batches of 1,000 stETH through centralized exchanges and market makers. At current prices, the program could retire approximately 8% of circulating LDO supply.
Activation conditions include: ETH price above $3,000, annual revenue exceeding $40 million, and annual buyback capped at $10 million under the anti-cyclical model. An automated mechanism was targeted for Q2 2026 deployment.
Yield Basis, a protocol built by Curve Finance founder Michael Egorov to eliminate impermanent loss, activated its fee switch after investors deposited over $130 million in Bitcoin. The governance vote passed unanimously. Holders of vote-escrowed YB (veYB) now receive proportional shares of trading fees, with 17.55 BTC in accumulated fees awaiting initial distribution.
The critical question is whether buybacks absorb enough supply to offset ongoing token emissions. The answer varies significantly by protocol.
Hyperliquid presents the strongest case: with $1.16 billion in cumulative buybacks and 45.7 million HYPE removed, buybacks have outpaced emissions. Aave's $30 million annual buyback covers roughly 1.28% of supply annually, a modest rate against the protocol's $402 million in revenue. Uniswap's annualized buyback-and-burn, extrapolated from its $23 million in seven months, runs at approximately $39 million per year — meaningful against a ~$5 billion market cap but not transformative.
Ethena illustrates the structural challenge. The protocol generates $50–$60 million in monthly fees and has activated a fee switch for sENA holders, with projected yields of 4.5–15% annualized. However, over $300 million in ENA emissions remain scheduled for 2026. A $26 million annualized buyback does not compensate for that dilution.
The protocols where buybacks demonstrably exceed emissions — Hyperliquid, Meteora, and a handful of others — represent the minority. For most, buybacks slow dilution rather than reverse it.
| Protocol | Mechanism | Revenue Source | Annual Rate | Automation | |----------|-----------|---------------|-------------|------------| | Uniswap | Burn-only (TokenJar/Firepit) | Swap fees (0.05%) | ~$39M est. | Fully automated | | Hyperliquid | Market buyback (Assistance Fund) | Trading fees (97%) | ~$700M+ | Continuous, automated | | Aave | Automated buyback (Aavenomics 3.0) | Lending/GHO revenue | $30M | Automated, no committee | | Jito | Buyback & burn (Rev Splitter) | JTX fees (80%) | TBD (pre-launch) | Automated per epoch | | Lido | Treasury buyback (anti-cyclical) | Staking revenue | $10–20M | Conditional triggers | | Yield Basis | Fee distribution (veYB) | Trading fees | Variable | Governance-activated | | Ethena | Staking yield + buyback | Derivatives/staking | ~$26M | Governance-activated |
Three distinct models have emerged: burn-only (Uniswap), where tokens are permanently destroyed; market buyback (Hyperliquid, Aave), where tokens are purchased on open markets; and direct distribution (Yield Basis), where fees flow to locked token holders. Each carries different implications for supply dynamics, price impact, and regulatory classification.
Price response to fee-switch activations has been mixed. UNI hit a new cycle low despite the fee switch going live, according to Bitget analysis. Hyperliquid's HYPE token performed strongly, but Forbes attributed this primarily to trading activity and structural demand rather than the buyback alone.
EtherFi's ETHFI token doubled after boosting buybacks, according to The Defiant. Aave's AAVE token bounced on the initial $50 million buyback proposal in October 2025 but gave back gains as the budget was later reduced.
The data suggests buybacks are necessary but not sufficient for price appreciation. Trading volume, broader market conditions, and emission schedules remain dominant factors. As one industry analysis noted: "Buybacks can reinforce a positive flywheel, but they cannot create one from thin air."
There are also governance risks. Marc Zeller's ACI, the entity that designed Aave's buyback program, announced plans to exit the Aave DAO by July 2026 amid governance tensions. Aave's governance proposal to reduce the buyback budget from $50 million to $30 million demonstrated that buyback commitments, even when described as "permanent," remain subject to DAO votes and revenue volatility.
The DeFi fee-switch wave of 2025-2026 marks the sector's most significant tokenomics restructuring since the initial yield-farming era. Protocols that collectively process trillions in annual volume are no longer content to let that value accrue solely to liquidity providers and treasuries. The shift toward buybacks, burns, and direct distributions represents an attempt to make governance tokens function more like equity — with claims on protocol cash flows rather than voting rights alone.
The structural change is real. Whether it is sufficient depends on the specific protocol. Hyperliquid's 97% fee allocation and $1 billion-plus buyback program sets a standard few can match. Most others operate at margins where buybacks slow token inflation without eliminating it. The test for 2027 will be whether protocols can sustain these programs through a revenue downturn, or whether "forever buybacks" prove as temporary as the emission schedules they were designed to replace.