In 2026, token buybacks have become one of the main tools of tokenomics: according to Cointelegraph Magazine, crypto projects have spent around $640 million on the practice. That is roughly 17% more than during the same period a year earlier and an order of magnitude more than the $366,000 spent throughout all of 2024. Nearly 90% of current spending comes from just two protocols — Hyperliquid and Pump.fun.
The mechanics are simple: a protocol directs part of its revenue toward buying its own tokens on the open market. A buyback creates additional demand, while a burn reduces supply — together, this creates upward pressure on price. But the key question remains open: does this financial engineering create long-term value, or does it merely make tokens look more valuable than they really are?
Key figures: $640 million spent on buybacks since the start of 2026; Hyperliquid directs 99% of its revenue to buying back and burning HYPE; Pump.fun has removed $446.65 million worth of PUMP from circulation; Spark has bought back more than 143 million SPK, but does not burn them and instead holds them in its treasury.
Orest Gavryliak, chief legal officer at DEX aggregator 1inch, explains the logic this way: “When projects introduce revenue-funded buybacks and burns, they are usually pursuing one of two goals: either reducing the circulating supply of tokens or demonstrating the rationale for investing in protocol revenue.” In other words, a buyback links the success of the product to the value of the token and gives holders a clear narrative.
Spending dynamics show explosive growth: $366,000 in 2024, followed by a jump to hundreds of millions in 2025 and $640 million in 2026 (+17% year over year). At the same time, the market is extremely concentrated: Hyperliquid and Pump.fun account for nearly 90% of all spending. In scale, the practice is beginning to resemble corporate share buybacks in TradFi, where S&P 500 companies repurchase hundreds of billions of dollars worth of stock every year, but in crypto the emphasis is not on dividends, but on supply scarcity.
| Protocol | Model | Scale |
|---|---|---|
| Hyperliquid | Buyback + burn, 99% of revenue | ~90% of the market together with Pump.fun |
| Pump.fun | Buyback + burn, 50% of revenue | $446.65M removed from circulation |
| Spark | Buyback without burn, held in treasury | More than 143M SPK |
| Uniswap | Buybacks under UNIfication | Price gave back ~50% of its gains |
Hyperliquid and Pump.fun take the idea to the extreme: tokens are bought back and permanently burned, irreversibly reducing supply. Each remaining token theoretically becomes more valuable, while holders receive a “hidden dividend” through scarcity. For memecoin and perpetuals protocols with strong cash flow, this is a way to convert speculative demand into structural demand.
Spark chose a different model: the protocol bought back more than 143 million SPK through open-market purchases funded by surplus funds, but did not burn them. The tokens remain in the treasury and are intended to reward long-term participants in the ecosystem. Spark co-founder and CEO Sam MacPherson explained: “Token holders should participate in the protocol’s long-term economic success, rather than simply receiving a distribution every time it generates revenue.”
Every dollar spent buying back a token has an opportunity cost: it is a dollar not spent on hiring developers, expanding the business, strengthening the balance sheet or improving the product. MacPherson formulates the principle this way: “The question should be: what is the most valuable use of the next dollar of surplus?” If a protocol can reinvest capital at an attractive return, that may be “far more valuable” than simply distributing revenue as it comes in.
There is no ironclad guarantee that price will rise. Pump.fun has aggressively bought back and burned PUMP since July 2025, yet the token is still trading roughly 50% below its September 2025 all-time high. UNI has also given back about half of the gains it recorded after the UNIfication proposal was announced in November 2025, despite the stated buybacks.
A sustainable protocol with real surplus funds can reasonably conclude that buying back its own token is the best use of part of its capital. But a project that is “barely hanging on” may use buybacks simply to move the price. MacPherson summarizes it this way: “A buyback does not make an unsustainable protocol sustainable.” Financial engineering cannot replace product quality, revenue and demand.
For the industry, the buyback boom is a test of maturity: protocols are learning how to return capital to holders, but they risk substituting development with the appearance of value. Investors should focus not on how many tokens are burned, but on whether the project still has surplus after all obligations and reinvestment needs are covered. It is this remainder, not headline burn programs, that will determine whether the tokenomics survive the next bear market.
“When an asset trades below its intrinsic value, a buyback is the best thing management can do for owners. But if the business is weak, a buyback merely delays recognition of the problem.”
— Warren Buffett, investor, on the logic of buybacks as applied to tokens
