DWF Labs’ November 9 report says token value-accrual programs are producing different results because of how revenue reaches the token, rather than because the underlying protocols are necessarily unprofitable. Of 15 tokens with an active or announced value-accrual mechanism, 13 outperformed Bitcoin this year, according to the market maker and investor. The report identifies SYRUP and LDO as tokens whose protocols either reduced buybacks or made them conditional.

Revenue must pass through four checkpoints before reaching a token: the revenue figure used by the rule, the rule that converts it into a budget, whether that budget is spent, and what happens to the purchased tokens afterward. Maple and Lido encounter problems at different points in that process. The same framework doubles as a checklist for evaluating any buyback announcement: which revenue figure a rule uses, whether the resulting budget is actually spent, and where the purchased tokens end up.

Maple’s business grew while SYRUP fell

Maple’s lending business expanded during a year in which the broader decentralized-finance lending market contracted. The October Maple Memo reports $4.8 billion in assets under management and $1.67 billion in loans outstanding, up 12% from a year earlier. Maple said the wider DeFi lending market declined 29%. Its financials page shows $21.37 million in protocol revenue over the past 12 months, while DWF Labs estimates that SYRUP fell about 45% year to date.

For much of the year, little of that revenue reached the token. Under MIP-019, a Maple Improvement Proposal, the Syrup Strategic Fund received 25% of monthly revenue. Chairman Joe Flanagan explained in a July post that the allocation covered both buybacks and reserves and did not establish a fixed monthly purchase schedule. The fund bought 8 million SYRUP in 2025. Maple’s SYRUP transparency then showed no purchases from November 2025 through June 2026. Flanagan attributed the pause to a decision to build reserves for a prolonged downturn.

Maple responded with MIP-021, which replaced the discretionary allocation with a fixed formula. The formula directs 10% of monthly revenue below $1.5 million to buybacks, 20% of revenue between $1.5 million and $2 million, and 30% above $2 million. DWF notes that the change initially reduced the share directed to SYRUP from 25% to 10%. Unlike the previous allocation, however, the new amount must be spent. Purchases followed in July, August and September.

A 7% revenue increase doubled the SYRUP buyback

The transparency dashboard records $147,098 spent on 676,294 SYRUP in August and $314,116 spent on 1,371,166 SYRUP in September, a 113.5% increase in dollars spent. Maple’s October memo links the increase to September revenue crossing the $1.5 million threshold. Because the MIP-021 rate applies to the entire month, crossing the threshold doubled the share taken from every September dollar, including the first dollar earned.

Working backward from the purchases produces estimated revenue of about $1.47 million in August and $1.57 million in September, an increase of close to 7%. If July’s $136,768 purchase also reflected the 10% rate, July revenue was approximately $1.37 million and quarterly revenue was roughly $4.41 million. That matches the quarterly total of just over $4.4 million cited by DWF.

The same step works in reverse. If September revenue had been about $75,000 lower and remained below the threshold, the formula would have funded less than half of Maple’s actual purchase. DWF describes Maple’s revenue as being close to the threshold and expects buybacks to move between tiers from month to month.

The $21.37 million figure is already the narrower revenue measure used for the calculation. The same financials page lists $122.42 million in gross revenue for the period. A reader using the headline gross-revenue figure to estimate SYRUP’s buyback would therefore overstate the base nearly sixfold. Coindoo examined a similar distinction in its report on who benefits from the $5.9 billion users paid in fees.

The arrangement is also temporary. MIP-021 went to a SYRUP-holder vote from July 13 to 17 and applies for six months. Purchases are made after each month-end, once revenue has been finalized.

Lido earns millions, but NEST has bought no LDO

Lido’s core business also continued to generate revenue. The DAO’s first-half 2026 report shows $15.71 million in net staking revenue from January through June and a $6.7 million positive result for the staking unit. DWF nevertheless lists LDO among this year’s decliners and describes both of Lido’s buyback programs as slow and conditional.

Buybacks are only part of the explanation. Lido’s share of staked ETH fell from 23.93% to 21.18% in the first half, according to the same report, as new capital went to custodians, exchanges and institutional stakers. A token can decline while its business loses market share, and a buyback design does not by itself offset that pressure.

NEST, short for Network Economic Support Tokenomics, went live in mid-August. Each day, it compares the DAO’s share of staking revenue with a baseline of $40 million per year, or approximately $109,000 per day, and adds half of the difference to a running balance. On a day when revenue falls short, half of the deficit is subtracted. Under Lido’s launch overview and the LIP-36 specification (a Lido Improvement Proposal), NEST buys LDO only while the balance is above zero. Purchases are subject to a $50,000 daily cap and a $10 million rolling annual ceiling.

Lido’s revenue is below that baseline. The first-half staking figure works out to roughly $87,000 per day. At that rate, the running balance would decline by about $11,000 each day, and the accumulated deficit would have to be rebuilt before NEST could make its first purchase.

Lido’s own assessment is direct. At current staking levels and market share, its report estimates that revenue reaches the daily baseline only with ETH near $2,730. It states that “buybacks are not expected soon.” DWF reports that NEST has not executed a single buyback.

The caps are ceilings, not targets. Reaching the $50,000 daily maximum every day would use the $10 million annual limit in 200 days. Lido’s backtest using 2024 and 2025 revenue produced $7.09 million in purchases over two years, or approximately one-third of the annual ceiling per year.

NEST launched in treasury-only mode, sending every purchased LDO to the DAO treasury. A second mode would divide each allocation between LDO and wstETH (wrapped staked ETH) and deposit both assets as DAO-owned liquidity in a Curve pool. Activating that mode would require another onchain vote.

Lido says the strict design is intended to prevent buybacks from consuming treasury funds when the DAO has no genuine surplus. The first-half figures illustrate that risk: an operating surplus of $1.6 million before one-off items became a $4.5 million loss after costs related to the Kelp DAO incident. A Maple-style percentage would have continued spending treasury funds on LDO during a six-month period in which the DAO lost money.

Lido’s LDO purchases run through a committee

Lido purchases have continued through another. On April 13, the DAO approved a committee-run accumulation program authorizing up to 10,000 stETH (staked ETH) from the treasury for LDO purchases, executed in batches. The first-half report says the first two batches exchanged 1,591 stETH for approximately 10.03 million LDO at an average price near $0.30.

A September 25 execution report covers a third batch. It exchanged 680 stETH for 4,583,559 LDO between August 31 and September 23 at a volume-weighted price of $0.37 before fees, or about $1.7 million. An onchain transfer confirms that the LDO returned to the DAO treasury. After the batch, 7,000 stETH remained available under the authorization.

The arrangement reverses the design debate. The discretionary, human-run route that Lido rejected as a model for NEST is currently the mechanism moving LDO, while the automated system waits for its accounting balance to recover.

Hyperliquid sends fees directly to a burn

Hyperliquid removes both the threshold and the human decision. Its fee documentation divides trading fees among HLP, the exchange’s community liquidity vault, the Assistance Fund and market deployers. Spot and HIP-3 deployers can retain up to half of the fees generated by their markets.

The Assistance Fund automatically converts its share into HYPE as part of the chain’s execution. The HYPE it holds is burned, removing the tokens from both circulating and total supply. DWF estimates that Hyperliquid directs 95% to 99% of its revenue into this programmatic buying. It compares HYPE’s gain with a Bitcoin decline of about 8.5% over the same period and says Hyperliquid and Pump.fun together accounted for nearly 90% of a record $638 million spent on buybacks between January and late August.

DWF also identifies a limitation in its own finding. Some of the performance may reflect a crowded trade into a small group of large tokens with buybacks, rather than a lasting change in how the market values tokens. The report argues that the shift is structural but does not quantify how much of the move is attributable to either explanation.

Bought tokens do not always leave the market

Flanagan wrote that SYRUP transferred into Maple’s treasury leaves circulation. The dashboard counts 126.14 million SYRUP held by Maple outside its 1.14 billion circulating supply. MIP-021, however, places buyback SYRUP in the Strategic Fund, whose listed functions include market making, DEX liquidity positions and strategic transactions such as acquisitions. The tokens are outside circulation by classification, but the fund can return them to market-related uses.

Lido treats treasury-held LDO similarly. Its first-half report presents the 10 million LDO from the first two committee batches as a circulating-supply reduction of approximately 1% of total supply. Under NEST, Lido rejected an outright burn, describing it as “a one-time signal, not a durable mechanism.” All NEST purchases remain DAO-owned.

Hyperliquid’s burn is the only one of the three mechanisms that permanently reduces total supply. It also removes an option retained by Maple and Lido: burned HYPE cannot later fund contributors, liquidity incentives or an acquisition.

In each model, the money goes to the holders who sell. Holders who retain their tokens receive no direct payment; their potential benefit is indirect, through additional demand or a smaller float. DWF distinguishes this from equity ownership: a shareholder whose dividend is reduced still has a claim on a company’s profits and assets, while a token holder whose buyback stops has no equivalent economic link to the protocol. New issuance can also outpace purchases, as Coindoo reported when Aave and JUP buybacks fell behind new supply.

Where each buyback leaves a public trail

Maple: Each monthly row on the SYRUP transparency page lists dollars spent, tokens purchased and the average price paid: $0.2175 per SYRUP in August and $0.2291 in September.

Lido NEST: NEST fills orders through CoW Swap and sends LDO to the DAO’s Aragon Agent contract. Any purchase should therefore appear as an LDO inflow to that address.

Lido committee: Before each 1,000-stETH batch, the price cap is posted on Lido’s research forum. The Easy Track motion releasing the funds can be halted by tokenholder objections.

Hyperliquid: The Assistance Fund uses the system address 0xfefefefefefefefefefefefefefefefefefefefe. HYPE held at that address counts as burned, making its balance the running burn total.

Signals that revenue is reaching the token

Maple’s October purchase had not appeared on its dashboard as of November 9. Its size relative to October revenue will show which tier applied and whether the month-to-month swings anticipated by DWF have begun. MIP-021’s six-month term also places a renewal, amendment or expiry decision before SYRUP holders around the turn of the year.

For Lido, the relevant signal is NEST’s first purchase. ETH near $2,730 is the level at which Lido estimates revenue reaches the baseline, but crossing that level would only begin repaying the accumulated deficit. Buying starts after the balance returns above zero. Until then, the committee has 7,000 stETH remaining under its authorization.

Hyperliquid has no threshold monitor. The variable is fee volume, since the only claims between a trading fee and a burned HYPE token are the portions retained by HLP and market deployers.