LDO Inflation Analysis · September 2026 · Supply shrinking, projected to keep shrinking
Lido DAO creates no new LDO and destroys none — the count of LDO in existence read exactly 1,000,000,000 at both ends of the last 90 days — and the Pressure Framework still reads LDO at −1.25% over the trailing 90 days and −0.53% over the next 90. All of it is one mechanism: a token-holder-approved treasury buyback delivered 10,336,446 LDO into the Lido DAO Aragon Agent across three settlements, and that wallet sits outside the classified circulating float, so the coins genuinely left the market rather than being shuffled inside it. Sell pressure is 0, buy pressure is 10.4M LDO, and the DAO still holds 7,680 of the 10,000 staked ether it authorised itself to spend.
The verdict, in one paragraph
Against a circulating base of 834.2M LDO, the framework books 0 of sell pressure and 10.4M LDO of buy pressure over the trailing 90 days — a net of −1.25% — and projects −0.53% for the next 90 days on the one buyback tranche that carries a published deadline. The inflation monitor reads −1.23% for the same window, a gap of 0.02 percentage points, comfortably inside the framework's 0.5pp tolerance, so no data-conflict flag ships on the overview page. That agreement is structural rather than lucky: both numbers measure the same physical event from opposite ends — the framework reads the treasury wallet filling up, the monitor reads the classified float draining — and the small residual is nothing but the base convention, since the monitor divides by the 90-day-old supply while the framework divides by today's. The label for LDO is a fixed-supply governance token deflating through a treasury buyback that actually removes float.
Sell pressure: where new LDO comes from
It does not come from anywhere. Sell #1, protocol inflation, is 0. Lido DAO has no issuance mechanism running: the LDO contract reported a total of 1,000,000,000 LDO at both ends of the window, identical to the eighteenth decimal place, and Lido pays its validators and node operators out of staking rewards denominated in staked ether, never in LDO. That flat reading is worth one extra sentence, because a flat number can be an artefact of the code rather than a measurement. It is not one here. The LDO contract stores its supply in a checkpointed history rather than as a constant compiled into the code, and that history has genuinely been written five times — including a real destruction of 3,691,500 LDO in March 2022 and a real re-creation of the same amount in May 2022. The door is therefore still open: LDO is an Aragon governance token whose mint and destroy functions both exist and answer to a controller a token-holder vote drives. Nothing here is permanent.
Sell #2, vesting unlocks, is 0. The original December 2020 allocation to the Lido DAO treasury, investors, validators and founders ran a one-year lock followed by a one-year linear release, and it finished on Dec 16 2022. Every unlock tracker shows zero scheduled events remaining, and the chain agrees: all nine identified early-allocation wallets were read at both ends of this window and every one held an identical balance, 246,948,984 LDO between them. There is one live contributor programme, and it did move — its multisig fell by 822,837 LDO across the quarter. It contributes nothing to this row for a reason that governs the whole page, explained in the next paragraph: that multisig is already counted as circulating, so LDO leaving it was tradable before it moved and tradable after.
Sell #3, Foundation and unscheduled unlocks, is 0, and this one is measured rather than assumed. We swept every LDO transfer touching the Lido DAO treasury across all 645,601 blocks of the window: five arrivals, and zero departures. The treasury did not release a single LDO. Sell #4, long-term locked or bankruptcy, is 0 as well — LDO has no bankruptcy estate, no trustee and no court-ordered distribution attached to it.
Buy pressure: where new LDO goes
The whole page is Buy #3, Foundation buy, at 10,336,446 LDO. In April 2026 Lido DAO token-holders approved spending up to 10,000 staked ether from the DAO treasury buying LDO on the open market, drawn one thousand at a time, each tranche carrying a published LDO-to-ether price ceiling, a 3% slippage bound and its own three-day objection window before it can run. Execution runs through the Lido Ecosystem Foundation under a Growth Committee mandate, which makes it a discretionary programme rather than a rule the protocol enforces — hence the Foundation row rather than the programmatic one. Three tranches settled inside this window and each is matched to a transaction on chain and to the DAO's own published report: 1,715,992 LDO on Jun 11 2026, 6,472,316 LDO on Jul 8 2026, and 2,148,138 LDO on Aug 25 2026. Spending to date is 2,320 of the 10,000 staked ether authorised.
The reason that number counts is the part most buyback pages get wrong. Bought-back LDO is held in the Lido DAO treasury, not burned — the count of LDO in existence never changes. A buy-and-park only removes supply if the parking place sits outside the float the market counts, and here we could prove that it does rather than assume it. Reconstructing the excluded bucket wallet by wallet, the classified circulating figure fell from 844,550,636 to 834,151,323 LDO across the window while the Lido DAO treasury rose from 106,514,674 to 116,913,987 LDO. Those two moves are the same size to the unit. Subtract the treasury from the excluded bucket at each end and what remains is 48,934,690 LDO on both dates — static to 0.15 LDO across 90 days. Every unit the classifier removed from the float is the treasury filling up, and nothing else in the excluded bucket moved at all. The classifier's own series even steps down on the exact settlement dates and by the exact settled amounts. That is also why the contributor multisig's 822,837 LDO outflow books nothing: had that wallet been outside the float, the static remainder would have fallen by that amount instead of holding still.
Buy #1, programmatic buyback, is 0 — and it is the more interesting zero. Lido DAO has a second, automated buyback called NEST, with real on-chain contracts, a $50,000 daily cap and a $10 million annual cap, funded from the surplus of staking revenue above a floor. Token-holders enacted it on Aug 10 2026 and it went fully live on Aug 14 2026, in a treasury-only mode where everything it buys goes straight to the DAO treasury rather than into a liquidity pool. It has bought nothing. The trigger needs roughly $109,000 a day of protocol revenue and Lido earned about $75,000 a day through August, so it has not cleared once — and none of its contracts appear anywhere in the treasury's five-event arrival log.
Buy #2, protocol fee burn, is 0. Lido DAO has no burn of any kind: the protocol takes its fee in staked ether, so revenue never touches the LDO supply. We checked both surfaces anyway, because the two are independent — a transfer to a dead address leaves the total untouched, while the contract's destroy function cuts the total without touching any dead address, and Lido DAO has used that function before. The dead address held 5.2496 LDO on both dates and the total held 1,000,000,000 on both dates. Neither moved, so this is a verified zero rather than a hole in the check. Buy #4, new long-term lock, is 0 — LDO earns no staking reward, has no vote-escrow and gives no claim on protocol revenue, and repeated 2026 proposals to add staking and a revenue share have not been enacted. A fifth row carries 62,867 LDO of other treasury arrivals: a contributor allocation returned on Jul 31 2026 and one small on-chain purchase settled on Aug 13 2026, both crossing the same boundary and both removing float.
Foundation and overhang
The dominant overhang is the destination of the buyback itself: the Lido DAO treasury, holding 116,913,987 LDO at the close of the window against 106,514,674 at the open. It is read from the chain at every rebuild, it grew every month of this quarter and shrank in none, and it is spendable only through a token-holder vote or an Easy Track motion with an objection window. The second item is the contributor programme multisig at 3,998,971 LDO, down from 4,821,808 — already inside the counted float, so its movements change the reading only if the classification changes. The third is the nine identified early-allocation wallets holding 246,948,984 LDO between them, every one read at both window ends and every one flat to the token; their vesting is long finished, so this is an overhang of ownership rather than of unlock. The fourth is the remaining 48,934,690 LDO outside the classified float that the classifier does not itemise — genuinely opaque as to which wallets it names, though provably static across the window to 0.15 LDO.
One further item is dated and worth watching precisely because it runs the other way. A proposal published on Aug 31 2026 would authorise up to 4,000,000 LDO from the treasury as a recallable facility for centralised-exchange market makers, to be disbursed to a liquidity multisig if LDO pairs risk delisting on thin volume. It has not been voted and nothing is deployed, so it contributes zero today — but if it activates it would move LDO out of the excluded treasury and into a wallet inside the float, which is the mirror image of the buyback and would add roughly half a percent of supply back to the market. The trigger sentence applies to every item above: if any of these balances falls between refreshes, the outflow enters Sell #3 at the next refresh.
How LDO compares to other governance tokens with treasury buybacks
LDO belongs to the class of pure governance tokens — fixed supply, no fee accrual, no staking yield — that have turned to buybacks to give the token something to do. That puts LDO alongside the exchange and lending tokens running repurchase programmes, and against the uncapped continuous-emission chains where a staking-linked issuance of 5% to 15% a year is normal. On the issuance axis LDO is about as strict as a token gets: nothing has been minted since 2020.
The comparison that actually matters is not cap versus uncapped but burn versus park, and this is where most buyback programmes quietly fail the test that LDO passes. A burn is unambiguous: the coins stop existing and every measure of supply falls. A buy-and-park is only a removal if the wallet doing the parking is excluded from the float the market counts — and on a great many tokens it is not, so the programme spends real money moving coins from one counted wallet to another and the supply reading does not budge. Lido DAO parks rather than burns, which sounds like the weaker version, and yet it passes cleanly because its Aragon Agent sits outside the classification. The proof is arithmetic rather than assertion: the excluded bucket minus the treasury held still to 0.15 LDO over 90 days while the float fell by 10,399,313, so the removal is real and complete.
Where LDO is weaker than its peers is durability. A fee-funded burn scales with usage and never runs out; a treasury buyback spends a finite pot. Lido DAO authorised 10,000 staked ether, has deployed 2,320 in roughly five months, and the automated programme meant to make the buying permanent has never cleared its revenue trigger. Tokens with fee-linked removals do not have to decide each quarter whether the price is right; Lido DAO's Growth Committee does, and it has revised its price ceiling three times and left the most recent tranche only a third filled. LDO is deflationary today by decision, not by design.
What to watch in the next 90 days
First, the remaining 680 staked ether of the current tranche, which carries a published execution deadline of Sep 29 2026 and was confirmed as proceeding by the Growth Committee on Sep 4 2026; it is the only firing with a date, and it is the whole of the −0.53% forward reading. Second, whether a fourth tranche is published after that deadline — none exists yet, so none is projected, which means the forward number carries upside rather than downside if the programme continues at its recent pace. Third, the NEST trigger: it needs Lido protocol revenue above roughly $109,000 a day against about $75,000 in August, so a recovery in staking revenue or in the ether price would switch on a second, automatic bid that the framework currently books at zero. Fourth, the exchange market-making proposal of Aug 31 2026, which would send up to 4,000,000 LDO the other way, out of the treasury and back into the float. Fifth, the treasury balance itself at 116,913,987 LDO — it has only ever grown, and the day it falls is the day this page changes sign.
Summary
The MrNasdog Pressure Framework reads LDO at −1.25% over the trailing 90 days and −0.53% projected forward: supply shrinking, projected to keep shrinking. The structural mechanism is neither issuance nor burn — Lido DAO does neither, and the count of LDO in existence has been 1,000,000,000 since 2020 — but a token-holder-approved treasury buyback that delivered 10,336,446 LDO into a wallet the market does not count as circulating, verified by reconstructing the excluded bucket and finding everything else in it static to 0.15 LDO. The key risk is that this deflation is discretionary and finite: the pot is 10,000 staked ether of which 7,680 remain, the committee sets its own price ceiling and has lowered it three times, and the automated programme built to replace that judgement has not triggered once since going live on Aug 14 2026. The ceiling of 1,000,000,000 LDO is the one fixed thing here, and even that is a governance decision rather than a law of the code.
MrNasdog Pressure Framework analysis of LDO, Metric 1 — Inflation. Data + explanation only. Not financial advice. Updated Sep 8 2026.