AALGO · Algorand
ALGO overview
MrNasdog Pressure Framework · Inflation Analysis

ALGO Inflation Analysis · September 2026 · Supply growing, projected to keep growing

Algorand created all 10,000M ALGO at its 2019 genesis, has no instruction that mints another and no instruction that destroys one — and the Pressure Framework still reads ALGO at +1.27% over the trailing 90 days and +0.94% over the next 90. Every unit of that comes from Algorand Foundation custody through two taps: 19.0M ALGO of per-block proposer rewards paid out of the fee sink, and 96.3M ALGO transferred out of the 74 wallets the Foundation publishes. Buy pressure is 123.2K ALGO of transaction fees and nothing else. The ceiling is real; the float underneath it is not fixed.

The verdict, in one paragraph

Against a circulating base of 9,039M ALGO, the framework books 115.3M ALGO of sell pressure and 123.2K ALGO of buy pressure over the trailing 90 days — a net of +1.27% — and projects +0.94% for the next 90 days, the difference being a single 30.0M ALGO payment on Jul 17 2026 that fired once and has no schedule behind it. The inflation monitor reads +1.28% for the same window, a gap of 0.003 percentage points, far inside the framework's 0.5pp tolerance, so no monitor-gap warning ships on the overview page. That agreement is worth a sentence of its own, because the two sides count different things: the monitor reads a market classifier that treats ten billion less the Foundation's own wallets as circulating, while this ledger counts the individual transfers that actually left those wallets plus the block payouts that actually landed in accounts the Foundation does not run. The label for Algorand is a hard-capped chain that is still distributing its own genesis: ALGO cannot inflate, and its tradable float does anyway.

Sell pressure: where new ALGO comes from

It does not come from minting, because Algorand has no mint. The genesis file allocates exactly 10,000,000,000 ALGOacross 102 entries and the protocol has no path that adds to that number. What Algorand has instead is a reward account — the fee sink — that the Algorand Foundation keeps topped up out of its own reserves, and which pays every block's proposer a bonus. Sell #1, protocol inflation, is 19.0M ALGO, and the way that number is reached matters more than the number. Algorand's published bonus is 8.35 ALGO a block, decaying 1% every million blocks; multiplied across the 2,827,501 blocks this window actually produced, that implies 23.9M ALGO. The chain paid 19.5M. The reason is that a block pays nothing at all when its proposer is not incentive-eligible, and 18.25% of blocks in a 1,200-block uniform sample paid exactly zero — a fact that only reading block headers reveals, and that no tokenomics page states. A further 0.5M ALGO of the payout went to nodes the Algorand Foundation runs itself and never left its custody, which is why the row ships at 19.0M rather than 19.5M.

Sell #2, vesting unlocks, is 0.Algorand's original sale, grant and team schedules finished in 2024, and because ALGO is a native chain coin rather than a contract token there is no escrow and no lock contract anywhere for a cliff to release from. All 74 published Foundation wallets were read this session and every one is an ordinary spendable account. That is exactly why the pressure sits in the next row instead: the remaining reserve moves by decision, not by calendar.

Sell #3, Foundation and unscheduled unlocks, is 96.3M ALGO, and it is four fifths of this page. Every payment out of all 74 published wallets was walked transaction by transaction across the window, with wallet-to-wallet moves inside the set cancelled out so the row counts only ALGO that genuinely crossed into other hands. Of the 96.3M, 46.4Mwent to the single trading address the Foundation's disclosed rules-based selling programme uses; 30.0M went out in one payment on Jul 17 2026 to an account that is not on the published registry and that now reads zero, fully dispersed; 5.5M went to one further recipient; and 14.5M moved in dated grant, ecosystem and xGov batches. Nothing came back the other way. Sell #4, long-term locked or bankruptcy, is 0: ALGO has no bankruptcy estate, no trustee and no court-ordered distribution.

Buy pressure: where new ALGO goes

Buy #1, programmatic buyback, is 0, and the evidence is positive rather than an absence of announcements: every payment into the 74 published wallets was checked for the full 90 days and zero ALGO arrived from outside the set. The Algorand Foundation is a disclosed net seller, not a buyer. Buy #3, Foundation buy, is 0on the same measurement, and the Foundation's own quarterly report lists selling, grants and staking rewards on the outgoing side with no purchases opposite them.

Buy #2, protocol fee burn, is 0, verified on both surfaces at both ends of the window rather than on one. Algorand's all-zero account holds 39,465 ALGOand received no payment at all in 90 days, so the dead-address surface did not move; and the count of ALGO in existence is a protocol constant with no instruction able to reduce it, so the supply surface cannot move either. Fees do not burn on Algorand — they route to the fee sink, and half of each block's fees is handed straight back to that block's proposer. The v5.0.0 consensus upgrade that activated on mainnet on Aug 22 2026 changed how fees are priced, adding resource-based charges, but not where they end up; measured fees ran 0.0424 ALGO a block after it against 0.0438 before, so nothing about that upgrade shows up in this ledger. Buy #4, new long-term lock, is 0 because Algorand consensus has no bond, no unbonding wait and no slashing: 2,028M ALGO is online and every coin of it is spendable the same second. The one non-zero entry on this side is a tracked extra, Buy #5, the 123.2K ALGO of transaction fees that left the float into the fee sink. It is real and it is measured, and it is 0.0014% of supply.

Foundation and overhang

The Algorand Foundation's 74 published wallets held 931.0M ALGO when they were enumerated on Sep 8 2026, the single largest of them holding 140.6M, and the fee sink that funds every block reward holding 16.2M. None of that sits on a release schedule. The sink is refilled by hand rather than by rule — 10.3M arrived on Jul 10 2026 and 9.0M on Aug 19 2026, both from a Foundation treasury wallet — and at the current drain of roughly 0.25M ALGO a day it holds about 65 days of reward funding between refills. Two further overhangs are tracked: the trading address the structured-selling programme feeds, which holds 0.8M ALGO and passes the rest on; and a 29.6M ALGO residual that the market classifier treats as non-circulating but that sits in no wallet the Foundation publishes, which is the one genuinely opaque item on this page. All of these are re-read from the chain on every rebuild. If any of those balances falls between refreshes, the outflow enters Sell #3 at the next refresh.

How ALGO compares to other capped proof-of-stake chains

Algorand is usually filed next to Bitcoin as a hard-capped coin, and on the mint question that comparison holds: neither chain can create a unit beyond its ceiling, and Algorand goes further because its entire 10,000Mwas created at once in 2019 rather than being released by a halving schedule. But the mechanism that matters for a holder is different. Bitcoin's cap is approached from below by issuance to miners, and every coin not yet issued is genuinely unowned. Algorand's cap was reached on day one, and the difference between the cap and the float is not unissued supply — it is 961M ALGO owned by a single identified entity with discretion over when it sells. A cap protects against dilution by issuance. It does not protect against distribution, and distribution is the entire ALGO story.

Against uncapped continuous-emission chains such as Cosmos Hub, ALGO reads better on structure and only slightly better on the number: those chains mint new units into staking rewards at a rate a governance vote can raise, whereas Algorand's rewards come out of a finite pre-minted account that must be topped up by hand and that shrinks by 1% every million blocks. The forward path is downward by construction. Against exchange tokens that run quarterly buybacks and burns, ALGO has no offset at all — no burn path exists, and the 123.2K ALGO of fees are recycled rather than destroyed, so there is nothing on the buy side that can turn negative into positive as supply grows. And against fully-distributed fair-launch coins, the distinguishing feature is custody concentration: 10.3%of all ALGO sits in one organisation's published wallets, which is both the risk and, because those wallets are published, the thing that makes this page measurable at all.

What to watch in the next 90 days

The block bonus steps down 1% at each million-round boundary, which on the measured 2.75-second block time falls on about Sep 13 2026, Oct 15 2026 and Nov 16 2026— three small, certain reductions in Sell #1. The fee sink's next refill is the more consequential event: it has arrived roughly monthly and the account holds around 65 daysof runway, so a missed or reduced refill would cut block rewards sharply and a larger one would signal the Foundation extending the subsidy past its original roughly 24-month commitment. The Algorand Foundation's Q3 2026 Transparency Report, due in late October, will report holdings at Sep 30 2026 and is the independent check on whether the structured selling that ran at 46.4M ALGOthis quarter is holding, accelerating or pausing. The monthly Algo Insights posts carry the same series at higher frequency. Finally, watch for a second payment resembling the 30.0M ALGO Jul 17 2026 transfer: this build projects it forward at zero because it fired once, and a repeat would move the next-90-day reading from +0.94% back above +1.27%.

Summary

Algorand is a hard-capped proof-of-stake chain that cannot mint an ALGO and cannot burn one, and whose tradable float still grew +1.27% in 90 days and is projected to grow +0.94% in the next 90. The structural mechanism is distribution rather than issuance: 115.3M ALGO of already-minted coin left Algorand Foundation custody, 19.0M as per-block proposer rewards paid out of the fee sink and 96.3M as ordinary transfers out of 74 published wallets, against 123.2K of fees as the only flow in the other direction. The key risk is that 931.0M ALGO — a tenth of everything that will ever exist — remains in those wallets with no release schedule and no obligation to publish one in advance. The ceiling is 10,000M ALGOand it is real, but a ceiling reached in 2019 constrains the mint, not the seller.

MrNasdog Pressure Framework analysis of ALGO, Metric 1 — Inflation. Data + explanation only. Not financial advice. Updated Sep 8 2026.