XRP Inflation Analysis · September 2026 · Supply growing, projected to keep growing
XRP has no protocol inflation at all — every one of the 100B XRP was created in the XRP Ledger's genesis ledger, and the network has no operation that issues more. The pressure comes from Ripple's escrow instead: 1.00B XRP unlocks on the first of each month, but reading the escrow contracts directly shows the locked balance falling only from 32.60B to 31.70B across this window, so the realised release was 900M XRP, not the 3.00B the calendar promised. Against a fee burn of just 30.0K XRP, the MrNasdog Pressure Framework reads XRP at +1.43% net over the last 90 days versus a supply-monitor reading of +0.66% — a gap of 0.78 percentage points, which ships with a monitor-gap flag. XRP is non-inflationary by protocol and schedule-inflationary on the float.
The verdict, in one paragraph
For the 90-day window ending Sep 9 2026, the Pressure Framework reads XRP at +1.43% net. The sell side is the Ripple escrow release, measured on-chain as a realised outflow rather than as a calendar entitlement: the locked XRP escrow balance fell by 900M XRP across the window, even though 3.00B XRP unlocked on paper across three monthly firings and 2.10B XRP went straight back into fresh escrow contracts. The buy side is a mere 30.0K XRP of transaction-fee burn — real, permanent, and a few thousandths of one percent of the release. The independent supply monitor reads +0.66% over the same window, so the gap is 0.78 percentage points and sits outside the framework's half-point tolerance; the page therefore carries a monitor-gap flag, and the primary on-chain reading is the one that ships. XRP is best described as non-inflationary by protocol but schedule-inflationary on the active float — the total supply of XRP can never grow, yet the monthly escrow unlock keeps expanding the tradable share of a fixed pie.
Sell pressure: where new XRP comes from
Sell #1 — protocol inflation — is zero, and it is zero for a reason no vote can undo. The XRP Ledger has no minting operation in its transaction set. The entire 100B XRP supply was created in a single genesis ledger, and every read of the ledger's total-supply field since has been lower than the one before it. Two direct on-chain reads bracket this window: 99,985,648,777 XRP at the start and 99,985,618,817 XRP at the end, reproduced exactly on a second, independent public node. XRP supply fell. That falling read is also what proves the field is a live measurement rather than a hardcoded constant — a flat number would have proved nothing. There is no emission curve, no block reward, no staking issuance and no inflation parameter to tune.
Sell #2 — vesting unlocks — is the whole XRP supply story, and it is where the published schedule and the tradable float pull apart. Ripple wrote 55B XRP into dated on-ledger escrow contracts in 2017, and 1.00B XRP expires out of escrow on the first of each month. Three such unlocks fell inside this window, on Jul 1 2026, Aug 1 2026 and Sep 1 2026, which is 3.00B XRP of calendar entitlement. The escrow contracts are readable on-chain, and reading them settles it: total escrowed XRP was 32.60B at the start of the window and 31.70B at the end. Ripple locked 2.10B XRP straight back into fresh escrow contracts. The realised release — the number the Pressure Framework ships — is therefore 900M XRP, or roughly 300M XRP a month. Booking the gross 3.00B would have invented 2.1B of XRP sell pressure that never touched a market.
The two legs are one operation, not two, and the framework counts them once. The escrow balance cannot fall without the monthly release, and the re-lock is funded out of that same release, so treating the unlock as sell pressure and the re-lock as buy pressure would double-count a single monthly event on both sides of the ledger. The arithmetic closes with nothing left over: 3.00B XRP of escrow finished, 2.10B XRP of escrow created, and a measured escrow balance 900M XRP lower — the sweep and the independent balance read agree to zero. Ripple's own published distribution figure for the start of the window, 32.60B XRP escrowed, matches the on-ledger enumeration exactly.
Sell #3 — Foundation and unscheduled unlocks — is zero for this window, on measurement rather than on assumption. Ripple's identified non-escrow XRP wallets held 3.68B XRP at the start and 3.78B XRP at the end — a net inflow of about 106M XRP, so those wallets absorbed part of the escrow release rather than draining into a market. That leg of the sweep is partial by construction, since Ripple's own classification of non-escrow company holdings runs larger than the wallets that can be identified by label, so it is used to establish direction and never as a row value; the complete leg is the escrow one. Sell #4 — long-term locked or bankruptcy supply — is zero too: no estate holds XRP on a trustee schedule and no long-dated XRP lock is unwinding.
Buy pressure: where new XRP goes
Buy #1 — programmatic buyback — is zero. Nothing buys XRP back. The $750M repurchase Ripple announced in March 2026 bought the company's own private shares at a $50B valuation, not the coin, and it closed in April, before this window opened. Buy #3 — foundation buying — is zero for the same structural reason: Ripple is a seller of XRP into its own operations, never an accumulator, and no on-chain flow this window shows XRP being bought.
Buy #2 — the protocol fee burn — is the only thing on the XRP buy side, and it removed 30.0K XRP across the window, about 334 XRP a day. The XRP Ledger destroys every transaction fee outright rather than paying it to a validator, which is why the chain has no block reward and no dead address to watch: the coins leave the supply meter, and the four unspendable accounts on the ledger moved less than 1 XRP in ninety days between them. Reading both surfaces is what caught this — the dead-address surface says nothing happened, the supply surface says 30.0K XRP is gone, and the disagreement is the finding. The burn was closed against the transaction log twice: across two separate slices of the chain, the sum of every fee paid equalled the fall in total supply to the drop, with a residual of zero. All-time, that mechanism has destroyed 14.38M XRP since launch — under 0.015% of genesis, and against a 900M release it is a rounding error. Buy #4 — new long-term locks — is zero, because the 2.10B XRP re-escrowed this window is already netted inside the Sell #2 figure.
Foundation and overhang
XRP has the largest single tracked overhang in the framework, and it is fully readable. The Ripple escrow now holds 31.70B XRP, just over half of the circulating float, spread across eight on-ledger accounts and released on a published monthly rule with roughly seventy percent locked straight back. Alongside it sit Ripple's identified non-escrow wallets at 3.78B XRP, which grew rather than drained this window. Both balances are re-read on every refresh rather than taken on trust. Exchange custody wallets, the seven US spot XRP funds and unlabelled large holders are excluded by rule — those XRP belong to depositors, to fund shareholders, or to no identified group. If the escrow balance falls faster than the monthly rule implies, or if the non-escrow wallets drain between refreshes, the extra outflow enters Sell #3 at the next refresh.
How XRP compares to other payment-rail chains
XRP sits in an unusual structural position: a hard-capped, zero-issuance asset that nevertheless prints a positive inflation reading. A proof-of-work chain with a hard cap still pays a block subsidy, so its supply rises toward the ceiling while its float and its supply move together. An uncapped proof-of-stake Layer 1 mints validator rewards continuously and hopes a fee burn offsets them. XRP does neither — it mints nothing at all, and its total supply falls by a few hundred XRP a day. What grows is the tradable share of a fixed 100B, released from escrow on a calendar. That is a genuinely different mechanism from issuance, and it has a definite end: at roughly 300M XRP a month of realised release, the 31.70B XRP still escrowed is about nine years of drip before the schedule runs out and the sell side goes permanently to zero.
Against the exchange-linked chain tokens the framework also tracks, the contrast is in what funds the buy side. A chain that routes a slice of every gas fee into destruction shrinks its float continuously, and a chain that burns a reserve on a quarterly schedule shrinks it in steps. The XRP Ledger does neither at scale, because its fee is priced as an anti-spam floor of a hundredth of a US cent rather than as revenue: a day of several million XRP transactions destroys a few hundred XRP. The fee economy on the XRP Ledger runs at roughly $175,000 a year against a market capitalisation near $89B — several orders of magnitude below the smart-contract chains whose burns actually move their float. That design choice is deliberate and it makes XRP cheap to use; the trade-off is that nothing on the XRP buy side can offset the escrow drip, so the sell side sets the reading on its own.
What to watch in the next 90 days
First, the three monthly escrow unlocks on Oct 1 2026, Nov 1 2026 and Dec 1 2026: each releases 1.00B XRP gross, and the whole forward reading turns on the re-lock ratio holding near seventy percent — a month with no re-escrow would triple the realised release. Second, the listed XRP treasury vehicle whose shareholder vote lands on Sep 30 2026, targeting a launch balance of at least 473M XRP; open-market buying by that vehicle moves existing float between holders, but the Ripple contribution disclosed in its filing lands only on closing and would be new float, entering Sell #3 if it fires. Third, a regulatory filing this summer flagged that Ripple could release XRP from escrow beyond the monthly rule if pending US market-structure legislation passes — the single biggest tail risk on this page, and the one event that would break the 300M-a-month pace. Fourth, the quantum-resistance upgrade whose validator activation window opened on Aug 28 2026 and could take effect from Sep 11 2026: it changes signatures, not issuance, so the forward reading was left alone.
Summary
The MrNasdog Pressure Framework reads XRP at +1.43% net over the trailing 90 days and +1.43% over the next 90, on a sell side made entirely of one row. The structural mechanism is a chain that mints nothing — the full 100B XRP existed at genesis and total supply only falls — combined with a monthly escrow release whose realised outflow was 900M XRP against a 3.00B XRP calendar entitlement, and a fee burn of 30.0K XRP far too small to offset it. The key risk is that the re-escrow is discretionary rather than contractual: Ripple chooses how much to lock back each month, so a decision to stop would triple the sell side overnight without any change to the published schedule. The ceiling is the escrow itself — 31.70B XRPstill locked, roughly nine years of drip at the current realised pace, after which XRP's sell side is permanently empty.
MrNasdog Pressure Framework analysis of XRP, Metric 1 — Inflation. Data + explanation only. Not financial advice. Updated Sep 9 2026.