Exchange reserves have fallen to a seven-year low of about 1.6 billion XRP, half what they were at the October 2025 peak. ETFs have absorbed nearly a billion tokens. Ripple still holds roughly 36 billion in escrow. This is the full map of where XRP’s supply actually sits in mid-2026, what moved, what it means, and why a shrinking float has so far failed to move the price.
Summary
- XRP exchange reserves have fallen to a seven year low while spot ETFs have accumulated nearly one billion tokens and long term holders continue moving coins into private wallets.
- Ripple still controls about 36 billion XRP in escrow, but steady monthly releases and relocks have not stopped exchange balances from shrinking to multi year lows.
- The report says tighter supply alone has not lifted XRP’s price, with weak market demand continuing to outweigh the effects of a declining tradable float.
Something unusual is happening to XRP’s supply, and it is happening quietly, underneath a price chart that has spent 2026 telling a story of decline. Exchange reserves, the pool of tokens sitting on trading venues ready to be sold, have fallen to roughly 1.6 billion XRP, the lowest level in seven years and down about 50% from the October 2025 peak of 3.76 billion. On Binance alone, the largest venue for the asset, reserves have dropped 20% since November 2024 to about 2.6 billion tokens across its wallets, pushing a metric called the Scarcity Index to its highest reading in more than two years. Meanwhile the seven US spot ETFs have quietly accumulated more than 970 million XRP, locked in custody on behalf of fund holders, after nine consecutive weeks of net inflows.
Tokens are leaving the places where they can be sold and accumulating in the places where they tend to sit still. In most assets, that migration is the textbook setup for a supply squeeze. In XRP, the price has fallen anyway, trading near $1.13, down roughly 70% from its July 2025 peak of $3.65, through the entire period in which the float was tightening.
That contradiction is the story. This piece maps the full distribution of XRP’s supply as of mid-2026: what sits on exchanges, what the ETFs hold, what Ripple controls in escrow and operational wallets, and what the remaining tens of billions in private hands are doing. It then works through why a halving of exchange reserves has not produced the price response the squeeze thesis predicts, the competing explanations for the gap, and the specific conditions under which a tight float starts to matter. The supply side of XRP has rarely been this interesting; the demand side is the reason nobody has noticed.
The map: 100 billion tokens, five buckets
XRP’s supply structure is unlike any other major asset, and the map has to start from its founding fact: all 100 billion tokens were created at launch in 2012. There is no mining, no issuance schedule, no future supply beyond what already exists. About 14 million XRP have been permanently destroyed as transaction fees since then, a rounding error, leaving total supply just below 100 billion. Everything else is a question of where the existing tokens sit, and in mid-2026 they sit in five buckets.
The first bucket is Ripple’s escrow, the largest single concentration of XRP in existence at roughly 36 billion tokens, about 36% of total supply. These are time-locked on-chain contracts releasing one billion XRP on the first of each month, of which Ripple typically relocks 600 to 800 million and keeps a net 200 to 300 million for operations, a mechanism this publication has explained in full. In July, Ripple relocked about 70% of the monthly billion, releasing 300 million into circulation. The escrow is the structural overhang critics cite and the transparency mechanism defenders praise, and either way it is the slowest-moving bucket: at current net-release rates, depletion is roughly nine years out.
The second bucket is circulating supply proper, about 62 billion tokens, and the remaining buckets are subdivisions of it. Exchange reserves, the third bucket, are the sellable edge of the market: roughly 1.6 billion tokens across venues, the seven-year low. The fourth bucket is the ETF complex: seven US spot funds holding a combined 970 million or so tokens, a bit over $1 billion in assets, tokens held by custodians and effectively removed from trading circulation for as long as fund investors stay put. The fifth bucket, by far the largest slice of circulating supply, is everything else: private wallets, corporate treasuries, whale cold storage, and long-term holders, somewhere near 59 billion tokens whose owners have, on the evidence of on-chain data, been net withdrawers from exchanges for over a year.
Two things stand out from the map. First, the actively tradable float, the exchange reserves, is now under 3% of circulating supply and under 2% of total supply, remarkably thin for a top-six asset by market value. Second, the two fastest-growing buckets, ETF custody and private cold storage, are both one-way doors in the short term: tokens flow in easily and come back out only when holders make an affirmative decision to sell.
What moved, and why
The reshaping of the map over the past eighteen months has three drivers, each visible on-chain.
The first driver is the ETF complex, which did not exist before November 2025. Since the first spot XRP fund launched, the products have absorbed roughly $1.5 billion in cumulative inflows, and because they hold the underlying token, every dollar of inflow is a market purchase moved into custody. The funds have now recorded nine consecutive weeks of net inflows, adding $17 million in the latest week even as Bitcoin and Ethereum funds bled, a rotation this publication has tracked. Nearly a billion tokens now sit in ETF custody, and the mechanism only reverses if fund investors redeem at scale, which, so far, they have done on exactly one notable day, the quarter-end outflow of June 30.
The second driver is whale and institutional withdrawal. CryptoQuant data shows the Binance drawdown accelerating recently, from about 2.8 billion tokens in May to 2.6 billion in early July, exactly the window in which the Scarcity Index broke out to 0.77. Large-holder activity has strengthened while retail stays cautious, new-wallet creation hit a three-month high, and Korean venues have recorded repeated multi-million-token outflows. The pattern, tokens moving from hot exchange wallets to cold private ones, is the classic signature of accumulation by holders with no near-term intention to sell.
Notably, this is the reverse of December 2024, when the Scarcity Index collapsed because holders were depositing XRP onto Binance in bulk to sell the rally to $3; today’s flows run the other way, out of the venues, into storage, at prices two-thirds lower.
The third driver is the escrow’s steady arithmetic. Ripple’s net release of 200 to 300 million tokens a month adds roughly 4-6% to circulating supply annually, a bounded, scheduled inflation the market can model years ahead. In 2026 the company has if anything leaned conservative, relocking 70% in recent months, and part of what it does release goes to institutional counterparties off-exchange, never touching the tradable float at all. The escrow is a source of supply, but it is a metered one, and its pace has not changed while the exchange drawdown accelerated, which means the drawdown is demand-side behavior, not a supply-side trick.
The puzzle: a tightening float and a falling price
Here is where the story stops being simple. Every element above, reserves halved, ETFs absorbing, whales withdrawing, metered issuance, belongs to the standard playbook of a supply squeeze, the setup in which shrinking availability meets steady demand and the price ratchets upward because sellers become scarce. XRP has instead spent 2026 falling, from $2.41 in January to near $1 in late June, before the modest recovery to $1.13. The float tightened; the price halved. Any honest supply analysis has to explain that, and there are three serious explanations, not mutually exclusive.
The first is that scarcity on exchanges measures potential, not pressure. A thin order book amplifies whatever demand arrives; it does not create demand. Through 2026, demand has been the missing side: derivatives open interest collapsed from last year’s highs, retail participation stayed weak, funding rates flipped decisively negative as price approached $1, and ETF inflows, while persistent, ran at a pace of tens of millions per week, roughly the same order of magnitude as Ripple’s monthly net escrow release in dollar terms. Australian lawyer and longtime XRP commentator Bill Morgan has made the sharper version of this point: neither the supply-squeeze thesis nor the older escrow-dump fear explains XRP’s price well, because the dominant variable is simply Bitcoin, which fell through the same months and dragged the whole market with it. On this reading, the tight float is dry tinder, and 2026 has been a year without a spark.
The second explanation is that the headline reserve numbers may overstate the tightness. Skeptics of the squeeze thesis note that measured exchange reserves depend on which wallets analysts attribute to which venues, that internal transfers can masquerade as outflows, and that estimates of total platform-held XRP across all venues and custodians run far higher than the headline 1.6 billion, with some placing 14 to 16 billion tokens within fast reach of order books. The February-March episode in which roughly 350 million XRP dipped and rebounded on Binance, likely internal wallet reshuffling rather than organic flow, illustrates how noisy the data is. If the true sellable supply is several multiples of the visible reserve, the squeeze is further away than the dashboards suggest.
The third explanation is structural: the sellers who matter are not on exchanges yet. Millions of tokens were accumulated between $1.50 and $1.90 during the spring’s failed rallies, and holders underwater at those levels represent a standing wall of supply that will migrate back onto exchanges precisely when price approaches their break-even. Add Ripple’s monthly release and the possibility of ETF redemptions in a risk-off shock, and the tight float is best understood as tight at current prices, with reinforcements waiting at higher ones. Santiment’s MVRV data showing holders at their deepest unrealized losses in the token’s history cuts both ways: it signals capitulation-grade sentiment, and it also marks exactly where the exit orders cluster.
How to read the metrics without fooling yourself
Because the supply story runs on a handful of dashboards, and because those dashboards are routinely misread in both directions, a short field guide to the metrics is worth the space.
Exchange reserves are an attribution exercise, not an audit. Analytics firms tag wallets they believe belong to venues and sum the balances, which means the headline number moves when tagging improves, when exchanges reorganize custody, and when internal transfers cross the tagged perimeter, none of which involves a single token changing owners. The 350 million XRP that appeared to leave and re-enter Binance across February and March was almost certainly internal wallet management, and any single week’s reserve print should be read with that episode in mind. The signal is in the trend across months and across independent data providers, and on that standard the 2026 drawdown is robust: the direction has been consistent since late 2024, it appears in CryptoQuant, exchange-published data, and third-party trackers alike, and it has accelerated instead of mean-reverting.
The Scarcity Index is a ratio, and ratios have two moving parts. The index compares available supply on Binance against demand conditions, so it can rise because tokens leave, because buying absorbs, or both, and it can whipsaw, as it did on the round trip from 0.80 in spring to 0.34 in June to 0.77 in July, without the underlying reserve base moving anywhere near as violently. Its historical extremes are more informative than its level: the deeply negative readings of December 2024 marked holders flooding coins onto the venue to sell a top, and the current two-year high marks the opposite regime, coins leaving into weakness. As a regime indicator it has value; as a timing tool it has embarrassed everyone who used it as one this year.
ETF holdings are the cleanest series in the entire picture, because fund custodians disclose and the products file, which is why the roughly 970 million tokens across the seven funds is the number this piece leans on hardest. Even here, one habit matters: distinguish flows from assets. Net assets fall when the price falls even while inflows continue, which is exactly what happened through the spring, deposits arriving as valuations shrank, and reading the AUM decline as investor exit inverted the truth. Flow data, positive for nine consecutive weeks, is the demand signal; asset data is mostly a price echo.
Escrow figures, finally, come with the strongest health warning of all, because the number that matters is not the billion that unlocks but the net that stays out, and the net is only knowable after the relock lands days later. Ripple’s own quarterly reports, the on-chain escrow contracts, and the monthly relock transactions are all public, and the discipline is to compute the net against the trailing 200-to-300-million average before drawing any conclusion. A month in which the net spikes above the band is a genuine signal about the company’s cash needs; a month of headlines about a billion-token unlock that ends in a 70% relock, like this July’s, is a signal about headlines. Every metric in this story is public, which is XRP’s genuine advantage as an object of analysis, and every one of them rewards the reader who checks the denominator before repeating the numerator.
What history says about tightening floats
The squeeze thesis is not being invented for XRP in 2026; it has a track record in this asset and others, and the record is worth consulting because it cuts both ways.
The supportive precedent is 2024. Exchange outflows through that year preceded the powerful multi-month rally that carried XRP from under a dollar to its January 2025 highs above $3, with Korean regional demand and shrinking sell-side reserves amplifying the move once the SEC settlement and ETF approvals supplied the demand spark. The structure of that episode maps closely onto today’s: months of quiet withdrawal, a scarcity metric stretching to extremes, skeptics dismissing the data, and then a catalyst arriving into a market with far fewer sellers than buyers expected. Holders who lived through it read the current seven-year-low reserves as the same picture at an earlier frame.
The cautionary precedents are just as instructive. The Scarcity Index itself has whipsawed within 2026: it climbed to nearly 0.80 in the spring, sagged to 0.34 by late June amid heavy long liquidations, then broke out to 0.77 in the first week of July, and the price fell through the entire sequence. A metric that can round-trip that violently inside one quarter is measuring flow conditions, not destiny, and the June reading arrived alongside more than $13 million in single-day long liquidations, a reminder that leverage positioning can overwhelm spot scarcity on any given week. December 2024 offers the mirror lesson: reserves ballooned precisely at the top, as holders raced to deposit and sell the $3 rally, which is to say the metric is at its most bullish after prices have already fallen and its most bearish after they have already risen, a lagging emotional gauge as much as a leading structural one.
The broader crypto record adds a final nuance. Bitcoin’s great supply-squeeze narratives, the 2020-21 exchange exodus, the post-ETF custody absorption of 2024, each eventually mattered, and each mattered on the demand side’s schedule, not the supply side’s. Assets have sat at multi-year reserve lows for quarters while prices drifted, and then repriced in weeks once flows arrived, because a thin float does nothing until someone leans on it, at which point it does everything at once. That asymmetry, long stretches of irrelevance punctuated by sudden amplification, is the honest historical summary, and it is why the traders who take the supply map seriously express the view through patience and position sizing, the same execution discipline any thin market demands, rather than through timing calls the data cannot support.
There is one more structural actor worth watching that previous cycles lacked: the corporate and fund treasuries. Beyond the seven ETFs, a growing roster of listed companies has adopted XRP treasury strategies, and the ETF custodian wallets themselves have become the single most legible accumulation channel in the asset’s history, absorbing roughly 750 million tokens in their first two months alone. Treasury demand is slower and stickier than trader demand, it neither chases rallies nor panics in drawdowns on the same timescale, and its growth quietly raises the floor beneath the float. Whether it grows fast enough to matter against escrow issuance is, like everything in this story, a race whose lap times are published monthly.
What would make the float matter
The supply map becomes decisive only when demand shows up, so the forward-looking question is what could supply the spark, and the candidates are concrete.
The nearest is legal. The CLARITY Act’s commodity classification for XRP, if enacted, is the gate behind which the large conditional forecasts sit: JPMorgan and Standard Chartered have each projected $4 to $8.4 billion in first-year ETF inflows under passage, an order of magnitude above the current run rate.
Flows of that size, arriving into a float of under two billion exchange-held tokens, are the scenario in which the scarcity math stops being academic; the Senate’s three-week window is therefore as much a supply-side story as a regulatory one. The second candidate is institutional adoption converting to token demand through collateral and settlement use, the slow path whose honest accounting runs through Ripple Prime, and the third is simply the market cycle: XRP has historically fallen harder than Bitcoin in downturns and snapped back harder in recoveries, and a thin float mechanically steepens the snapback.
Against these, the checkable risks: a CLARITY failure pushing institutional flows past 2027, ETF inflows decelerating or reversing for consecutive weeks, or reserves rebuilding as underwater holders redeposit into any rally. The dashboard for all of it is public. Exchange reserves, the Scarcity Index, weekly ETF flows, and the monthly escrow relock are each published within days, and together they will show the squeeze forming, or failing, in close to real time.
The conclusion the map supports is narrower than either camp’s slogan. XRP’s tradable supply has genuinely, measurably contracted to multi-year lows while long-horizon buckets absorbed the difference, and that contraction has been irrelevant to price for a year because demand collapsed faster than the float did. Scarcity is not a catalyst; it is a multiplier waiting for one. The honest position is that XRP enters the second half of 2026 with the most squeeze-prone supply structure it has had since at least 2019 and no evidence yet of the demand that would trigger it, which makes the supply map neither bullish nor bearish on its own, but the single best lens for judging how violently the price will move when the demand question, one way or the other, finally resolves.
One final frame is worth carrying away, because it reconciles everything above into a single sentence: XRP in mid-2026 is an asset whose company is accumulating credentials, whose long-horizon holders are accumulating tokens, and whose traders have spent a year accumulating losses, and the supply map is the ledger on which all three behaviors are legible at once. The reserves data records the holders’ conviction, the ETF flows record the institutions’ patient entry, the escrow relocks record the company’s restraint, and the price records the absence, so far, of anyone forced to compete for a shrinking float. Markets in this configuration tend to resolve abruptly rather than gracefully, because thin floats do not permit gradual repricing in either direction: the same scarcity that would turbocharge an inflow shock also means a demand collapse finds few bids on the way down, which is the double edge the squeeze narratives rarely mention. The map says the stage is set. It has never claimed to know the play.
For readers who want to run the numbers themselves, the recipe is short. Take the circulating supply of roughly 62 billion, subtract the ETF custody balance published in the funds’ daily disclosures, subtract the aggregated exchange reserves from at least two independent trackers, and treat the remainder as the private-holder bucket whose behavior the withdrawal trends describe. Cross-check the month’s escrow arithmetic against the on-chain relock, and note the week’s ETF flow direction. Fifteen minutes of public data, repeated monthly, reproduces every structural claim in this piece and will catch the turn, whichever way it breaks, well before the headlines do.
The last variable, as always with this asset, is the one no dashboard tracks: how much of the withdrawn supply belongs to hands that will actually hold through the next stress test. Cold-storage balances built at $1.10 by buyers who watched the token at $3.65 carry a different resolve than balances built chasing a rally, and the 2026 drawdown has, if nothing else, transferred an unusual share of the float to owners who bought weakness deliberately. That is not a prediction. It is the one qualitative fact the quantitative map quietly implies, and the one that will decide whether the next demand shock meets a wall of break-even sellers or an empty room.
Disclaimer: This article is for informational purposes only and does not constitute investment advice. Digital asset markets are volatile and you can lose your entire investment. On-chain and market figures are estimates current as of July 8, 2026, and may change. Always do your own research.
Feature,XRP#XRP #vanishing #exchanges #supply1783518128
