RLUSD supply on the XRP Ledger fell by about $284.4 million on Friday 30 October, from 991.7 million tokens to 707.4 million, according to XRP Ledger data read by Coindaro. That is a 28.7% drop in a single day, the largest in dollars in the DefiLlama series for Ripple’s stablecoin on that chain. Total RLUSD supply fell by far less, because Ethereum supply rose by about $92.7 million over the same period. Ripple has not said why.
CryptoSlate reported the fall on Monday 2 November, working from DefiLlama figures and ledger readings by NOSHASHI. We then went through the issuer account ourselves, transaction by transaction. That reading gives the exact hours and shows one wallet handling every large return. DefiLlama’s own daily figures add a month-end pattern going back to January.
The pattern does not settle the question. It is an observation about the calendar, no one at Ripple has confirmed what it means, and October’s drop is about twice the size of any earlier one. At 06:24 UTC on Monday the XRP Ledger supply stood at 711.8 million, so the rebound that followed earlier month-ends has not appeared.
RLUSD supply on the XRP Ledger, hour by hour
The figures below come from the gateway_balances method of the XRP Ledger, which gives the total an issuer has issued, queried on a public server for Ripple’s RLUSD issuer account. Each payment was then found in its own ledger. All times are UTC on 30 October.
| Time (UTC) | Movement (RLUSD) | Supply after (million) |
|---|---|---|
| 00:20 | 10.0M returned | 981.7 |
| 02:11 to 02:15 | 10.4M returned | 971.3 |
| 05:54 to 05:57 | 17.0M returned | 954.3 |
| 13:02 to 13:18 | 30.0M returned | 924.4 |
| 14:45 | 20.0M issued | 944.4 |
| 14:52 | 20.0M returned | 924.4 |
| 16:08 | 106.5M returned | 817.9 |
| 16:31 | 42.9M returned | 775.0 |
| 19:11 | 50.0M returned | 725.0 |
| 21:22 | 19.6M returned | 705.4 |
| 22:16 | 2.0M issued | 707.4 |
So this was no single overnight event. It came in 12 large payments spread over 21 hours, totalling about $306.4 million, while the issuer put $22 million of new tokens out in two payments. Those two sets of numbers net to the $284.4 million fall. Two more returns of 1 million RLUSD each went through late on 29 October, at 22:42 and 22:48 UTC, which is why supply had already slipped from a high of 993.7 million reached at 19:41 UTC that evening.
The dates can confuse. DefiLlama stamps the fall on 31 October and CryptoSlate reports it that way, but the ledger puts it on the 30th. The difference comes from when the daily figure is recorded. The size matches to the token.
One wallet handled every large return
All 12 large payments came from the same XRP Ledger wallet. Coindaro could not identify its owner, and XRPScan, the explorer we checked, shows no label for it, so we do not attribute it. The wallet holds no RLUSD now. It works as a funnel: tokens arrive from other wallets and leave for the issuer within minutes or hours.
On 30 October five wallets fed it. One sent 106.5 million at 15:16 UTC and 42.9 million at 16:15 UTC, nearly half of the day’s total. Another sent 50 million at 15:49 UTC. Nothing in the ledger shows whose tokens these were or why they were returned.
The funnel is not new either. By our count it has sent RLUSD back to the issuer every month since April: $232.1 million in April, $228.3 million in May, $162.8 million in June, $224.9 million in July, $218.7 million in August and $339.3 million in September. October’s total is $1.08 billion across 101 payments, about three times the previous high.
RLUSD supply at each month-end of 2026
Since April, and before 30 October, the wallet’s six busiest days of returns were 30 April, 29 May, 30 June, 31 July, 31 August and 30 September, between $114 million and $146.7 million each. Every one of them was the last working day of its month. So was 30 October: the 31st fell on a Saturday.
DefiLlama’s daily series for the XRP Ledger shows the same shape, and it also shows what came next. Supply dropped at each month-end in 2026. In every case most or all of the loss was back in the DefiLlama stamp that follows the first working day of the new month.
| Month-end | Before (million) | After (million) | Change | Rebound level (million) |
|---|---|---|---|---|
| January | 338.0 | 260.8 | -22.8% | 348.0 |
| February | 345.0 | 256.4 | -25.7% | 345.0 |
| March | 317.9 | 189.8 | -40.3% | 340.3 |
| April | 372.6 | 253.3 | -32.0% | 376.3 |
| May | 705.3 | 613.3 | -13.1% | 734.4 |
| June | 813.1 | 676.9 | -16.7% | 846.4 |
| July | 881.2 | 759.7 | -13.8% | 850.7 |
| August | 1,040.6 | 913.5 | -12.2% | 1,049.4 |
| September | 1,124.7 | 1,013.5 | -9.9% | 1,235.7 |
| October | 991.7 | 707.4 | -28.7% | not seen at 06:24 UTC, 2 Nov |
Read that way, October is the largest month-end fall in dollars but not in percentage terms: March and April were steeper. The size in dollars still stands out. The previous largest was about $136 million, at the end of June.
The cycle has been noticed before. In May, The Crypto Basic reported that XRP Ledger validator Vet, who built a community tracker of RLUSD, said large month-end burns tend to be followed by equally large mints early in the next month. It listed $93.2 million burned on 30 January and $102 million minted on 2 February, among other pairs. That is a third-party reading. We found no Ripple document that describes such a routine or gives a reason for it.
We can see the timing. We cannot see the purpose, and calling this housekeeping would be a guess. Monday 2 November is the first working day of the month, so the same calendar would put a large mint on the ledger from today. It may not come, and if it does it will show timing again, not a reason.
Ethereum gained a third of the XRP Ledger fall, and the move goes back weeks
Tokens moving from one chain to the other is the obvious reading, and the numbers do not support it on their own. According to DefiLlama, Ethereum supply went from 1,617.2 million to 1,709.8 million across the same two daily stamps, a rise of $92.7 million. That is a third of the XRP Ledger fall, and it was the largest Ethereum rise of any month-end this year. Combined supply fell from 2,608.9 million to 2,417.2 million, a net drop of $191.7 million, or 7.3%.
Ethereum was growing anyway, by about $300 million between 18 October and 2 November. As CryptoSlate points out, a rise on Ethereum does not prove that the funds migrated. A stablecoin issuer can burn on one chain and mint on another, and the ledgers will show two separate events.
The wider shift is easier to measure. On 2 October, the XRP Ledger held 49.2% of the supply on the two chains, 1,235.7 million against 1,274.0 million on Ethereum. By 2 November it held 29.4%. Supply on the XRP Ledger had already fallen by $244 million during October, before the burns of 30 October.
Between the 31 October and 2 November stamps, XRP Ledger supply rose by 4,470,449 tokens and Ethereum supply fell by the same number, to within a token. Total supply did not move. That is what a chain swap would look like, but only for $4.5 million, and it tells us nothing about the $191.7 million that left the total.
What a burn is, and what it does not prove
On the XRP Ledger a token such as RLUSD is a balance owed by its issuer. When a holder sends it back to the issuer, that balance disappears: the tokens no longer exist and the issued total falls. This is what the market calls a burn. The closest comparison is a bank cancelling a deposit.
The step that matters to a holder happens off-chain. Sending tokens to the issuer does not show that dollars went out in exchange, as CryptoSlate also notes. A customer may have redeemed for cash. A market maker may have returned spare inventory. Ripple may have reduced tokens it held for its own purposes. The ledger looks identical in all three cases, and we do not know which applies.
Ripple’s own figures do not help yet. Its transparency page, read on 2 November, gives $2,627.0 million of circulating RLUSD and $2,758.6 million of reserve funds as of 29 October, a ratio of about 105%. Those numbers predate the burns. The page gives no split by chain and says nothing about this episode, and the latest attestation it lists is for September. How RLUSD is issued and supervised is set out in our RLUSD project page, and the OCC trust bank approval covers who will hold the reserves.
Nothing in the price points to stress. DefiLlama’s daily RLUSD price was $1.0005 on 31 October and $1.0003 on 2 November, and its lowest reading of the past week, $0.9993, came on 29 October, before most of the burns. These are daily readings, not tick data. Moody’s A2 and P-1 ratings, published days earlier, were assigned before the fall and say nothing about it.
XRP pools barely moved because the tokens never touched them
For anyone trading XRP against RLUSD on the ledger, the main finding belongs to CryptoSlate and NOSHASHI. On NOSHASHI’s reading, as published by CryptoSlate, the XRP/RLUSD automated market maker held 1,883,641.61 XRP and 2,327,587.15 RLUSD at 23:59 UTC on 29 October. We read the same smart contract holding reserves of two or more tokens, supplied by users, that a decentralized exchange uses to execute trades automatically at a price set by a formula.">pool at ledger 107990150, at 23:57 UTC, and again at ledger 108023839, at 13:11 UTC on 31 October, and got the same reserves as NOSHASHI at both points. The second reading was 1,871,106.09 XRP and 2,335,993.08 RLUSD. That is 0.67% less XRP and 0.36% more RLUSD, after a fall of nearly 29% in supply.
The reason is mechanical. A pool holds its own deposits, and the burned tokens came from other wallets. This pool held about 2.34 million RLUSD at 06:24 UTC on Monday, about a third of 1% of supply on the ledger. A reader new to these pools can start with our guide to how DEXs and AMMs work.
Order books moved more, on NOSHASHI’s figures as reported by CryptoSlate, which we did not reproduce. XRP-funded bids fell from 71,292 RLUSD at the 29 October close to 40,871 at the 30 October close, and were back at 59,593 on 31 October. Offers selling RLUSD for XRP rose from 62,953 to 112,163 over the same dates. These are small numbers, and a thinner side of a book can have many causes. NOSHASHI estimated a 10,000 RLUSD buy at 0.14% above the mid price and a sale at 0.08% below it, on 31 October at 13:11 UTC. Those are estimates from resting orders, not executed trades.
XRP traded near $1.25 around 05:54 UTC on Monday. We see no way to link that level to the burns from these data.
RLUSD and MiCA: what European readers can and cannot conclude
Nothing in the ledger or in Ripple’s page connects the fall to Europe. The regulatory position is a separate matter. Under MiCA, a dollar stablecoin offered in the EU has to be an e-money token from an authorised issuer. On 2 November we downloaded the e-money token file of ESMA’s interim register, EMTWP.csv, last modified on 30 October, and searched its 50 rows for RLUSD, Ripple and Standard Custody. There were no hits.
That shows one thing: this version of the file has no entry for the token or for those names. It does not prove more. ESMA says the register is built from what national authorities report and is refreshed weekly, so a recent notification could lag. The file is not a ruling on RLUSD and says nothing about any pending application.
Ripple does hold EU licences. Its release of 6 July says Luxembourg’s CSSF authorised it as a crypto-asset service provider and refers to an EU e-money licence. That release does not mention an e-money token authorisation for RLUSD.
ESMA’s opinion of 8 October tells MiCA-authorised firms to stop services linked to non-compliant stablecoins for EU clients, with existing exposures to be dealt with by about 8 January 2027. The opinion names no token. Treating RLUSD as covered is our reading, not a statement by ESMA or Ripple. None of the sources we read ties the 30 October burns to that deadline, and we do not suggest a link.
What is still unknown, and the next dates
Four questions remain open. Who owns the funnel wallet, and whose tokens passed through it. Whether dollars left Ripple’s reserves, and how many. Why October’s month-end fall was about twice the usual size. And whether the $191.7 million drop in total supply will reverse, as the falls before it did.
Coindaro found no statement from Ripple on its site or on X by the morning of 2 November. Ripple’s transparency page has yet to show figures from after the burns, and the October attestation, when it appears, would be the first third-party report to cover this period. The next data point is on the issuer account, which anyone can read through the same public call.