What an XRPL Account Has Cost, 2013-2026
The reserve has been cut four times, from 50 XRP down to 1. In dollars it went the other way first: opening an account cost $0.17 in 2013 and $55.60 at the January 2018 peak.
Same ledger, same purpose, same one-line transaction. The reserve has only ever been cut — from 50 XRP to 1 — yet the price of using the XRP Ledger swung 800× along the way, because the reserve is only half of what sets it.
Every reserve change, read off the ledger
Reserve levels live in FeeSettings, a singleton ledger object that validators update by vote. Bisecting it across the full history — ledgers 32,570 to 105,876,803 — gives the exact ledger and date of every change. There have been four.
| Date | Ledger | Account | Per object |
|---|---|---|---|
| 2013-04-24 | 562,177 | 50 XRP | 12.5 XRP |
| 2013-12-05 | 3,715,073 | 20 XRP | 5 XRP |
| 2021-09-19 | 66,462,465 | 10 XRP | 2 XRP |
| 2024-12-02 | 92,508,417 | 1 XRP | 0.2 XRP |
| today | — | 1 XRP | 0.2 XRP |
| proposed | — | 0.1 XRP | 0.01 XRP |
Before April 2013 there was no FeeSettings object on the ledger at all — the earliest available ledger, 32,570, predates it. The 50 XRP entry is the first value the ledger ever recorded.
What each cut was worth on the day
The same reserve priced in dollars the day it changed — what an account and a single object cost immediately before the cut, and immediately after.
First value the ledger ever recorded — nothing to compare against.
Not enacted. The account is cut 10x but each object 20x.
The December 2024 cut is the one that changed the character of the ledger: it landed with XRP near $2.73, taking an account from $27.28 to $2.73 and a trustline from $5.46 to $0.55 overnight. Each cut has arrived after the dollar cost had climbed, not before.
Since then the reserve has not moved at all — but the cost has still fallen by more than half, because XRP did. That is the whole point of this article in one line: the dollar cost of an XRPL account has two inputs, and the reserve is only one of them.
The dollar cost peaked in the middle
The reserve has only ever gone down in XRP terms. In dollars it rose for a decade first, because XRP appreciated far faster than the reserve was cut. Every month since 2013:
159 months, 2013–2026, both panels on the same log scale so they can be read against each other. The first 21 months are priced from the XRP Ledger's own DEX. Line colour on the lower panel tracks cost: cool when cheap, hot when dear.
Peak: 7 January 2018. The base reserve was 20 XRP and XRP closed at $2.78, so opening an account cost $55.60 and every trustline, offer or NFT page cost another $13.90. The cheapest it has ever been was 2016-2017, around $0.12 — the same 20 XRP, when XRP traded under a cent.
What that meant for a real wallet
Reserves compound, because every object costs one. A wallet holding ten tokens needs the base reserve plus ten trustlines:
Same wallet, same ten tokens, 59× cheaper. That gap is most of the reason holding a broad basket of XRPL tokens stopped being a serious capital decision.
The proposed cut to 0.1 XRP
The proposal cuts the base from 1 XRP to 0.1 and the per-object reserve from 0.2 to 0.01 — ten times for the account, twenty for each object. At today's price that is $0.11 to open an account and $0.011 per trustline. Both are record lows; the object figure is less than half the previous floor of $0.029 set in 2016.
The asymmetry is the interesting part, and it has two consequences that fall straight out of the arithmetic.
Every era has set the base at 4–5× the per-object reserve — 50:12.5, 20:5, 10:2, 1:0.2. At 0.1:0.01 that becomes 10×, double the long-standing norm. It would be the first time an account is priced at ten trustlines rather than five.
Because objects are cut twice as hard as the account, the saving grows with what you hold: 90% for a bare account, 93% at ten tokens, 94.8% at a hundred. A proportional cut would have saved the same everywhere.
It also moves the number that has actually preceded cuts. The two modern ones landed when a single object cost $5.24 (September 2021) and $5.46(December 2024) — a tight band, where the account-cost figures at those same moments were scattered, $20.98 and $27.28. At 0.01 XRP per object that band would not be reached again until XRP is around $525–546, against roughly $26–27 today — twenty times the headroom.
That is a departure from precedent rather than a repeat of it. December 2024 cut both halves by exactly ten, 10→1 and 2→0.2, holding the ratio steady. This one does not, and the effect is a deliberate tilt towards wallets that hold many tokens.
That band is inferred from two data points, not published policy. Reserves move when validators vote them, and nothing obliges them to hold any particular dollar level.
It is locked, not spent
The reserve is not a fee. It is your own XRP, made unspendable while the object exists — and released when the object goes away.
Close a trustline and the 0.2 XRP comes back. Delete the account entirely with AccountDelete and the base reserve is returned to a destination you choose, minus a fee of one owner reserve. So the dollar figures above are the capital locked up at the time, not money burned — which is exactly why the January 2018 peak mattered so much to anyone running many accounts.
How that compares to Bitcoin, Ethereum and Solana
Most chains do not charge you to exist, so the comparison is usually made against a transaction fee — and that is the wrong comparison, because a fee is spent and a reserve is not.
There are really two models. Bitcoin and Ethereum charge a fee: the money leaves and nobody gets it back. The XRP Ledger and Solana take a deposit: the money is immobilised while you occupy space, and returned when you stop.
AccountDelete costs one owner reserve (0.2 XRP), so 0.8 of the 1 XRP base comes back. Closing a trustline returns its full 0.2 XRP.
Refundability improves to 90%: AccountDelete costs one owner reserve, and the proposal cuts that 20x against the base’s 10x, so proportionally less is withheld.
Rent-exempt minimum. Close the account and the rent is returned in full, less the signature fee.
Nothing to activate — an address exists the moment someone pays it. The cost is the sender’s fee for a typical 1-in-2-out transaction at 1 sat/vB, and every UTXO carries a further fee to spend later.
Addresses exist implicitly. A first ERC-20 transfer writes a new storage slot (~65,000 gas at 0.045 gwei); EIP-3529 removed almost all storage refunds, so none of it returns.
Measured 27 July 2026 with Bitcoin at 1 sat/vB and Ethereum at 0.045 gwei— both near the floor. That is the point: a fee is whatever the network charges the day you pay it. At a routine 50 sat/vB and 50 gwei the same two actions cost $4.61 and $6.33— fifty times more, and still nothing comes back. The deposits on the other two chains do not move with congestion at all; they move only with the token price, and you keep them.
Bitcoin is the cleanest illustration of the difference. It has no account to activate at all — an address exists the moment someone pays it. The closest thing to an activation cost is the fee the senderpays, and none of it is recoverable by anyone. The nearest analogue to a minimum balance is the dust limit, 546 satoshis, which is a floor on what can be received rather than a deposit held on your behalf. At today's 1 sat/vB that is about $0.09 for someone to pay you, gone the moment it confirms.
On the XRP Ledger the 1 XRP is never spent. Roughly 80% of it comes back—AccountDelete returns the balance to a destination you choose, less a fee of one owner reserve, so 0.8 of the 1 XRP is recoverable. Trustlines are cleaner still: close one and the full 0.2 XRP is released, costing only the 12-drop transaction fee.
Which makes the honest scoreboard mixed. Today XRPL asks 16× more than Solana to open an account ($1.11 against $0.07) and 1.4× more per token held($0.22 against $0.16) — though both are deposits you get back, unlike either fee chain.
The proposal changes that picture more than it changes the account figure. At 0.1 and 0.01 XRP, opening an account would still cost slightly more than on Solana ($0.11 against $0.07), but holding a token would cost $0.011 against Solana's $0.156— roughly fourteen times less. And because AccountDeleteis priced at one owner reserve, which falls 20× while the base falls 10×, the share you get back would rise from 80% to 90%.
One comparison deserves care rather than a victory lap. At this moment Ethereum's first token transfer costs $0.0057, which is less than the proposed XRPL trustline deposit of $0.011. But that is 0.045 gwei — near the floor — and the Ethereum figure is spent while the XRPL one is returned when the trustline closes. At an ordinary 50 gwei the same transfer is $6.33, some 570× the XRPL deposit, and still gone.
Method
Reserve levels were read from the FeeSettings object at index 4BC50C9B...A6A651 and every change point located by bisection across the full ledger history on s2.ripple.com, then dated from each ledger's close time. Prices before 2016 come from the XRP Ledger's own DEX — the best ask on the XRP/USD.Bitstamp book, read at the exact ledger — since no conventional market data reaches that far back. The two methods overlap at 1 January 2016 and agree to three significant figures: $0.005882 from the DEX against $0.0059 from market close. From 2016 the figures use daily closes; the live price is a median across 11 exchanges. Everything here is reproducible against public infrastructure.