◈ XRPL Network Analysis · onledger.net

XRPL Account Reserve Policy
Evidence Against the State Bloat Argument

Published July 2026 Data source: db.xrplwin.com · xrplcluster.com Ledger range: genesis → ledger 105,862,544
Position Statement

The prevailing argument against reducing the XRPL base account reserve is that a lower reserve will enable ledger state spam — a flood of cheap accounts creating objects that burden validators and full-history operators. This concern has been used to justify keeping the reserve at its current level of 1 XRP.

The on-chain evidence does not support it. The XRPL has 8 million accounts. 82.9% hold no owned objects. 422,439 accounts are already stranded under the current reserve. The network is under-utilised by every object-based measure available. The state spam risk being cited to block a reduction has not materialised even at the current reserve level — let alone at a lower one.

This document argues for a reduction of the XRPL base account reserve from 1 XRP to 0.1 XRP, supported by 13 years of on-chain evidence.

Summary. The primary argument against reducing the XRPL base account reserve — that lower reserves cause ledger state bloat — is not supported by on-chain evidence. This analysis examines 13 years of XRPL ledger state data and demonstrates that state object growth is driven by application-layer activity, not by account creation. The current reserve structure strands a significant number of active accounts while providing minimal protection against genuine state abuse. A reduction to 0.1 XRP is justifiable on both economic and network-health grounds.

1. Current Network State

As of July 2026, the XRPL has 8,020,650 active accounts. The following measurements were obtained directly from validated ledger state via ledger_data iteration across all AccountRoot objects.

Active accounts
8,020,650
isActive = 1 in accounts table
Accounts with zero owned objects
82.9%
6,647,022 accounts — OwnerCount = 0
XRP locked in reserves
9.64M
0.014% of circulating supply
Already stranded today
422,439
Balance below current reserve requirement
Avg owned objects per account
1.39x
Among accounts that hold any objects: 3.57x
Total ledger state objects
19.2M
~11.96 GB raw · ~8 GB compressed (NuDB)

Key finding: Of 8,020,650 active accounts, 6,647,022 (82.9%) hold no owned objects whatsoever. They pay the base reserve solely to maintain account existence. The reserve increment charged per owned object (0.2 XRP) is irrelevant to the overwhelming majority of XRPL accounts.

2. Owner Count Distribution — The Full Picture

The following distribution was obtained by reading the OwnerCount field from every AccountRoot object in the current validated ledger. This represents the exact number of owned ledger objects (trust lines, offers, escrows, payment channels, etc.) per account.

XRP available to spend
67.53B
99.986% of all XRP in accounts is spendable
XRP locked in reserves
9,638,888
0.014% of all XRP in accounts
Total XRP in all accounts
67.54B
Base reserve + owner reserve + available
Owned objects Accounts % of total Reserve locked (XRP) Note
0 objects 6,647,022 82.9% 1.00 XRP each Base reserve only — no activity
1 object 688,762 8.6% 1.20 XRP each Minimal usage — 1 trust line or offer
2 objects 266,341 3.3% 1.40 XRP each
3 objects 104,829 1.3% 1.60 XRP each
4 objects 61,933 0.8% 1.80 XRP each
5 objects 52,463 0.7% 2.00 XRP each
6–10 objects 81,748 1.0% 2.20–3.00 XRP each Active DEX / token users
11–20 objects 39,583 0.5% 3.20–5.00 XRP each Multi-token holders
20+ objects 62,127 0.77% 5.00+ XRP each Power users — exchanges, market makers

Observation: 94.8% of XRPL accounts hold 2 or fewer owned objects. The owner reserve mechanism — designed to price in the cost of additional state — primarily applies to 0.77% of accounts holding 20 or more objects. The base reserve is the dominant cost for virtually all users, and reducing it to 0.1 XRP would benefit 99.23% of accounts without materially changing the incentive structure for heavy state users.

3. The State Bloat Argument — Examined

The claim is that reducing the account reserve will cause a surge in account creation which will in turn cause ledger state to grow unsustainably, harming validators and full-history operators.

This argument has three testable components:

2.1 Does reserve reduction cause account growth?

The most recent reserve reduction — from 10 XRP to 1 XRP on 2 December 2024 (ledger 92,508,417) — provides direct evidence. Account growth did accelerate after the drop, consistent with lower barriers to entry. However, the rate of owned object creation did not accelerate proportionally. The ratio of owned objects per account continued its long-term decline.

3.2 Does account growth cause state bloat?

Storage cost per account (measured)
~1 KB
AccountRoot + DirectoryNode + avg trust lines · NuDB compressed
Cost of 1 million new accounts
~800 MB
At current object ratios and NuDB compression

At 1 KB per account on disk, one million new accounts adds approximately 800 MB to ledger state. Ten million new accounts adds 8 GB — equivalent to doubling the current state tree. Modern server hardware handles this trivially. State storage is not a binding constraint at any plausible account growth rate.

3.3 Was the historical state explosion caused by reserve reductions?

The largest ledger state growth event in XRPL history — the trust line explosion of 2021 — provides the clearest evidence. Examining daily object creation data from daily_object_metrics:

◈ XRPL Network Data

Ledger State Object History

Cumulative objects in XRPL ledger state · January 2020 – July 2026 · Source: daily_object_metrics via db.xrplwin.com

Accounts (AccountRoot)
8.02M
1.88M existed before Jan 2020
Trust Lines (RippleState)
6.86M
Peaked at ~12.9M — Jan 2022
NFT Pages
435K
Launched Oct 2022
NFT Offers
315K
Active secondary market
Cumulative ledger objects over time
Trust lines and NFTs drove more state growth than accounts
RippleState
AccountRoot
NFTokenPage
NFTokenOffer
XRPL object history
Aug 20 2021 — Trust line explosion begins (10,099 net new in one day) Sep 16 2021 — Reserve drop 20→10 XRP (ledger 66,405,889) Nov 2021 — Peak ~350,000 net new trust lines/day Dec 18 2021 — Trust line cleanup begins Oct 2022 — NFTs launch on XRPL Dec 2 2024 — Reserve drop 10→1 XRP (ledger 92,508,417) Faint dashed lines show 1 year prior to each reserve drop
Monthly net new objects
Objects created minus objects deleted each month
RippleState
AccountRoot
NFTokenPage
Monthly delta
Owned objects per account — the state bloat myth
Average owned ledger objects per account · most accounts hold zero · ratio falling back toward pre-boom baseline
Avg objects per account
Today: 82.9% of accounts have ZERO owned objects
Object ratio chart
Jan 2022 peak: 3.2 objects per account — trust line mania Jul 2026: 0.98 objects per account — back near baseline 82.9% of all 8M accounts hold ZERO owned objects today
19 Aug 2021
Trust line creation spikes from ~200/day to 1,252/day. The token issuance craze begins. Reserve is 20 XRP at this point.
20 Aug 2021
10,099 net new trust lines in a single day. The explosion is underway — four weeks before any reserve change.
16 Sep 2021
Reserve drops from 20 XRP to 10 XRP (ledger 66,405,889). Trust line creation is already at 21,307/day and accelerating. The reserve drop did not cause the explosion — it was already in full effect.
Nov 2021
Peak: 350,000+ net new trust lines per day. 4,751,884 cumulative owned objects across 3.75M accounts. Average 4.82 objects per account that held any objects at all.
Jan 2022
Absolute peak: 12,479,589 total owned objects across 4M accounts. Average 6.16 objects per holder. This was a speculative token craze, not a structural consequence of reserve policy.
Dec 2021 onward
Mass cleanup begins. -77,143 net trust lines on 24 December 2021 alone. Users deleted trust lines organically as the token craze subsided.
Jul 2026
11,158,666 owned objects across 8,020,650 accounts. Average 1.39x per account overall, 3.57x per holder. The ratio continues to decline.

The 2021 trust line explosion preceded the September 2021 reserve drop by four weeks. Causal attribution is reversed in the common argument. Reserve reductions did not cause state bloat — speculative token issuance did, and it self-corrected without any policy intervention.

4. Owned Objects Per Account — Network Activity Indicator

The average number of owned objects per account is a direct measure of how actively the network is being used for its intended purposes: holding tokens, placing offers, maintaining payment channels, using escrow, and so on.

Date Accounts Owned objects Avg (all accounts) Avg (holders only) Accounts with objects
Aug 2021 (pre-boom) 3,239,601 1,512,760 0.47x 2.72x 557,159
Nov 2021 (mania) 3,749,110 4,751,884 1.27x 4.82x 986,390
Jan 2022 (peak) 4,040,656 12,479,589 3.09x 6.16x 2,025,893
Jan 2023 (cleanup) 4,563,186 10,363,079 2.27x 4.14x 2,505,512
Dec 2024 (reserve drop) 5,817,668 8,863,173 1.52x 3.57x 2,485,375
Jul 2026 (current) 8,020,498 11,158,666 1.39x 3.57x 3,129,561

Despite account numbers growing from 4M to 8M between Jan 2022 and Jul 2026 — a doubling — the average owned objects per account has fallen from 3.09x to 1.39x. The network has twice as many accounts but proportionally far less active usage per account. Accounts are being created but not actively utilised.

Interpretation: The data indicates that XRPL account creation significantly outpaces genuine network utilisation. The average account holds fewer than 1.4 owned objects. Among the minority of accounts that hold any objects at all — 3.1M of 8M — the average is 3.57 objects. The remaining 4.9M accounts exist as bare-minimum reserve holders with no active on-chain presence beyond the AccountRoot itself.

5. Reserve Impact on Existing Accounts

Owner Count Distribution — 8,020,650 Active Accounts
OwnerCount field read directly from each AccountRoot in validated ledger state · log scale
Zero objects
1–19 objects
20+ objects
82.9% — 6,647,022 accounts hold ZERO objects · base reserve only 8.6% — 688,762 accounts hold exactly 1 object 94.8% of accounts hold 2 or fewer objects 0.77% — 62,127 accounts hold 20+ objects · power users only
Accounts Stranded at Each Reserve Level
Active accounts that would be unable to transact · measured against current balances and OwnerCount
422,439 already stranded at current 1 XRP reserve (5.3%) 1.0→1.1 XRP strands 1.18M additional accounts instantly 1.1→10.0 XRP only adds 0.71M more — damage is front-loaded at the floor The 1 XRP level is an extreme density point — 1.39M accounts at exactly 1.0–1.1 XRP
Pre-Drop Account Balance Distribution (0–10 XRP)
Accounts that existed before the Dec 2024 reserve drop · current balance in 1 XRP bands · shows the cliff at the reserve floor
728,946 accounts sit at 1–2 XRP — the bare reserve floor with zero buffer 142,108 accounts at 9–10 XRP — withdrew to just below the old reserve 112,638 accounts below 1 XRP — already stranded under current rules The floor has become the target — users treat 1 XRP as their operating balance
Risk Profile — Pre-Drop vs Post-Drop Accounts
Accounts created before vs after the Dec 2 2024 reserve drop · by current balance exposure to a return to 10 XRP
Pre-drop (5.34M accounts)
Post-drop (2.68M accounts)
Post-drop accounts: ~50% hold 1–10 XRP · they have only ever known the 1 XRP reserve Pre-drop accounts: 18.3% at risk · chose to draw balances to the new floor 2.68M accounts post-drop — a third of the entire network created under 1 XRP regime
XRP Locked in Reserves vs Available to Spend
Measured directly from Balance and OwnerCount across all 8,020,650 AccountRoot objects
Available to spend
67,530,966,017
XRP · 99.986% of total
Locked in reserves
9,638,888
XRP · 0.014% of total
Proportion of total XRP in accounts
99.986% — Available
◀ 0.014% locked
The entire reserve system locks less XRP than a single mid-size exchange holds in a hot wallet
67,530,966,017 XRP — 99.986% — is freely available to spend 9,638,888 XRP — 0.014% — is locked in base and owner reserves The reserve system locks a negligible fraction of circulating supply

The following analysis uses direct ledger state measurements from all 8,020,650 AccountRoot objects, reading Balance and OwnerCount fields to compute exact available balance for each account.

Already stranded (balance < reserve)
422,439
5.3% of active accounts · cannot transact today
Of which: fee erosion (0.9–1.0 XRP)
201,559
93.98% of stranded · nibbled below floor by fees
At risk if reserve raised to 1.1 XRP
1,604,098
20% of all active accounts · 0.1 XRP increase

The sensitivity of the account base to reserve changes is extreme. A 0.1 XRP increase — from 1.0 to 1.1 XRP — would immediately strand 1.6 million accounts, representing 20% of all active accounts. This concentration reflects the mass adoption of the 1 XRP floor as an operating norm.

Reserve level Accounts stranded % of active accounts Change vs current
1.0 XRP (current) 422,439 5.3% baseline
1.1 XRP 1,604,098 20.0% +1,181,659
1.2 XRP 1,720,947 21.5% +1,298,508
2.0 XRP 2,094,283 26.1% +1,671,844
5.0 XRP 2,290,715 28.6% +1,868,276
10.0 XRP (pre-Dec 2024) 2,316,555 28.9% +1,894,116

Notable: The stranding curve is front-loaded. The jump from 1.0 to 1.1 XRP strands 1.18M additional accounts — nearly as many as the entire remaining increase from 1.1 to 10.0 XRP combined (1.18M vs 0.71M). The 1 XRP level is an extremely dense concentration point.

6. Pre-Drop Accounts — Historical Context

Accounts created before the December 2024 reserve drop from 10 to 1 XRP experienced two different reserve environments. Of these pre-drop accounts still active today:

Pre-drop accounts still active
5,339,703
Created before ledger 92,508,417
At risk from return to 10 XRP
976,819
18.3% · balance 1–10 XRP
Active in last 12 months
427,144
39% of at-risk pre-drop accounts

427,144 pre-drop accounts — accounts that operated under a 10 XRP reserve and chose to draw their balance below that level after the drop — have been active within the last 12 months. These are not dormant wallets. They are active users who have normalised to the 1 XRP reserve floor. A return to 10 XRP would strand them without warning.

8. State Objects as a Measure of Adoption — The Fundamental Contradiction

The argument that ledger state growth justifies higher reserves contains a fundamental internal contradiction that the data makes explicit.

Ledger state objects are not a problem to be minimised. They are the evidence of network use. A trust line represents a user holding a token. An offer represents a user participating in the DEX. An escrow represents a user settling a payment. An NFT represents a user engaging with digital ownership. Every owned object in the XRPL ledger is the on-chain record of a user doing something with the network.

By this measure, the current data presents a clear picture of where the XRPL stands:

82.9% of XRPL accounts — 6,647,022 of 8,020,650 — hold zero owned objects. These accounts exist. They pay a reserve. They do nothing. They are not using the network for any purpose beyond maintaining an address. This is not a sign of a thriving ecosystem. It is a sign of adoption that has stalled at the account-creation stage and has not progressed to actual use.

The owned-objects-per-account ratio tells the same story over time. In January 2022, when the XRPL was experiencing genuine speculative activity, the ratio reached 3.09 objects per account across all accounts, and 6.16 objects per account among those actively holding anything. By July 2026, with twice as many accounts, the overall ratio has fallen to 1.39x — and 82.9% of accounts contribute zero to that figure.

The network has grown in account count. It has not grown proportionally in use.

The policy contradiction

If state object growth is the concern, the correct response is to encourage the kind of network activity that generates state objects — not to raise barriers to account creation. The two goals are in direct opposition:

High reserve argument
"State object growth is dangerous. Keep reserves high to deter new accounts."

Result: Fewer accounts. Fewer users. Fewer trust lines, offers, and escrows. Lower state object counts. The network shrinks, not grows. The stated concern — state bloat — does not materialise, but neither does adoption.
Lower reserve argument
"Adoption requires accessible entry. Lower the barrier. Let the network prove its value."

Result: More accounts. More potential users. If the network has genuine utility, those accounts become active — generating the trust lines, offers, and payment channels that represent real adoption. State objects grow because the network is being used.

The data shows that even at 1 XRP — the current reserve — the overwhelming majority of accounts are not generating state objects. The barrier is not the only explanation for this, but it is a contributing factor. Users who cannot afford to experiment with trust lines and DEX offers because their balance is already at the reserve floor are users who cannot participate in the network's core functions.

What the evidence actually shows

Proponents of higher reserves often cite the 2021 trust line explosion as evidence of what happens when activity increases. But that event demonstrates precisely the opposite of what is claimed:

The claim
Reserve drops cause state bloat. The 2021 explosion was caused by the September 2021 reserve drop from 20 to 10 XRP.
The evidence
The explosion began on August 20, 2021 — 27 days before the reserve drop. 10,099 net new trust lines were created on that single day. By September 16 when the reserve dropped, trust line creation was already at 21,307 per day and accelerating. The reserve change did not initiate, accelerate, or sustain the explosion. It was irrelevant to the event.
The resolution
The cleanup required no policy intervention. Trust lines peaked in January 2022 and declined organically as the speculative activity subsided. The market corrected itself. On December 24 2021 alone, 77,143 net trust lines were deleted — not because of any reserve change, but because users chose to close them.

The 2021 event was not a failure of reserve policy — it was the XRPL being used. State objects were created because users were doing things. That is the correct outcome. The fact that it subsequently reversed is evidence that the ledger self-regulates through user behaviour, not through reserve levels.

The adoption measurement problem

If the goal is XRPL adoption, then state objects are the metric of success — not the metric of risk. A network where users hold tokens, trade on the DEX, use payment channels, and engage with NFTs will have more state objects than a network where users merely hold addresses. The former is a healthy network. The latter is the current situation.

Using the potential for state object growth as a justification for keeping account creation expensive is equivalent to arguing that a city should not build new roads because roads attract traffic. The traffic is the point.

The contradiction in a single observation

The XRPL has 8 million accounts. 82.9% hold no owned objects. 422,439 are already stranded under the current reserve. 2.3 million would be stranded if the reserve returned to 10 XRP. The network is a ghost town by every object-based measure.

Arguing to keep the reserve high — let alone raise it — to prevent state over-utilisation is arguing to lock the doors of an empty building. The evidence does not support the concern. It supports the opposite conclusion: the reserve should be reduced to 0.1 XRP to give the network a genuine chance at adoption.

9. The Case for 0.1 XRP

The current data provides affirmative support for a further reduction to 0.1 XRP:

State cost is not a constraint. At 1 KB per account, one million new accounts costs 800 MB. Even 100 million accounts — far beyond any plausible near-term scenario — would require approximately 100 GB of additional state storage. This is not a meaningful barrier for any serious infrastructure operator.

Network utilisation is low. 82.9% of accounts hold zero owned objects. The owned-objects-per-account ratio has fallen from a 2022 peak of 3.09x to 1.39x despite account numbers doubling. The network has capacity — it is not being heavily used.

Historical bloat events were application-driven. The 2021 trust line explosion — the largest state growth event in XRPL history — was caused by speculative token issuance and began four weeks before the reserve drop that is often cited as its cause. It self-corrected without policy intervention.

The current reserve harms active users. 422,439 accounts are already stranded under the current 1 XRP reserve. Any increase strands significantly more. The reserve's protective function against state abuse is not supported by evidence; its harmful effect on existing users is directly measured.

0.1 XRP still creates a cost barrier. At current XRP prices, 0.1 XRP is approximately USD $0.02. This is not zero. It remains a non-trivial cost for mass automated account creation at scale, while being negligible for legitimate individual users. The anti-spam function is preserved.

Conclusions

The state bloat argument against reserve reduction is not supported by 13 years of XRPL ledger data. State growth correlates with application-layer activity, not with reserve levels.
The largest state growth event in XRPL history — the 2021 trust line explosion — began four weeks before the reserve drop that is cited as its cause, and self-corrected without policy intervention.
82.9% of XRPL accounts hold zero owned objects. The network is significantly underutilised relative to account count. The reserve is not preventing state abuse — it is preventing user access.
422,439 accounts are already stranded under the current 1 XRP reserve. A return to 10 XRP would strand 2.3 million active accounts, 94% of whom have been active in the last 24 months.
A reduction to 0.1 XRP adds approximately 800 MB of state per million new accounts — a trivial infrastructure cost — while removing a barrier that demonstrably harms existing network participants.
The owned-objects-per-account ratio, falling from 3.09x in January 2022 to 1.39x in July 2026 despite account numbers doubling, indicates the network is becoming proportionally less active per account over time. Lower barriers to entry are warranted, not higher ones.

Data sources: ClickHouse full-history database at analytics.xrplwin.com (daily_object_metrics, transactions, accounts, activations tables) · Direct ledger_data iteration via xrplcluster.com public RPC and local rippled node · All figures measured from validated ledger state, not estimated. Ledger scan conducted July 2026 across all 8,020,650 AccountRoot objects reading Balance and OwnerCount fields directly.

Published by: onledger.net · July 2026