Exchange wash trading
Exchange wash trading is a trading venue reporting volume that did not genuinely occur, or permitting others to generate it, in order to appear more liquid and rank higher than it is.
How does exchange wash trading work?
The technique is ordinary wash trading. What makes it a distinct entry on this site is who benefits and what verifies it.
In a regulated market, a venue’s volume is a fact reported to a regulator, audited, and reconstructible from a complete order audit trail. In much of the digital asset market, a venue’s volume is a number the venue publishes about itself, with no independent verification of any kind.
That changes the incentive completely. Reported volume drives rankings on aggregator sites. Rankings drive listings, because token projects choose venues by apparent reach. Listings drive fees. And traders route orders to where they believe liquidity is.
Volume, in other words, is not a measurement of the business. It is the marketing.
There are three ways a venue produces it.
Directly. Internal accounts trading against each other. Cheapest, most complete control, and most straightforward to characterise if discovered.
By invitation. Permitting or encouraging market makers and large participants to generate volume, sometimes as a condition of favourable terms.
Structurally. Setting a fee schedule where the maker rebate exceeds the taker fee, so that round trips are costless or profitable. Nobody needs to be asked; the participants do it because the venue pays them to. This is the most elegant version, because the venue can truthfully say it did not instruct anyone.
A worked example with real numbers
A venue reports $1.9 billion of daily volume in a token pair, which would rank it among the largest markets for that asset.
What can be observed independently.
| Indicator | Observed | What $1.9bn of genuine volume implies |
|---|---|---|
| Order book depth within 1% of mid | $340,000 | Tens of millions |
| Daily net deposit and withdrawal flow | $6.2m | Hundreds of millions |
| On-chain settlement attributable to the venue | $11m | Far higher |
| Volume change during a 30% market drawdown | −4% | Genuine venues saw +200% to +400% |
The last row is the most informative. Genuine markets convulse under stress: volume spikes, spreads widen, depth evaporates. A venue whose reported volume barely moved during a violent drawdown is not reporting a market’s behaviour.
The economics from the venue’s side.
Reported volume $1,900,000,000
Genuine volume (estimated) ≈ $45,000,000
Fee revenue on genuine volume at 10bp ≈ $45,000
Listing fees attributable to ranking ≈ $180,000/day equivalentThe venue earns little from the fake volume itself — round trips at zero net fee generate no revenue. It earns from what the ranking buys: listing fees from projects that want access to an apparently large market, and genuine order flow from traders who believed the statistics.
Who pays. A trader routing a $2 million order to this venue expecting deep liquidity encounters $340,000 of depth and moves the price several per cent against themselves. That slippage is the real cost of the false statistic, and it is borne by the person who trusted it.
Why is exchange wash trading unlawful?
Where the venue’s products fall within a regulator’s perimeter, the analysis is settled. The Commodity Exchange Act prohibits wash sales, accommodation trades and fictitious sales by name, and CFTC Rule 180.1 reaches manipulative and deceptive devices in commodity markets. The CFTC has brought actions against digital asset platforms on exactly this basis. Where a token is a security, Exchange Act § 9(a)(1) and Rule 10b-5 apply.
Wire fraud does not depend on any of that. Publishing false trading statistics to attract customers and listing fees is a scheme to obtain money by materially false pretences over interstate wires, whatever the asset is. This is the charge that reaches venues outside every other perimeter.
Where the venue is entirely offshore, the conduct is plainly deceptive and the enforcement question is jurisdictional rather than substantive. A venue incorporated in a jurisdiction with no market regulator, serving customers everywhere, publishing statistics nobody audits, occupies a gap that no amount of doctrinal clarity closes.
Two consequences deserve emphasis.
The first is that prices from a fake market are not prices. Volume-weighted references, index calculations and oracle feeds that ingest data from venues with inflated volume are propagating a number that does not describe anything. This is how venue-level wash trading connects to oracle manipulation and to derivatives pricing generally.
The second is that the measurement problem is real and unresolved. Studies of this question produce widely varying estimates depending on method, and any single confident figure about “how much crypto volume is fake” should be treated with suspicion — including figures cited approvingly by people making the case that the practice is widespread.
| Provision | Citation | Primary text |
|---|---|---|
| CFTC Rule 180.1 — fraud-based manipulation | 17 C.F.R. § 180.1 | Read the text |
| Commodity Exchange Act — wash sales prohibition | 7 U.S.C. § 6c(a)(2) | Read the text |
| SEC Rule 10b-5 | 17 C.F.R. § 240.10b-5 | Read the text |
| Wire fraud | 18 U.S.C. § 1343 | Read the text |
Which real enforcement actions have alleged exchange wash trading?
This library holds 3 enforcement actions tagged exchange wash trading. The table shows the largest by civil penalty together with the most recently filed. Every row links to a page carrying the regulator's own release and, where one was published, the complaint.
| Action | Agency | Filed | Penalty | Status |
|---|---|---|---|---|
| CFTC v. unnamed respondents (cash vs derivatives schemes, 2022) | CFTC | 2022-10-20 | $41m | judgment |
| CFTC v. Coinbase Inc. (exchange wash trading, 2021) | CFTC | 2021-03-19 | $6.5m | judgment |
| CFTC v. FTX (exchange wash trading, 2024) | CFTC | 2024-12-04 | — | judgment |
How does exchange wash trading get detected?
Detection is done from outside, by reconciliation, because the venue’s own data cannot be trusted.
Depth reconciliation. Reported volume against observable order book depth. A venue reporting enormous turnover through a book that could not absorb a moderate order is reporting something other than trading.
Flow reconciliation. Deposits, withdrawals and on-chain settlement attributable to the venue. Genuine trading requires assets to arrive and leave; volume without flow is volume without customers.
Stress-response analysis. How reported volume behaved during periods when every genuine market convulsed. This is the hardest signal to fake, because it requires anticipating market-wide events.
Trade size distribution. Genuine trading shows clustering at round numbers, because humans and algorithms both use round sizes. Generated volume designed to look random frequently lacks these artefacts, and the deviation is measurable.
Fee schedule analysis. Whether the venue’s own pricing makes round trips costless or profitable. A venue that pays participants to trade with themselves has arranged the outcome without instructing anyone.
- Reported volume that cannot be reconciled with order book depth, withdrawal flows or on-chain settlement.
- Trade size distributions that lack the rounding artefacts human and algorithmic trading both produce.
- Volume that does not vary through periods when every genuine market experienced stress.
- Fee schedules that make round trips costless or profitable, removing any economic barrier to the practice.
- Internal accounts operated by the venue itself appearing on both sides of trades.
What penalties does exchange wash trading actually attract?
The numbers below are computed from this site's own case records at build time, not quoted from a secondary source. They change whenever a new action is added to the library.
- Actions recorded
- 3
- Median penalty
- $23.8m
- Largest penalty
- $41m
- Criminal parallel
- 0%
- Median sentence
- —
What are the red flags?
- A venue ranking highly by reported volume with visibly thin order books.
- Reported turnover that is large relative to the venue's total customer assets.
- Volume statistics published by the venue with no independent verification.
- Spreads that stay tight during periods when deeper markets widened.
The check available to any trader in thirty seconds: look at the order book, not the volume figure. A venue reporting billions with a few hundred thousand dollars of depth within one per cent of the mid has answered the question. Depth is much harder to fake than volume, because faking it means standing behind orders that can be hit.
What exchange wash trading is not
It is not market making. Genuine market makers take real risk on both sides and end sessions with real positions.
It is not high turnover. Some venues genuinely trade a great deal.
It is not incentive programmes as such. Rebates and maker incentives are used by regulated exchanges too. The question is whether the schedule makes self-trading profitable in itself.
It is not proof about any particular venue. The indicators above are inferential. They identify inconsistency between claims and observations, which is a reason to investigate rather than a finding.
Frequently asked questions about exchange wash trading
- Why would an exchange inflate its own volume?
- Because rankings drive listings and users. Aggregators rank venues by reported volume, token projects choose where to list based on those rankings, and traders go where they believe liquidity is. Volume is the venue's marketing.
- Does the exchange do the trading itself?
- Sometimes directly through internal accounts, sometimes by permitting or encouraging others, and sometimes through fee schedules that make round trips free or profitable so that participants generate the volume without being asked.
- How much of reported crypto volume is affected?
- Multiple academic studies have found a substantial proportion of reported volume on some unregulated venues to be inconsistent with observable indicators. Estimates vary widely and depend heavily on method, so no single figure should be quoted with confidence.
- How can it be measured from outside?
- By reconciling reported volume against things harder to fake: order book depth, withdrawal and deposit flows, on-chain settlement, and the statistical fingerprint of genuine trade sizes, which shows rounding artefacts that generated volume typically lacks.
- What is the trade size fingerprint?
- Genuine trading produces characteristic clustering at round numbers, because humans and algorithms both use round sizes. Machine-generated volume designed to look random often lacks that clustering, which is detectable statistically.
- Is this illegal?
- Where the venue offers products within a regulator's perimeter, yes — the CFTC has brought actions against digital asset platforms for wash trading. Where a venue is entirely offshore, the conduct is plainly deceptive but enforcement is a jurisdictional problem.
- Who is harmed?
- Traders who route orders to a venue believing it is liquid and receive poor execution; token projects that pay for listings based on false statistics; and anyone using the reported prices, since prices from a fake market are not prices.
- Does this happen in traditional markets?
- Wash trading does, but venue-level volume inflation does not, because regulated exchanges report to regulators, are audited, and operate under a complete order audit trail. The absence of those three things is what makes this a crypto-specific problem.
What techniques are related to exchange wash trading?
Terms defined on this page
Sources
- CFTC Rule 180.1 — Electronic Code of Federal Regulations
- Commodity Exchange Act § 4c — prohibited transactions — Cornell Legal Information Institute
- CFTC — digital assets — Commodity Futures Trading Commission