The Economics of Wash Trading is my solo-authored research paper on manipulation in non-fungible token (NFT) markets, written in October 2023 and available on SSRN as abstract 4610162. Across 85 pages it asks a blunt question: when someone trades an asset with themselves to print volume, what are they actually buying, and does it work?
The question
NFT markets in 2021 and 2022 reported extraordinary trading volumes, and a meaningful share of those prints never involved two independent people. A wash trade is a transaction in which the buyer and the seller are, economically, the same actor. On a public blockchain these trades sit in plain sight once you connect the wallets, which makes NFT markets an unusually clean laboratory for studying a manipulation that is as old as organised exchanges.
There are two candidate motives, and they imply different economics:
- Advertising.Wash trades inflate a collection's apparent volume and price history to make it look liquid and in demand, hoping real buyers follow the signal.
- Incentive farming. Some marketplaces paid traders in their own tokens in proportion to trading volume. Self-trading then becomes a direct claim on token emissions, no victim required at the moment of the trade.
The paper measures which motive dominates, and what wash trading does to the variables a manipulator would want to move: real volumes and average prices in subsequent days.
What the data shows
The findings, in the order the paper builds them:
- There is a nuanced interplay between wash trading volumes, prices, and subsequent genuine activity, but no significant relationship between wash trading volumes and real trading volumes on future days. As advertising, fake volume largely fails.
- Wash traders predominantly exploit token-based incentives on the exchanges that offer them. Where volume is directly rewarded, as on LooksRare, wash trading concentrates, and the rewards appear to influence real trading activity on those venues.
- The results support reading most NFT wash trading as rational farming of poorly designed incentives rather than as successful demand manufacturing, which is exactly where policy and marketplace design should aim.
The interpretation matters for how seriously one takes headline NFT statistics. Reported volume is not a measurement of demand; it is a measurement of what the fee and reward structure made profitable to print. I walk through the scale estimates on the NFT wash trading page and the market plumbing on the NFT markets page.
Method, in brief
The study identifies suspicious activity from public blockchain data, then applies statistical tests to the relationship between wash volumes, real volumes, and prices at the collection level over time. The design addresses the obvious confound that active collections attract both genuine and fake volume, and it exploits differences between marketplaces with and without token rewards as a source of identification. If you work on wash trading detection, the methodological sections describe the filters and their failure modes.
Abstract
In this study, I conduct a comprehensive examination of the efficiency of non-fungible token (NFT) markets by analyzing the intricate relationship between wash trading activities, real trading volumes, and prices. Employing rigorous statistical methods, the research investigates the financial motivations of wash traders, assessing whether such practices are employed to increase the attractiveness of a collection or to capitalize on token-based incentives provided by specific NFT exchanges. My findings reveal a nuanced interplay between wash trading volumes, prices, and their subsequent effects on genuine trading volumes and average prices, with no significant relationship observed between wash trading volumes and real trading volumes in future days. Furthermore, the results demonstrate that wash traders predominantly exploit token-based incentives on exchanges, such as LooksRare, where wash trading rewards appear to influence real trading activity. This research contributes to a deeper understanding of NFT market dynamics and the role of wash trading, providing valuable insights that can inform policy discussions and aid in the development of a more transparent and efficient NFT ecosystem.
Download the full paper on SSRN ↗ or use the permanent DOI link: dx.doi.org/10.2139/ssrn.4610162. The paper is also indexed on ResearchGate and Google Scholar, which tracks the live citation list.
The talk
I presented the research at EthCC 2023 in Paris as Advait Jayant - The Economics of Wash Trading in the NFT Markets. The talk is a 20-minute version of the argument: how wash trades are identified on-chain, why token incentives rather than buyer deception explain the volume, and what that means for how seriously to take NFT market statistics.
▶ Watch the EthCC talk on YouTube ↗
Where the paper sits in the literature
Two strands of prior work frame the study. On centralised crypto exchanges, Cong, Li, Tang and Yang's Crypto Wash Trading found that the majority of reported volume on many unregulated venues was wash traded. On NFTs specifically, von Wachter, Jensen, Regner and Ross quantified suspicious trading behaviour at the transaction level. The Economics of Wash Trading extends this line by asking not just how much wash trading exists but what it is economically for, and by measuring whether it achieves anything its practitioners might want. The legal backdrop, from the Commodity Exchange Act to MiCA, is covered on the legality page.