Unveiling Wash Trading in Popular NFT Markets
Yuanzheng Niu, Xiaoqi Li, Hongli Peng, Wenkai Li
TL;DR
The paper tackles wash trading in NFT markets by performing a large-scale, cross-market analysis and introducing a graph-based wash-trading detection algorithm. It combines transaction-network topology with behavioral signals to identify illicit patterns and demonstrates that incentive-bearing markets (notably LooksRare and X2Y2) exhibit markedly higher wash trading, up to 94.5% of ETH volume in some cases. The findings highlight how reward structures can distort market signals and raise concerns about long-term sustainability, while offering a scalable detection framework and open data/code for replication. This work informs policymakers and market designers about the effects of incentives on market integrity and provides a practical method for ongoing surveillance of NFT ecosystems.
Abstract
As emerging digital assets, NFTs are susceptible to anomalous trading behaviors due to the lack of stringent regulatory mechanisms, potentially causing economic losses. In this paper, we conduct the first systematic analysis of four non-fungible tokens (NFT) markets. Specifically, we analyze more than 25 million transactions within these markets, to explore the evolution of wash trade activities. Furthermore, we propose a heuristic algorithm that integrates the network characteristics of transactions with behavioral analysis, to detect wash trading activities in NFT markets. Our findings indicate that NFT markets with incentivized structures exhibit higher proportions of wash trading volume compared to those without incentives. Notably, the LooksRare and X2Y2 markets are detected with wash trading volume proportions as high as 94.5% and 84.2%, respectively.