In:
Management Science, Institute for Operations Research and the Management Sciences (INFORMS)
Abstract:
We present the first systematic approach to detect fake transactions on cryptocurrency exchanges by exploiting robust statistical and behavioral regularities associated with authentic trading. Our sample consists of 29 centralized exchanges, among which the regulated ones feature transaction patterns consistently observed in financial markets and nature. In contrast, unregulated exchanges display abnormal first significant digit distributions, size rounding, and transaction tail distributions, indicating widespread manipulation unlikely driven by a specific trading strategy or exchange heterogeneity. We then quantify the wash trading on each unregulated exchange, which averaged more than 70% of the reported volume. We further document how these fabricated volumes (trillions of dollars annually) improve exchange ranking, temporarily distort prices, and relate to exchange characteristics (e.g., age and user base), market conditions, and regulation. Overall, our study cautions against potential market manipulations on centralized crypto exchanges with concentrated power and limited disclosure requirements and highlights the importance of fintech regulation. This paper was accepted by David Simchi-Levi, special issue of Management Science: Blockchains and crypto economics. Funding: This research was partly funded by the Ewing Marion Kauffman Foundation [Grant G-201907-6995], the National Natural Science Foundation of China [Grants 72192802, 72192800, and 72192801] , Ripple’s University Blockchain Research Initiative (UBRI), and the FinTech at Cornell Initiative. Supplemental Material: The online appendix and data are available at https://doi.org/10.1287/mnsc.2021.02709 .
Type of Medium:
Online Resource
ISSN:
0025-1909
,
1526-5501
DOI:
10.1287/mnsc.2021.02709
Language:
English
Publisher:
Institute for Operations Research and the Management Sciences (INFORMS)
Publication Date:
2023
detail.hit.zdb_id:
206345-1
detail.hit.zdb_id:
2023019-9
SSG:
3,2
