Only 10 out of 83 cryptocurrency exchanges, analyzed by crypto investment company Bitwise Asset Management, passed a test aimed to show if they were reporting systematically inflated bitcoin trading volumes.
The 10 that passed, in the order of their average daily volume during April 2019 from the highest to the lowest, are:
Exchange (Daily volume in million USD):
South Korean exchanges once again were excluded from the analysis, as Bitwise explains, due to their volumes being isolated from the globally connected bitcoin market because of the capital controls.
The authors of the analysis aimed to identify the exchanges that were reporting systematically inflated volumes, that is, any reported trading volume that does not reflect legitimate price found in the market. They have put the exchanges through three tests:
- time period of visualized data (one week in this case was found to be the most suitable for visualizing data to detect anomalous or artificial patterns),
- trade size histograms (a data visualization technique that allows people to see the percentage of trading volume on an exchange that occurs at particular trade sizes over a specified period, as defined by the paper),
- and volume spike alignment (given the global nature of bitcoin, all exchanges should respond to the same developments, and charts should demonstrate similar patterns).
The paper, which is addition to the research that Bitwise presented to the U.S. Securities and Exchange Commission in March, 2019, confirms their earlier findings that about 95% of all reported volume, or to be more precise, USD 10.5 billion out of the USD 11 billion in reported average daily spot bitcoin volume, is either fake volume or wash-trading.
In the paper presenting these findings, published on May 24th 2019, titled “Economic and Non-Economic Trading In Bitcoin: Exploring the Real Spot Market For The World’s First Digital Commodity”, authors claim that public perception of bitcoin’s spot trading market as being disorderly, inefficient, “with wildly variant pricing and strange volume patterns” is wrong. They find that “the modern bitcoin spot market is both significantly smaller and significantly more efficient than commonly understood”, and that “effective arbitrage keeps prices on real bitcoin spot exchanges around the world in lockstep, with meaningful pricing discrepancies eliminated in a matter of seconds,” especially given the massive changes after December 2017.
Methods and motivations for exaggerating volume
There are quite a few ways in which exchanges can inflate volume, and have so far suffered little to no consequences for doing so, including but not limited to practices such as:
- fraudulent printing (printing more volume to their tape; using algorithms that post trades that never happened);
- exchange-level wash trading by placing orders (doing wash trading on their own platforms, simultaneously buying and selling a single asset with itself or an affiliated party);
- paid third-party wash trading (paying market makers to engage in wash trading);
- trade mining (happens when exchange pays traders to trade, typically in an exchange-specific coin, that way economically incentivizing trading activity),
- and fee-tier wash trading (offering benefits such as lower fee tiers or preferential trading to traders that attain high volumes of trades).
Reasons to inflate volume are usually twofold, according to the paper’s authors, the first being trader lead-generation through league table dominance – major data aggregators, such as CoinMarketCap, are an important source of lead-generation for new exchanges that want to build volume, as they are more likely to attract attention if they are at the top of popular lists. The second motivator to exaggerate volume is listing fees from ICOs – to attract listing fees from initial coin offerings.
Gray area of exchanges
After one of the comments to their previous study pointed to a probable gray area, the paper too states that there likely is a gray area between exchanges with 100% real volume and those with 100% fake volume. The authors took Gate.io as an example, stating: “We could reasonably argue with someone about whether Gate.io is a fake volume exchange or not, or whether some percentage estimate of its volume should be applied to our total volume statistics.” They add: “The reality is, however, that Gate.io does not have enough volume for this question to meaningfully alter our conclusions: Gate.io reported [USD] 12 million in average daily volume in the month of April where the total daily volume of the 10 real volume exchanges was [USD] 554 million. Even if we counted all of Gate.io’s volume as real, none of our analysis would change.”
In an interview with Cryptonews.com in April, Gate.io claimed that their “volume is all real, and right now we’re averaging about USD 50 million per day.” The company also said that they’re certified by Blockchain Transparency Institute, a group of blockchain data researchers and enthusiasts, that is known for their reports on wash trading (however, the website of the project does not disclose who the members of this for profit group are.)