Further hints of fake volume on major cryptocurrency exchanges
Recent analysis of crypto-currency exchange order book data showed evidence of wash-trading via unseemly large price slippage in comparison to published volume. Here the Merkle Team shows how Intra-Exchange Velocity (IEV), which measures the average number of times an asset changes hands inside an exchange can provide an additional method to benchmark and investigate crypto-currency exchange behavior. IEV is derived from exchange published volume and blockchain level analysis to total token holdings of users inside an exchange.
An unregulated market that traffics in speculative assets is bound to attract questionable actors and practices. Among crypto exchanges, one of the tricks frequently deployed is the faking of transaction volume via wash trading. Such behavior is damaging to the functioning of a healthy financial system and leads to an environment unsafe for investors, leading to potential stagnation of the ecosystem.
Though illegal in regulated markets, wash trading is not uncommon among crypto exchanges. Writing for CryptoSlate.com, Joseph Young reports that OKCoin, Huobi, BitForex and others are widely suspected of deploying bots that trade among each other to inflate transaction volumes. The incentive for an exchange to fake its transaction volume is obvious. Exchanges benefit from a feedback loop such that more users result in more trades, which yields more liquidity and strengthens the marketplace — thereby attracting more users, rinse and repeat. For startup exchanges, this feedback loop is nonexistent; trapped in a chicken and egg dilemma, they fake their volume to attract new users, and hope those users will stick around long enough to create a legitimate marketplace.
Beyond Young, others have called attention to this issue. In a Medium article published in March of 2018, crypto trader Sylvain Ribes shares the findings of a study he did on various crypto exchanges. The headline: “more than $3 billion of all cryptoassets’ volume [is probably] fabricated.”
To conduct this study, Ribes measured the effect that market selling $50,000 worth of each cryptocurrency would have on its price, dubbing this metric the “slippage” of a coin on a given exchange. Ribes hypothesized that, on balance, slippage would decrease as volume increased on an exchange. Since elevated transaction volume should provide enhanced liquidity, the greater the volume on an exchange the more easily the market could absorb a large order, so he reasoned.
However, the results did not confirm his hypothesis. As Ribes writes, “where I had expected mild differences between currencies, I [instead] found massive discrepancies between exchanges.” He presents the example of OKex, which at the time was ranked #1 by volume ($1.7 billion daily) but produced disproportionality deep slippage on large market orders, so much so that Ribes estimated its transaction volume was inflated by as much as 95%. Other culprits include HitBTC and Binance.
Ribes analysis relied on measuring price movements from large trades to estimate slippage. Here we propose an additional metric Using on-chain analytics, an alternative passive metric such as Intra-Exchange Velocity (IEV) can be computed continuously and across all asset classes providing similar insight. IEV captures the movement of assets inside an exchange, and is defined as the average number of times a token is traded on an exchange. Specifically, it is the ratio of total traded token volume across all trading pairs by the token balance on the exchange. To illustrate this, we compared the IEV of ETH on different exchanges.
Looking at IEV from September, 2015 to August, 2018. We compared token velocity across several exchanges, including Exchange-1, Exchange-2, Exchange-3 , Exchange-4, and Exchange-5. Tokens on Exchange-4 and Exchange-5 routinely trade at velocities that are approximately 15 times higher than the averages on other exchanges like Exchange-2, and Exchange-3. We note that the token velocities on Exchange-5 exploded suddenly in late 2017, and have continued to significantly outpace the rest of the market. In the case of Binance, token velocities outperformed from September, 2017 to January, 2018, but have since subsided somewhat. At its peak, Exchange-2 token velocity is around 20% while that of Exchange-4 and Exchange-5 is around 280%, which means that if none of the volume is fake, each token gets traded about an order of magnitude more times more on Exchange-4 and Exchange-5 than on Exchange-2.
This statistic by itself does not prove fake volume on Exchange-4 and Exchange-5, and we will not draw such a strong conclusion. Any number of factors can cause higher velocities on these exchanges, including lower transaction fees or just innate popularity due to various factors. However, it does hint at orders of magnitude higher activity, especially in these more recent months, which is quite suspect given that we are in a crypto-winter.
Nonetheless, we consider wash trading and volume faking to be a problematic feature of crypto markets. We caution traders and investors against using exchanges that report fake volumes. For one, tokens with inflated velocities deceive buyers into thinking their markets are more liquid than they actually are. It will be difficult to get market prices on illiquid tokens in the event of a sudden evacuation from the market. Moreover, if an exchange is willing to fake its volumes, it forces us to question what other unethical practices the exchange may conduct.
Taking a step back for a second, we should point out that unlike in traditional markets, information about brokerage accounts on all investing and trading entities is never known. However in crypto-currencies, this information can be reconstructed by intelligent analytics on the public blockchain. So although much of the crypto-world is unregulated, we believe that on chain insights can bring a level of transparency to the ecosystem and lead to a healthy ecosystem. At Merkle Data, our mission is to be the world leader for on-chain analytics and are committed to bringing transparency to the cryptocurrency and virtual asset ecosystem.
