A New Wolf in Town? Pump-and-Dump Manipulation in Cryptocurrency Markets

Pump-and-Dump (P&D) schemes to manipulate the prices of cryptocurrencies are unlike the P&D schemes found in the equity market.  They produce very large price distortions on the order of 65%, very large trading volumes of 13.5x the average, and generate very large profits to cryptocurrency manipulators.  They target illiquid coins but only have temporary, short-term impact on prices.  In contrast to equities, the authors of this article argue P&Ds in the crypto market are more like gambling, where the players attempt to be the first to buy a pumped-up currency and then sell before the price of the currency implodes. In sum, while P&D crypto schemes are short-term in nature, they do generate extremely large trading volumes as well as extreme distortions in price.

A New Wolf in Town? Pump-and-Dump Manipulation in Cryptocurrency Markets

• Anirudh Dhawan and Talis J. Putnin
• Review of Finance
• A version of this paper can be found here
• Want to read our summaries of academic finance papers? Check out our Academic Research Insight category.

What are the research questions?

The analysis was conducted using data from two cryptocurrency exchanges (Binance and Yobit), plus the chat history from 72 Telegram pump-and-dump channels. Data included the prespecified date, time and exchange of each pump signal, and name of the pumped coin, number of groups participating, and number of previous pumps conducted by each group. This data set guaranteed the pump group administrator made an explicit communication that the pump was intended.

1. Why are the characteristics of investors who participate in pump-and-dump crypto schemes?
2. What are the patterns of coin prices and trading volume in a typical P&D scheme?
3. What factors affect pump-and-dump outcomes?

What are the Academic Insights?

1. The authors describe these investors or players, in theory, as overconfident with a tendency to overestimate their own skill at timing the peaks of crypto prices.  To this set of investors, P&Ds seem to be a profitable opportunity.  A similar assessment would be applied to those who are simply attracted to the probability of a large payoff, especially since P&Ds have been generated from a distribution of payoffs that are skewed to the right. A rational investor would avoid P&Ds unless they were in possession of some type of skill or trading advantage.  Citing the notion that pumps represent zero-sum games where investor wealth is simply redistributed between the P&D players, excluding manipulators of the scheme, P&D players in the aggregate lose money. So, how do these schemes maintain participation? Pump-and-dump schemes appear to be a function of two behavioral factors: overconfidence and gambling characteristics.
2. A quick snapshot of the price dynamics: As manipulators build their initial positions, prices of the targeted cryptocurrency tend to rise by 10%, approximately 15 minutes before the pump signal. As non-manipulators or players enter the scheme, prices rise 40% further after the signal. After the peak, prices begin to fall slowly until they reach the price just prior to the initiation of the scheme.  Trading volume follows a similar pattern. Fifteen percent of trading volume also occurs within the first 15 minutes, peaking just after the signal and gradually becomes more subdued during the “dump” phase.
3. Regression analysis was conducted to examine the relationship between pump outcomes and various characteristics of the crypto market.  Here are the findings:
   • A positive and significant relationship between manipulator profits and the size of pump participation.  An increase of 1% in participation is associated with almost a 1% increase in profits.
   • A weak relationship existed between profits and illiquidity.  The relationship was negative but not significant.
   • A positive and significant relationship between participation and the pre-pump inventory held by the manipulator.  An increase of 1% in participation was associated with a 0.44% in inventory.
   • A 1% increase in participation was also positively associated with a 0.24% increase in return.
   • The duration of the pump scheme was negatively related to the level of participant experience. A 1% increase in experience reduced duration by 0.33%.  Groups with 3 or more pumps conducted previously were 16.5% faster, on average than groups with 2 previous pumps.

Why does it matter?

The major contribution from this research aligns with the regulatory environment that is developing around the use and abuse of cryptocurrencies. Cryptocurrency and the various associated forms of infrastructure are currently in a stage of rapid innovation.  If the manipulation embedded in P&Ds is widespread then the confidence in and integrity of the crypto market will suffer.  We would expect a loss of confidence to be associated with the growth of illiquidity in coins and a reduction in participation by large financial institutions and the general investing public.  The realization of the economic benefits from this aspect of the financial infrastructure will be subject to the cost of uninformed and inordinate regulation.

The most important chart from the paper

The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained.  Indexes are unmanaged and do not reflect management or trading fees, and one cannot invest directly in an index.

Abstract

We investigate the puzzle of widespread participation in cryptocurrency pump-and-dump manipulation schemes. Unlike stock market manipulators, cryptocurrency manipulators openly declare their intentions to pump specific coins, rather than trying to deceive investors. Puzzlingly, people join in despite negative expected returns. In a simple framework, we demonstrate how overconfidence and gambling preferences can explain participation in these schemes. Analyzing a sample of 355 cases in 6 months, we find strong empirical support for both mechanisms. Pumps generate extreme price distortions of 65% on average, abnormal trading volumes in the millions of dollars, and large wealth transfers between participants.