Daily Liquidity Jump and Diffusion, and Treatment on Crypto Wash Trading

Qi Deng,Zhong-Guo Zhou

2024-10-28

Abstract:Wash trading among crypto assets induces short-term price jumps, which manifest as liquidity fluctuation. We develop a model to decompose asset liquidity into two components: liquidity jump and liquidity diffusion, which quantify the size and probability of wash trading. Using the trading data from US stock markets as a benchmark, we establish that the combination of high liquidity diffusion and high liquidity jump indicates wash trading. On the other hand, the majority of large-volume trades with high liquidity jump but low liquidity diffusion (<1) are actually legitimate ones. Our model provides advice to investors in making informed decisions in crypto investments, and to regulators in taking targeted actions to combat wash trading. While we use crypto assets to exemplify the utilities of our model, the model should remain effective for other asset classes with extreme liquidity.

Risk Management,Computational Finance,Statistical Finance,Trading and Market Microstructure

What problem does this paper attempt to address?

The problems that this paper attempts to solve are:

1. **How to accurately measure the probability and scale of wash trading**: Wash trading refers to traders manipulating market prices by simultaneously buying and selling the same asset to obtain short - term profits. This behavior is particularly prevalent in the cryptocurrency market but difficult to detect and quantify.

2. **What specific information should investors analyze to make informed cryptocurrency investment decisions on a daily basis**: Due to the existence of wash trading, investors need to understand which indicators can help them identify potential manipulation behaviors and adjust their investment strategies accordingly.

3. **What targeted measures should regulatory agencies take to combat wash trading**: Regulatory agencies need effective tools and techniques to identify and curb wash trading in order to maintain market fairness and transparency.

To answer these questions, the author proposes a model that decomposes asset liquidity into two components:
- **Liquidity jump**: Measures the magnitude of liquidity fluctuations.
- **Liquidity diffusion**: Reflects the probability of liquidity fluctuations.

These two indicators are used to quantify the scale and probability of wash trading respectively. By using trading data from the US stock market as a benchmark (these markets are strictly regulated and there is almost no wash trading), the author finds that the combination of high liquidity diffusion (≥1) and high liquidity jump (≥1) usually indicates the occurrence of wash trading.

In addition, this model also provides the following insights:
- For most large - volume transactions, although the liquidity jump is high (≥1), if the liquidity diffusion is low (<1), these transactions are usually legal.
- Investors can make more informed investment decisions by analyzing the liquidity jump and liquidity diffusion distributions of high - market - capitalization cryptocurrencies.
- Regulatory agencies can use these liquidity indicators to identify and combat wash trading, especially in extreme - liquidity scenarios.

In conclusion, this paper aims to provide an effective method to identify and quantify wash trading and provide practical suggestions for investors and regulatory agencies.