Liquidity in Cryptocurrency Markets

Liquidity represents one of the key concepts one needs to understand when investing in any market. At a general level, liquidity represents the degree at which a given asset can be quickly sold or bought in the marketplace with a limited price impact. In simplest terms, liquidity refers to the ability of an asset to be converted into cash easily. The most liquid instrument is typically cash, alongside the whole variety of money market instruments since they tend very stable and can be readily liquidated and therefore easily spent to meet short-term financial needs.

A common prerequisite of liquid assets is that they all have a ready and open market to be traded on. That is, a liquid asset should be actively traded on different exchanges with stable prices. On the other hand, illiquid assets are normally traded on private markets, i.e., over-the-counter (OTC). Prices can therefore vary by a huge margin and trades can take a significant amount of time to execute. Essentially, the harder and more expensive it is for an asset to be turned into cash, the more illiquid it is.

The importance of liquidity in crypto markets

Liquidity assumes a particular relevance in the cryptocurrency space given the fragmented and multi-platform structure of crypto trading. Higher liquidity is preferred as it brings the following advantages: First, in a liquid market, prices are a much fairer representation of the underlying fundamental value of a given asset due to the large number of buyers and sellers who compete against each other. In other words, a competitive, liquid market increases the likelihood of reaching an equitable market equilibrium for all. This results in a much more stable market and ensures that market participants are not disadvantaged above a priori, regardless of their contingencies, capital constraints and investment strategies.

Second, and related to the presence of a fairer market equilibrium, higher liquidity ensures that prices are more stable and less prone to large and unexpected price swings. For instance, in highly illiquid markets with little trading activity it is relatively easy for large institutional investors to significantly influence prices. The price impact of their trading orders can be quite substantial and at the expense of smaller investors. In such situations, a single large buy or sell order may heavily affect both returns and volatility. On the other hand, in a stable liquid market, trading activity is deep enough for prices to withstand large orders due to the presence of multiple market participants and orders.

Third, trading in a liquid market is normally not only quicker - so transactions can be executed faster due to a large number of competitive buyers and sellers - but also cheaper. As a matter of fact, one of the main outcomes of a high liquidity environment is the presence of a small distance between bid and ask prices, the so-called bid-ask spread. Quick and cheap execution is critical in the fast-paced cryptocurrency markets.

Factors that affect liquidity in crypto markets​

There are a variety of factors that can affect liquidity in cryptocurrency markets beyond the endogenous trading motives of investors. First and foremost, the presence of large and reliable exchanges is key. An exchange is a marketplace where assets are traded freely between buyers and sellers. A large number of exchanges could signify increasing trading activity and therefore market liquidity. That implies that a higher number of cryptocurrency exchanges signifies greater trading volume at an aggregate level.

Trading Volume

Trading volume represents a core factor affecting liquidity (see, e.g., Bianchi and Dickerson 2018). It typically refers to the amount of coins that have been traded on exchanges, usually on average, over a given time period. Essentially, the volume reflects the market activity of a specific coin; a higher volume indicates that more people are buying and selling it. The increase in frequency and volume of trading helps to enhance market liquidity.

Exchanges

There are currently over 200 cryptocurrency exchanges, in addition to a small universe of both decentralised exchanges and a handful of peer-to-peer (P2P) trading platforms, at the time of writing. The number of active exchanges has significantly increased over the last two years, signalling an increasing interest by retail and large investors in cryptocurrency trading. Major exchanges such BKEX, MXC, Biki and IDAX trade hundreds of millions of USD in cryptocurrency on average per day. A higher number of exchanges goes hand in hand with increasing market activity and aggregate trading volume. As a result, not only trading per se becomes more liquid, but there are also more venues for professional and retail investors to participate in the marketplace.

Adoption

Another important factor that affects liquidity in cryptocurrency markets is the adoption of cryptocurrencies as a mean of payment. This is linked to the fact that the success and viability of any currency - and therefore its appeal as an investment -  depends on the acceptance of the masses or at least a network of individuals that can actually use it for some concrete purpose.

Regulatory interventions

The laws and regulations of different countries can also impact cryptocurrency liquidity. There are certain countries that have banned either cryptocurrency trading or the use of cryptocurrencies in dealings. For instance, China banned cryptocurrency exchanges in September 2017. This would have impaired liquidity since a ban on exchanges is implicitly a ban on cryptocurrencies per se. Anyone looking to own cryptocurrencies would have to privately find a seller, or rely on peer-to-peer platforms. Therefore, the liquidity in that country would be extremely low. This usually results in higher prices and a larger price impact, which gives sellers and large speculators more power to affect prices.

Measuring liquidity in crypto markets​

The gap between bid and ask prices is fundamentally linked to the dynamics of liquidity in financial markets. The bid-ask spread is just the difference between the best ask price and the best bid price for a given security. The larger the spread between bid and ask prices the more illiquid and thus more costly it is to trade a given security. In this respect, large bid-ask spread securities may also be more difficult to offload in the future, as only few market participants could be willing to provide liquidity at an adequate price. On the other hand, small bid-ask spreads typically indicate greater liquidity in a given security. These securities are normally easier to trade. As a result, the price formation process is usually correlated with the exchange features and especially with the analysis of spread determinants between the bid and ask prices.

The information content of the bid-ask spread in cryptocurrency markets is the same as in traditional asset classes. That is, it measures the extent to which it is “easy” to trade a given cryptocurrency on a given exchange. The fact that there are hundreds of different exchanges makes the measurement of the bid-ask spread at the aggregate level for a given security a rather difficult task. If anything, this is due to the fact that different exchanges may have a different bid-ask spreads on the same cryptocurrency depending on their size and trading activity.

Corwin and Schultz (2012) and Abdi and Ranaldo (2017) showed that a synthetic version of the bid-ask spread could be derived from intraday high, low, and closing prices. We implement their proposed methodology to approximate the bid-ask spread in cryptocurrency markets by aggregating high, low and closing prices across the most liquid 150 exchanges. The aggregation is made on a volume basis. That is, the aggregate bid-ask spread is a volume-weighted version of the exchange-specific bid-ask spreads.

Figure 1: Aggregate bid-ask spreads

Source: Aaro Capital Research
Notes: Aggregate bid-ask spreads. This figure shows the time series of the Corwin and Schultz (2012) (blue bar) and Abdi and Ranaldo (2017) (red bar) bid-ask spread measures. The sample period spans from 1st April 2017 to 30th May 2019.

Figure 1 gives a first impression of the dynamics of liquidity in cryptocurrency markets by showing an equally weighted average of the crypto-specific liquidity measures over the sample period. Recall that the measures represent a proxy for the bid-ask spread, that is a lower bid-ask spread indicates increasing liquidity, and vice versa.

The evidence points towards improved liquidity over the sample period. The approximated bid-ask spread (aggregated across exchanges) steadily decreases to almost half by the end of the period. Price impact steadily decreases as well, although remains about a half of the initial value towards the end of the sample. As a whole, Figure 1 suggests that aggregate liquidity for major cryptocurrencies improved over time, consistent with the hypothesis of a more mature market and the assumption that the creation of more exchanges tends to be driven by increasing trading demand and turnover.

Although informative, such measures dilute, by construction, the cross-sectional heterogeneity in the liquidity across different cryptocurrencies. As an example, Figure 2 shows the dynamics of the Corwin and Schultz (2012) measure for two of the most liquid crypto in the sample, namely ETH (right panel) and LTC (left panel).

Figure 2: Ether and Litecoin bid-ask spreads 


Source: Aaro Capital Research
Notes: This figure shows the time series of Corwin and Schultz (2012) bid-ask spread for Litecoin (left panel) and Ethereum (right panel). The sample period spans 1st April 2017 to 30th May 2019.

The liquidity of both Litecoin and Ethereum steadily improved over time. As a matter of fact, the bid-ask spread went from an average of 2% from July 2017 to January 2018, to an average 1% from February 2018 to May 2018. LTC and ETH are two of the most traded cryptocurrencies, which justify the relatively similar dynamics in liquidity.

Figure 3 shows the dynamics of the bid-ask spread for two less known, and less traded, cryptocurrencies, namely DigixDAO (left panel) and Monacoin (right panel).

Figure 3: DigixDAO and Manacoin bid-ask spreads

Source: Aaro Capital Research
This figure shows the time series of Corwin and Schultz (2012) bid-ask spread for DigixDAO (left panel) and Monacoin (right panel). The sample period spans 1st April 2017 to 30th May 2019.

The dynamics of the bid-ask spreads are now quite heterogeneous. While DigixDAO (DGD) shows steadily increasing liquidity over time - similar to ETH and LTC – Monacoin (MONA) shows a spike in illiquidity from April 2018 to August 2018.

The heterogeneity in the dynamics of the bid-ask spreads do not come as a surprise. As explained above, there are a variety of factors that could affect liquidity, and these factors do not necessarily align across different cryptocurrencies. For example, the level of adoption for one cryptocurrency may be different from another.

To understand better how much heterogeneity there is in terms of liquidity, we consider bid-ask spreads across the main cryptocurrencies. Figure 4 shows the sample means of bid-ask spreads for the top 25 cryptocurrencies in terms of market capitalization as of September 2019, excluding Bitcoin (BTC).

Figure 4: Mean bid-ask spread for the top 25 crpyotcurrencies