Oct 3, 2021
THORChain Data Digest: Lower Bound of Liquidity to Prevent High Slip Fees
How much liquidity is enough to keep slip fees low on THORChain? In this report, I dive into the numbers to find out. I show that in times of regular trading volume, as little as 2 million USD in liquidity pool depth can protect users from high slip fees on their transactions. However, when the markets are in periods of high greed or fear, even 20 million in depth might not be enough.
The Importance of Deep Liquidity:
THORChain implements a Continuous Liquidity Pool model (CLP), where fees are proportional to the share of liquidity a swap demands. When liquidity is low, this can make swap fees prohibitively high, and push users away from the platform.
We can see evidence of this when plotting volume per pool against average slip %. Figure 1 shows average slip in red and total volume to the pool (Rune) in green. As average slip % increases, volume to the pool decreases. Swappers trade less volume if they have to pay a high slip fee for it.
Plotting slip % against average transaction size by pool corroborates this story. Figures 2 and 3 show the average slip fees vs average transaction sizes for ETH/ERC-20 and non ERC-20 pools respectively. As average transaction size increases, the average slip % decreases. Intuitively, you might expect the opposite. Larger transactions put pools more out of balance, creating a larger slip fee. However, what we see in the data is that swapper behaviours are responsive to the available liquidity. If slip fees are high, swappers react by only trading small amounts. Swappers are discouraged from making large transactions, which limits volume to the pool. Slip fees dictate transaction sizes, rather than the other way around.
Low Liquidity is the main driver of Slip Fees:
Rather than transaction sizes, it is pool depth that drives slip fees on THORChain. Plotting total liquidity depth of all pools on THORChain against the average slip % on transactions in the pool, the trend is clear. Deeper liquidity means lower slip fees. BTC, with over $30 million USD total liquidity, has 0.13% average slip. Meanwhile the Raze pool, with just under $200k liquidity, has an average slip of 2.1% per transaction.
The high volatility seen in the average slip graph is due to the different properties of each chain on THORChain. ETH/ERC-20 pools have much higher gas fees, which drives up the minimum size of a trade for it to be economically worthwhile. A user will not swap $100 USD of ETH if the gas fee is also $100 USD. This pushes up the average slip % for all ETH/ERC-20 pools. BEP2 pools by contrast have the lowest slip fees, since gas is so cheap. Average slip in ERC-20 pools is 0.82%, while it is just 0.1% in BEP2 pools.
Lower Bound for Sufficient Liquidity:
Figure 5 shows the historic average slip for Bitcoin against Rune Depth in the pool. Average slip is shown in green, and Rune Depth in black. In the first several weeks, we see a rapid decrease in average slip as initial liquidity is added. However, from week 8, average slip flattens out and stabilises below 0.1%. Further increases in Rune Depth have diminishing returns. The Rune Depth at week 8 to achieve this was 803,760, corresponding to a total pool depth of $19,880,000 USD (value of Rune Depth + value of Asset Depth).
Looking across all other pools on THORChain, a similar patten can be seen across the deepest 8 pools (figure 6). Rapidly decreasing average slip for the initial caps raises, before reaching a stable point of consistently low slippage. For ETH/ERC-20 pools, the smallest pool to reach this point and achieve consistently low slippage was USDT. It achieved this with $3,883,329 USD total liquidity. The next smallest ERC-20 pool is ALCX with a little under $3 million USD total liquidity. However ALCX still has high average slippage, ranging from 0.3 to 0.4%. From this we can establish an initial lower bound of just under $4 million USD to achieve low slippage for ETH/ERC-20 pools.
For non ERC-20 pools, as little as $1.9 million USD liquidity is sufficient to achieve consistently low slip. This can be seen in figure 7 with the Bitcoin Cash pool. By week 7 a total of $1,925,699 USD liquidity had been added to the pool, and this corresponded to a drop in average slip from over 0.6% just a couple of weeks earlier to around 0.15%.
Greed and Fear:
Returning to our BTC graph, we can see that while $20 million USD liquidity was sufficient to bring down average slip to below 0.1% for several weeks, it was still vulnerable to a spike in slippage on week 12. This was due to high fear in the market. In this week, Rune price dropped from just over 8 USD to under 6 USD. $20 million USD in liquidity is sufficient to keep slippage low in periods of normal trading volumes, but is still insufficient in the peak periods.
An idea I have discussed in a previous Data Digest report is that tolerance for swappers to pay slippage is highly sentiment based. The baseline average slip % across all pools for all time is 0.09%. In periods of extreme fear this increases to 0.136%. In periods of extreme greed it is 0.2165%. When there is high fear in the market, tolerance to pay slippage massively increases as swappers panic sell assets to stable coins to protect their portfolio. Similarly, when the market is highly greedy, swappers accept a higher slippage as they fomo into the next must have asset. The lower bound for liquidity to prevent high slip fees in periods of market turbulence is significantly higher than in stable market conditions, and specifically is somewhere in excess of $20 million USD.
Conclusion:
When markets are stable, as little as $2 million USD liquidity can be enough to provide low slip trades for swappers on THORChain. For ETH/ERC-20 pools, this number increases to about $4 million USD. However, in periods of extreme market sentiment, whether it be extreme greed or extreme fear, upwards of $20 million USD liquidity might still not be enough.
