AAVE is a decentralized lending platform where users can collateralize assets to receive loans, with the loan-to-value (LTV) ratio determining the amount received. To maintain solvency, loans must remain overcollateralized, and if the collateral value drops, liquidation occurs. Liquidators can profit by paying off undercollateralized loans and receiving a portion of the collateral plus a bonus. This report by Amberdata's Director of Product Mgt, Pat Doyle, and Amberdata's Research Analyst, Noah Swerhun analyzes AAVEv2 liquidations, particularly for stablecoin loans collateralized by ETH, and reveals that liquidations offer a low-risk profit opportunity. The profitability is influenced by factors like gas fees and the size of the loans, with top liquidators earning substantial profits.

Overview

AAVE is a decentralized lending platform, where users can lock up an asset as collateral, and in exchange, receive another asset based on AAVE’s loan-to-value (LTV) ratio. For example, consider Alice who wishes to take out USDC on loan. Alice can give AAVE 1 ETH, and in return receive the equivalent USDC (based on 0.86 ETH, since AAVE’s LTV for USDC is 86%). (You can read more about AAVE’s specific LTV parameters here.)

We will be looking at liquidation calls for the AAVE protocol. All the code to recreate this analysis can be found here.

To keep the protocol solvent, loans must always be overcollateralized, as demonstrated in the case of Alice. But, the value of assets is always changing. If the market value of the collateral decreases, the loan can become undercollateralized. This presents a problem for AAVE because the only way for the contract to remain solvent is for loans to remain collateralized. The solution to this problem is liquidation. Liquidation is when another user of AAVE pays off part of an undercollateralized loan in return for a proportional amount of the collateral with a bonus. Back to the example above: for Alice’s USDC loan, LTV must not exceed 88%, or she will be at risk of liquidation. Suppose the price of ETH drops sufficiently that LTV reaches this threshold. Bob, a liquidator, will see that Alice’s loan is unhealthy, and pay off part of the principal (in USDC) and receive, in return, an equivalent amount of ETH (the collateral), plus a bonus 5% as his reward for keeping AAVE solvent. This can also be interpreted as Bob receiving ETH at a discount, because he gets more ETH than he “should” for the amount of USDC he covered, reducing the effective exchange rate. Once again, see the docs for more information. See the chart below for a concrete example.

Interest payments not shown for simplicity’s sake

Anyone can be a liquidator on AAVE, and there are several other protocols with similar lending/liquidation functionalities. Being a liquidator on AAVE, or another protocol, presents an opportunity to potentially make low-risk gains. This report investigates historical liquidation data on AAVEv2, specifically on loans of stablecoins (USDT, DAI, USDC, BUSD, TUSD, sUSD, GUSD, and PAX) collateralized by ETH.

Market Dynamics

Chart showing liquidations (bubbles) and the price of ETH (black line). The size of the bubble represents the amount of debt covered. The bubbles are color-coded by token. The red line in the center marks a 5% return.

The above chart shows liquidations plotted alongside the price of ETH, from January of 2021 to January of 2023. The y-value of each bubble (liquidation) is the percent return the liquidator made by performing the liquidation. This chart shows several interesting characteristics. First, there are distinct vertical “bands” of liquidation events that correspond to drops in the price of ETH. This makes sense because a drop in ETH’s value could cause loans backed by ETH to become undercollateralized. What this means, is that one could predict a major liquidation event by observing the volatility of ETH over time. More volatile periods have more opportunity for liquidations because there is greater chance that more loans will become undercollateralized. Liquidations present a way to potentially profit off of a volatile market with relatively low risk, because of the guaranteed bonus provided to the liquidator. Furthermore, knowledge of the correlation between volatility and liquidations may present opportunities for options traders to construct strategies to take advantage of these conditions. Investigation of these possibilities is left to the reader. 

The return on every liquidation does not seem to always be exactly 5%. Although the average return is indeed 5%, percentage gains as measured by Amberdata fluctuate between 0% and 10%, distributed normally.

Liquidator Return Distribution

We observe this likely due to the fact that we do not know the exact price of ETH at the moment of liquidation; we work only with minute-to-minute data. To calculate liquidator gains, we used the market price of ETH at the nearest minute, as the closest approximation. Regardless, the data demonstrates that a 5% return can certainly be expected, with a small margin of error. 99.8% of all liquidations we measured had a return between 0 and 10%. 98.7% were between 3% and 7%. Liquidation returns prove to be consistent and expectable.

12-16 May 2022. The black line shows the price of ETH, red line shows a 5% discount. Dots represent liquidations and are placed at the “effective” discounted price of ETH the liquidator received

The above chart displays the gap between the discounted price of ETH the liquidator received and the actual market price of ETH. This clearly visualizes the spread liquidators make.  Once again, the actual liquidation returns are normally distributed around the red line, showing the predictability. The “gap” between the red and black lines also grows larger for other assets offered by AAVE. As previously mentioned, USD stablecoins give a collateral bonus of 5%: liquidating wrapped Bitcoin yields 6.5%, and Curve DAO tokens grant 8.5%. For other assets, the potential return is even greater.

Gas

The analysis so far has not considered gas prices and transaction fees. In reality, gas has a noticeable effect on profitability. Examining the return distribution gives the clearest picture of how this works:

Return distribution, taking into account transaction fees. Black line at 0%, solid red line at 5%

Taking all fees into account, the mean return drops to just 2%, and 19% of all liquidations actually result in a loss for the liquidator. However, liquidation still presents a potentially highly profitable opportunity, as seen in the performance of the top liquidators.

Performance & Top Liquidators

To give concrete dollar values to the opportunities around liquidation: from 1/1/21 to 1/1/24, liquidators on AAVEv2 made a total of $16,006,566, or $5,335,522 per year. The top 10% of liquidators made 88.3% of the profits. These numbers all include transaction fees.

Statistics by liquidator. Blue: Average profit per liquidator ($). Red: Total profit ($). Green: Number of liquidations executed. Filtered to show only liquidators who made at least ten liquidations.

Investigating the performance of top liquidators can give an idea of the level of returns that can be expected in the long term. We chose profitable liquidators, who made at least 10 liquidations, and analyzed their performance. Some liquidators liquidated fewer but larger loans, while others liquidated many small loans. Those who had very low profit per liquidation (only liquidating small loans), seemed to find it difficult to compensate their total profit with sheer quantity. This is likely due to the fact that the amount of loans able to be liquidated at any given time is limited, so one should not assume that they will always be able to liquidate a large number of loans. Additionally, gas takes up a higher percentage of small transactions than large transactions, because it is based on computational complexity, not transaction size. Therefore, it is beneficial for a liquidator to target large loans, because they will be paying the same amount of gas as they would on a small loan, thus increasing profits. The liquidators with the highest total profit are those who either balanced size with quantity, or those who focused primarily on size. The 5 liquidators who made over $1 million followed one of these strategies. 

These liquidators had an average profit per liquidation of $2,507, and an average total profit of $242,077 over the three year period, or $80,692 per year. Once again, we reiterate that this profit is relatively low-risk, because the bonus is guaranteed by AAVE. The risk mainly comes into play when considering gas prices. However, the above data shows that it is possible to target large loans to effectively cover transaction fees and turn a significant profit nonetheless. For a trader to replicate these gains, time and manpower would have to be devoted to developing the necessary infrastructure. However, once this initial up-front investment is made, theoretically the profits should start “rolling in.” If the infrastructure developed could be generalized to other lending platforms such as Maker and Compound—which have similar features and shouldn’t require drastically different business logic—then these profits could be multiplied.

Conclusion

DeFi lending liquidation presents a large profit opportunity to those with the expertise and time to take advantage of it. The data investigated in this report represents just a small subset of the entire DeFi lending ecosystem. Scaling a liquidation strategy over multiple protocols, collateral assets, and L2s drastically increases profit opportunities.

Legal and Disclosures

Disclaimers

The information contained in this report is provided by Amberdata solely for educational and informational purposes. The contents of this report should not be construed as financial, investment, legal, tax, or any other form of professional advice. Amberdata does not provide personalized recommendations; any opinions or suggestions expressed in this report are for general informational purposes only.

Although Amberdata has made every effort to ensure the accuracy and completeness of the information provided, it cannot be held responsible for any errors, omissions, inaccuracies, or outdated information. Market conditions, regulations, and laws are subject to change, and readers should perform their own research and consult with a qualified professional before making any financial decisions or taking any actions based on the information provided in this report.

Past performance is not indicative of future results, and any investments discussed or mentioned in this report may not be suitable for all individuals or circumstances. Investing involves risks, and the value of investments can go up or down. Amberdata disclaims any liability for any loss or damage that may arise from the use of, or reliance on, the information contained in this report.

By accessing and using the information provided in this report, you agree to indemnify and hold harmless Amberdata, its affiliates, and their respective officers, directors, employees, and agents from and against any and all claims, losses, liabilities, damages, or expenses (including reasonable attorney’s fees) arising from your use of or reliance on the information contained herein.

Copyright © 2024 Amberdata. All rights reserved.