Summary:
A proposal to increase stablecoin Interest Rate parameters across all Aave deployments.
Motivation
Following the implementation of this AIP to increase stablecoin borrow rates across Aave V2 and V3 deployments, we continue to observe volatility in borrow rates across Aave deployments, given the upward market with various speculative investment strategies, leading to an incredible demand for leverage. We recommend a more aggressive increase of the Slope1 parameter to stabilize borrows under the UOptimal point.
Stablecoins Total Supply and Borrow: The data from the charts reflects an overall increase in both the supply and borrowing of stablecoins on the Ethereum V3 platform. From December 9th to March 6th we observed the following increases in supply and borrowing across USDC, USDT and DAI:
| Asset | Total Supply (M) - Dec 9th | Total Supply (M) - Mar 6th | Total Supply Increase (M) | Supply Increase (%) | Total Borrow (M) - Dec 9th | Total Borrow (M) - Mar 6th | Total Borrow Increase (M) | Borrow Increase (%) |
|---|---|---|---|---|---|---|---|---|
| USDC | 614 | 833 | 219 | 35.6% | 550 | 761 | 211 | 38.3% |
| USDT | 557 | 712 | 155 | 27.8% | 476 | 654 | 178 | 37.3% |
| DAI | 147 | 161 | 14 | 9.5% | 129 | 149 | 20 | 15.5% |
| Total | 1,318 | 1,706 | 388 | 29.4% | 1,155 | 1,564 | 409 | 35.4% |
We’ve observed an overall $388M increase (29.4%) in supply across USDC, USDT and DAI on Ethereum V3, and a $409M increase (35.4%) in borrows. This signifies a substantial demand for leverage, while concurrently enticing suppliers with consistently high and stable supply rates.
Interest Rate Analysis: Over the last few months, stablecoin borrow rates have hovered above the slope1 by quite a large margin, which, given the lack of a base_rate, effectively implies the optimal borrow rate. Since the AIP which increased slope1 from 5% to 6%, given the upward market with various speculative investment strategies, has led to an incredible demand for leverage.
Below we depict borrow rates over a highly granular timeframe, with snapshots taken approximately every 3 hours, followed by a simple moving average (MA) converted to the daily equivalent (approximately 8 snapshots/day) and a 7-day MA. Observation reveals that the rates have seldom aligned with the parameterized slope1, except for the period immediately following the slope1 adjustment from 5% to 6%, instead consistently scaling considerably higher.
The extent of this upward deviation is illustrated in the histogram below, showing the differences between APR(t) and the slope1 value. The histogram peaks at around a 3-4% difference for USDC and DAI. However, USDT exhibits a broader range of variance, resulting in a fatter tail in upward deviations from the slope1 value.
Computing both the geometric and arithmetic mean over varying timeframes for the stablecoins mentioned above, we observe a significant recent uptick, accompanied by sustained rates priced well above slope1 by a substantial margin.
| Mean | Last 7 Days | Last 14 Days | Last 30 Days | Last 60 Days | Last 90 Days |
|---|---|---|---|---|---|
| USDC Interest Rate (Arithmetic) | 12.29% | 11.04% | 9.89% | 8.94% | 8.91% |
| USDC Interest Rate (Geometric) | 12.08% | 10.83% | 9.87% | 8.58% | 8.63% |
| USDT Interest Rate (Arithmetic) | 14.46% | 10.93% | 10.34% | 8.94% | 9.04% |
| USDT Interest Rate (Geometric) | 14.16% | 10.05% | 9.87% | 8.26% | 8.35% |
| DAI Interest Rate (Arithmetic) | 9.64% | 9.35% | 9.28% | 8.46% | 8.73% |
| DAI Interest Rate (Geometric) | 9.16% | 9.07% | 8.97% | 7.89% | 8.22% |
Utilization Rate Analysis: Our continuous monitoring of the utilization rates post-AIP, setting Slope1 at 6%, shows sustained demand, with the utilization rate consistently maintaining a value above the 90% Uoptimal threshold.
| Mean | Last 7 Days | Last 14 Days | Last 30 Days | Last 60 Days | Last 90 Days |
|---|---|---|---|---|---|
| USDC Utilization | 92.10% | 91.68% | 89.17% | 91.04% | 90.72% |
| USDT Utilization | 92.82% | 91.22% | 88.76% | 89.58% | 88.77% |
| DAI Utilization | 90.92% | 90.97% | 88.21% | 88.33% | 88.69% |
Upon analyzing the percentage of time where utilization exceeded UOptimal, alongside the aggregate utilization rate across various timeframes, it’s apparent that the slope1 increase to 6% did not succeed in stabilizing utilization at or below UOptimal. This result can be attributed to evolving market dynamics, notably the recent surge in crypto asset prices.
| Mean | Last 7 Days | Last 14 Days | Last 30 Days | Last 60 Days | Last 90 Days |
|---|---|---|---|---|---|
| USDC Time Above Uoptimal | 100.00% | 100.00% | 88.79% | 81.97% | 87.23% |
| USDT Time Above Uoptimal | 100.00% | 76.24% | 75.70% | 62.06% | 66.98% |
| DAI Time Above Uoptimal | 88.24% | 94.06% | 83.64% | 68.38% | 75.86% |
Plotting various theoretically higher slope1 values as scalar values to accommodate the historical “time above uOptimal” relative to the rolling interest rate allows us to evaluate how “time above uOptimal” decays alongside a scaling slope1, based on the market-priced interest rate. Notably, the considerable surge in current interest rates (last 7 days) is evident, suggesting logarithmic decay as slope1 increases with recent rates, contrasting with a more linear or exponential decay observed over longer timeframes.
Recommendations
The analysis indicates a necessity for further refinement of stablecoin interest rates. To align with the market conditions, we propose two potential slope1 alterations:
Conservative:
Adjust Slope1 of stablecoins to 8%
The proposal entails increasing Slope1 to 8% for stablecoins across all Aave deployments. This adjustment assumes that current rates will regress to a longer-term mean, as seen over the last 60-90 days. Factoring in the impact of interest rates above a certain Uoptimal with a higher Slope1 value, an interim Slope1 value of 8% is expected to strike a balance with historically determined priced-in borrow rates. Despite the recent trend of rates skyrocketing and stabilizing above this value over the last month, a more gradual approach will enable us to thoroughly assess the situation going forward.
Aggressive:
Adjust Slope1 of stablecoins to 10%
We are proposing a significant increase in Slope1 to 10% for stablecoins across all Aave deployments. This additional adjustment aims to establish a more predictable and stable borrowing rate, ensuring an equilibrium utilization under the UOptimal point. This decision is made considering the recent substantial spike, with the assumption that rates will either remain sustained or continue to grow in the current risk-on market environment.
Following either of these increases, we will continue monitoring the usage and equilibrium rate and make additional recommendations as necessary.
| Market | Asset | Current Slope1 | Rec Slope1 Conservative | Rec Slope1 Aggressive |
|---|---|---|---|---|
| Ethereum V2 | USDC | 6% | 8% | 10% |
| Ethereum V2 | USDT | 6% | 8% | 10% |
| Ethereum V2 | FRAX | 6% | 8% | 10% |
| Ethereum V2 | sUSD | 6% | 8% | 10% |
| Ethereum V2 | GUSD | 6% | 8% | 10% |
| Ethereum V2 | LUSD | 6% | 8% | 10% |
| Ethereum V2 | USDP | 6% | 8% | 10% |
| Ethereum V3 | USDC | 6% | 8% | 10% |
| Ethereum V3 | USDT | 6% | 8% | 10% |
| Ethereum V3 | FRAX | 6% | 8% | 10% |
| Ethereum V3 | LUSD | 6% | 8% | 10% |
| Avalanche V2 | USDC | 6% | 8% | 10% |
| Avalanche V2 | USDT | 6% | 8% | 10% |
| Avalanche V2 | DAI | 6% | 8% | 10% |
| Avalanche V3 | USDC | 6% | 8% | 10% |
| Avalanche V3 | USDT | 6% | 8% | 10% |
| Avalanche V3 | DAI | 6% | 8% | 10% |
| Avalanche V3 | MAI | 6% | 8% | 10% |
| Avalanche V3 | FRAX | 6% | 8% | 10% |
| Polygon V2 | USDC | 6% | 8% | 10% |
| Polygon V2 | USDT | 6% | 8% | 10% |
| Polygon V2 | DAI | 6% | 8% | 10% |
| Polygon V3 | USDC | 6% | 8% | 10% |
| Polygon V3 | USDT | 6% | 8% | 10% |
| Polygon V3 | DAI | 6% | 8% | 10% |
| Polygon V3 | MAI | 6% | 8% | 10% |
| Polygon V3 | USDC.e | 7% | 9% | 11% |
| Optimism V3 | USDC | 6% | 8% | 10% |
| Optimism V3 | USDT | 6% | 8% | 10% |
| Optimism V3 | DAI | 6% | 8% | 10% |
| Optimism V3 | sUSD | 6% | 8% | 10% |
| Optimism V3 | LUSD | 6% | 8% | 10% |
| Optimism V3 | MAI | 6% | 8% | 10% |
| Optimism V3 | USDC.e | 7% | 9% | 11% |
| Arbitrum V3 | USDC | 6% | 8% | 10% |
| Arbitrum V3 | USDC.e | 7% | 9% | 11% |
| Arbitrum V3 | USDT | 6% | 8% | 10% |
| Arbitrum V3 | DAI | 6% | 8% | 10% |
| Arbitrum V3 | LUSD | 6% | 8% | 10% |
| Arbitrum V3 | FRAX | 6% | 8% | 10% |
| Arbitrum V3 | MAI | 6% | 8% | 10% |
| Base V3 | USDbC | 7% | 9% | 11% |
| Base V3 | USDC | 6% | 8% | 10% |
| Metis V3 | USDC | 6% | No Change | No Change |
| Metis V3 | USDT | 6% | No Change | No Change |
| Ethereum V3 | crvUSD | 5% | 8% | 10% |
| Ethereum V3 | pyUSD | 6% | No Change | No Change |
| BNB Chain V3 | USDT | 6% | 8% | 10% |
| BNB Chain V3 | USDC | 6% | 8% | 10% |
| BNB Chain V3 | FDUSD | 6% | No Change | No Change |
| Scroll V3 | USDC | 6% | 8% | 10% |
Next Steps
- Following community feedback, submit the ARFC for a snapshot vote for final approval.
- If consensus is reached, submit an Aave Improvement Proposal (AIP) to implement the proposed updates.
- We believe this adjustment will align the protocol’s rates more closely with current market conditions and borrower behavior. Based on the outcomes, potential future steps include:
-
Increasing UOptimal:
This move would signify a higher threshold before Slope 2 interest rates are triggered. With the increase in Slope1, the underlying interest rate volatility is expected to diminish, leading to decreased demand above the kink. Consequently, by raising UOptimal, we can anticipate a similar utilization rate outcome as today, while offering a relatively higher incentive for new deposits if UOptimal is reached.
We will conduct a comprehensive analysis of the historical concentration and distribution of supplied stablecoins used as collateral (generally quite minimal). Subsequently, we will determine an appropriate increase in Uoptimal, ensuring an adequate buffer to facilitate theoretical liquidations and supplier withdrawals. Opting for the more aggressive approach of 10% is likely to yield a greater increase in the Uoptimal value compared to 8%, as the higher Slope1 value is expected to foster a more stable utilization rate, thereby minimizing fluctuations driven by market-priced rate movements.
-
Further increase of Slope 1: Depending on the market’s and community’s response to the initial adjustment, a subsequent increase in Slope 1 could be considered to further optimize the IR curves.
As always, our priority remains to monitor these developments closely and provide timely, data-driven recommendations to maintain Aave’s competitive edge and market responsiveness.
Disclaimer
Chaos Labs has not been compensated by any third party for publishing this ARFC.
Copyright
Copyright and related rights waived via CC0
