Gauntlet’s BASE/Moonbeam/Moonriver Recommendations (2024-01-02)

Gauntlet’s BASE Recommendations

Simple Summary

Risk Parameters

A proposal to adjust 5 risk parameters:

Risk Parameter Current Value Recommended Value
wstETH Collateral Factor 76% 77%
rETH Collateral Factor 76% 77%
cbETH Collateral Factor 76% 77%
USDC Borrow Cap 8.5M 12M
wstETH Supply Cap 800 1000

*Cap Recommendations will be implemented via Guardian

IR Parameters

A proposal to adjust 1 IR curve:

cbETH IR Parameters Current Recommended
Base 0 0
Kink 0.45 0.45
Multiplier 0.07 0.06
Jump Multiplier 3.15 3.15

Rationale:

Risk Parameters

Our recommendations have an estimated VaR at $0 and remains unchanged. LaR increased to $1.17M from $1.16M with our current recommendations. Based on simulation results, Gauntlet recommends increasing the collateral factor for staked ETH asssets cbETH, rETH, and wstETH in order to improve capital efficiency.

Top wstETH Supply Accts

The largest 2 wstETH suppliers are not incurring any debt.

Top rETH Supply Accts

The largest rETH accounts either have no debt or have highly correlated or recursive positions within the BASE market.

Top cbETH Supply Accts

Excluding the 2nd largest position, all large cbETH collateralized positions are borrowing correlated assets or cbETH. The 2nd largest supplier of cbETH 0x3f7c10cbb has a 1.78 health factor with USDC & USDbC borrow positions.

BASE Liquidity

Asset Borrow Cap Supply Cap Borrow Cap Usage Supply Cap Usage DEX 25pct Slippage Token DEX 25pct Slippage USD
USDC 8,500,000 15,000,000 72.41% 50.99% 5,243,976 $5,244,032.88
cbETH 1,000 6,000 59.89% 51.00% 1,242 $2,995,018.64
USDbC 4,000,000 5,000,000 10.85% 11.73% 3,970,725 $3,970,768.44
WETH 8,000 10,500 59.07% 62.93% 2,428 $5,539,460.11
DAI 6,000,000 7,500,000 70.05% 69.77% 1,342,398 $1,343,645.43
rETH 100 600 51.00% 57.92% 318 $797,179.97
WSTETH 160 800 43.30% 63.05% 441 $1,158,799.06

BASE Circulating Tokens and Supply Cap

Asset Circulating Supply Supply Cap ($) Supply Cap / Circulating Supply Supply Cap Liquidity
USDC 44,122,810 $15,000,000.00 34.00% $7,351,500.00
cbETH 19,492.00 $15,070,560.00 30.78% $7,384,574.40
USDbC 108,071,282 $5,000,000.00 4.63% $4,413,500.00
WETH 51,178 $24,930,045.00 20.52% $9,241,567.68
DAI 9,531,400 $7,425,000.00 78.69% $2,244,577.50
rETH 803 $1,567,512.00 74.72% $659,609.05
WSTETH 2,319 $2,186,560.00 34.50% $807,933.92

IR Parameters

cbETH IR Parameters

Utilization rates for cbETH have been relatively low since the initial listing of the asset on the protocol with utilization staying at or below 20%.

cbETH Utilization


Link to chart

Gauntlet intends to lower the Borrowing rate in order to promote more borrowing of cbETH and generate further reserves for the protocol.

Recommended cbETH IR Curve

Recommended APRs

Utilization Borrow APR Supply APR
0 0 0
45 2.7 0.91125
100 175.95 131.9625

Current APRs

Utilization Borrow APR Supply APR
0 0 0
45 3.149999999 1.063125
100 176.4 132.3

The Borrower APR at the kink is anticipated to drop by 45 basis points (bps) with the suggested interest rate (IR) curve. Depending on supplier and borrower elasticity, this move could potentially boost borrowing by $317,000, potentially increasing annualized reserves by $1,000. Other LST assets are still in the process of reaching their utilization equilibrium since their introduction in the Base market. Gauntlet will monitor their utilization and consider additional actions if their levels remain low.

Moonwell BASE Projected Revenue KPIs

Projected Annual Reserves as of Forum Post

Methodology

This set of parameter updates seeks to maintain the overall risk tolerance of the protocol while making risk trade-offs between specific assets.

Gauntlet’s parameter recommendations are driven by an optimization function that balances 3 core metrics: insolvencies, liquidations, and borrow usage. Parameter recommendations seek to optimize for this objective function. Our agent-based simulations use a wide array of varied input data that changes on a daily basis (including but not limited to asset volatility, asset correlation, asset collateral usage, DEX / CEX liquidity, trading volume, expected market impact of trades, and liquidator behavior). Gauntlet’s simulations tease out complex relationships between these inputs that cannot be simply expressed as heuristics. As such, the input metrics we show below can help understand why some of the param recs have been made but should not be taken as the only reason for recommendation. To learn more about our methodologies, please see the Helpful Links section at the bottom.

Supporting Data

The below figures show trends on key market statistics regarding borrows and utilization that we will continue to monitor:

Top 10 Borrowers’ Aggregate Positions & Borrow Usages

Top 3 debt positions on Base market are recursive.

Top 10 Borrowers’ Entire Supply

Link to Chart

Top 10 Borrowers’ Entire Borrows

Link to Chart

Utilization Rate of Assets - Timeseries

Link to Chart.

Borrow Cap Usage

Supply Cap Usage

Balances by User Strategies

The percentage of borrower balances utilizing LST Yield Farming continues to grow since our last market recommendation. The percentage has increased from 15.85% to 18.06%.

Stablecoin 25% Slippage on BASE

Non-Stablecoin 25% Slippage on BASE

Risk Dashboard

The community should use Gauntlet’s Moonwell Base Risk Dashboard to better understand the updated parameter suggestions and general market risk in Moonwell BASE.



Moonbeam Recommendations

Simple Summary

A proposal to adjust 4 total risk parameters:

Parameter Current Value Recommended Value
xcDOT Collateral Factor 60% 58%
USDC.wh Reserve Factor 15% 20%
xcUSDT Reserve Factor 15% 20%
xcUSDC Reserve Factor 15% 20%

A proposal to make a IR curve adjustments for xcUSDC, xcUSDT, and USDC.wh:

xcUSDC IR Parameters Current Recommended
BASE 0.02 0
Kink 0.8 0.8
Multiplier 0.1 0.075
Jump Multiplier 1.9 6.0
xcUSDT IR Parameters Current Recommended
BASE 0 0
Kink 0.8 0.8
Multiplier 0.065 0.075
Jump Multiplier 4.0 6.0
USDC.wh IR Parameters Current Recommended
BASE 0 0
Kink 0.8 0.8
Multiplier 0.065 0.075
Jump Multiplier 4.75 6.0

Rationale:

Risk Parameters

Our recommendations maintain an estimated VaR at $0, unchanged from previous estimates. LaR decreased to $836k from $856k with our current recommendations. Based on simulation results, Gauntlet recommends decreasing the collateral factors for xcDOT in order to mitigate risk to the protocol. The collateral factors for WGLMR, WBTC.wh, xcUSDT, WETH.wh, and USDC.wh effectively balance risk with capital efficiency.

Frax Liquidity Pool

As mentioned in previous forum post, the FRAX market is experiencing a liquidity issue that the community is working to resolve. Gauntlet has put up previous proposals to mitigate the impact from the high utilization and we continue to monitor the market as the community works through the roadmap on the Nomad Incident Recovery.

xcDOT Top Supply Positions

4 out 5 of the top suppliers of xcDOT have borrow usage below 60%.

Gauntlet recommends decreasing the collateral factor for xcDOT from 60% to 58%. This decrease to collateral factor should not impact any users or cause any liquidations.

Moonbeam Liquidity

As we make recommendations through our risk models, we keep a constant check on the market liquidity and concentration risk to the Moonbeam protocol. In this regard, we would like to present some key liquidity figures for Moonbeam assets to share with the community. Since our last post, liquidity has decreased across all listed assets.

Asset Borrow Cap Relative to Supply Total Circulating Supply 25% Depth Token 25% Depth USD 25% Depth Token - Prev Month
WETH.wh 20% 2,517 25 $55,724 19
USDC.wh 80% 2,983,792 135,000 $135,000 180,000
WBTC.wh 26% 192 1.4 $58,499 1.5
xcUSDT 136% 957,399 140,000 $140,000 145,000
xcDOT 91% 936,083 27,000 $150,390 18,000
FRAX 99% 5,319,872 140,000 $140,000 145,000
WGLMR* 184% 12,252,369 500,000 $148,589 400,000

*WGLMR circulating supply doesn’t include non-wrap tokens.

IR Parameters

USDC.wh and xcUSDT Utilization

USDC.wh and xcUSDT Borrow Rates

Since MIP-M11 was submitted on-chain, borrow rates have averaged between 14% to 16%. Gauntlet recommends further IR Curve and reserve factor adjustments to help mitigate high utilization. We will be adjusting the Jump Multiplier, Multiplier, and Reserve Factor higher. The Jump Multiplier increase will push borrower APR at full utilization to 126% from 100%. The borrow APR at kink will increase from 5.19% to 6%. The recommended values will be the same across all stablecoins within the Moonbeam markets. We will monitor the impact of these adjustments and make any further adjustments if necessary to maintain utlization at the kink.

IR Curve for USDC.wh, USDT, and xcUSDC

Recommended Borrow & Supply APR for USDC.wh, xcUSDT and xcUSDC

Utilization Borrow APR Supply APR
0 0 0
80 6 3.84
100 126 100.8

Current Borrow and Supply APR for USDC.wh and xcUSDT

Utilization USDC.wh Borrow APR USDC.wh Supply APR xcUSDT Borrow APR xcUSDT Supply APR
0 0 0 0 0
80 5.19 3.53 5.19 3.53
100 100.2 85.17 85.2 72.42

The current incentives for USDC.wh and xcUSDC yield approximately 11% APY, making recursive positions profitable in the Moonbeam market even with high utilization. Gauntlet recommends increasing reserve factors to reduce Supply APY by 6%, encouraging users to maintain utilization at the kink. As shown below, the top borrow positions across xcUSDT and USDC.wh are recursive.

Additionally, this increase in reserves will support the community’s decision to enhance liquidity in the FRAX market as voted in the Options for Enhancing Liquidity in the FRAX Market on Moonbeam snapshot.

xcUSDC IR recommendations implementation was delayed from our previous recommendations in December. This proposal will work to rectify the curve to match it with other stablecoins.

Moonwell Moonbeam Projected Revenue KPIs

Projected Annual Reserves

Methodology

This set of parameter updates seeks to maintain the overall risk tolerance of the protocol while making risk trade-offs between specific assets.

Gauntlet’s parameter recommendations are driven by an optimization function that balances 3 core metrics: insolvencies, liquidations, and borrow usage. Parameter recommendations seek to optimize for this objective function. Our agent-based simulations use a wide array of varied input data that changes on a daily basis (including but not limited to asset volatility, asset correlation, asset collateral usage, DEX / CEX liquidity, trading volume, expected market impact of trades, and liquidator behavior). Gauntlet’s simulations tease out complex relationships between these inputs that cannot be simply expressed as heuristics. As such, the input metrics we show below can help understand why some of the param recs have been made but should not be taken as the only reason for recommendation. To learn more about our methodologies, please see the Helpful Links section at the bottom.

Supporting Data

The below figures show trends on key market statistics regarding borrows and utilization that we will continue to monitor:

Top 10 Borrowers’ Aggregate Positions & Borrow Usages

Top 10 Borrowers’ Entire Supply

Link to chart

Top 10 Borrowers’ Entire Borrows

Link to chart

Utilization Rate of Assets - Timeseries

Link to chart

Borrow Cap Utilization

Moonbeam 2% Market Depth for non-Stables

xcUSDC Check-in

xcUSDC liquidity on Moonbeam ecosystem has not ramp up since initial listing.

Collateral Usage - Timeseries

Risk Dashboard

The community should use Gauntlet’s Moonbeam Risk Dashboard to better understand the updated parameter suggestions and general market risk in Moonbeam.

Value at Risk represents the 95th percentile insolvency value that occurs from simulations we run over a range of volatilities to approximate a tail event.

Liquidations at Risk represents the 95th percentile liquidation volume that occurs from simulations we run over a range of volatilities to approximate a tail event.



Gauntlet’s Moonriver Recommendations

Simple Summary

A proposal to adjust 3 risk parameters:

Current Borrow Cap Recommended Borrow Cap
xcKSM 11,000 14,000
FRAX 300,000 420,000
WMOVR 76,000 40,000

Rationale:

WMOVR Cap Recommendations

Gauntlet’s risk model estimates VaR at $2.7k and LaR at $57.8k based on current user positions and market conditions. FRAX, WMOVR and xcKSM’s collateral factors are effectively balancing risk and capital efficiency.

With the recent appreciation of MOVR and the outflow of MOVR borrow positions from the protocol. Gauntlet has recommended to reduce the borrow cap of WMOVR to 40k. This reduction is meant to help to protect the protocol from the varying liquidity within the ecosystem.

The Moonriver market experienced a large increase in TVL because of recent asset price appreciation and large inflow of supply to the FRAX LP.

Supply & Borrow Balance USD


Link to Chart

Supply & Borrow Balance


WMOVR and xcKSM supply slightly decrease in supply during the rally while FRAX supply experienced a large increase. Relative to our last recommendation, supply across all 3 markets have increase. xcKSM and FRAX have experience a 80% and 97% increase in supply, respectively.

Collateral Usage

Collateral Usage experienced a large increase during the recent market rally.

Collateralization Ratios

Weighted average collateralization for assets have increased with relative price appreciation and changes in user positions during the market rally.

xcKSM Cap Recommendations

Gauntlet recommends increasing xcKSM Borrow Caps to 14,000 driven by the recent increase in supply.

xcKSM Circulating Supply and Local Liquidity

Furthermore, existing local on-chain liquidity are likely sufficient to support the increase in borrow caps.

Top xcKSM Suppliers

Most of the top users who use xcKSM as collateral haven’t borrowed anything or are recursively borrowing for the most part.

FRAX Cap Recommendations

Gauntlet recommends increasing Borrow cap to 420,000 FRAX in order to match increasing borrow demand for FRAX.

FRAX Local Liquidity

Furthermore, existing local on-chain liquidity are likely sufficient to support the increase in borrow caps.

Top FRAX Suppliers

The largest FRAX Suppliers are borrowing uncorrelated assets or are non-recursive in their borrows. Gauntlet will continue to monitor these positions.

IR Parameters

When Gauntlet analyzes interest rate parameters, we prioritize two main objectives:

  • Mitigating the risk of reaching 100% utilization in a pool.
  • Maximizing the growth of the protocol reserve to provide coverage for potential insolvencies or future expenses.

For this recommendation, Gauntlet advises against making any adjustments to the interest rate curves, as they are already optimized to achieve these objectives.

Moonwell Moonriver Projected Revenue KPIs

Projected Annual Reserves

Methodology

This set of parameter updates seeks to maintain the overall risk tolerance of the protocol while making risk trade-offs between specific assets.

Gauntlet’s parameter recommendations are driven by an optimization function that balances 3 core metrics: insolvencies, liquidations, and borrow usage. Parameter recommendations seek to optimize for this objective function. Our agent-based simulations use a wide array of varied input data that changes on a daily basis (including but not limited to asset volatility, asset correlation, asset collateral usage, DEX / CEX liquidity, trading volume, expected market impact of trades, and liquidator behavior). Gauntlet’s simulations tease out complex relationships between these inputs that cannot be simply expressed as heuristics. As such, the input metrics we show below can help understand why some of the param recs have been made but should not be taken as the only reason for recommendation. To learn more about our methodologies, please see the Helpful Links section at the bottom.

Supporting Data

The below figures show trends on key market statistics regarding borrows and utilization that we will continue to monitor:

Top 10 Borrowers’ Entire Supply

Link to chart

Top 10 Borrowers’ Entire Borrows

Link to chart

Utilization Rate of Assets - Timeseries

Link to chart

Borrow Cap Utilization

Liquidations on Moonriver

Link to chart

During the market rally, there was approximately $40k of collateral liquidated.

Risk Dashboard

The community should use Gauntlet’s Moonriver Risk Dashboard to understand better the updated parameter suggestions and general market risk.

Value at Risk represents the 95th percentile insolvency value that occurs from simulations we run over a range of volatilities to approximate a tail event.

Liquidations at Risk represents the 95th percentile liquidation volume that occurs from simulations we run over a range of volatilities to approximate a tail event.

Quick Links

Please click below to learn about our methodologies:

Gauntlet Parameter Recommendation Methodology
Gauntlet Model Methodology

By approving this proposal, you agree that any services provided by Gauntlet shall be governed by the terms of service available at gauntlet.network/tos.

2 Likes

Thanks for all the great work!!!

Can you please provide definitions for BASE, Kink, Multiplier, and Jump Multiplier and describe how they effect the IR curves?

I think I saw somewhere that VaR means Value at Risk and LaR means Liquidity at Risk, but can you confirm that and tell us how you calculate them, please?

It would be great to get these definitions up on the Moonwell Gitbook for everyone to reference.

Thanks again!

2 Likes

Great suggestion @Curly! I’m having my definitions reviewed now for accuracy and will add to the IR Curve section of the Moonwell Docs ASAP.

1 Like

Hey @Curly! Just wanted to let you know that IR Curve definitions have been added to the Moonwell Gitbook. Thanks again for the great suggestion.

2 Likes

Hi @Curly,

The definitions have been included by @majin in the Gitbook. For convenience, they are reiterated below:

Base

The base interest rate in a lending and borrowing protocol. It’s the minimum interest rate that borrowers pay and lenders receive.

  • Effect on IR Curves: In an interest rate curve, the Base is the starting point. As demand for borrowing increases or decreases, the actual interest rate moves away from this base rate according to preset rules or algorithms.

Kink

The point at which the interest rate model changes slope.

  • Effect on IR Curves: At the kink, the behavior of the interest rate curve changes. Below this point, the platform aims to encourage borrowing with lower rates. Above this point, the rate increases more steeply to discourage additional borrowing and to encourage repayment.

Multiplier

The multiplier in interest rate models is a factor that determines how steeply the interest rate increases with utilization rate.

  • Effect on IR Curves: The multiplier affects the slope of the interest rate curve. A higher multiplier means that interest rates increase more rapidly as the utilization rate approaches and surpasses the kink.

Jump Multiplier

The jump multiplier is an additional factor that comes into play when the utilization rate surpasses the kink.

  • Effect on IR Curves: This multiplier causes a significant jump in interest rates beyond the kink point, creating a steep increase in rates to prevent liquidity shortages and manage risk.

What is VaR?

In quantitative analysis, Value-at-Risk (VaR) is a popular metric for the overall riskiness of a system. Given a range of possible profit or loss outcomes, VaR represents the losses incurred in the worst percentiles of the distribution. As a straightforward way to quantify tail risk scenarios, VaR is widely used by market participants ranging from DeFi protocols to major traditional institutions.

For risk management customers, we use VaR to communicate levels of market risk and changes to the level over time. Gauntlet’s dashboards display daily VaR calculations as a topline metric for community information. Frequently, we also highlight changes in VaR due to parameter adjustments or market events that materially affect protocol risk.

What is LaR?

Liquidations at Risk are very similar to VaR in formulation, except instead of measuring the net insolvent amount, we instead measure the amount of liquidations at the 95th percentile.

Liquidations, while healthy for the protocol, can adversely affect Borrower UX. Specifically, when lowering Collateral Factors/LTVs/Liquidation Ratios for given collaterals some users’ could be forcefully liquidated.

2 Likes

Beautiful!!! Thanks so much for laying all that out so clearly…I feel like I gained a full IQ point

1 Like

Review of Gauntlet’s BASE recommendations.

Warden has reviewed Gauntlet’s recommendations and supports the proposal. We also propose further updates on Base deployment to unlock more rETH borrow activity and reduce protocol exposure to DAI market.

Review

  • USDC borrow cap increase - USDC on-chain liquidity on Base is healthy enough to support a borrow cap increase from 8.5M to 12M without significantly increasing risk for the protocol to accumulate bad debt.
  • cbETH IRM update - We also acknowledge that utilization for LST markets is currently not optimal. We think it is a good measure to apply the proposed IRM change only to cbETH to start with. It will be an opportunity to run an A/B experiment with other LST markets as control group in order to better measure rate elasticity for such market. If results are positive, next step will be to apply the same IRM updates to wstETH and rETH.
  • LST collateral factors increase - Proposed collateral factor changes do not significantly increase the risk for the protocol to accumulate bad debt in a worst case scenario. Time available to liquidate each asset a worst historical price drawdown scenario is still well over Warden’s recommended buffer of 60 minutes. The proposed collateral factor increases are a good opportunity to improve markets utility while not significantly increasing risk for the protocol to accumulate bad debt.

Additional Recommendations

  • Reduce DAI market usage - Reduce DAI supply-side rewards in order to reduce max recursive yield to 20% APY at 5.5M total supply (current recursive yield is 25% APY at 6.6M total supply, which is more than all other stables). We will follow-up with specific parameters to achieve this target.
  • Tighten DAI caps - Reduce DAI supply cap to <75% circulating supply (<5.5M) once usage decreases after above change is rolled out.
  • Increase rETH borrow cap - Increase rETH borrow cap from 100 to 200.
  • wstETH and rETH IRM updates - If IRM change for cbETH proposed by Gauntlet drives positive impact, we proposed applying the same IRM updates to rETH and wstETH markets as well.

Detailed Analysis

Notable market conditions changes since last update

  • rETH borrow cap reached - rETH borrow cap has been filled over the last 3 days by users entering recursive rETH positions. We recommend increasing rETH borrow cap from 100 to 200 to accomodate the sudden increase in demand for borrowing. The proposed borrow cap change does not significantly impact risk for the protocol to accumulate bad debt.

  • DAI low circulating supply / high caps usage - DAI circulating supply on Base has decreased by 2M. (9.5M → 7.5M). DAI supply cap is now equal to circulating supply, which is not optimal from a risk perspective. Also, DAI supply and borrow caps are currently used at 90% and 85% respectively. High cap usage hinders our ability to tighten the caps. We will reduce DAI reward rates to reduce recursive yield, which should help decrease cap usage. Once usage starts diminishing, we recommend to gradually lower DAI supply and borrow caps to target <75% circulating supply.

  • Improved Native USDC DEX liquidity - USDC DEX liquidity has now reached roughly the same level as USDbC. Increased tradeability allows for more USDC supply and borrow positions to be onboarded with relatively unchanged risk exposure for the protocol.


  • Reduced rETH liquidity - rETH liquidity depth on Base has diminished by 50% during the last 30 days. This change does not have significant impact on ability to quickly liquidate sizeable rETH collateral positions. In fact, current liquidity level allows liquidating 10% of the supply cap ($150k out of $1.5M) profitably in a single transaction.

  • Low LST markets utilization - Market utilization has been constantly low relative to IRM kink for LST markets over the past month (~15% vs. 45% kink).

Source: Search Markets | Warden Finance - Risk & Analytics for DeFi

2 Likes

Thanks @WardenFinance,

Gauntlet used Cap Guardian to modify DAI and rETH cap (see here).

  1. DAI Cap Adjustments:
  • Supply Cap: Decrease from 7.5 million to 6.5 million.
  • Borrow Cap: No change, remaining at 6 million.
  • Reasoning: Despite a high borrow cap usage, the circulating supply of DAI has dropped significantly. The decrease in supply cap aims to reduce risk by not accumulating bad debt.
  1. rETH Cap Adjustments:
  • Supply Cap: No change, remaining at 600.
  • Borrow Cap: Increase from 100 to 200.
  • Reasoning: This is due to an increase in rETH’s circulating supply and holders, suggesting a healthier market condition.
2 Likes

In order to help mitigate risks associated with low DAI circulating supply, we have reduced DAI supply-side reward speeds. The update aims to reduce overall demand for DAI lending and borrowing, especially for recursive lending which occupy a large portion of the DAI supply and borrow caps.

The applied update aims to keep DAI total supply on Moonwell below 5.4M, or ~75% of DAI circulating supply.

DAI rewards have been shifted towards USDC and ETH markets. Both markets have sufficient on-chain liquidity to sustain growth.

Epoch # Total rewards Start timestamp Start date Duration
6 15,754,807.69 WELL
36,812.46 USDC
1703887200 Fri Dec 29 2023 22:00:00 GMT 28 days
Reward distribution (% of allocation) cbETH DAI USDbC USDC ETH wstETH rETH
WELL supply 20% 18%→12% 0% 20%→22% 36%→40% 4% 2%
WELL borrow 0% 0% 0% 0% 0% 0% 0%
USDC supply 20% 18%→12% 0% 20%→22% 36%→40% 4% 2%
USDC borrow 0% 0% 0% 0% 0% 0% 0%
2 Likes