CRV-long LlamaLend Market Next Steps

Summary

The CRV-long LlamaLend market became illiquid today after having accumulated substantial bad debt during a volatility event on October 10, 2025. Due to the shortfall, it is likely to remain illiquid. Therefore, mitigating action is being taken to minimize the accumulation of additional bad debt in the market by reducing borrowing rates to 0. Also included here are immediate deprecation items for this market and for similarly affected markets. A subsequent proposal will present a repayment plan, which may be considered and enacted at the DAO’s discretion.

Current Market State (April 17, 2026)

The CRV/crvUSD one-way lending market currently carries ~754K crvUSD in bad debt (net shortfall) across 38 underwater positions. Those positions together owe ~1.67M crvUSD in gross debt — 49.4% of the market — but still hold ~919K crvUSD of collateral (crvUSD accumulated in the AMM from soft-liquidations) that will be applied against repayment. On April 17, a large lender withdrawal of ~728K crvUSD pushed utilization to 100%, spiking borrow rates to the max rate of 293%.

Metric Value
Vault 0xCeA18a8752bb7e7817F9AE7565328FE415C0f2cA
Controller 0xEdA215b7666936DEd834f76f3fBC6F323295110A
AMM 0xafca625321Df8D6A068bDD8F1585d489D2acF11b
Current monetary policy 0x8a7138cF49AffCd35D05dBE5a160Dac870E4B6F5
Max borrow rate 293%
Total debt ~3,389,576 crvUSD
Active loans 99
Utilization 100%

Background

The CRV lending market experienced substantial bad debt of ~$500k during the extreme volatility event on October 10, 2025. It has since experienced a gradual build-up of bad debt driven primarily by CRV price decline and accumulating borrowing rates. The bad debt in the market now stands at around $754k.

Bad-debt trajectory

On-chain scans at specific historical Ethereum blocks give a clean before/after picture of the Oct 10 crash and the subsequent six months:

Date Block CRV Price Underwater Bad Debt (shortfall)
Oct 10, 2025 00:00 UTC (pre-crash) 23,543,616 $0.7327 9 (dust) ~$1
Oct 11, 2025 00:00 UTC (post-crash) 23,550,763 $0.5202 53 ~$491K
Apr 17, 2026 13:21 UTC (today) 24,899,083 $0.2334 38 ~$755K

The Oct 10 crash created ~$491K of bad debt essentially instantly. Benny’s Oct 2025 governance post references ~$539K for the CRV LlamaLend market, consistent with a few days of additional drift after our Oct 11 snapshot. From there, the shortfall has grown ~54% to over $750K today — driven primarily by CRV’s continued decline from $0.52 to $0.23, with a meaningful amount also coming from interest accumulation.

Sources of the bad debt growth

Using the arithmetic identity shortfall = debt − collateral_value, the delta between Oct 11 and today for this cohort decomposes cleanly:

Component Δ Share
Interest accrual on outstanding debt (debt(user) grew $1.59M → $1.67M) +$83K ~30%
Value lost to CRV price decline +$198K ~70%
Total shortfall growth ($473K → $755K) +$282K 100%

The collateral mechanism — and why soft-liquidation actually helped. On Oct 11 these 37 positions still held 744,838 CRV worth ~$387K at the prevailing $0.52 oracle. As CRV fell from $0.52 → $0.23, the LlamaLend AMM progressively soft-liquidated those tokens band-by-band, selling each band’s CRV at that band’s price as the oracle crossed through it. Total crvUSD proceeds realized: ~$189K (effective average sale price ~$0.254).

The $198K in the table above is the gap between “744,838 CRV valued at Oct 11 oracle ($387K)” and “proceeds actually realized ($189K)”. It is not a cost of soft-liquidation — it is the cost of CRV price declining 56%, merely crystallized through soft-liquidation as the price fell. Critically, soft-liquidation was a net benefit here: if it did not exist and those 744,838 CRV had simply stayed in the position, they would be worth only ~$171K at today’s $0.23 oracle — $18K less than what the AMM actually realized. By selling into each falling band, the AMM locked in higher prices along the way down.

The collateral side is now exhausted. These 37 positions hold zero CRV collateral remaining — all of it has been converted to crvUSD. No further price risk can be realized because there is no more CRV to sell. From this point forward, the shortfall can mainly grow through interest accrual on the outstanding debt.

No new borrowers are contributing to bad debt

Tracing individual addresses across the Oct 11 and Apr 17 snapshots:

  • 39 new borrowers have opened positions since Oct 11. All 39 are healthy today, carrying ~$1.49M in debt. The tightened loan discount and liquidation discount from the prior actions effectively prevented any newly-originated position from ending up underwater, even as CRV price roughly halved.
  • Of the 53 positions underwater on Oct 11: 13 closed (repaid or fully exited), 3 actively added collateral and recovered to healthy status, and 37 remain stuck underwater today. Those 37 account for essentially the entirety of the current ~$755K shortfall (the 38th underwater position today is a dust account with $0 debt).

Loan Breakdown

Category Loans Gross Debt (crvUSD) Share
Healthy 61 ~1,715,759 50.6%
Underwater 38 ~1,673,818 49.4%

The 38 underwater positions hold ~919K crvUSD of collateral (crvUSD-in-AMM from soft-liquidations) against their ~1.67M gross debt. The resulting bad debt (shortfall) is ~754K crvUSD — the amount that cannot be recovered from collateral at current prices. See the appendix for the full per-position breakdown.

Previous Governance Actions

Over the past months, several governance actions were taken to manage the situation:

Prior Actions (October 2025 – March 2026):

  1. Increase Loan Discount to 20% (Oct 2025) — Following the October 10 market crash, loan discount was increased from 11% to 20% to reduce maximum borrowing power against CRV collateral. Accompanied by LlamaSimulator-based re-parameterization analysis suggesting A≈68.
  2. Modify Liquidation Discounts (Oct 2025) — Coordinated parameter update across LlamaLend markets to improve risk alignment post-crash. Increased liquidation discount to ensure positions enter liquidation at higher LTV, providing more buffer for successful liquidation processing. Changes apply on next user interaction (no retroactive liquidations).
  3. Temp Check: Use Treasury Funds to Repay Bad Debt (Oct 2025) — Benny proposed using DAO Treasury funds to repay bad debt. LlamaRisk recommended a targeted ~100K crvUSD repayment across 7 addresses to bring positions to P_i ≤ 1, making them liquidatable at ~$1 CRV, with monthly reassessment rather than full treasury drain.
  4. Prepared 0% monetary policy contract (0x8b1c4d2718ba8d1be499d679687454e1db5ace19) (Mar 2026) — A standby monetary policy contract was deployed that outputs a ~0% rate, to be activated in the event of a bank run to prevent bad debt from compounding.
  5. Bad debt banner added to the lending market frontend.

Next Steps

Step 1: Set monetary policy rate to ~0% (eDAO — immediate)

The eDAO has been notified to switch the monetary policy to the prepared 0% contract (0x8b1c4d2718ba8d1be499d679687454e1db5ace19).

Rationale: With rates spiking above 290% APR, the ~1.67M in gross debt owed by underwater positions compounds rapidly — inflating the unrecoverable shortfall (~754K today) with every block. Since these positions are deeply underwater (health as low as -78%), they will never be repaid regardless of rate. Halting interest accrual prevents the shortfall from growing further, which only worsens the solvency ratio for lenders.

Tradeoff: With rate at 0%, the 61 healthy borrowers (~1.72M crvUSD debt) have no cost incentive to close their positions. However, this is accepted as a necessary step given the severity of the bad debt situation.

The cost of delay: At the current 138% borrow APR, the ~$1.67M gross debt owed by underwater positions accrues roughly $1.67M × (1.38 / 365) ≈ $6,300 per day in interest. Interest flows directly into the unrecoverable shortfall:

Duration Additional bad debt
1 day ~$6.3K
1 week ~$44K
30 days ~$190K

Switching the monetary policy to ~0% caps the shortfall at approximately its current ~$755K level. Every day the change is deferred adds roughly $6K of bad debt that lenders will have to absorb.

Step 2: Set loan discount to maximum (DAO vote)

Minimizes borrowing against CRV collateral. This is a procedural market deprecation action to minimize further exposure, although it does not have an immediate impact on existing user positions.

Step 3: Kill the CRV/crvUSD gauge (DAO vote)

Remove gauge emissions to stop incentivizing deposits into a market being wound down. This follows the same deprecation procedure used for previous market wind-downs.

Other Markets

Alongside the CRV-long deprecation, we reviewed the broader set of LlamaLend markets carrying bad debt. Cross-referencing the Curve bad-debt endpoint we recommend taking the same course of action for the following markets.

Chain Market Bad Debt USD AMM Gauge set is_killed to true set loan_discount to 50% set borrow rates to 0
fraxtal FXS-long 144,436.61 0xe8d7Ce159EFB40dD332902B5B23400Cba5938445 0xfc6891c8482aef6aef71a09a09ce14432617a403 yes yes yes
fraxtal SQUID-long 130,812.17 0x8feCf70B90ED40512C0608565EB9A53af0Ef42Ad 0x0bd2980cdc585b489dfbfce6254a2ee6ed51bccb yes yes yes

Appendix: Full Position List

Snapshot at block 24,899,083 (April 17, 2026). Sorted: underwater (by debt desc), then healthy (by debt desc). Rows where Status = bad are underwater positions contributing to bad debt.

Price source for collateral conversion: “Total Collateral (crvUSD)” converts CRV to crvUSD at $0.230566 using the AMM’s on-chain price_oracle() — the protocol’s canonical EMA feed over external CRV/USD DEX liquidity. This is the same rate the LlamaLend market itself uses to compute position health and trigger soft-liquidations, so the shortfalls shown here match the market’s internal view. Using the raw AMM spot instead would understate collateral for positions already mid-liquidation.

Borrowers Table
# Status Borrower CRV Collateral crvUSD Collateral Total Collateral (crvUSD) Debt (crvUSD) Health Shortfall (crvUSD)
1 bad 0xD76AA47f1Ad6ce7f78ddCD4643A85C87B7e7C047 0.00 196,972.58 196,972.58 374,247.01 -51.58% 177,274.43
2 bad 0x82925fbee7060FB6B23600A92Ff1C04735DE0b99 0.00 188,369.15 188,369.15 334,053.95 -48.12% 145,684.80
3 bad 0xe69A49F8F9C1d66fefA66Dfb61155c06606986D1 0.00 114,031.68 114,031.68 222,913.61 -52.94% 108,881.93
4 bad 0x6bE91cbC96071e3F1Cb79b50deF8D59a016Fa1c6 0.00 131,314.29 131,314.29 183,322.78 -34.10% 52,008.49
5 bad 0x6e22c49c85de053107fEA6E09ffFbC4b5029C5f1 0.00 64,525.85 64,525.85 128,338.65 -53.74% 63,812.80
6 bad 0x6620912AC168458ED84D5D3d6c83E728abf9941F 0.00 49,286.55 49,286.55 94,921.83 -52.23% 45,635.28
7 bad 0xFfb3Cad58bb14eeDDfDd077c99c73e96Feb1602a 0.00 44,797.04 44,797.04 84,600.68 -51.28% 39,803.64
8 bad 0x1cABC2a3E7fD58B16502bF7c60c5a5e41b9Db638 0.00 33,438.85 33,438.85 63,962.27 -51.90% 30,523.42
9 bad 0x3b96f28e3B45a19FA3e89232D0680E54872D3d58 0.00 28,588.87 28,588.87 56,322.37 -53.30% 27,733.50
10 bad 0x1Fdf62394aBDa67BE5346f9eA426A447Bf602Fcc 0.00 13,211.82 13,211.82 26,244.68 -53.69% 13,032.86
11 bad 0x72Ec1Fd9929A854719d9A1a2aFc7fFcd7C0C7DeE 0.00 11,134.08 11,134.08 21,394.60 -52.12% 10,260.52
12 bad 0x2400a45243e6B13dbaF549FCAc88Eda203df80D2 0.00 10,734.49 10,734.49 21,105.47 -53.21% 10,370.97
13 bad 0x802d8Db785bD165125D5907A5F67Ce3E4012C384 0.00 6,986.05 6,986.05 13,698.26 -53.08% 6,712.21
14 bad 0x0fc6144F9eA35EE67Ee4e38c5Ea47B2960C13Ccc 0.00 6,153.00 6,153.00 10,274.85 -44.91% 4,121.84
15 bad 0x21F4aA70F4C375b6Ee91570A1e1D86FDE4Cad098 0.00 5,020.24 5,020.24 9,649.79 -52.14% 4,629.54
16 bad 0xA903f50Fe59B8939443e818b2451f4f1e021BcE4 0.00 3,226.91 3,226.91 6,192.36 -52.06% 2,965.45
17 bad 0x28949AEdec812949Ff872bF13dA323Ce5bc82728 0.00 2,819.18 2,819.18 5,611.75 -53.78% 2,792.56
18 bad 0x36bF825269bCD6988a491c04053e0855d445db97 0.00 2,553.85 2,553.85 5,175.93 -54.61% 2,622.08
19 bad 0x38F4362d888985a8ef4B9Cc9E51277C13Aca5C37 0.00 1,927.99 1,927.99 3,683.99 -51.85% 1,756.00
20 bad 0xbbc84a064d7FD00F2383A6b0faFf8c78C7517eB2 0.00 1,319.49 1,319.49 2,666.99 -54.48% 1,347.50
21 bad 0x50de06228FB5b6ba51e4d40AA93d16B9BeFBaE7E 0.00 1,307.00 1,307.00 2,484.17 -51.60% 1,177.17
22 bad 0xcbcE52b5576771C7c8F5C21E29640D29E9636a8F 0.00 1,230.72 1,230.72 2,484.16 -54.42% 1,253.45
23 bad 0x1b7A9C831B4f2088FA6eaE337088ae2Fc9913ef5 0.00 197.19 197.19 369.27 -50.87% 172.08
24 bad 0x7Ae37200fbc08e1cb786059934f671fbE62d7c0f 0.00 35.95 35.95 71.19 -53.54% 35.24
25 bad 0x9Fc1a0601872762B5C8B2c602C2457A55Bb0d80f 0.00 5.88 5.88 11.52 -53.02% 5.64
26 bad 0xEbD31ba78255244ff183BdeE468311074533149e 0.00 5.81 5.81 11.40 -53.09% 5.58
27 bad 0xaC090a9513D5f8654262C327aD554EB5979B6C62 0.00 0.32 0.32 1.21 -75.83% 0.89
28 bad 0x127656ba749DA2Cc8A41a1a87C73c2c5ea801728 0.00 0.44 0.44 0.90 -55.42% 0.46
29 bad 0x7319aC39755c0FB8afF281aB07aE50FCF67FD3CA 0.00 0.28 0.28 0.73 -64.63% 0.45
30 bad 0x1a1051c39652C0FF81A819DA4283871c3f076a3A 0.00 0.38 0.38 0.55 -37.28% 0.18
31 bad 0xF80b3C8343E4763B89B3B3915faECF1E853C5F7a 0.00 0.17 0.17 0.53 -71.04% 0.37
32 bad 0x9858218c248864698e9F3454c5F828761094D190 0.00 0.08 0.08 0.18 -59.52% 0.10
33 bad 0x13e0f5bEff1AB88Be9A8735e75a46DAB85D5C17A 0.00 0.04 0.04 0.14 -73.86% 0.10
34 bad 0xB2A8beB4bcb2A526AC358407F2994278fc5A2C5D 0.00 0.01 0.01 0.02 -59.51% 0.01
35 bad 0xf35FD0AeC586bd543ff48ca113C7BDaF68152213 0.00 0.01 0.01 0.02 -49.62% 0.01
36 bad 0x764B3aAf71891c351B83fE3D4C8Fbcb7609AA56b 0.00 0.00 0.00 0.00 -50.45% 0.00
37 bad 0xFac75d2e2FC279e4d311a9E61B0D7c5C245A6Ba0 0.00 0.00 0.00 0.00 -77.27% 0.00
38 bad 0xC729770162F63dF2b04dD2864eEFA4471EF9a088 0.00 0.00 0.00 0.00 -30.90% 0.00
39 ok 0xc724505Fe1BcEaEb8b7D00899b834460F07f3f6E 8,097,587.66 0.00 1,867,029.65 958,616.87 56.96% 0.00
40 ok 0x235eD541655d8A13D27c5a1Da1Cd7C9863C03453 900,074.36 0.00 207,526.68 125,454.17 27.38% 0.00
41 ok 0x85522E08937910B77eF7B46dbba0f8425AaD58Ff 952,007.51 0.00 219,500.71 104,278.68 72.32% 0.00
42 ok 0x1Aa53CB3400B7733B32Fd8405e3784803eFF5012 441,289.79 16,659.10 118,405.59 67,892.44 9.64% 0.00
43 ok 0x79e77Ff9041A8a4ccDdEf45Dfa98f0d5127871E2 500,000.22 0.00 115,283.13 56,986.71 75.65% 0.00
44 ok 0x96F03285532392Df68C9332B8DB2c30de167617D 616,250.32 0.00 142,086.47 56,579.80 112.68% 0.00
45 ok 0xc0fc9fFA2510e3F1AaB463dF2e8D58dED5Ae7919 417,866.41 0.00 96,345.85 41,639.34 83.50% 0.00
46 ok 0x923D2f38fd86Bc66478b990168b8E0D4D64d1FC8 272,902.96 0.00 62,922.19 27,795.59 88.01% 0.00
47 ok 0x7451d7Dc1570f2Be4aA3ECA8d482dFF117570AB7 452,436.24 0.00 104,316.48 26,235.37 259.75% 0.00
48 ok 0x3112aFc6Cb984eC88E3eDBdFbc32727a15f8A592 250,218.06 0.00 57,691.82 23,913.44 117.16% 0.00
49 ok 0x4810a072DF31cf62915f10cE0a2fAb8CFfFf8DDa 231,542.00 0.00 53,385.75 20,995.44 116.64% 0.00
50 ok 0xAea0884559deA9fA3f80bDD9f045c5658B91Adb0 312,755.38 0.00 72,110.80 20,021.18 106.68% 0.00
51 ok 0x0E4c3DC2B13Ef1B9b79d940CD64Cc6C54a4C7b43 284,471.77 0.00 65,589.56 19,923.84 190.77% 0.00
52 ok 0xd6474c019c95894Da396Fd7a89335A972E93Fd3b 149,102.79 0.00 34,378.06 18,458.56 48.85% 0.00
53 ok 0xAF1f459738f427247617E139C3c0ca4311DAF14A 89,558.13 92.26 20,741.34 14,651.74 5.11% 0.00
54 ok 0x16343015dDc0c11a3bBc3EA6Bf52b91493DDd55d 202,000.00 0.00 46,574.36 13,750.67 200.49% 0.00
55 ok 0x75d32cA496B239ee9E7F3D6e6FBb3813Ab14aAaC 102,093.97 0.00 23,539.41 11,818.93 60.51% 0.00
56 ok 0x37223d079dEDc6A96d6415EE4850748D226eE5BF 119,283.27 0.00 27,502.68 11,687.59 96.76% 0.00
57 ok 0xe5Ae5AAc9A19Ec43aBdd161e95640A1cc12aE348 100,000.00 0.00 23,056.62 10,066.85 104.56% 0.00
58 ok 0x66B0d6cD3A7704baEeB42A78122311949ac3B743 58,115.60 0.00 13,399.49 7,615.84 52.10% 0.00
59 ok 0xB4d44660d47d6Acd276FcDC64d29d064a66a7D90 76,079.75 0.00 17,541.42 7,574.36 93.65% 0.00
60 ok 0x44Fa1B7B7aCc00f6C81539A8Fe255794edf7C03c 42,285.34 0.00 9,749.57 5,770.43 31.40% 0.00
61 ok 0xb0f2ac3A0C526dB4f01AA19726c1703EBF52CB0c 45,823.71 0.00 10,565.40 5,769.81 45.08% 0.00
62 ok 0x24F41E16170A05419769FE84963688D1c1B4DCB5 45,400.00 0.00 10,467.70 4,433.78 98.70% 0.00
63 ok 0x48309a3fDdC71b16169e80795ce20A81fbDAbDc6 34,997.36 0.00 8,069.21 4,338.65 56.99% 0.00
64 ok 0x09a6973B9571e1F6fd02E9749EEA3c1D3607154F 67,199.76 0.00 15,493.99 4,167.89 247.85% 0.00
65 ok 0x2c3ed4d8dD57c725EC5E2775E2eEa1E9B8dD66d0 43,113.22 0.00 9,940.45 3,969.76 112.07% 0.00
66 ok 0x1adb5493758495dF95192372D7AA604E7095d394 47,597.28 0.00 10,974.32 3,691.80 158.85% 0.00
67 ok 0x68d692b40dbe8170fb8DAa6935e1cB05979e126A 63,470.00 0.00 14,634.03 3,568.08 157.52% 0.00
68 ok 0x9c8cAf318DB77d2D760988eFD0B3edEC5618Bc66 21,508.40 0.00 4,959.11 2,814.91 52.30% 0.00
69 ok 0xFd085E6d5ADF648D9B68C7F679F28873EB3D2FdB 29,178.34 129.14 6,856.68 2,783.25 6.82% 0.00
70 ok 0xda986d4B5dDC40045C7DCCe6605c96663FC7a814 0.00 3,250.38 3,250.38 2,775.63 3.05% 0.00
71 ok 0x2618F4c64805526a3092d41f25597CcfE4Dd8216 22,479.38 0.00 5,182.99 2,605.62 60.31% 0.00
72 ok 0x4aEE1ff9b46882af5efa2bF4bc177f770c27E51F 32,374.75 0.00 7,464.52 2,568.39 153.07% 0.00
73 ok 0xC3d09017A4D34006bB6269f164E41d3088a5cE03 30,000.00 0.00 6,916.98 2,338.96 157.51% 0.00
74 ok 0xA78E67E364918B3C719200b85B80D0509d8220C5 36,705.10 0.00 8,462.95 2,319.01 114.60% 0.00
75 ok 0x3aE13Eca1f6d3B1bFB2d8bF1D537eBae2FBFf071 21,253.94 0.00 4,900.44 2,128.33 92.52% 0.00
76 ok 0x1112770b24396dB6fd45a9548FD14c95B068Ec21 50,124.04 0.00 11,556.91 2,087.55 301.77% 0.00
77 ok 0xf1c7f8e68C594A7b87dd3d22057FED15d7fB63d7 41,234.18 0.00 9,507.21 2,037.09 329.19% 0.00
78 ok 0xc3c33437B840Fee4e5C9F52d16D67A3569fF4885 8,737.68 0.00 2,014.61 1,119.81 41.25% 0.00
79 ok 0x71c7033aa9f6C67da295975133dE1fc44edb6310 10,908.62 0.00 2,515.16 910.94 137.92% 0.00
80 ok 0x06be41998571cf35f11d8Ea6F8eB53Eb5f0829CB 8,908.86 0.00 2,054.08 907.92 63.75% 0.00
81 ok 0x9A9AB4d9f84d9c6797a963445D0A369A4796E0fF 11,297.66 0.00 2,604.86 833.47 60.00% 0.00
82 ok 0xe41Af4FAAAE12df9EEf7D09A2B14bD6721536fF5 8,978.84 0.00 2,070.22 814.80 127.68% 0.00
83 ok 0x1c0C0169b46c37A826fCD8F40aB76FF5f85AcA8a 6,481.26 0.00 1,494.36 700.26 75.98% 0.00
84 ok 0xe5b2a2459914B78F52bEF1d294605EfC91E311a3 4,589.21 0.00 1,058.12 680.08 18.22% 0.00
85 ok 0xB8884F6E8f676c1CA8531A7acb56982a6FD5a8de 10,150.91 0.00 2,340.46 652.94 105.69% 0.00
86 ok 0xd36aC1B46CB03AeDf8604507301d7f3FF6146566 7,398.82 1.52 1,707.44 652.92 9.88% 0.00
87 ok 0x29de71aD7Bb4427E632656698D3926e634e43e88 8,482.96 0.00 1,955.88 542.13 222.74% 0.00
88 ok 0x440d880905a7402986C8D8593ECD41f92Cf47aa2 7,790.58 0.00 1,796.24 510.01 178.36% 0.00
89 ok 0x5d8f75fcA7Db569BAbf7b4ff4C7F49f3f8784E97 8,000.00 0.00 1,844.53 507.07 113.91% 0.00
90 ok 0xEd477e6921D6b9F90984B072D77a9Cf1856783F9 3,000.00 0.00 691.70 361.16 61.43% 0.00
91 ok 0xa7dD77a2EA51f1A1fFEdbac99e0a2882906A02Cd 3,442.63 0.00 793.75 323.37 107.88% 0.00
92 ok 0xC2A0B6Ed78843a3F91fb5BA00B4CD139131E16AD 934.53 0.00 215.47 68.41 176.94% 0.00
93 ok 0x67D12edbC28384E9dd8E3211F4464e421A3935F9 398.73 0.00 91.93 45.73 77.25% 0.00
94 ok 0x7C1EF124eA6b78FB6E0b82DF2f32Fdd2208f8636 10.00 0.00 2.31 1.23 57.92% 0.00
95 ok 0x158BF6bE282000DEC7820a5Ec90c59767FBc527B 51.62 0.00 11.90 0.00 >1000% 0.00
96 ok 0x4137aFcC5E6dDFc5380298bB5A753AbaD908C21F 5.00 0.00 1.15 0.00 >1000% 0.00
97 ok 0x83F4E92d60d3b9160a39f95106587EC3dF293F67 0.06 0.00 0.01 0.00 191.44% 0.00
98 ok 0x6058b5031A34D7F1135df1402eB2081D1Cd68815 0.01 0.00 0.00 0.00 77.29% 0.00
99 ok 0x2693eacb6Aa92B3Cad7Dfd9E057d00ee756Bf5cA 3.08 0.00 0.71 0.00 >1000% 0.00
2 Likes

The eDAO has set the monetary policy so borrow rates in the market are now 0%, preventing additional bad debt accumulation.

A vote is live to minimize borrow power in the market (loan_discount → 50%) and to kill CRV emissions to the market.

1 Like

I’m one of the OK borrowers in this Market. I meant no harm in opening my position and am sorry for any difficulty it may have caused. How will this vote to adjust the loan_discount value and to kill CRM emissions effect my position? Will it just sort of sit there until I get out of it?

as a lender, should’ve withdrawn when i had the chance. :frowning:
what’s going to happen next? will the lenders made whole?

loan_discount only affects new loans, so your position won’t be affected, you will essentially get a interest free loan for the forseeable future.

1 Like

When are lenders going to know when and how they will be able to recover their funds?

1 Like

Thanks to LlamaRisk for moving quickly on this — the 0% policy, loan_discount reduction, and gauge kill together look like the proper market wind-down. Worth noting #1391 now has ~117M veCRV FOR with no opposition and just crossed quorum after Stake DAO’s vote today; on current pace it should pass cleanly before the 25th.

To sevenz’s question — a few honest observations from reading the vault state on-chain:

  • After #1391 passes, the market is effectively frozen in “run-off” mode: no new borrows, 0% interest, existing positions can self-repay or sit. saintrat is right that current borrowers
    are unaffected (and yes, effectively get a free option).
  • Recovery for lenders therefore depends entirely on (a) whether CRV rallies enough to push borrowers back above water, (b) voluntary repayments, and (c) any post-wind-down cleanup the
    DAO decides to fund. (a) and (b) are passive; only (c) is actionable.
    On (c), I’d like to float a question to LlamaRisk and the broader community: would a revenue-linked cleanup, sized and scoped to avoid bailout dynamics, be in scope for a follow-up vote?

Concretely, something like:

  • A fixed allocation (e.g. 5-10%) of crvUSD revenue for N months, routed to a purpose-built repayment contract.
  • Capped at a specific USD figure (~$100K range feels proportionate to the residual gap).
  • Targeted narrowly at the underwater addresses already identified in LlamaRisk’s Oct 2025 temp-check.
  • Auto-terminating once the cap or duration is reached — no open-ended commitment.

The framing that seems to meet Michwill’s stated “salvaging should not be guaranteed… done in a way to NOT set up a precedent” bar (Oct 24 comment in the Temp Check thread): this isn’t treasury-funded, isn’t precedent-setting for live markets, and only activates after the market is formally wound down.

It treats cleanup as a discretionary, bounded, post-mortem action — not a guarantee

I also have some on-chain data that might help frame sizing — per-band recovery curves vs. CRV price, borrower distribution, liquidator activity since the freeze — happy to share or write up a formal proposal draft if there’s appetite. Also genuinely interested to hear LlamaRisk’s view on whether a mechanism like this is worth pursuing or whether the pure run-off path is preferred.

2 Likes

Lender here, I’m Ok with proposal #1391 being pushed, but what will happen next?

Looks unbalanced to me that borrow rate is set to 0% indefinitely - this gives no incentive to borrowers to repay their position.

Given that time is money, a solution to this shouldntt be a slow-burn. I’d like to access & reinvest my leftover cash, so I propose a liquidation of all positions (good or bad) and a 76% repayment to all lenders asap. You can sort the remaining 24% later, if ever.

Otherwise, how do you plan for lenders to ever exit this market?

3 Likes

Next steps look good to get this under control. I would support.

First, I am a lender, and in light of the latest developments around Proposal #1391, I would like to share a few observations: 1)Should we reconsider the current risk management framework—specifically the soft liquidation mechanism? In practice, there were multiple opportunities to trigger liquidations when CRV was trading around $0.7 or even $0.5. However, liquidation was ultimately executed at an extreme low level of approximately $0.25. whether the current mechanism is too passive or delayed in responding to deteriorating collateral conditions. 2) Under these circumstances, any potential compensation from the DAO should be understood as having two components: One part reflects support for CRV as the platform’s governance token, given its systemic importance within the Curve ecosystem. The other part represents compensation for shortcomings in the soft liquidation mechanism itself, where the design may have contributed to suboptimal liquidation timing and increased bad debt.

2 Likes

For the future market parameters: I propose to explore new parameters for LlamaLend v2 taking into account 10/10 crash. Most importantly - we need to scan ma_time - it makes LTV compromise much less severe

2 Likes

Based on my knowledge, the liquidation of all positions is impossible, and more , this action would also harm the price of CRV.

The key issue right now is that 49.4% of the loans are effectively “dead” — “dead body” are not generating any interest.

Given this situation, why don’t we consider an alternative approach:

using a portion of the Treasury funds (for example, 200k) to subsidize interest payments for the “dead body”?

Assuming an annual interest rate of around 15% (which is likely an overestimate in practice), 200k would be sufficient to cover approximately two years of interest.
We can get the system running again.