Other Presentations and Updates
Hugh Karp
Nexus Mutual Insurance
Thread: MakerDAO Protocol Cover using Nexus Mutual
50:45
- Thanks to everyone. It’s good to be here. I’m Hugh Karp, and I’m the founder of Nexus Mutual. I suggested a potential coverage for MakerDAO as a protocol, so It’s basically building a tailored policy for Makerdao and Nexus. The main objective is to address the potential for MKR’s pro-cyclical issue when it comes to bad debt. Basically, you have this issue where MKR is burned at high prices when you’re accruing fees. Then if there’s an event, usually the price may drop. Then you mint more MKR at that low price to pay off the bad debt. So we’re aware that Nexus could, in the future, cover a larger section or a more massive chunk of potential bad debts that might accrue to the protocol. Still, Maker is currently bigger than Nexus at the moment. So there are limits that Nexus has, so the primary use would be for a dampening of MKR’s pro-cyclical nature. The idea is that if you look over a cycle of paying costs in terms of the cover versus the claims that you get back, assuming that that’s negative EV over the life cycle, there’s still a significant positive benefit from the actual value of MKR that’s net, burned or minted. I can go through that.
- So, the proposal in summary and all of this stuff can be tweaked. It’s just something to put out there, is that you could buy tranches of cover on Nexus, say three million DAI worth. If bad debt was to exceed five million DAI, then one trench would claim as paid. And, at 10 million, then you paid two trenches, so six million in total, and so up to sixty percent of the bad debt is covered. I think it’s essential that not all of the bad debt is covered because we still want alignment of interest between Maker’s risk management. So I think that that’s important, but this is really a potential dampening mechanism. Regarding limits and pricing and stuff like that, the total maximum cover right now on Nexus would be about 12 million DAI. It’ll grow as Nexus grows, so you could, for example, purchase four-three million tranches. The minimum pricing actually got dropped last week to 2.6% per annum. The actual price is determined by staking the number of people that want to back it, so it is kind of unknown at this point, but that gives you a range of what the pricing might be if this was to proceed. In general, if there’s a known decent amount of demand, I would expect a substantial amount of taking on Nexus. That’s what we’ve seen in the past, so there’s obviously a bit of a range. It depends on the staking, but it kind of gives you a flavor for where it might end up.
- To give you an example of what could happen using black Thursday as an example, apologies, but I didn’t quite get the numbers right I’ve tried to get the right ones. I believe there’s about 5.3 million of bad debt accrued, resulting in nearly 21 000 MKR being minted. If you had the protocol cover, you would have avoided minting something close to 12,000 MKR. Essentially you pay that at the MKR price pre-event, which is the higher price rather than, the lower price, then you still save, in terms of MKR, minting over the whole cycle. So that’s it. I wanted to keep it short. That’s the main proposal.
- In terms of other key questions that other people have asked, claims assessment, how claims assessments are done because that’s a pretty important one. The Nexus Mutual members vote on claims, and I guess one of the key things is why they would approve to pay a claim, and that’s the fundamental thing we’re doing at Nexus. All of the incentive mechanisms are designed to ensure that legitimate claims are paid. I won’t go into massive detail about all of that right now. Still, you can definitely look it up, and there are two layers to the incentive structure to pay genuine claims. It’s probably more of a corporate financial type decision here rather than ensuring tail risk or something like that because it’s a bit different right now given Nexus’s size versus Maker’s size. However, I still think there’s potentially a lot of value here in it because the pro-cyclical nature of MKR has, I guess, been discussed quite a few times in the past. So yeah, I might just stop there, and if anyone has any questions, I’m more than happy to answer them.
Discussion
57:30
- LongForWisdom: I would say that when I initially read this, I kind of didn’t understand the benefits before I kind of understood the effect as in the kind of cyclic nature of implementing and burning, so I thought that was so once again I realized how it would help, I thought that was good, I know a couple of you in the forum were quite positive and others were quite negative. If any are here and would like to comment, I’d appreciate it.
- Frank Cruz: So Hugh, I guess the downside is that Nexus Mutual doesn’t survive the next 5 or 10 years or even less? Suppose Nexus Mutual governance doesn’t approve the payout. What would be the difference between just going out to Barclays and getting insurance from them instead of Nexus Mutual, in your opinion?
- Karp: I think the first one is you’d struggle to get any cover just because it’s so tailored, and they don’t understand the details right now. The second one is the payment could be made on-chain directly into the right contract. Still, then that would directly avoid the minting of MKR. Hence, there’s a direct level of integration there that can provide a significant benefit in terms of voting for actual auctions taking place. Honestly, I think you’d spend 6 months going around various insurance brokers and not really getting anywhere because this isn’t a standard product at all.
- Frank Cruz: Got it. I guess you’re saying that for someones like Lloyds to cover one of these protocols is way out there, maybe in the future but not now.
- Hugh Karp: Yeah, exactly, and they give you cover in US Dollars, and then you’d somehow have to get the claim payment into the protocol. There are quite a lot of details behind it. Even if they were to come up with something like this, they’re quite risk-averse. The regular insurance industry is very risk-averse on this stuff right now. They just don’t understand it. Anyway, long story short, I don’t think it would be worth while.
- LongForWisdom: You mentioned something that gave me a thought. Usually, the way that the protocol works in bad debt is that flop payment is triggered in a fixed amount of time after the bad debt happens. I think it may be a week, So I question how quickly would Nexus Mutual governance kind of approve or deny payment. It would need to happen before that time would expire.
- Karp: Exactly, so 36 hours minimum payment should be finalized within 72 hours. It could go a bit longer than that, so it’s not absolutely guaranteed to get under your kind of week, but in nearly all circumstances, it would.
- LongForWisdom: Ok, thanks.
- Kordez: Are those prices fixed & static, or If you fixed the long portfolio of MKR, would the pricing change, or is it fixed for a determined time?
- Karp: The pricing is fixed for whatever covered period you purchased, and then let’s say you bought cover for a year that would be a fixed price, and then let’s say that expired and you wanted to renew it and over time have different risk parameters than Nexus mutual stakers may have a different opinion and the renewal rate may be higher or lower depending on the staking at that point, but it’s a fixed rate and doesn’t vary.
- Kordez: Ok, Thanks. How do you manage that the capital is locked inside Nexus, I read something about it but could you just explain?
- Karp: Yes, we have this concept called the MCR, which is our minimum capital requirement, required to back the covers that we’ve written. No one can take the funds out of Nexus unless we have funds above our MCR. It’s not absolutely guaranteed because insurance companies aim to be undercollateralized to diversify risk. We don’t always have more money than potential liabilities, but the whole point is that we have a certain amount of money that has a very high likelihood of paying all legitimate claims.
- Kordez: Ok, thanks, and who calculates those MCR parameters?
- Karp: It’s kinda done with a flat factor based on the cover outstanding, and then we run a more complicated model off-chain.
- Kordez: Ok, Thanks.
- Karp: This is tailored to cover bad debt accruing in the protocol, so it’s a bit different from our standard product that we just offer right now.
- LongForWisdom: So I guess you’re saying that it would cover bad debt regardless(of the source.
- Karp: Yes, it is debt regardless of how it perpetuates.
- LongForWisdom: Any other questions before we move on? Thank you so much for coming and doing the presentation.
Vishesh Choudry
The State of the Peg
MCD System Stats
MCD System Stats Alt
DAI 24hr VWAP Graphs
Maker Vault Stats
01:06:15
- For the last week, we’ve been sitting at $1.01 for DAI. There’s been some movement in ETH price, so there’s been some fluctuation as well. That did coincide with the sudden drop in ETH price. Around the 26th, we saw a bit of a spike and some increased volume, nothing too significant. It’s still holding pretty steady. The interceding volumes between the 24th and 26th were very light, so there wasn’t much movement.
- You could see here the trades that have been executed are centered around $1.009. A majority of that volume being on Uniswap, which is the latest steady equilibrium point, is to have most of that volume being on Uniswap.
- If we look at the DAI-USDC pair, we see here between Swerve, Curve, and DYDX that curve accounts for most of the DAI-USDC trading volume. In the last week or so, we’ve seen that price centered around $1.01 with a bit of a dip and with a bit of a spike around the 22nd, 26th, and 27th. Then we saw a bit of a dip around the 24th when the trading volumes were a bit lighter.
- DAI-USDT is looking very similar to DAI-USDC. Now we’re sitting at around 937 million total DAI, 337 million of that is from ETH and the other 600 million is from other assets. Obviously, a huge chunk of that is in Compound–about 290 million in terms of the amount that’s not locked up in the DSR at the end of the day. Although that’s a little bit of an irrelevant statistic because it’s not really out of circulation in the same way.
- Uniswap v2 pool is at the top of the list, followed by the BarnBridge curve and Binance. The majority of the largest positions are in staking protocols or other yield farming, which is expected. We can sort of inspect the positions for some of these larger vaults. If we look at USDC-A, Roughly 360 million is below 101% collateralized, an additional million of that is below 100% collateralized, but the largest chunk is between 101 and 100% collateralized. Pretty much the entirety of the USDC pool is some fraction of 100% collateralized.
- The 3 or 4 largest positions are all 100.6 collateralized. These positions will be below 100% collateralization in, say, December. That’s basically the continuing growth of these debt positions with the stability fee they currently have and the rate at which that fee will accrue. If you look at the last updated dates for these vaults, they, by and large, haven’t been touched in a month, and then that’s true for the vast majority of stable coin vaults.
- Some haven’t been touched in the last 20 days or so, even that is few and far between. You could see the curve here where these prices above $1 are irrelevant because you’re already there. It’s basically a question of how much of this debt is ahead of this line or below this line.
- If we look at some of the other positions, BAT, for example, the largest position is ~2.5 million DAI in debt, and the next one is 600 thousand. Those are, respectably, 240 and under 98% collateralized, which is pretty decent collateralization. That has a projected liquidation wall around .12 for 3.2 million DAI. We now look at ETH, which is another large position. We got a big liquidation wall at ETH price of $318, so that’s about 90 million Dai that we liquidated. You can see that basically, a massive chunk of this debt position is 180% collateralized, and that’s about 85 million.
- If you look at some of the larger positions, you’ll see that the largest 2 vaults are 280-380% collateralized, which is very solid. Still, they only account for between the two of them approximately 60 million of the overall DAI portfolio. It’s interesting that, in this case, looking at a couple of the largest positions isn’t necessarily representative of the majority of the risk profile for the ETH vaults.
- The ETH vaults have been updated quite a bit more regularly. It appears that ETH vault users keep up with their collateral a bit more than stablecoin vault users, which is pretty well expected.
- If we look at the LINK vaults, we’ll see one large vault with 4.2 million dai debt 300% collateralized. Again, depending on your expectation of market volatility, that’s a comparatively larger buffer. At a price of 6.5, you’d have 4.5 million DAI being liquidated.
Conclusion
1:18:11
- LongForWisdom: I think we’ll finish up. It’s reasonably late, so any final questions or general reminders anyone wants to make? Nothing? Ok, alright, I guess we’ll stop there. Thank you, everyone, for coming, we’ll have another one of these next weeks, and I’m sure I’ll see you all, thank you, Hugh.
Links from Chat
Common Abbreviated Terms
MCD: The Multi-Collateral Dai system
CR: Collateralization Ratio
DC: Debt Ceiling
ES: Emergency Shutdown
EV: Executive Vote
GF: Governance Facilitator
GP: Governance Poll
SF: Stability Fee
DSR: Dai Savings Rate
MIP: Maker Improvement Proposal
OSM: Oracle Security Module
LR: Liquidation Ratio
RP: Risk Premium
RWA: Real-World Asset
Credits
- David Utrobin produced this summary.
- Artem Gordon produced this summary.
- Jose Ferrari produced this summary.
- Dennis Mitchell produced this summary.
- Everyone who spoke and presented on the call, listed in the headers.
