Episode 107: August 27, 2020

--- tags: G&R --- # Episode 107: August 27, 2020 ## Agenda - [00:00](https://youtu.be/Yxzpggi6B68): Intro with LongForWisdom - [01:44](https://youtu.be/Yxzpggi6B68?t=103): Announcements from Cyrus Younessi and Mariano Conti - [15:24](https://youtu.be/Yxzpggi6B68?t=924): Smart Contracts Team Update with Wil Barnes - [16:50](https://youtu.be/Yxzpggi6B68?t=1010): Oracle Team Update with Nik Kunkel - [20:35](https://youtu.be/Yxzpggi6B68?t=1235): Risk Team Update by Cyrus Younessi - [21:45](https://youtu.be/Yxzpggi6B68?t=1310): Governance at a Glance with LongForWisdom - [25:46](https://youtu.be/Yxzpggi6B68?t=1546): MIPs Update with Charles St. Louis - [32:00](https://youtu.be/Yxzpggi6B68?t=1920): Governance Cycle Review with LongForWisdom - [37:21](https://youtu.be/Yxzpggi6B68?t=2241): State of the Peg with Vishesh Choudry ## Video <https://youtu.be/Yxzpggi6B68> ## Introduction ### LongForWisdom #### Agenda and Preamble [00:00](https://youtu.be/Yxzpggi6B68) - Hello, everyone, welcome to 107th MakerDAO Scientific Governance and Risk meeting. Today is Thursday, August 27th at 4 pm UTC. My name is LongForWisdom. I am the current MakerDAO Governance Facilitator. Forty people join me at the moment. I will briefly review the agenda, and then we'll start at the top. Anybody is free to interrupt and ask questions during any point of this call. - Discussions happen in the [Forum.](https://forum.makerdao.com/) ### Cyrus Younessi #### Leaving the Foundation [1:44](https://youtu.be/Yxzpggi6B68?t=103) - I've got some sad personal news to share. Unfortunately, my time with the MKR foundation will be coming to an end over the next few weeks. It's a really sad announcement for me because I truly love this project and all the people behind it, from the community to the foundation. This was one of the first projects that got me interested in crypto. - A few quick words about what's' going on and what is coming next. First of all, a very sincere thank you to everyone in the community that has lent me support over the last 1.5 years. You guys are truly awesome. This is probably the most battle-hardened community in all of DeFi after all that Maker has been through in the past year and survived. This makes me optimistic for its future. Many may not know this, but I have been a member of the Maker community since 2016. It has been four years since I started following the Maker project, and it is absolutely crazy to see how much stronger and more sophisticated and mature the community is today vs. back then. When I first started getting involved in the foundation, it was because I believed that I had a lot of value to contribute. I had spent a few years doing fundamental research and trading for crypto. I thought that skillset fit very well for Maker risk. Today, I can look at the community and see 20 other people with the same level of knowledge and skills. If you look at the forums, you can see how the community has stepped up its game; we constantly have community members banging on the doors and trying to contribute. This is very awesome to see. - Along those lines, there is some good news: the operational transition for the Risk team. Currently, I'm working on getting the rest of my team up and running to take over my role. Primoz and Marco, my two Slovenian teammates, have been running the whole show for some time now. I have nothing but the most confidence in them. One of my dark and shameful secrets is just how much of the credit I have been stealing; thank you and sorry, Primoz. They have been doing all of the collateral evaluations and monetary policy. They do an amazing job keeping track of everything going on in the broader DeFi ecosystem, which, of course, very important. These guys really understand Maker risk, maybe even better than I do at some point. I hope everyone in the MKR community will get to know them more closely. I'm excited for them to become more public-facing, so give them a warm welcome. Also, a huge shoutout to Vishesh, he has been instrumental and life-saving to the Risk success. I'm sure he will continue doing amazing work on data analysis and risk modeling. The best thing I've ever done was to bring these three guys on board. - As for me, I've been in this project for four years, and I want to try something new. But I can't just walk away, and I'll still be continuing to give to the community as a community member for some time. As to what's coming next, I accepted a role as Head of Risk - just joking - no, I'll be taking some time off to wind down and figure out what's next. Thank you, everyone. If anyone has any questions, I'll be around for a few more weeks. - Nik Kunkel: It's been an absolute pleasure to work with you, Cyrus. As someone burning the midnight oil, I see you rarely getting any sleep. You've been focusing 100% on Maker, and I don't know if it's inspirational or depressing. It was good to have a fellow spirit. A big part of why we were able to get as far in the governance, risk, and collateral onboarding was because of you. A huge thank you from the entire community and from me. We're all the better for your involvement in this project. - LFW: It's been awesome working with you, Cyrus. - Chris_p: You've done a great job, dude. - Cyrus: Thank you all, I appreciate the kind words. ### Mariano Conti #### Also Leaving the Foundation [9:14](https://youtu.be/Yxzpggi6B68?t=554) - I'm leaving too. Bye, everybody! - Of course, that's not all. Not sure how to follow Cyrus, since the Governance and Risk meetings have been mostly his thing along with LFW and Rich (who already left a few weeks ago). I agree. This is probably going to be the most depressing GnR call. To say a few things, I joined the Maker project about four years ago. My crypto life has been tied to Maker, from building the first Single Collateral Dai and being paid with those proto-SAI tokens to the MCD launch and deploying MCD from this very room and this very computer. It has been a huge honor to have been a part of it. - Like Cyrus said, I will be stepping away from my role in the foundation but will continue to be a big part of the Maker protocol and community as a voter, community member, and MKR holder. I want the best for the protocol. You will still see me around, and I'll be taking it easy for the next couple of weeks. I will be spending less time on twitter and hopefully sleeping more. I have no immediate plans. I won't be going to another protocol and want to relax for a bit. The smart contracts team is amazing, and you probably already know all of them. The fort is well secure in that regard. I want to thank everyone. You will probably see me in a few weeks on the governance calls. I'll be joining again, and I think there will be a few things that I will be able to bring the project as a community member that I couldn't as a member of the foundation. I'm really looking forward to it. - LFW: Thank you very much, Mariano. You've always been a super positive influence on all of us. - Nikkunkel: Without Mariano, there would not have been a Maker. Remember Sai? The first dashboard with the RGB buttons and two dozen buttons that you would have to press in the right order. You can thank Mariano for that. Supreme example of back-end engineers cobbling together a usable front-end for testing. At some point, it was testing, but then we went into launch. It takes a certain kind of person to be able to be ok with that (lol). On a personal level, Mariano, you've been a mentor to me. You've relented to a lot of questions from me, which have made me so much better. Mariano also got me into the Oracles; he was the original Oracle guy. From creating his own baby and being comfortable enough to hand it off. I love man, and I'll see you in Argentina in December. ## Domain Team Updates ### Wil Barnes #### Smart Contracts Team Update [15:24](https://youtu.be/Yxzpggi6B68?t=924) - A handful of bullet points. Liquidations 1.2 spell is slotted to be deployed tomorrow. - Next week, we have office hours modifiers, so it won't be able to be cast until Monday, between Monday and Friday. - The spell for PAXUSD and tether for September 4th, the same view, but there are office hours modifiers on that which will be created to not cast on Labor Day so between Tuesday and Friday. - Dark spell, spoken about in the last governance call, is up for comment until September 9th. ### Nik Kunkel #### Oracles Team Update [16:50](https://youtu.be/Yxzpggi6B68?t=1010) - In the past week, we had an application to get whitelisted by wiring on the ETHUSD OSM for their delegated Vaults. I had initiated a fast-track signal request on the forum so we could get that into this week's polling cycle. It looks like the vote passed, so that will be included in this Friday's executive vote. - I just wanted to touch on, really quickly, the reason why I initiated this signal request to fast-track it. It's because I think we've gotten a bit too comfortable with the Peg being above $1, and it's not even about the magnitude anymore; it's just the fact that it's still above $1, and we haven't been able to push it down. To me, anything that has the potential to generate very large amounts of Dai, I think, is worth fast-tracking. I think it's imperative that we get this under control, and the potential that I see with the delegated ETH Vaults is huge. Right now, on Yearn, you can lock up your yCRV, which is just a basket of stable coins, and you are earning somewhere close to 100% of your APY, which in crypto terms maybe some people yawn at, but in traditional finance, that's an amazing return, right? You have a lot of people right now who have very big ETH bags, and they're borrowing against their ETH using like a Vault and then using that Dai to yield farm, but it's kind of risky because they constantly have to monitor their position, so they're either collateralized very highly, so they can get some sleep, or they're losing a lot of sleep by being on the edge, like Mr. Mariano here who posted a twitter picture the other day of being like a 152% collateralized. So, because this is risky, there's probably a lot of people on the sideline who don't want to take on that risk. Having this Vault, where you can just lock up your ETH, and it's automatically taking out a bunch of Dai and making sure programmatically that that Vault is not getting liquidated, I think that's going to be huge. If this doesn't generate 40-50 million Dai within a month or two, I would be disappointed. So, I just wanted to give some color on why I thought that this proposal was urgent because I didn't want it to seem like there was any kind of favoritism with wiring or anything like that. I'm purely looking at it from a Dai supply type of perspective. ### Cyrus #### Risk Team Update [20:35](https://youtu.be/Yxzpggi6B68?t=1235) - We spent last week working a lot on liquidations 1.2 parameters, working on analysis there. We also did a research analysis piece on cTokens as potential collateral types. We are doing some final revisions there, and then we'll be posting that to the forums as well, some interesting ideas around cTokens. We've been working on ETH-a and ETH-b risk premium calculations. Those are done and going through final revision, so a few more days for that, unfortunately, but will be posted very soon. Same with Chainlink. The evaluation is also completed, but we just need to do some final sign-offs. Hoping all of these can come next week. ## Forum Recap ### LongForWisdom #### Governance at a Glance [21:45](https://youtu.be/Yxzpggi6B68?t=1310) - [Governance at a Glance - Forum Thread](https://forum.makerdao.com/t/governance-at-a-glance/84): If you have the time to read over and vote in active threads, that would be much appreciated. ##### Discussions [32:00](https://youtu.be/Yxzpggi6B68?t=1920) - [Using Snapshot to Improve Signaling and Polls](https://forum.makerdao.com/t/using-snapshot-to-improve-signaling-and-polls/3810) - There appears to be a way to vote using tokens off-chain to sign transactions, which avoids gas costs. Used by Balancer. - [Incentivized Attention: MakerDAO Governance Prediction Markets](https://forum.makerdao.com/t/incentivized-attention-makerdao-governance-prediction-markets/3789) - Prediction markets to benefit MakerDAO governance. Personally thought it was interesting, a lot of potential. - [Feature Analysis of Crypto Governance Systems](https://forum.makerdao.com/t/feature-analysis-of-crypto-governance-systems/3811) - A comprehensive description of the various governance systems being used by the various protocols with pros and cons for each one. Improvements are mapped and related to Maker. ##### Signal Requests - [Signal Request: Increase WBTC Debt Ceiling](https://forum.makerdao.com/t/signal-request-increase-wbtc-debt-ceiling/3731) - Moving on-chain on Monday. - [Signal Request: Should we change the RPs for various collateral types?](https://forum.makerdao.com/t/signal-request-should-we-change-the-rps-for-various-collateral-types/3484) - wBTC risk premium changed from 2% to 4%, passed with 34k MKR. ## MIPs ### Charles St. Louis #### Weekly MIPs Update [25:46](https://youtu.be/Yxzpggi6B68?t=1546) ![](https://i.imgur.com/5IHxubL.png) ![](https://i.imgur.com/p7N0vkm.png) ![](https://i.imgur.com/8qXkBUy.png) ![](https://i.imgur.com/4oEPoOI.png) ![](https://i.imgur.com/0o4zz1W.png) ### LongForWisdom(GF) #### Governance Cycle Review [32:00](https://youtu.be/Yxzpggi6B68?t=1920) - This a discussion on how the last Governance Cycle went and essentially what is going to happen in the next one. - We had three proposals submitted this month, which were the Declaration of Intent (forward Guidance), the domain work requirements for onboarding LRC and the sub-proposal called for personnel off-boarding Rich. - All three of those were accepted into the Cycle, but only two passed the inclusion polls, which was the Declaration of Intent (forward Guidance) and the sub-proposal to remove Rich. So we lost LRC at that stage, which was a little bit of a surprise for some. I know many of the people I spoke to expected it to be included. I think that might be the first time we've seen something that hasn't been included, or at least it's definitely the first time we've seen something like a collateral onboarding proposal not being included. That's potentially worthy of notes. I think the plan is to resubmit it next Cycle. We'll get a better idea of whether not being included was a result of a lack of participation or was an actual objection to the collateral type itself. A couple of people were saying that there were similar gas prices around that time, which is true, and it was one whale who voted against and pretty much carried the poll. Both of those are true. Hopefully, it will encourage more people to participate if they want to see it onboarded. - As Charles mentioned, the Declaration of Intent and the subproposal to remove Rich officially are both still up for Executive and that Executive has until around this time Friday to pass before it expires, so there's still a chance that it will pass, which is good. - In terms of participation in the various actual sections, similar to what we've seen previously for the inclusion polls, the same 17K Maker participating again. And in the Governance Poll bundle,, we saw 3.5K Maker participate. I think last month we had 18K, but the two months prior, we had like 5K or 4K, so it's potentially a return to average, based on the high gas prices. - In conclusion, we see what we've seen for the last two months, some amount of MKR is participating in the Governance Cycle, while a larger portion is not. I don't think that's necessarily a bad thing because the point of the monthly Governance Cycle is that if you're only paying attention at the end, you can vote and have it, in fact, right, and you can use the previous votes to inform your decision at that point. Obviously, it would be great to have more participation, but I don't think we are at troublesome levels yet. I think we'll hopefully see more participation next governance cycle because there's a fair few things that are going to be submitted. I think it'll be a fuller Cycle, so hopefully, they'll attract more attention. ## Analysis ### Vishesh Choudry #### Relevant Links [MCD System Stats](http://daistats.com) [MCD System Stats Alt](https://catflip.co/) [DAI 24hr VWAP Graphs](http://dai.descipher.io/) [Maker Vault Stats](http://makervaults.descipher.io/) #### The State of the Peg [37:21](https://youtu.be/Yxzpggi6B68?t=2241) ![dai.descipher.io](https://i.imgur.com/1fbsoNp.png) - Broad strokes in terms of Dai<>ETH trades, in the last few days, the prices have been fairly volatile and trending slightly upward in the $1.01 range and pushing into $1.02. There has been a fair bit of spread on those trades lately. Part of that will likely be contributed to the volatility in ETH price as there hasn't been an insane amount of movement in terms of farming opportunities and the yield that is available there. ![](https://i.imgur.com/6ZvIRgc.png) - In terms of the Dai<> USDC pair, the trend is a bit clearer, with trades dominated by Curve. About two weeks ago, the Dai peg dropped down a bit due to the end of a few farming yield opportunities, and now it has kind of stayed flat and then risen a bit in the last day or two. This may be in part due to market volatility and not just farming behavior. ![](https://i.imgur.com/QklH9ri.png) - For Dai<>USD pair, there is a bit more spread than the Dai<>USDC paired. There is a lot more variability in the sizes of trades on the Dai<>USDC pair. This chart is always a bit messier. But you can see more clearly the moments of run-up and liquidity crunch. ![](https://i.imgur.com/06IaOQW.png) - Dai<> USDT is similar to Dai<>USDC. ![](https://i.imgur.com/GcB0jqA.png) - Overall, Dai supply is about 440 million. ![](https://i.imgur.com/DNYNlLz.png) - 210 million Dai in DSR, primarily through Compound, and 230 million Dai spread across other addresses, holders, and contracts. ![](https://i.imgur.com/qrPpgCW.png) - In terms of collateral, the largest group is, of course, ETH. Here we see that there is a continuing trend that the Dai supply minted from ETH is continuing to grow. Obviously, at an increased pace since July, but that pace appears to be slowing down a bit. It will be interesting to watch whether there continues to be a significant supply of Dai minted from ETH. I believe we will see that curve accelerate during times of more frothy yield and slow down or resume pace of growth in a more steady-state environment, especially if ETH price continues to appreciate. - It will be interesting if there will be a drying up of yields or, if ETH price drops a bit, it will be interesting to see the amount of Dai minted from ETH. For now, it appears to be continuing growth. ![](https://i.imgur.com/mKq9yuK.png) - USDC<>DAI is a more instanced vault, where there are a few large individual users minting and repaying. When this happens, there is more impact on the supply curve and a jankier graph. This has not been as heavily used in the last couple of weeks. It is sitting more or less in the 10-20 million Dai range. ![](https://i.imgur.com/fHzkCnC.png) - WBTC has continued to cap out as the goal-posts have continued to move. With the recent increase, it took a few days to reach around 80 million. It appears there is a fair amount of demand for utilization, but again, similar to USDC; it is primarily dominated by a few large users. This will be a discussion point for the community to determine how each collateral type is used. ![](https://i.imgur.com/f6Gq6Xo.png) - A few large movements for BAT, with a jump to about 5 million Dai minted. Partially this is due to a decreased capability due to using/generating yield on other platforms. Presumably, someone long on BAT may want to get more leverage out of that capital, and Maker is the way to do this. This is to be expected for this particular collateral pool, although it is not a significant share of the collateral pool. ![](https://i.imgur.com/iPAmagb.png) ![](https://i.imgur.com/il2JPa4.png) ![](https://i.imgur.com/jfjrMVy.png) - In terms of what the collateral looks like, there has been a hop-scotching back and forth in terms of collateral utilization. As the ETH price has appreciated, we are seeing an increase in the collateralization. As it decreased a little bit, we say that the CR ratios held constant. Over the last couple of days, we have seen a little bit of a re-increase in collateralization for some of these larger vaults. They are still sitting in a medium risk scenario around 200-250% collateral ratios. This is definitely something to keep an eye out, and there are signs of life in that people do seem to be managing their collateralization. I think we see more of a movement in the CR with the ETH price moving than when it is more or less constant. This means that people aren't necessarily managing their CR as much as just letting it stay somewhat constant and then fluctuating with the ETH price. - Some of these larger vaults are 200%,250%,300%. If the ETH price were to hit $280, you would see a huge chunk of Dai liquidated, which is something to be aware of in terms of risk parameters and liquidation structures. This is in terms of the liquidation 1.2 structures, and what would happen in the instance of these liquidations; just something to be aware of. - USDC is rather boring to talk about. It is where you would expect and is close to the line. ![](https://i.imgur.com/ij2AwV8.png) - WBTC is interesting to note, as some of the larger vaults are maintaining a bit more collateral, although there are at least three large WBTC vaults that are sitting around 160% collateralization. It's always going to be a bit different than ETH, but it is important to note since we do not discuss this one very often. ![](https://i.imgur.com/H67HntB.png) - To look at Dai supply, a lot of that sits in Compound, with 210 million Dai sitting excess in Compound, which finds its way back in the DSR. ![](https://i.imgur.com/pWUW20d.png) ![](https://i.imgur.com/M8Mhl2C.png) - This is roughly half of the Dai that has been minted. The remainder, at a glance, shows a decent chunk of the remaining supply in Curve, Balancer, Aave, and Zombie finance pools. These are different opportunities for users to obtain yield. Obviously, there is a much larger stake in some of these previous yield farming opportunities that have cropped up and have flown out, as expected. The remaining 230 million Dai supply is definitely a moving target, and there is a fair amount of volume flow. ![](https://i.imgur.com/pAzRFxp.png) - We still see 25 million or so of that in Curve, which is a very decent chunk. - Comparing the volume, USDC has a significantly larger balance share in Curve, but the amount of trading volume per coin on those platforms is roughly comparable to Dai in USDC. You can see that Dai is moving around quite a bit even though it may be holding smaller shares than other tokens on these platforms. ![](https://i.imgur.com/Z7Rw7wn.png) - Here we see Balancer, with about 20 million in that particular Dai pool. #### Discussion [50:20](https://youtu.be/Yxzpggi6B68?t=3020) - Nik Kunkel: Earlier, you showed a chart that showed that if ETH price goes up, collateralization increases, and as the ETH price goes down, it decreases. That all makes intuitive sense. But is the magnitude of those changes, especially when ETH is going down, is the magnitude proportional to the price change in ETH, or are we actually seeing people lever up when the price is going down? Are they falling into the drops, or is it just price movement? - Vishesh: I don't even know if that's necessarily the right phrasing. But, it's definitely basic math that as the ETH price goes up, collateralization goes up. What we've seen historically with Dai is that, after that collateralization has gone up, people tend to, in different degrees at different times, do some sort of re-leveraging where they end up lowering their collateralization ratio in response either by minting more Dai or by removing a little bit of that unnecessary collateral. In the recent environment, it tends to be minting more Dai, and then, when the ETH price comes down a little bit, I think we tend to see a compensatory behavior where people re-collateralize in an anticipatory kind of way to protect against liquidation. In the past, that tended to be adding more collateral back; lately, I think either there's a decent amount of utilization of people's capital and that it's deployed in a lot of different places or some other potential explanation. But, I think that people tend to manage their Dai balance more than their collateral balance, possibly. ## End-Discussion ### yETH Vault Launch [53:20](https://youtu.be/Yxzpggi6B68?t=3202) - Cyrus: The basic idea is that there is this DeFi protocol called yEarn. One of their upcoming products will be ingesting ETH. Using a strategy inside their yVault, as they call it, and yVaults are not to be confused with Maker Vaults, very different concept, similar names. This yVault, which takes ETH, called yETH, will be opening up a Maker Vault and minting Dai at a 200% collateralization ratio, I believe. They will take that Dai, and they will deposit it into the curve yPool, which in return, they would get an LP token on that yPool, which is called yCRV. Essentially, yCRV acts like a stable coin, but in actuality, it's like a claim or a liquidity provider token for this yPool. And then, it will yield farm with that. That's the summary from the high level, although I think there is definitely a lot of confusing terms in there. - One of the cool things about this strategy is that they want to be able to read from the OSM so that they can protect their Maker ETH-A Vault if the ETH price starts to fall. This makes it safer for them, and I believe they've requested whitelist access; that's what Nik was referring to earlier. I think the things that this community needs to consider are: - Any blowback risks, anything that might be contagious from yEarn into Maker. I think from the perspective of reading off the oracles, that's fairly harmless; I'll let Nik comment on that. On the risk side, the benefit is that an enormous amount of Dai could be created depending on the popularity of this product. Dai deposited into the curve yPool has a direct downward pressure on the Dai price, which is obviously something we are all interested in. Over the last couple of weeks, we saw the price of Dai get pretty close to $1, which was a direct result of the CRV farming from the yPool. I think that's the main positive point. - The main negative point might be, in theory, as a community, we've never come to a consensus on how we feel about users taking advantage of the OSM in the one-hour delay. Of course, it's totally possible, and plenty of individuals do it. But, this could be more from like an institutional level scale, it could be hundreds of millions, and we have to explore what the consequences there are. Basically, on average, the collateralization ratio for Maker as a protocol would be lower because people can take advantage of the OSM, but I've already mentioned the upsides, so that is something that the community should probably discuss from a risk angle. If there's a huge ETH drop, I'm actually not sure of what the strategy would be to deal with that. But, it may not be to MakerDAO's favor. ### Upward pressure on Dai when unwinding yETH Vault [58:47](https://youtu.be/Yxzpggi6B68?t=3527) - Akash: It's great that on the first side, it's adding a bunch of Dai to the y-curve Pool. That should decrease the Dai price, and it's a great thing, right? What I'm a little worried about is, when ETH drops, and then they come to buy Dai to close out their position, aren't they going to be putting a bunch of upward pressure on Dai when they are trying to close their position? - Cyrus: Well, they wouldn't be buying it, they would just be removing it from the yPool, which skews the pricing on Curve, but it's not nearly the same as going into Coinbase or Oasis base and lifting the offer. - LongForWisdom: The Dai isn't guaranteed to be there on the curve pool, and that's part of the issue, I think. - Akash: I think the potential solution is, if the Dai is there, they're removing it and obviously closing out their ETH-A Vault. But, if the Dai is not there, then somehow being smart enough to use the USDC Vault to mint Dai to help temporarily close out their ETH Vault. - Cyrus: Yes, I think there's a number of ways these strategies could be improved. This is just like a first pass, right? I think this is the most complicated strategy to date. Given there's no leverage built into it, doing something like USDC could be something that, in theory, is entirely unrelated to this. That would be more along the lines of those Keeper DAO strategies, where you have a bunch of people deposit USDC or Dai or whatever into a smart contract, and its job is to be on Auctions or stuff like that. So, I think that's definitely a smart way of doing things but also maybe somewhat distinct from this strategy directly. - Akash: I guess my other concern is if it's audited. I mean, that a concern for the initial yEarn and ETH person, right? - Cyrus: That's why I was saying that we need to think about potential blowbacks, so that's one right there. The community discussed it over the years, but if there's a hundred million Dai sitting in some exchange or some smart contract or whatever, and then it gets hacked, what effect does that have on MakerDAO? I don't think it has ever really been this community's mandate to oversee what other users do with Dai, and I think it's a bit unclear on what impact it would have anyways. - Akash: Again, this is my personal opinion. It sounds like something that could help the Dai Peg, so generally, I'm in favor of it. Obviously, there are risks for the individual user who goes to yEarn. - Cyrus: I think a good way of thinking about it is that maybe they could be considered like a superuser. Because, potentially, they could dwarf any other single individual Vault, even the largest ones that we've ever seen. So, it could be interesting. I don't think we'll be able to consider all of the cases at this stage, but for now, the proposal focuses on the Oracle whitelist, and so I think that could be the primary concern for now. I'm not sure. - LongForWisdom: It's worth mentioning that this isn't something we can stop from happening. This is the usage of Maker Vaults. We need to be aware of this happening, right? - Cyrus: Even without the Oracle whitelist, this strategy could still work; it just wouldn't be as effective on their part. I think it's a question of, on the Oracle side, does MakerDAO want to help transition this? Is that accurate? - Nik Kunkel: Yes, that's what I was going to say. They are still going to do this, right? They can still look at the Oracle prices; they just do it in a less trustless way. So, if we want to make this decentralized, then they need access to the OSM. But, this thing is happening either way, so it's best to at least do it in a controlled manner. We don't want to risk surprise liquidations, so I think it's in our best interest to do this in a legit way. - LongForWisdom: I feel it may be good to focus the conversation on the Governance considerations that this is going to bring up. If this proves to be really popular, then we potentially want to go up the debt ceiling fairly rapidly, right? Just other considerations that we need to be aware of as a result of this coming into existence. - Cyrus: Taking a step back from Maker and thinking from a broader yield farming DeFi perspective, this is how I've been viewing things. The yCRV yEarn strategy, which is called yyCRV, has been very popular. I think upwards of a hundred million in there. A lot of that comes from borrowed stable coins either from Maker or Compound, so the ultimate source of collateral is ETH for certainly a good percentage of this. For anyone who's tried this strategy, you know that there is a considerable amount of maintenance and anxiety and headache, trying to manage your liquidation risk and liquidity risk actually for the strategy. If that can all be automated away and the yEarn strategy can even manage the liquidation risk by reading off the OSM, then I think this is going to be enormously popular. Because, even with all the current difficulties, it's already ingested upwards of a hundred million. So, this could go a lot higher, and I wouldn't be surprised if this easily tacks on a couple of hundred million Dai over the next couple of months after this strategy is live, assuming the contract is battle-hardened and audited and all of that. People start to understand exactly how the liquidation management works. The downside is we are going to have to wrestle with very quickly ballooning Dai supply off of the ETH collateral. Over the years, I've started to realize that it's just going to be maybe one of those endlessly no-lack of consensus debates. Some people love ETH as collateral and think it's super safe; others feel strongly in the opposite way. Obviously, a big consideration for the circuit breaker and liquidations 1.2 and liquidations 2.0 has been to allow MakerDAO to more efficiently process large collateral liquidations, especially with ETH. So, I think this is going to be one of those ongoing conversations. - Nik Kunkel: Yes, I think this is definitely going to increase disproportion of the Dai portfolio that's backed by ETH, there's no question about that. The trade-off here is for Peg stability, which is only going to be achieved by minting a bunch of Dai, which this enables us to do. But the trade-off is that we are more relying on ETH, and, even with the liquidation changes that we're adding, we definitely increase the long-tail risk for a Black Thursday type of scenario because we are now even more exposed to ETH. If you look at USDC, it was at one point a very significant chunk of the collateral portfolio, and that's since fallen by the wayside. I think it's worth it, but it's definitely a serious risk to consider the externalities. And this is all assuming that this yETH delegated Vault thing is super successful; if it's not, then it's not a big risk. But, to be honest, I don't see the latter scenario happening. ## Outro ### LongForWisdom #### Closing Comments - Thank you, everyone, for coming, and I will see you next week. ## 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 ## Credits - David Utrobin produced this summary. - Gala Guillen produced this summary. - Anatta produced this summary. - Everyone who spoke and presented on the call, listed in the headers.