- 00:00: Intro with Rich Brown
- 02:55: Governance at a Glance with LongForWisdom
- 5:38: Weekly MIPs Update with Charles St. Louis
- 11:12: MIPs Submission Review with Governance Facilitators
- 12:17: Liquidations v1.1 with Mariano Conti
- 13:47: TUSD with Mariano Conti
- 15:35: New Oracle Proposals with Niklas Kunkel
- 19:30: New wBTC merchants with Kiaresh
- 21:30: PSM Advantages and Discussion with Wouter Kampmann
https://www.youtube.com/watch?v=Z2KzVB460Zc
- Hello, everyone, welcome to the July 9th edition of the Scientific Governance and Risk meeting.
- This meeting will be special for two reasons:
- We have some weighty concepts to discuss.
- This is the 100th Governance Call.
- How did we get here? Where are we going? What does that mean? We are not going to be getting into any of those topics today. I was hoping to do skits, perhaps a virtual cake, a retrospective, etc. The ecosystem has not given us that space, so we will leave the congratulations and celebrations until another time.
- It is important that we get this show on the road.
- We desperately want to hear from the ecosystem; a lot of smart people are on this call. Do not be shy, if you have a question, type your questions in the chat or jump on the mic. Please keep in mind that conversations happening in the sidebar don't make into the public record or historical memory of MakerDAO, so if you have something smart to say, please consider participating on the microphone.
- There's going to be a timer for each one of the presentations, so please try to keep your segments to five minutes or less if you can.
- Discussions happen in the Forum.
Governance at a Glance - Forum Thread
- Peg Stabilization Modules - Forum Thread
- I guess that the most urgent discussion is the discussion around the Peg Stabilization Modules(PSM.) We'll be getting into that a bit later.
- Governance Chief Contract Redesign - Forum Thread
- In addition to that, we've had Derek, who has posted on the forum talking about the DSChief contract redesign. Fairly soon, there's going to be a chance to replace DSChief. Derek is going to talk more about that in a future meeting, but there is a thread on the forum if anyone is interested prior to that.
- SourceCred Trial 1 Month Review - Forum Thread
- We also had Seth from SourceCred post the first-month review of the SourceCred trial, which has got some interesting data on it. So if people are not keeping track of that, I would suggest checking that out, giving it a read.
- GSM delay increase - Forum Thread
- We had Joshua_Pritikin asking whether we should consider increasing the duration of the GSM delay up to 24 hours to help protect against Governance attacks.
Signal Request Should MakerDAO signal that it prefers for COMP rewards for DAI suppliers and borrowers to be disabled?
- COMP rewards signalling - Forum Thread
- We've also had Lev, or equivrel on the forum, asking if Maker should signal to Compound that we would prefer COMP rewards to be turned off for Dai because of the effect that it's having on the peg. So those are both up on the forum currently.
- WBTC Debt Ceiling and Risk Premium - Forum Thread
- The previous signal requests to adjust the WBTC debt ceiling and risk premium had been passed, and Debt Ceiling has already been used up. This signal request thread is over.
- Compensation Plan Forum Poll - Forum Thread
- The informal Compensation Plan poll is out. Eventually, we will do one on-chain. Even if we move at a slow pace, we can move forward, one way or another.
- Last week was this gap week between June and July, where we had a pause for people to collect themselves, and for proposals to get finalized and drafted up.
- Last week was quite calm. This week we had the formal submission period from July 6th to July 8th, ending yesterday. We saw seven proposals come in; I'll review them on the next slide.
- We also saw two new Pre-MIP discussions posted to the "Proposal Ideas" subcategory.
- I urge everyone to go and check those out or constantly review them; I know the Peg Stabilization Module forum thread is quite busy.
- The Community Greenlight Polls ended on Monday, the Collateral Status Index was updated to reflect the results. These will eventually lead into the collateral onboarding discussions later on.
- MIP19: I believe Mariano will be speaking briefly on it later in the call.
- MIP18c4-SP1 aims to remove the DSR spread from weekly Polls.
- Lastly, we have a MIP12 amendment proposal for improving the efficiency of overall collateral onboarding for both the domain team and the overall process, finally clearing up everything that should allow us to use all the monthly cycles for collateral onboarding going forward.
- MIP15 Dark Fix Mechanism is one month away from entering a Governance cycle, once the author Wil Barnes has officially added it or the edits from the community that he thinks are reasonable.
- MIP14 I believe it's still in the request for comments because it was rejected from formal submission and is going to be reviewed for the community.
- The Collateral Onboarding Community Greenlight Polls will be ending by the end of the cycle as opposed to the formal process that we use, which may be carried over to the Governance cycles and reduce the efficiency of overall collateral onboarding.
- That's my bit for today. I urge you to check out all of the proposals that have been formally submitted in the Forums, and I will be posting my weekly MIPs update this week as well, recapping everything I just spoke about.
- Charles just covered the seven proposals that are going to be submitted for July. There are four MIPs: MIP 16, 17, 18, and 19, and there are a few sub-proposals.
- I don't think there are any issues with the currently proposed MIPs that would prevent them from continuing as the inclusion Poll, so they will continue.
- I would ask people to at least read the summaries of what was proposed because there are a few that changed things or introduced new things.
- Liquidation 1.1 finally got into the Governance Cycle, so that's going to be voted on at the end of the month.
- You all should know about this by now, I've talked about it plenty of times. It fixes a couple of things, the double liquidity needed for all of our auction types, plus fixing the bug that we saw in March, where debt auctions could be stuck. There is a lot more information in the MIPs write-up.
- This is not a big change, but it will mean that there are going to be new deployments for the smart contracts. Anybody running a keeper, things like DeFi Saver, even our internal UIs, will all need to be updated to the new addresses, so we are going to have a more aggressive message in the coming weeks, giving out the deployed addresses and everything so people can switch on time.
- If you have any questions, ask me in Chat, @mariano.conti, or in the Forums.
MIP 19 - Liquidation System Upgrade - Forum Thread
- I do want to give an update on TrueUSD. We have Lucas Manuel from the Smart Contracts' team working on reviewing the latest implementation. So now, we are in touch with the TrueUSD team, and they are going to do another upgrade to their token in less than two weeks. This time they've already deployed the code for us to give it a look before and give our recommendation about whether we feel that it's safe. And if we do, we are going to recommend that the community reactivates TrueUSD using a weekly governance cycle, maybe next week or the week after. This will also guide us in the future for how we are to work with other tokens that feature this proxy upgrade system.
- That was an announcement. We don't have anything concrete with TrueUSD yet, we are going over it, but I think we'll have more information next week.
- I want to give a quick overview of the oracle proposals that we've seen on the forum recently. Keep in mind that because we haven't done the refactor of MIP10: Oracle Management yet, it's a similar situation with what we're doing with LEND where we go through a weekly cycle instead of a monthly one. We will try to align the timing of the Oracle related proposals with the monthly cycle to keep up a consistent cadence.
- We have an application from Argent to become a light feed. I'll be doing the due diligence on the information that they have provided on that proposal and posting it early next week.
- We have a proposal from Kyber to create a KNC-ETH oracle for them, that they will use for their buy-and-burn model for the KNC token. We will be doing that as well.
- We're starting to add a lot of feeds, so the number of feeds is growing a lot. In
SCDwe had 14. WithMCDwe launched with 20, and now we're at 24. 25 if Argent gets ratified to be a feed as well, and I'm working with a couple of other groups in the background that are interested in applying. Part of having new feeds is that you have more data to work with, but the risk increases as well, there are now more parties that can be malicious. I want to create a proposal to raise the quorum. Right now, it's at 13, so if we want to keep a similar ratio, I think it should be 15, and possibly even 17, moving forward. I don't think that's very controversial, but it needs to be discussed. I'll post the proposal early next week with the reasoning, and we'll talk with Charles if it needs to go through the weekly or the monthly cycle, in which case it will get pushed into the August cycle.- Rich Brown: Thanks, Nik. Looking forward to those discussions. Maybe when you kick them off, you can add some indication of the urgency to help us discern in which cycle to include them.
- Bittrex is launching a wBTC-BTC, ETH, and USDT pairs, I think, later today.
- Also Alameda Research, Swipe Wallet, and Three Arrows Capital are becoming wBTC merchants, so we're expecting that to help with liquidity quite a bit.
- Rich Brown: Interesting. I think this speaks to the
DCof the wBTC. If that could be added to the ongoing discussions in the forums, so it's a bit stickier than this call, that'd be great.
- We can't decentralize if we can't become self-sustainable because we can't scale.
- It will take some time to get there, and we need the necessary tools to get there.
- The state of the peg has been less than ideal.
- How can we develop the tools to handle this?
- This is becoming quite urgent.
- Governance is getting questions about what we're doing to deal with this.
- This is quite a simple concept, but it might be hard to wrap your head around its consequences.
- In the second line, it should say
PSM(as in Module) instead of PSN.
- In the first line, it should say
PSM(as in Module) instead of PSN.
- MIP19: Liquidations System 1.1 Upgrade - Forum Thread
- Peg Stabilization Modules: A Pre-MIP Discussion - Forum Thread
- Forum Poll: Deciding the Peg Stabilization Modules Implementation Timeline - Forum Thread
- Christopher Mooney: I have mixed feelings about the
PSM. But, if I can convey this one idea, it would alleviate a lot of my concerns. Vault holders have a very strong market incentive to maintain the peg. ThePSMwould allow us to create an emergency mechanism on the edges of where we would find the peg to be so far out of bound that we would want this Module to spin up and take over. So the red is where thePSMwould be running. The green on top and below the red is to indicate that thePSMwould have a debt ceiling.- Any time the peg would go over the limit (in the diagram, 1.03), the
PSMwould be this natural mechanism to dump some stablecoin collateral into the system and get Dai out. This should, in theory, outside of these ranges, push back down to the 1.03 mark. Once you hit thatDC, you go back to the market incentives that the vault holders have, which appear very strong. A Vault holder in this area will create a vault and mint Dai to arbitrage the peg. - And on the low end, you could unload the
PSM. We have a lot of control mechanisms to control the low end. We can more likely unload this thing closer to a dollar, so we can use other mechanisms around this lower area.
- Any time the peg would go over the limit (in the diagram, 1.03), the
- Christopher Mooney: Sorry if the diagram doesn't convey the meaning that well, I just made it. The more we tighten this band and get to 1, the more I think it will remove the very natural market mechanism we have with vault holders. There's an entire mechanism for balancing the peg that we would snipe out the closer we get to that dollar mark.
- Christopher Mooney: I like the
PSMas a tool that sits on the edges and prevents situations like the one during Black Thursday; I think at some point, we saw 1.14 on an exchange. That would allow this instant, natural, cycle down, very low friction, permissionless, way to do that. I still believe in the core mechanism of the Maker Protocol: vault holders being the ones that arbitrage the peg. I want everyone to be extremely aware that if we kill that mechanism (the green area on the diagram), that could be disastrous for the protocol. The same way as being off the peg is disastrous.
- Wouter: Looking at the peg today, if it starts trading outside of that range you indicate, then it shows that the mechanism that's supposed to take care of it is not working well enough, so you're getting to a point where you would get into a negative spiral or negative feedback loop. So pushing back to that area, supporting the market makers, to me, it makes a lot of sense, but I'm not a risk analyst.
- Rich Brown: I'm trying to wrap my head around how much downward pressure the
PSMcould possibly exert before it exhausts itself. How much utility can be brought to bear here? If there's a significant deviation from the peg, people are piling in, and we don't want that to happen; How much of a buffering effect can this thing have? Or is the extent of the utility here that once the buffer is expanded, that's the end of the utility of thePSM?
- MakerMan: My reaction was, "have a spread of PSMs." The "wide" is the liquidity provider of last resort. The narrow is that you want to provide some liquidity. That's the one you put the lightest
DCon. Your job is good when these things aren't being tapped at all, and it's living on the narrowest range. And you put a lightDCthere to provide liquidity. You take the bet. If it's down, we'll buy it, because we're going to be back at one and we know it. If it's high, you sell it and get back eventually. It's the wide range where we experience heavy hands. If we hit 1.02 or 1.03, we come in with the heavier hand, and we take that bet because we know that it's going to go back to 1, and it's our job to do it. We need to provide liquidity to markets when they need it.
- MakerMan: The question is, how much? Someone asked me about how much. We talked about 50 million. How do you get it? It's a question of tools you apply to the system. I like the idea of at least starting with a wide
PSM. Maybe at 10 or 20 million USDC as an experiment? Along with reducing rates and making sure our debt ceilings are high so that it's used. We can take a stab at this market with 50 million Dai and see what happens. Compound is offering two times the Dai market cap as a yearly return on a billion dollars. That's a big driver. It beats us hands down. How much cap do you have to have to be able to absorb this? You can do the math on that and put it out. From what I'm seeing, it's decaying, though.
- Christopher Mooney: I suggested something similar back in April, and fell out of love with the idea because I was worried about its effects on market behavior. Currently, today we expect the vault holder to see the market incentive: high peg means an arbitrage, but they have to lock their capital up and hope or have faith that the peg will eventually return back down so that they realize the profits. While they're doing that, they're taking the risk. With the
PSM, market makers can instantly turn that profit, MKR holders bear all the risk until the peg drops back down. I want to highlight this because I have a concern, the same way we added USDC-A as an emergency facility for liquidity for liquidations, that governance will see this and eventually will want a half a percentage deviation from the peg, or 25 basis points and lock this down, and strangle off the free market or more virtuous cycle that would exist there.- Christopher Mooney: When we do that, we shift the entire mechanism where we charge this rate for holding these assets over time, and that accrues in the surplus buffer, and that compensates MKR holders with the burn. That burn exists because we have a risk on the other end of the market, where we would need to mint MKR to recapitalize the system. The risk of the long-hold mechanism is borne entirely by the Protocol and MKR holders. We give the profit to the market makers. The closer you get to the 1:1 peg, the more we kill our existing incentive mechanism, and we create this weird perverse incentive that could be disastrous for the protocol. It's hard to say. I'll let that idea stand. I have other objections that I think we should think about.
- Monet Supply: I think that looking at other entities that have started market-making, as Japan in their government bonds, it's really tough to step back from this market engagement once you start it. We agree that it will be tough to pull back the support if we start doing this.
- Rich Brown: The question of how do we unwind is important. Do people have thoughts on how to mitigate that risk?
- Akash: This is my own personal opinion, but I'll just throw out what I think might be interesting. Having the USDC-A vault at 4% is interesting, but that's for market makers that want to take on risk and are willing to ride it out for a year and pay 4% for the year. The problem is that there are not enough market makers with infinite liquidity, so at a certain point, the Protocol providing liquidity at a very high payment rate is maybe OK. So what I'm suggesting is having the USDC-A vault at 4%, as it currently is. But also having the
PSMthere, you switch 1:1, but you pay a 4% fee, let's say, to immediately convert your USDC to Dai, for example. So that a 4% fee is immediate, and it goes to the Protocol. So now you have two mechanisms battling to get the peg stabilized. This is my personal opinion; I have no idea if it's correct. I'm curious about what other people think.- Christopher Mooney: The 4% would go instantly to the protocol and MKR holders. So it compensates for that risk over the next year. The problem is, "what happens to the Protocol if it holds the asset longer than a year?"
- LongForWisdom: We have a similar problem with vaults, right? They pay 4% a year, but they pay it gradually over the year. If the collateral type explodes before the year goes out, we're still on the losing side. It switches from long term to short term. It's not fundamentally different. It just switches that around. It could be useful to have one set of tools that allows you to capture that immediately and the other one that allows you to do it over time. That could be useful.
- Akash: To be clear, I suggest having both. Having as many tools as possible to bring the Dai peg back down to a dollar.
- Brian McMichael: I like the idea of having a limit at 1.03 to 1.05. That puts us in a situation where we're saying to the community that Dai will not go beyond that as measured against USDC. That is an open question for the community to determine the level we're comfortable with. We were at 1.02 for a few weeks there, and that was not very comfortable for many people.
- Wil Barnes: This might change the current price situation, just hanging out at 1.02.
- Cyrus Younessi: The real question is not what we set the bands to or even what we set the fee to but rather what is the max tolerable debt ceiling for the collateral type. The most important takeaway is that this is an increase in collateral. The notion that we went from one model of incentivizing market makers to mint Dai and wait for the peg to converge back, versus now swapping that out immediately on the fly. I don't think that's as important, because the end state for Maker in both situations is the same: Maker has a bunch of collateral, and there's a bunch of Dai outstanding. The New
PSMmakes that more profitable for people who interact with thePSM. You can view it as a subsidy or a negative rate. The real issue is, "how big would we be willing to go with the USDC-ADCand how big with thePSM." The bands and the fees and all that are, for me, are second-order details.
- Brian McMichael: We're using USDC as a measuring stick, which has five times the market cap that Dai currently has. I am of the opinion that US citizens are using Coinbase to on-ramp dollars, where they can do that for free into USDC. They can off-ramp Coinbase onto Ethereum Mainnet, and at that time, they need to dump USDC for something else. If they want to keep in dollar-denominated terms, they'll go to Dai or else they'll go to ETH, but we're not in a market where people are buying USDC to trade back into dollars. If this were a Forex market, we would be seeing fluctuations between competing currencies, and I think we need to understand the macro situation of the fact that we're going into another bull market, and people are trying to throw dollars into Ethereum. We're just a more desirable place to be right now, but we don't have the supply to get there. This may be a great way to go if we need to increase the supply to meet demand. I'm concerned with how long we have to carry this and the risks that that entails.
- Cyrus Younessi: That's the same as asking how long would we be willing to carry the current USDC collateral? If the current USDC is under our tolerable max debt ceiling, then we don't care for how long. Once you go above what you're comfortable with, is when you start asking how long do we have to keep this on the books for? And then you start getting into questions about how big is Maker's capacity to take on USDC and if they're willing to hold it beyond a reasonable timeline. I think the most important question is, how big is the community willing to let the
PSMget? I think that needs to be answered before we start talking about ranges and fees and 1.03, or 1.02.
- Cyrus Younessi: That's the same as asking how long would we be willing to carry the current USDC collateral? If the current USDC is under our tolerable max debt ceiling, then we don't care for how long. Once you go above what you're comfortable with, is when you start asking how long do we have to keep this on the books for? And then you start getting into questions about how big is Maker's capacity to take on USDC and if they're willing to hold it beyond a reasonable timeline. I think the most important question is, how big is the community willing to let the
- LongForWisdom: I'm not sure I agree. I think that the bands are one of the more important issues. I feel that there's a difference of opinion about how far off-peg is too far? We're currently at 1.015 or something, which is one and a half percent. Do people think that's too much? The kind of message that we're getting from the Foundation is that it is too much, based on this presentation.
- Wouter: You cannot express this as 1% is too much, 2% is too much. It's more about the average price over the longer period of time as long as its fluctuation is around the dollar. Right now, it's above.
- LongForWisdom: Sure, but even if we turn on
PSM, that can't push it below peg, right? Maybe push it to the peg? So we're never going to drop below peg, unless we get supply from somewhere else.
- Wouter: The nuance is that you're creating improved confidence in the system again, so people will start acting in a way that assumes that the peg is fixed again, and then you will see that fluctuation again. I think it's a matter of getting away from that negative feedback loop.
- LongForWisdom: I don't entirely disagree with you. I think that the definition of the negative spiral is possibly incomplete. You're essentially saying that this negative spiral will reduce the amount of Dai, which is obviously not something we want in general. But the system does have measures that when the amount of Dai reduces, then the price of Dai drops below one, and we can start increasing the
DSRand other things to prompt more demand. - Wouter: I don't think that's the case. A system that's out of balance overall has a confidence problem. It's not just Dai users. Otherwise, the peg would be restored. I think that the general confidence in the system is much broader than just Dai users.
- LongForWisdom: I don't entirely disagree with you. I think that the definition of the negative spiral is possibly incomplete. You're essentially saying that this negative spiral will reduce the amount of Dai, which is obviously not something we want in general. But the system does have measures that when the amount of Dai reduces, then the price of Dai drops below one, and we can start increasing the
- Gregory Di Prisco: LongForWisdom, you're referencing a very short term cycle where we have these levers to pull that affects the Dai peg, but I think that the spiral that Wouter is talking about is more long term and systemic to what we're doing. If you look at the way that Dai has propagated in the market, its utility is derived from network effects via our integration partners. I deal with the majority of our integration partners from the Foundation standpoint. Since March, we've seen six decentralized applications that decided to use USDC over Dai. That's troubling to me, and when I called them to ask why, they unanimously cite the instability of Dai. That's tough to overcome but not impossible if we take the peg seriously. I'm nervous if we start tapping them for providing demand in the first place. I think that's a really dangerous precedence to set. Imagine that one of the partners integrate and start being successful, and you come back to them and tell them that they're providing too much demand, stop. It's going to provide a massive deterrent to integrate Dai in the first place, and it's going to have negative effects on the overall network. If those network effects start to deteriorate, we lose our moat and competitive advantage.
- LongForWisdom: Broadly, I agree with that. As the natural demand decreases because the peg is too high, that pushes down the demand for Dai, which then allows us to raise rates, which increase the
DSR, which increases the demand for Dai again. At that point, we can offer theDSRagain, and all these integration partners are integrating this coin, which has benefits to holding it, holding Dai, and offering interest in addition to their regular functionality. I'm not saying you're wrong; I'm saying there is a chance that it will come back again. The system incentives push that demand to come back again. The reason we have this problem is that we have extended our demand-side and we can't keep it on the supply side. So the solution is to increase the supply side, or reduce the demand side until we can keep up with it, right?
- LongForWisdom: Broadly, I agree with that. As the natural demand decreases because the peg is too high, that pushes down the demand for Dai, which then allows us to raise rates, which increase the
- Wouter: I think that's where the reality is a little bit different. You need integration partners; you need business partners, who put a lot of work, time, and investment into building the system and supporting the product. You cannot tell them, "We're currently off-peg, we're going to more-or-less shut down your product because the use case is destroyed for some time. But come back in January, and things will be fine again." That's where the theory of the economic model clashes with reality. That is exactly the concern that we hear that as a governance community, they are not appreciative enough of this. You say that there is a chance that they come back, but if you look at the investment needed to support a product, I would say that there's a chance that they don't come back.
- Kurt: There's a very high cost to doing these integrations from the engineering and logistics perspective. Once a project decides between Dai versus USDC, and they go with USDC, it's not like they can flip a switch and go to Dai later. It's a very sticky thing, and you get these network effects that Wouter alluded to, and it becomes non-linear.
- Wouter: This is a very interesting phenomenon. This is very different from vault holders, for example. I think that people looking for leverage will better understand and know that they can change markets and refinance. Dai integrations and Dai users are very different. It points to the same thing that we need to consider these use cases and that perspective more in-detail, and realize that it's not a symmetric situation.
- Akash: My own personal comment, in terms of positive feedback loops, right now, it's a little bit negative. Nobody knows when to sell Dai. So they kind of step out of the way, which allows Dai to go higher and higher above peg. There are two negative feedback loops:
- One is, Natural market makers are hesitant to sell because it can blow up in their faces.
- The second is vault owners who are doing leverage trading are unsure too because when they're closing out their vaults, they're not sure where their worst-case scenario buying Dai could be.
- To add, speculators will come and buy Dai because the downside is 1, but the upside is asymmetric.
- If you have the
PSMwith a fee at 5%, people will actually step in front of thePSM, because they know that the downside is at 1.05, so they'll step in at 1.04, knowing that it will come back to 1 dollar. Those are asymmetric risks. Having thePSMand other tools where people are confident that there's a big market maker at 1.05, the MKR holders create a positive feedback loop to get the price down.
- MakerMan: If we take the outside of the trade around 1, you can use it to test your auction system. Particularly from the high-side down, you can buy Dai at 1.05 and basically offer the auction, basically take anything down to 1.0. And do it periodically. Just throw it out. You can literally test the auction by being the outside bidder in the market. And everybody will sit in front of us. Great! Let them! We want them to. Ideally, they will sit there with Dai, just holding it. Will it be a profitable trade? I don't know. But the fact that it limits the peg at 1.05 or whatever number you pick, I think that's a great thing, it kind of all works. And the idea of using the auction system to take the in-between, just offer it. Whatever the market will give you in there and wants to, let it because you don't care. You want to recycle that trade and be done with it. And get it out of the system ultimately. And don't let it go if it goes below 1. Be like, "I'll keep it. Thank you. We'll try again later". It's great from the auction system perspective. Doing this
PSMto collect collateral to throw at the markets on both sides. The low side wouldn't work, I guess. The high side definitely works.
- Will Remor in the chat: If this is used as a tool for monetary policy, would it make sense to add some clear boundaries on which kinds of situations this could be used, for what purposes, and at which DC/fees? The risk I see is that this becomes the default mechanism for price stabilization down the road because the system gets too accustomed to “easy fixes.”
- Rich Brown: I'm trying to understand where the incentivization mechanisms go in this Protocol when there is this soft landing or appeal to authority. Does this allow actors in the ecosystem to exploit the fact that they can play games with the peg, knowing that the Protocol will pick up the tab later? I'm not sure that there's a clear model of how things could go wrong with this.
- Akash: There's always a clear risk with USDC, or if we use TUSD, or whatever other stablecoin. There's always the underlying risk of that collateral.
- Rich Brown: We're changing some of the game theory here, so my first question is how could it be exploited for gain. Actors in the ecosystem to take advantage of this.
- Lev Livnev: inaudible
- Rich Brown: I think the question was, "Does the
PSMaddress the COMP risk?" - LongForWisdom: That's what I got. Does anybody want to address that? It seems like a reasonable concern to me with the COMP farming that if we just offer a large amount of Dai for USDC it will potentially get taken for the COMP farming.
- Rich Brown: This is not an infinite bag of stability. There's a hard limit to how much of a stabilization mechanism this thing can influence upward and downward pressure, right?
- Rich Brown: I think the question was, "Does the
- Kurt: It depends on the debt ceiling that you put on it. The gamble is that by providing this sort of upper limit, and you can absorb X tens of millions of Dai, once you tap out that
DC, there's no more ammunition left and the peg can proceed to deviate beyond that point, and then it becomes a moral hazard where Governance gets very tempted to expand that debt ceiling, and Dai becomes more and more a wrapper for whatever stablecoins are in thatPSMbucket. I think that is the scenario that a lot of people worry about. That Dai becomes a USDC wrapper.- Rich Brown: I'm trying to circle around that. I like numbers. Some kind of back-testing, or as an example, something saying "during Black Thursday, this would have lasted two weeks or seven hours". There's a risk-reward scenario here that needs to be understood. How much pressure can this thing absorb and does that additional window stability allow the ecosystem to come back to sanity and arbitrage the peg back down where it should be? It's a tool, how effective is that tool, and in which situations could it be deployed, and for how long would it be effective?
- Wil Barnes: We don't necessarily need to keep USDC as the only stablecoin. We can add more stablecoins, so we don't have exposure to just one collateral type.
- Wouter: I think it's a one-sided question. It's not really a question of "will it fix this?" It's more, "will it improve it or make it worse?" Same thing with new collateral types on vaults. You don't object to a new collateral type because you think, what happens if it bumps into the debt ceiling? It's just one more collateral type that helps. This is the same thing. It's one more tool that can help.
- Rich Brown: If it's a useful tool in the toolkit, that's great. I just want to get a sense of how useful it is and for how long that utility would last and what the risks are.
- Chris_p: It's the same class of risk. We can diversify stablecoins, but ultimately it will have the same effect. Calling it a USDC wrapper, or a wrapper for stablecoins is the risk. It seems there are other tools in the toolkit, as opposed to something that seems a bit new.
- Gregory Di Prisco: I'm trying to guess how much Dai we actually need to satisfy the COMP demand. I don't think that it's as much as we think it is, because we get so much COMP earning leverage with Dai there. There is a lower bound on the amount of yield that a COMP farmer will accept. If you can go with USDC to BlockFi and earn 8%, you're not going to mine COMP for 5%. I haven't done the calculations, but I imagine that there's a way to do this more accurately.
- LongForWisdom: Currently, we're dealing with COMP, but there are lots of other projects that are planning liquidity mining incentives as well. If we do have enough Dai to solve the issue with COMP, great, but I think that we'll find out that the problem comes back again soon.
- Gregory Di Prisco: Yes, COMP today, something else tomorrow.
- Sam MacPherson: The choice seems pretty clear to me. We either have to grow the stablecoin supply in whatever way, be the
PSMor USDC-A, or the peg is going to break. We just have to get the supply up. I don't see an alternative at this stage.- Rich Brown: Any idea here that the
PSMas part of the toolkit will help bootstrap and create some breathing room that allows more actors to come into the ecosystem, which increases the supply, and that's the mechanism. - Akash: I'm my opinion this is a virtuous cycle. The true solution is that we don't want to be the wrapper stablecoin. But you have to start off with that so that you can raise the ETH and wBTC debt ceilings because those people will have more confidence that when they need to close out, they can get a good Dai price. That is super important to them too. We want to be fully diversified, but I think this is a starting point, and it'll bring confidence in the other collateral types also. Just my opinion.
- Rich Brown: Any idea here that the
- Rune Christensen: I want to make a couple of arguments echoing what's already been said. The first one is that it's very important to distinguish between the question of whether the
PSMis a useful tool for Maker Governance, then what needs to be done about the peg emergency, the broken peg. It's been broken for three months, and it's a fundamental imbalance of the system. There are lots of different options here. One option that thePSMmakes even more efficient is massively increasing the stablecoin debt ceilings. There are other options, as well. There's adding MKR as collateral; there are negative rates; there's also letting the peg break. What's important to keep in mind is that even if you're against any of these particular options, it doesn't make sense to discount thePSMas a tool in all other situations. Unless there's actually a problem with it as a tool, but that's not what's being discussed, right? The discussion is that we're looking atPSM, but we're actually talking about stablecoin debt ceilings. I think it's very important to keep those two things separated. You're maybe against a high stablecoin debt ceiling, but there are actually cases where thePSMwill be a useful tool.
- Rune Christensen: One specific example is that if you have a
PSM, even with a low debt ceiling, let's say with 10 million, or 20 million, basically a debt ceiling that's not enough to actually fix the peg. We're not talking about stablecoin expansion. We're talking addingPSMand shuffling the debt ceilings around, so now there are some stablecoins sitting in thePSM. It still has a very remarkable effect on the entire structure of Maker and the entire paradigm of Dai liquidity. It means that now you're always able to cash out of Dai at extreme levels of liquidity. Potentially 20 million in one go, and you have a totally guaranteed price for it, right? Even when the peg is broken, just because the peg is right now trading above a dollar, that doesn't mean that you can actually come as a vault user and you can instantly generate, say, five million Dai, or 10 million Dai, or 20 million Dai, and actually cash that out at any sort of reliable value. You really need to have trading counterparties, and you need to be in every single DEX. It's quite complicated to actually cash out Dai at a good price, even when the peg is broken. ThePSMtotally changes that. And from that perspective it just, objectively, in my opinion, improves the user experience for any type of user of the system, but particularly it's important that it improves it for vault holders, because that's where it creates this positive effect that helps fix the real underlying issue, which is the imbalance of the system.
- Rune Christensen: The last comment I want to make is on the imbalance of the system itself. No matter what, Maker Governance has to fix the peg. It's complete suicide to actually signal an aversion to fixing the peg and preference to leave it broken because none of the options are pretty. At that point, if you signal that and the vault holders truly believe it, then you're basically telling them to run for the hills. Because then there's no guarantee that the peg will even hold, so you have no idea what it's actually going to cost you pay down your vault. It obviously generally increases uncertainty and even taxes the system on the path to emergency shutdown. I think that's very important. In that context, it's not OK to leave the peg broken. The one thing that the system is supposed to do is to keep the peg. That's always been the case. I think that there's a number of arguments to be made for different approaches. Like negative rates are the really classic idea of what you have to do. But that's also what's been pointed out earlier. In this early phase of the system is much more disruptive when you're eroding all the faith and all the entrenchment that has happened so far. When you look at things like adding MKR as collateral or adding a lot of stablecoins to the system, it's taking a risk in a different way, but it's a solution that actually has a future, in my opinion, as compared to basically saying "let's wind things down". It's really the question of whether Governance and the community expect Maker to move things forward or move backward as a response to this situation.
- Thanks, Rune, for the thoughts. This is a long and very complicated discussion. Probably this will not be the last one that we'll have about the
PSM. There are some very vibrant threads going on in the forums. I encourage people to get passionate about these particulars. There are decisions to be made about the next steps. I don't know what those are, and I would like the community to come back and make sure that its voice is heard, so we know how to move this thing forward. So please join us in the forum.- LongForWisdom: I already laid the next steps of what we're going to do here based on what happened last weekend. So just to show what's been said in the forums. We're going to spend July talking about this. There will be a segment in each Governance and Risk Meeting where we can discuss it. Hopefully, by the time we get to the third week of July, we'll have some idea of what we want the parameters to be so we can vote them on-chain at that point. And, by the end of the month, assuming that the on-chain polls are successful, we'll include them in the executive with the parameters that everyone's decided.
- Rich Brown: Thanks for that far more detailed recap than the one that I wasn't prepared to give. There are things to be explored, pros and cons, refinements, game theory, and its optimization mechanisms, we talked about the toolkit and the benefit of adding incremental improvement to the Protocol, so I very much encourage people to join in.
- Christopher Mooney: I thought I'd mention that this is such an amazing project, we're at the 100th governance call, and we're discussing this very coveted good problem to have, which is that there's a ton of demand for Dai. Not just from COMP, but from all over the market. And most startups, most companies, would die to be in this position, no pun intended. Or maybe pun intended. I just wanted to call out, and I know that we have hard decisions to make, and it's going to be difficult to figure out how we scale, but this is the right set of circumstances that we want to grow this project. It's exciting to be here. It's exciting to be with you guys.
- Rich Brown: Aw. Thanks, Chris. Ending this call in a positive spin. It didn't occur to me in the past, but maybe we should do that more often. Thanks, Chris. It's a great addition.
- Rich Brown: That's it for the call. Thanks, everybody. Let's continue the discussion in the forums. Thanks for joining us.
MCD: The Multi-Collateral Dai system
CR: Collateralization Ratio
DC: Debt Ceiling
ES: Emergency Shutdown
SF: Stability Fee
DSR: Dai Savings Rate
MIP: Maker Improvement Proposal
PSM: Peg Stabilization Module
- Tim Black produced this summary.
- David Utrobin produced this summary.
- Artem Gordon produced this summary.
- Galla Guillen produced this summary.
- Juan Guillen produced this summary.
- Everyone who spoke and presented on the call (listed in the headers.)