episode-98.md

Episode 98: June 25, 2020

Agenda

• 00:00: Intro with Rich Brown
• 03:40: Governance at a Glance with LongForWisdom
• 06:53: Vault Compensation Plan with Monet Supply and MakerMan
• 46:37: Weekly MIPs Update with Charles St. Louis
• 50:21: Governance Cycle Review with the Governance Facilitators
• 55:10: Liquidations v1.1 with the Smart Contracts team
• 58:17: State of the Peg with Vishesh Choudry

Video

https://www.youtube.com/watch?v=9U7NG8JRPM4

Introduction

Rich Brown

Agenda Summary and Talking Points

00:00

• Welcome to the June 25th edition of the Scientific Governance and Risk meeting, my name is Richard Brown. I am the Interim Governance Facilitator, and I'm joined in moderation duties with LongForWisdom, the community-selected Governance Facilitator.
• We have a jam-packed agenda as usual. We also have something happening today that has been on a slow boil for a long time, and I am interested in hearing this presentation. Monet Supply and MakerMan have been toiling in obscurity for a long time to work with a group of actors in the ecosystem who believe would like to be compensated for the zero-bid auction event. They have done a great deal of work with stakeholders to come up with a plan about how the Protocol could compensate the zero-bid auction individuals. They're going to be presenting that plan for the first time publicly today. I would like to encourage everyone on this call to pay close attention to this. This is a momentous event. This is one of the first sorts of a completely organic collaboration of ecosystem actors getting together to affect change in the system. We've had it on a smaller scale before, but this is governance in action. We talk about autonomy and agency, the ecosystem shipping the protocol, but this is the probably largest. There are lots of things happening here we need to understand. They put a great deal of attention and time and effort into this. Now it's time for the governors of the system to understand what this plan is and offer their feedback to keep the ball rolling. That is my framing of this situation. I am looking forward to it. Not totally sure what's in it. It will be very educational for all of us. I encourage people also to take notes of any questions that they might have about this process, what is it going to look like in the future, and the implications. Let's make sure we get these questions asked for the official public record.
• Discussions happen in the Forum.

Forum Recap

LongForWisdom

Governance at a Glance

03:40

• LongForWisdom: It's been a semi-active week with the whole liquidity mining/farming going on.

Threads

• Vault Compensation Plan
  • As Rich previously mentioned, Monet Supply and the Vault Compensation working group have posted the Vault Compensation Plan Proposal on the forum. Go check it out after the presentation. Review the details.
• Liquidity Mining on the Maker Protocol
  • Monet Supply also created an interesting discussion on liquidity mining on the Maker protocol and included several calculations on what people could be earning and how it compares to the rates of Maker.
• Wallet & CDP Liquidation-based Predictive Scoring Model
  • A user I hadn't seen before, ChristG87, created a thread proposing a predictive scoring model for liquidation risk, of how Maker could use a credit scoring mechanism to allow people to borrow at smaller collateralization ratios. Some interesting ideas.

Signal Requests

• Signal Requests
  • Created by Jiecut, to adjust the wBTC debt ceiling and risk premium. It should be moving on next Monday. The consensus results were 20 million debt ceiling and a 2% risk premium. So those are going to be the values going into the polls.

Vault Compensation

Monet_Supply and MakerMan

Vault Compensation Plan Presentation

06:53

• This is the initial presentation for the Vault Compensation Plan Proposal for Maker Governance. MakerMan and I haven't rehearsed this presentation before, so if you want to hop in at any moment, or if anyone has any questions, please let us know. Before we get into the details, there's a small disclaimer.

• I'll let everyone read independently.
• The intention here is that we're not working under the purview of the Foundation or another corporation. We want to make sure we're communicating the limits. That's included in the plan document itself.

• A little bit of background, although I assume most are aware.
• This plan is essentially the procedure for how we are going to determine those details as a community.

• We are trying to convey to the public and vault holders that this is an exceptional circumstance, and we don't want to create an expectation that these claims are going to end up falling on MKR holders in the future.
  • MakerMan: Let me add one thing. Go to the end of this slide: "The persons drafting or implementing this compensation plan maintain that the MKR holders have no obligation to issue any compensation..." This is a volunteer plan operated solely at your(MKR Voter's) discretion, and there is nothing in here that says that if you agree on wanting to do it that you will actually execute on that. We are very clear about this. This is an optional plan. Everybody opts-in, the governance opts-in, vault holders can opt-in by going yay or nay to terms and compensation.
  • Monet Supply: I think it's important to include this. We don't want to create expectations that we are not able to follow through on in the future. So we want to be super clear about where MKR holders feel their rights and responsibilities are.

• I'm going to walk you through the proposed stages of this compensation plan. It is going to involve some Governance action and some kind of project management and development work as well.

• The first phase is the current state - phase zero.
  • Develop a plan of action
  • Present it to governance and getting feedback here as well as in the forums
  • Eventually, we are going to put it in an on-chain governance poll for MKR holders to signal whether they support the plan as stated or not. The earliest that could run potentially, if there is support for it at this point, would be this coming Monday. But if people want more time to discuss, it could certainly be delayed from there.
  • If the poll passes, we are going to be moving to phase 1, the next phase.
  • If the poll fails, we basically need to figure out from MKR holders and the community overall what part of it was contentious, and then we can dig into how to fix it and gain the necessary support. That is mainly why we included the icebox section. It's stuff that, when we were drafting the plan, we thought of it as potential questions that governance might want to decide on but, in our opinion, it was simpler to narrow the scope of the governance voting, but that would be probably the course of action if this plan fails to gain support.
• Now assuming that it does gain enough support from MKR holders, we are going to be moving on to phase one.
• LFW: I think it's very unlikely that the poll will go up this Monday. We would like to give people more time to digest and understand the plan before voting on it.
  • Monet Supply: I think that is fair. There hasn't been a whole lot of discussion on the post, so I would love more feedback and people catching our blind spots before we move forward. I will scratch this when I post the deck, but we can push it another week or two or as long as needed.

• Once we get the governance poll up and assuming it does pass, phase 1, where we have two main things:
  • Firstly, we need to do a governance poll to determine the calculation method, which we call the method in reference price for determining compensation. Next slide, I will add more info on that, and I will probably have MakerMan pop in as well.
  • Parallel to that, the compensation itself is going to need to be distributed to vaults via smart contracts and then based on verified data about who suffered what losses and what-have-you. So, in parallel, while we are getting Governance to poll on this info or this question, we are also going to be trying to put out for tender(price discovery) some of that development work that is going to be necessary later in the process. So we can at least understand how much that will cost and how much governance is going to need to stump up for that.
• Maker Man: I will take over the next section. When talking about compensation, I came up with two different methods. The tab one, I was hoping I could just run across everybody and give a total number. It's just the data isn't there. So there are two ways we can look at this:
  • From the borrow perspective, which I call the tab, and what happened with them. When these vault users borrowed, they set their tab value, so you can use that as a really good measure of what they were thinking about their collateral.
  • Otherwise, you can look at it from what we think is a collateral deficit, and then it's what is the value that you want to put on it: at the time of the auction or ideally just one number for everybody.
• MakerMan: In this case, we have options A, B, and C. Is it the price on the day it happened that we pick or the price on the day we finally execute it or an average of those?
• MakerMan: Then, I had the tab change in relation to collateral change. Because when I started looking at these numbers even for the guy that self-liquidated and saved the system probably a half a million Dai-loss(vault 2288.) You look at him, and it works out for him too. He gets numbers, and they all line-up and so they all become manageable when you think about it from this perspective. So we just offer the different methods for choice for Governance.

• Monet Supply: This next page is for the less math-inclined people. Kind of a narrative version of how it would be calculated.
• Starting off by figuring out what percentage of the collateral they have gotten back and then the collateral rate percentage that MKR voters choose. You take what they have already received and subtract it from that. That's how you figure out the percentage of the collateral that they are going to be getting.

Nominal Return

22:22

• Basically, remember, I calculated this nominal return. The liquidation ratio and the fee determine the nominal return, and you can look at the auctions' stuff and determine where you think auctions were good and where you think they were bad and then what you want to do. So you just use the collateral deficit as the relevant means of dealing with the zero vaults. However, you want to classify your vaults, take a pick.

Tab Change

• And then I had done some work that you could also look at it from the tab change. Like if they are liquidated, their tab goes to zero, and their collateral goes up for auction, and you could look at it from that perspective as well and just compensate with respect to tab. And that actually works out, when I think about it, as an MKR holder, a much cleaner concept of what I'm trying to do here, why it's a separate issue. But those are the two ways to do it.

• Monet Supply: I'll post the slides as well, and then this next page also has more numerical examples of what those different options would look like.
• These are two vaults. One self-liquidated (vault 2288), the other who was liquidated at auction (vault 3834).
• Just to recall what those different A, B, C, and D options: (back to "Phase 1" slide)
  • A: nominal collateral return percentage and the price on Black Thursday
  • B: same and the future day when the compensation is awarded
  • C: the average
  • D: the tab method
• If you look back at what the numbers would be, assuming that price on Black Thursday is $100 and price the day that we issue the compensation is $240, you can see what the relative numbers would be.
• Which option we choose here will have a big impact. If we choose option B, which is using the ETH price whenever we actually issued the compensation, it's likely going to be a lot more than the other options. On the other hand, vaults which might have expected to receive some of their ETH back, so if we paid them back according to the price on Black Thursday, it might not seem as fair to them. So probably the biggest single decision that Governance has as part of this process is to decide what method we are going to be using to calculate the claims.
  • Rich Brown: Just to clarify, this is the opportunity cost of waiting for the compensation plan to finish.
  • Monet_Supply: Sort of. It's us, as MKR holders, taking on the risk of ETH price going up in the interim. Now, if the price went down, if it crashed again, potentially, we would be off the hook.
  • Rich: We're assuming that the price of ETH is always going to be better than whatever happened on Black Thursday. It's interesting.
  • Monet Supply: It's a pretty fair assumption, but you never know. Any other questions on this slide before I move forward?
• Rich Brown: let's clarify what self-liquidated vs. auction means. Actually, I think that we all understand what auction means. Can you talk a bit more about what self-liquidated is?
  • Monet Supply: Yes. In the case of vault 2288 but kind of all of them, they probably got paged on Black Thursday for their price alerts for ETH, and then they either just wanted to avoid the liquidation penalty, or maybe they realized that the auctions were not working as intended. So then they basically panic-bought back their Dai at really high prices, panic-sold their ETH just to be able to close up the leverage positions themselves.
  • MakerMan: They did this to salvage some parts of their vaults, like 2288, he salvaged 585 ETH rather than being totally liquidated and getting stuck sitting there with nothing and not doing anything. And the system losing another half a million Dai on this vault alone.
  • Rich Brown: It's not about motivation; it's about the mechanism. These people just chose to close out their positions, and they spent more than they should have in order to do that? Or these people attempted to close their positions and were liquidated at some point during that process?
  • MakerMan: He closed and sold like 2000 ETH at $76. It was like 2,200 at 76 or 75 just to raise enough Dai to salvage the ETH.
  • Rich Brown: I just want to be clear. So we have this category of people called self-liquidated. So these people that completely bought ETH in the open market to close-up their positions, or does this include people that attempted to rescue but were also liquidated at some point during that process?
  • MakerMan: This is only one example.
  • Money Supply: Theoretically, you could be in both groups. The self-liquidating is basically selling ETH to pay back Dai, and the auction is that when you hit the liquidation ratio and got sent to auction.
  • Rich Brown: Are those mutually exclusive groups, or is there a combination of the two?
  • Monet Supply: Theoretically, somebody could have paid back some of their debt, and then the price kept falling, and they got auctioned.
  • MakerMan: I already know from looking at examples of people who deposit their ETH and then got liquidated. I was looking through the first 500 vaults, and there were a number of examples where people tried to stave it off and then just got liquidated straight up and deposited ETH or whatever.
  • Rich Brown: That's an interesting distinction, but we should definitely keep going. Thanks, Monet.
  • Monet Supply: Definitely. And that is part of the icebox as well, whether or not we are going to include them. So if people have thoughts and they don't think they should be included, it's certainly not my opinion, but it's definitely open for discussion, and we can clarify further on the thread in the forum.
• That encompasses phase one, where we chose the compensation method and then put out some of the development work that we are going to need in the future for quotes.

29:29

• Phase two, the main governance decision here will be determining the compensation percentage.
  • So we recall from the previous slide. The compensation percentage determines what their collateral deficit is. Because in the example I selected 18%, the person already got 11% back and is going to get another 6%. That's going to have a big impact on how much compensation individual vaults receive.
  • One thing that will be made clear in the poll, of course, the plan we have is to pre-calculate all the compensation amounts for ETH vaults-holders that were liquidated at auction.
  • Since the BAT auctions and also individuals who self-liquidated require a little bit more manual review, we are going to ask Governance eventually for 150% of the token calculated for ETH liquidation. With the additional 50% to cover claims for BAT auctions and for self-liquidated vaults. Any unused funds, of course, are going to be going back to Governance, which I will describe a little bit further forward.
• So we will have a Governance poll to determine the compensation percentage, and from here, once we already have enough quotes for the necessary development work that we need to do to create the compensation contracts, we are going to put an on-chain poll up and an executive vote to request funding for that development work.
  • I know that just recently, the MIP14 Protocol Dai Transfer did not go through. So it seems pretty evident that Governance has some reservations about sending money outside the protocol. It is going to be 100% necessary to do that and to make this plan work, so there is no getting around it. So anyone who does have reservations on this, definitely make your thoughts known in the forum. Any ideas about how we can help Governance feel safe about moving funds out of the Protocol.
• Then the end of Phase two, once we've gotten funding approved and issued for the development work, that's when we basically get down to business, develop and audit the smart contract for dispersing compensation to users. We have some more specifications on that in a little bit.
• This is one bullet point, but this may well take a while. So I think I would note that. It could be a while to go from phase two to phase three before we actually have a working contract that's safe and ready to put funds into.

34:18

• Once we do have the smart contract created and ready to go, again: this is moving funds out of the protocol, and it's going to be a potentially large amount. But we are proposing to have an on-chain poll to include funding of the compensation contract into the next executive and then an executive vote to actually issue it to the compensation contract.
• Something to bear in mind, which is kind of obvious, but the compensation amount could be hundreds of thousands if not millions of Dai, so it could result in flop auctions if there is not enough Dai in the surplus buffer after the vote passes. So we didn't suggest to include any sort of strategy about how to collect the funds for the compensation, but it's something that governance might want to consider.
• Once the contract has been funded, then we go to the claims period where vaults are actually able to retrieve the compensation.
• Prospectively we are thinking of having the claim period extent for approximately three months, potentially a little bit longer at the discretion of governance. Then, any funds that haven't been claimed from the compensation contract at the end of that period, Maker governance would then have an executive vote to claim those back into the surplus buffer.
• Any ETH vaults are going to be already pre-coded into the contract with whatever their amount is based on the methods and percentage polls earlier so they'll just be able to go to either interact directly with a contract or go to a website to claim. It'll be simple for them.

• Then, for individuals who self-liquidated or BAT vaults who were auctioned, they aren't going to be pre-coded into the contract to receive funds. They will need to submit details of their claim within 30 days to a forum thread and then the compensation review team, the working group, is going to be manually reviewing those and determining, based on the method and the compensation percentage determined by MKR voters, how much each one of those vaults will receive.
• The perspective idea of how we thought we would add those individuals to the contract would be basically to have an admin function controlled by a multi-sig, and having a timelock of roughly a week so that if we approve adding a certain compensation amount, but it looks like it's not correct to MKR holders, they would have the ability to reclaim all the funds from the compensation contract before it's dispersed.
• At the end of the 90-day claims period, when anyone who wants to claim funds has, MKR governance would call the contract to self-destruct and return all funds.
• The flip side of the flop auctions earlier is that if there is a bunch of surplus that comes back, it could lead to immediate flap auctions.

33:37

• Phase four, where basically once the compensation has been claimed, MKR governance just takes all the remaining funds back.
• I know MakerMan had put this in the forum post on the plan as well, but there's a good possibility that only maybe 50%, or whatever, of the approved compensation, is going to be claimed by vaults, or certainly less than 100%. So that is something to bear in mind as far as the total cost goes.

• This is like the TL;DR version of that whole compensation process.

• The appendices section can be found towards the bottom of the initial post of this forum thread.

• These are different potential questions that we didn't include as poll votes or stuff that we're proposing to MKR voters, but potentially MKR governance might want to vote on these if they decide that the plan isn't really fit for purpose as of now.
• I encourage everyone to consider if they think there's anything missing in the plan that they want to vote on or just look over the icebox and see if there is anything there that sparks your interest.

• I'm just going to skip over that, but there is a github link in the plan which Joshua had made, specifying what we need the contract to do.
• Any feedback people have to offer on that would be appreciated. Just speaking for myself, I'm not super technical, so people with more experience, I would definitely love to hear your thoughts.

• As of about a week and a half ago, we had around 100 hours tally total. I was proposing a rate of compensation of 50 Dai per hour, it seems pretty reasonable, maybe cheap for professional services, but it's up to the discretion of MKR voters.
• I think none of us in the working group are doing this first and foremost to be paid, but that being said, I think it's good to set a precedence of paying for work.
• Based on this amount of time and that pay rate, it's about 5,000 of cost so far, and we would still need to determine the cost for these additional items, but I expect it to be a very small portion of the total amount of approved compensation that gets voted through.

Discussion 1

Analysis on the zero-bidders

19:05

• chris_p: One of the things that no one has mentioned yet about this compensation plan is, what about some analysis on the people that took advantage of the system by placing all the zero-bids?
  • Monet Supply: I think it's out of scope for compensating the vaults.
  • MakerMan: You need to define what "taking advantage is" because for people to borrow Dai, they have to put up collateral, right? And this is a contract, and they get to manage it. What's the point, Chris? I don't understand.
  • chris_p: If you're talking about compensation here, fund recovery is important. We should explore all paths on how to get that money back. If we could contact the zero-bidders and say "Hey, would you mind giving the money back?" if there was some kind of negotiation there, in addition to paying people out, that should be discussed.
  • MakerMan: Let me give you an example. Coinbase had a little price drop. My wife pointed this out: "Who are we to say what the right option was?" You could debate the way it worked. The same thing happened to Coinbase; what did they do? They ate those losses and made people whole, but that was their choice, right?
  • Rich: There's a lot to go over. We have a plan that took months to put together. There are a lot of things to digest. Let me do some moderation here. Chris has a point, and that's something that the community might be interested in getting into. It's not about judgment; it's about exploration. If the community is interested in aligning some efforts to filling the other piece of the puzzle, I don't see anything stopping them from doing that. So this is a conversation that can be continued in the forum. I don't imagine it takes much time to determine exactly where all these funds ended up. It's worth exploring. I would really like to focus on the actual details on these slides before we start diverging into philosophy about blame and who won and who lost.

MIP14 and Working Group Payment

32:26

• Rich Brown: MIP14, we have talked about this in the last call too, and it's worth revisiting, so if people have concerns or questions, please look at MIP14 again to get the necessary context.
• Rich Brown: Second question or observation; you mentioned briefly at the end of the slide previously about the compensation that is going to be needed for the work that went into doing all this evaluation. I want to make sure that that's not lost. Because you guys don't have a horse in this race, I'm assuming, and you all put in a significant amount of work here. So, are you guys going to get compensated as well?
  • Monet Supply: Yes. In the end, one of the appendix sections is "working group payment." We are going to be recording all of the various costs.
  • Rich Brown: Good, I just want to make sure that you guys are not shy.
  • Monet Supply: No. It's totally up to governance, of course.
  • MakerMan: We threw that in the end because it's such a small quantity in comparison to the total that it's going to be looked at here, that it felt appropriate.
  • Monet Supply: Yes, and I think it's really up to Governance whether they approve it.

Thanks to the working team

42:04

• Rich Brown: I want to thank you to Monet Supply, MakerMan, and anyone else whose name I forget that may have contributed to the effort. It was a tremendous amount of work. I encourage people to dig into the full document and the forum thread as well. We have a lot to cover today, but I don't want to end this conversation prematurely, so if people have questions, please jump on the mic.

Fix Amount and Issuing Equity

42:40

• Joseph Q: I just want to thank them because they did a great job on it. I'm ok with everything. My only question is, "why not instead of doing flop auctions, why can't we just say we agree to an amount?" It's in your appendices, the icebox. Just do it over time as the system makes money because the system is not making any money right now. And then we do a flop auction. The whole point of MKR being an asset that could be auctioned off, and have stability on it; we would crash the MKR price, I believe.
  • Monet Supply: My thought on that is that maybe we don't need to be as hesitant with issuing equity as we think we do necessarily. Compound just voted to dilute themselves by like 80%. It's not a direct comparison, but I think it would end up feeling stingy to the vault owners if we basically just said, "we'll get to it when we get to it."
  • Joseph: The VCs will sell, and I'll tell you that right now. Do you want MKR to be trading at $100 dollars? That is the blunt reality. My guess is that compensation would be like $2 million. MKR is trading at $470 right now. If this gets to governance, then they'll dump it, and they will bid on the flop auctions.
  • Rich Brown: I'm not sure about it. There's a spectrum for these things. It's obviously worth exploring, and maybe even a model of what we would have. First, people need to vote about which levels of compensations they're interested in, and then after that potentially scheduling. It seems like there's a pile of options and people who are more clever than I can come up with the best options that would minimize the impact on MKR. It's an interesting discussion.
• Rich Brown: Something that we need to understand sooner rather than later, for true agency to occur, there needs to be some method of funding, to control their own destinies. We have talked about MIP14; that's going to get us beyond helping the people that did the work. But at some point, there needs to be a process to pay for things. And this seems to be the instigator to make sure that we get to a plan that actually works. This is going to be an interesting couple of months. I'm going to ask people again. This is really interesting stuff. It has impacts on our ecosystem; it has impacts on our culture; it has impacts on the perception of the Foundation externally. It has an impact on the people that engage with our ecosystem as well and suffered a lot. There are lots of things to be discussed here. Please engage in the forums.

MIPS

Charles St. Louis

Weekly Update

Weekly MIPs Update Forum Posts

46:37

Executive Vote

• We're in the last week of June's governance cycle.
• The executive vote went live on Monday to officially ratify the six proposals which covered amendments, onboarding a risk team, and the Declaration of Intent MIP. I'll share the link right here.
• The important thing to know is that the executive vote has four days to pass, as of Monday, when it started. The proposals can only move to accepted status if the executive vote passes within that time limit. So, this is a reminder to go and vote. There are still approximately 11,000 MKR needed to pass(vote now passed.)

Governance Cycle Review

• As for today's business, after my update, Rich and LongForWisdom are going to do the Governance Cycle Review, which is a time to summarize and discuss June's governance cycle with the community. It can also be used to discuss any upcoming activity for the governance cycle in July, which includes any potential submissions. I'll definitely use some of that time to take a moment to speak briefly about a new MIP submission plan for July that we've been working on.

July's Governance Cycle deadline

• The formal submission period for the July governance cycle is July 8th. If you have any proposals in progress, post them in the forums and get the review sooner rather than later.

Light Feed Proposals

• In other news, the whitelisting oracles and new light feeds proposals that passed the governance poll stage will likely be put in an executive vote within the next week or so. I'm not sure about the exact date, but that will be included in the forum update that I'll post later.
  • This Executive Vote will be going up on Friday, July 03rd.

Collateral Onboarding

• In terms of collateral onboarding, the executive vote for KNC and ZRX will go live tomorrow. The new community greenlight polls are live and will run for approximately 11 more days. The polls are for gauging the sentiment about adding:
  • LEND
  • MATIC
  • PAXG
  • LYNC
  • ENJ
  • WKT
• I do want to note that if the amendment MIPs that pass this month exec are ratified, the community greenlight polls will start in the third week of the governance cycle and end at the end of the governance cycle as opposed to carrying over to the next month's cycle. This improves the flow of the overall collateral onboarding process, so it's important to get this in to improve the flow.

MIPs Forum Subcategory: Proposal Ideas

• The MIPs forum subcategory Proposal Ideas has an update. It has a new entry by Andy McCall, and it's for adding a poll to prioritize collateral applications. Here's the link.
• The proposal ideas category is where community members can post and discuss any proposal ideas to improve Maker, the protocol, or the governance system.
• I encourage everyone to vote and go to the forums.

Governance Cycle Review

LongForWisdom

Overview

50:21

• We started this month with nine proposals that we submitted in the initial submission period:
  • MIP13: Declaration of Intent
  • MIP14: Protocol Dai Transfer
  • Four amendment MIPs, amending parts of the collateral onboarding process
  • A subproposal to add Cyrus as a risk team
  • Two MIP12 subproposals to onboard KNC and ZRX Vault types
• In terms of stuff that we ended up accepting into the governance cycle, that ended up being everything except the MIP12 subproposals because they didn't contain all the domain work that was required to make them valid for the MIP process. So Charles said that's being pursued under the weekly cycle for this month.

Inclusion Polls

• Seven proposals went into the inclusion polls. The outcome of the inclusion polls is that six of those continued to the governance poll. The only one that did not make it was MIP14, as we discussed. So there are six up for vote in the executive out of the nine proposals, which is reasonable.

MKR Voters in Polls

• Comparing the first cycle to the second. We had one submission in the first month, versus nine submissions in the second. In the first month, we saw around 5k MKR voting in inclusion polls. This month, it was more like 17k, which is interesting.
• In the second poll, the governance poll bundle, the amount of MKR was around 5k the first month and 4k the second month. That indicates that MKR voters are finding that poll less important or less impactful. I encourage everyone to vote in it because if it gets voted down, on the stage, then it doesn't continue. It's potentially a signal that MKR holders view that this stage of Governance Polls are too much effort.

July's Governance Cycle

• Charles already commented on it briefly. There are currently two MIPs in Request for Comments:
  • MIP14: Protocol Dai Transfer
  • MIP15: Dark Spell Mechanism
• I don't believe any of these will be submitted for the next cycle, although I could be wrong.
• There's potentially some other stuff coming from the Foundation to be submitted as well.
• I encourage everyone that has MIP ideas or is in the process of writing MIPs to submit them as soon as possible.

Liquidations

Mariano Conti

Liquidations v1.1 with the Smart Contracts Team

55:10

Flopper contract fix and impacts

• As you know, when we did the flop auctions, some of them got stuck, so we discovered there is a small bug in the flopper contract.
• During that, we also realized that the three different auction contracts: the flip, flop, and flap (or collateral, debt, and surplus auctions) require double liquidity when interacting with them in certain cases. That is already fixed, tested, and formally verified.
• As well as the flopper with the stop-debt auctions. We plan on fixing those in the next governance cycle, and that would imply deploying new versions of all of these contracts.
• This is a change that impacts everybody running a keeper like DeFi Saver, for example, with their UIs. But, to be clear, the only impact is that they would need to point to new smart contracts, so nothing in the interface changes.

Timing

• We wanted to do this for these weeks, so that the new collateral types, KNC and ZRX, would deploy with the same fix. But we decided that we better put it out to a community.
• This is not an emergency but an upgrade. We are going to put it out next week in the forum, with all the information: the smart contracts and the result of the formal verification.
• Hopefully, it is not a controversial change, and we can get to vote on it at the end of next month.

Analysis

Vishesh

Relevant Links

MCD system stats DAI 24hr VWAP Graphs

The State of the Peg

58:17

• Total Dai is at $130 million, so it's not a significant change.
• USDC-B and TUSD have been added as empty buckets, with that being the plan for USDC-B, and TUSD having a surprise token contract upgrade that requires a fix before being usable.

• Dai from WBTC is sitting flat since it hit its capacity. It's probably not going to move since it's a few individual users. Unless they decide to change their plan or the DC gets moved.

• Dai from BAT down over the last few days. It's obvious why, but we can go into that. Also, it's not a huge margin to begin with.

• Dai from USDC hasn't moved much in the last couple of weeks, but it's come down since May. That utilization isn't expected to move, given the current ecosystem for lending (I don't think). That has basically come down since SF was moved off of zero. A lot of that utilization was either temporary or due to the fact that it was effectively free money.

• Dai from ETH has been growing; there were a few spikes. Those spikes were not particularly sustainable, so they leveled out, and it resumed the normal growth path. This was some refinancing to Compound. Then basically it's been a slow but steady increase since it dipped back down, about a week ago. It's still sitting at around $116 million Dai from ETH.

• In the context of the peg, trading volumes have been somewhat light but still sustained.

• The Dai peg had come back to about $1.01 after dipping down a bit lower, and then it's just slowly gotten closer to the $1 mark since then. Partly that's explained by the supply increase and partly from what's been going on with ETH price.

• Following the (blue) lines of the ETH price, you can see the (magenta) Dai price level off during fluctuations. Though during a spike in ETH, Dai price didn't change as quickly. These fluctuations have been too frequent and too small for Dai peg users to purchase back on the market. It didn't create a huge shift in market behavior with Dai.
• It's also good to keep in mind that during this time frame, there was a little bit more Dai minted, but significantly more Dai supplied to Compound in the same time frame. This pulls from circulating supply.

• If you look at the minting of ETH from June 15th up to today, you can correspondingly see that increase of supply on Compound. Even if the circulating supply has gone down, the volume of trades occurring at elevated prices did not increase.

• If we look at borrows on Compound, they've increased but not as much.

• In the same time frame, supply had a $13 to $16 million increase, depending on the day you pick. There's a bit more Dai that was supplied than borrowed. Even though liquidity should have decreased, the Dai price didn't increase, but actually decreased. Both when ETH went up and went down. Just a bit odd, a curious indicator.

Discussion 3

Double Dip

1:04:32

• MakerMan: There's the first dip on June 12th, and the peg actually went down. Just like it's doing on the first dip here. Whereas the second dip, it bounced back, June 14th, and then it went down again. That's when we really got the spike. The first dip didn't do anything. So, maybe, it's the double-dip.
  • Vishesh: Well, it's all relative. ETH had been going up from 200 all the way to 240. And then there are fluctuations, but there are small in comparison to the larger movement. So these can be considered everyday fluctuations. At that point, the market had not necessarily made any decisions about what to do with their Dai.
  • MakerMan: Well, you see what the trading did there. There is a spike down in the volume and a change around the average.
  • Vishesh: There always been small run up and down events. These are smaller volume trades. It sometimes helps to look at some the larger trades on this graphic:

• Vishesh: a lot of the trades are happening around peg even when the ETH price is elevated. Around June 14th, the price of ETH goes down, so people are less secure of what happens, and Dai purchasing occurs. It's not only about ETH-downward's movement, but also when that movement is sustained, and it could potentially lead to liquidations. That is the moment when people start to buy back in order to protect their positions. I don't want to belabor that point too much, but basically, it snapped back into this traditional inverse relationship, except in the last few days, where it broke from that a little bit. And it's been coming down anyway. That's a bit odd since there are large chunks of Dai that have been funneled straight into Compound's supply.

Yield Farming

1:09:06

• Rich: Was frothiness introduced by the yield farming craze? Do you think that's the explanation here, or is there something else?
  • Vishesh: The "yield farming craze" effect on Dai is what I'm talking about. The result of what's happening there is a bit contrary to what you might expect. Yes, there is more borrowing on Compound, but there's more supply, which is outpacing borrow. So theoretically, for the moment, it's as if there's effectively a higher interest rate for supplying on Compound. We know what happens in those situations because that's something that we've seen before. COMP is effectively another unit of account where people are earning a yield for this behavior. That could, economically, be thought of as an increased interest rate. If you're earning a higher supply rate on Compound if high enough, they can mint Dai on Maker and supply straight into Compound. It appears there's a lot of that going on. But also, generally speaking, they keep positions open, which continue to lever up and choose to eat any SF or Dai premiums in order to get access to those yields. That's the kind of stuff that we've seen in the past from Dai users. Even with all that going on, what you'd expect is the Dai price to rise, but it hasn't in that same time frame. So I think that's interesting.
  • Monet Supply: Depending on how Compound changes its incentive structure, that could have a big impact on Dai minting. There has been a discussion about incentivizing ETH deposits. That could change the game.
  • Vishesh: The one thing to note is let's say the yield for COMP isn't quite as good for Dai. That potentially reduces the amount of borrow and supply on Compound. It's not clear how that would affect Dai, if at all. It may not have an impact on the peg if they are symmetric effects. If they're asymmetric, they could have an effect either way. As it stands, it appears to be somewhat symmetric.
  • Monet Supply: Even if the peg doesn't get impacted, I think the open interest on ETH<>DAI borrowing could move and make everything more unpredictable at that point.

How COMP affects the peg

01:13:45

• MakerMan: I was going to follow up on Vishesh's analysis. It looks like the price on COMP is affecting rates. Does anyone see anything that could broadside us here? It looks like it's driving a rate seeking engine in a weird way that doesn't seem normal from a money supply perspective.
  • Rich: I second that question. I would like to hear more. This whole yield farming thing makes me nervous, and my main concern is how it affects Dai.
  • Vishesh: There are two separate questions there. One is, how does it affect Dai? The other one is, how does this affect lending markets? For the second one, we need another one-hour call to go over. Regarding Dai, as I said, at the moment, the impact appears to be balanced. There are lots of open interest on both sides there. The simple question is, is there risk? The answer is, "of course." Can we pinpoint how that risk mess with Dai markets? Very unpredictable right now. If something changes with those rates and they become more asymmetrical, then it could have a significant effect on the peg, in either direction. A not very realistic example, if it becomes more profitable to supply Dai on Compound than it is to borrow, then you would see liquidity for Dai dry up, pushing the peg upward. Until it becomes not so profitable anymore, and then you see a surge in liquidity and the peg goes the opposite direction, downward. So the point is that it adds a degree of peg risk and, in particular, a degree of dependency on the profitability of those behaviors on Compound. That's the impact that secondary lending markets have always had on Dai. Whatever is initially profitable will either dry up liquidity or dump a bunch of Dai on the market. It seems either way. Also, these kinds of yields don't seem to last. But it adds the potential for instability.
  • MakerMan: That's what I was going to say. If you double the number of opposing forces, you double the amount of volatility, loosely speaking—the pressure of volatility. So maybe we'll see higher volatility on the peg.
• MakerMan: Another thing, in terms of vaults, is; we've seen the BAT vaults and how it pulled BAT from the system. I'm wondering if they switch it to ETH, is it going to start pulling ETH? Is it about vault risk if someone has a better market for ETH? Think of it from all sides. It's not only the Dai risk. What happens to the business side too. I'm trying to digest it and see what's giving me, given gas prices and everything.
  • Vishesh: The types of users chasing yield on BAT are not necessarily the same that supply ETH to Maker for leverage on ETH. They're potentially two different types of users engaging in different types of behavior. Just because Maker is one system doesn't mean all users and collateral are the same. Actually, it's more the opposite. The types of users for USDC are very different from the types of users that use ETH. I would imagine the same is true for BAT. Yes, the BAT collateral supply was significantly reduced in the past week or so. But, it wasn't large, to begin with. So you have to ask yourself, are you going to use a percentage or a nominal basis? $400,000 worth of BAT is potentially not a large movement. So there's a question on how much to read into that.
  • MakerMan: It was just a note. I agree with you about the size of BAT. More about "what if that starts happening with ETH?" It was more of a general question of how it drives behavior. I was already a Compound user, and I posted on their Discord that it would be nice if they took the previous interest, for the last period, and figure that out for all their users, and as they're dispensing interest block-by-block for all the new stuff coming in, portion some chunk of it to the BAT users, so they get a forward payment on the COMP. It could be a proposal, but it's own thing, but looked wacky and would drive a lot of weird stuff. There's like 700 million in addition on Compound right now, just driving the whole thing.

Outro

Rich Brown

1:21:04

• I think we might be at the end. I'll wrap things up. I want to take this time and publicly thank Monet_Supply and MakerMan for the work they did on the Compensation Plan and compiling an intention into a project plan. Coming up with a project, largely on your own, is a thankless task. It's onerous. It requires a lot of work. I would like to thank you on behalf of the ecosystem. Please follow up in the forums, read the document that they've posted and comment if you have opinions. Thanks, everyone, for showing up. Talk to you next week!

Abbreviated Terms

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

flip: Collateral Auction

flap: Surplus Auction

flop: Debt Auction

Credits

• Tim Black produced this summary.
• David Utrobin produced this summary.
• Artem Gordon produced this summary.
• Gala Guillen produced this summary.
• Juan Guillen produced this summary.
• Everyone who spoke and presented on the call (listed in the headers.)