Ep. 26 Mar 14 - 2019¶
Video | Audio | Discussion
| References | Person | Text |
|---|---|---|
| # / 00:0 | Richard Brown | Hello, everyone. Welcome to the Thursday, March the 14th episode of the Scientific Governance and Risk call with MakerDAO. I'll do the preamble again. We are enormously interested in getting as much feedback and benefiting from the insights of the people in this call as we possibly can. So if you have a question, please type in the chat at the side. If you have a microphone, do not feel shy jumping right in and joining the discussion. It is welcome. Feel free to interrupt. I do it all the time, so precedence has been set. We have some tweaks to the process that we're going to be talking about today, but I'm not going to steal Steven's thunder. So I'm gonna hand it off to him, and then we can get this show on the road. Steven. |
| # / 00:50 | Steven Becker | Thanks, Rich. I just wanna first confirm, can you hear me? |
| # / 00:53 | Richard Brown | I can hear you. |
| # / 00:54 | Steven Becker | Awesome. So what I'd like to do today is, firstly, just set the broad aspects of what an agenda would look like for this kind of meeting. We would like to split this into three parts. First is looking at demand and supply imbalance of DAI, which obviously means stability fee decisions. The second part would be to look at ... And this is something we'll iterate into, into the future is a collateral type and portfolio considerations. Basically, the consideration of new collateral types, adjustments, and ramifications thereof. The third is to look at what I would to call exogenous risk, the consideration of the protocols that we are on. For now, it's obviously Ethereum. How would that affect us? All the way through to things like oracles, and what we need to consider as well. So just to review, I think this encapsulates pretty much the broad spectrum of everything we need to keep in mind. |
| # / 01:56 | Steven Becker | Firstly, the demand and supply imbalance. The second is collateral type and portfolio consideration, and the third is exogenous risk. And given that, I think today, we're gonna probably focus mostly on the demand and supply imbalance. We're gonna talk about the results and the next steps to possible stability fee changes. But the one thing I'd like to do before we get there is just address or rather go back and have a look at, this concept of continuous approval voting. The one thing that we found is that there are a lot of folks out there that still don't have it in their head yet as to what this actually means, and the use of it. I think there's two points that are kind of confusing. The one is the concept of continuity and the fact that you actually have to continuously be involved. And the second one is that this isn't a representative democracy type construct where you look at a vote and if it's going your way, you think, "Well, why do I need to vote?" Or if it's not going your way, you think you yourself, "Why do I need to vote?" |
| # / 03:13 | Steven Becker | Through your mechanism of voting, you establish value for yourself. And if you don't establish value for yourself, like our good man David [Strobin 00:03:22] says, you are then just proportionately handing it out to everyone else, and you're giving them the opportunity to do so. So let's go back and have a look at, firstly, this concept of continuity and continuous approval voting. The way I'd like to start there is to just look at something that is the absolute edge case. In the most ideal world with no friction and everything works perfectly, all one million MKR should be in the voting construct at all times. That is the absolute case. Why? Because that means all one million MKR and their representative holders are continually assessing the system every single moment of the day. Obviously, pragmatically speaking, that's not possible. And because of frictions and because of different point of views, you start moving away from that edge case very quickly. |
| # / 04:16 | Steven Becker | And if you think about it carefully, the other edge case is that if don't have all one million MKR in the voting construct, then you should have all your value in DAI. That is the other side of the construct. Now, why am I bringing up these two edge cases? Because it's somewhere in between where we get the balance between looking after the system from a continuous point of view, having enough people in there, making sure that there is sufficient representation, but at the same time, being very cognizant of the fact that this doesn't require quorum. So let me take another step back. It is really a part of the incentive structure of MKR token holders to be continuously involved. And if they're not continuous involved, then as I said before, they give up some of that value to other folks that are involved. The idea here is to be cognizant of this at all times, and know that, pragmatically, there may be times where you simply cannot vote, or there's just something that is hindering you from voting. |
| # / 05:24 | Steven Becker | At the same time, you might decide that the voting construct is not the safest place in the world for you right now to keep all your MKR forever. You'd rather have it in, obviously, your own storage facility. These frictions come in and these frictions start playing on the idea of what you consider to be the inherent value of using MKR to vote with. If you're holding MKR and it's not in the voting construct, then it really is a case of you are of the opinion that proposals are gonna be coming forth. They probably will be voted on and you will realize their value. But not voting, again, you give up that value to those that actually end up voting. I think the premise of this is very simple. You have to always be vigilant as to what the system is doing. I think the sort of game theoretic and opinions overhere push you into a corner that is obviously unrealistic and not really programmatic. But what it does, it pushes into the ideal stage. If you cannot reach the ideal stage, you always are tying to challenge it. So given that, it's really important to keep in mind that an executive vote is a continuous approval vote. It always requires that that particular vote that passes to always be there. If you find that the MKR is pulled out and the number of MKR holding that vote in starts diminishing, all that means is that you are creating a susceptibility to someone else voting in something new that you may not be aware. This obviously opens up so many possibilities. For the creative types out there, it becomes clear that, "Wait a minute. Why not have a messaging service that every time a new proposal comes on, it pings you on your computer or your mobile. Or you know something that basically is available to you that always gives you an alert in terms of where the vote is, and where the current construct lies. |
| # / 07:42 | Steven Becker | Given that little monologue that you've just put out right now, I think it's important to you to keep in mind that the incentive is for you to constantly be involved. MKR is a governance token. It is basically constructed by the very smart people that created this to ensure that you are always aware of the fact that it is a governance, and therefore, a voting. And therefore, the value component of it is in the ... so I wanna pause there and take a breath. |
| # / 08:23 | David Utrobin | You cut out a little bit there. |
| # / 08:27 | Richard Brown | I think we got the gist of the conversation, though. Ashley asks [crosstalk 00:08:32]. |
| # / 08:32 | Steven Becker | [crosstalk 00:08:32] incentives[inaudible 00:08:32]. |
| # / 08:32 | Richard Brown | Ashley, you have a mic, don't you? Do you wanna ask the question? |
| # / 08:40 | Ashley Schap | I was just wondering, how do you actually incentivize MKR holders to vote? Obviously, the way that stability fees work, they're basically used to [inaudible 00:08:51] MKR which [inaudible 00:08:56] MKR [inaudible 00:08:56] equally in theory. So is just talking of incentives and why someone would vote with their MKR [inaudible 00:09:06] think about that. |
| # / 09:06 | Steven Becker | Well, that's actually ... Well, firstly, a good question. The answer to that is first through an education process where folks have to realize that every time a proposal comes onto the voting construct, it inherently has value that needs to be released. So if you think about this in terms of collateral types, if we bring in collateral type A, B, C, and everybody votes for it to be included, you release obviously a potential supply of Dai which obviously has stability fees, which go into buy and burn. The same thing goes with an adjustment for stability fee, for instance. If you see that your demand and supply construct is out of kilter, there is value in voting it back into equilibrium again. So all these incentives are actually very much aligned with MKR token holders And the value, firstly, it is in the eye of the beholder. Everyone's gonna basically see their own potential value from these proposals. But I don't think it can be denied that each proposal inherently has value. |
| # / 10:23 | Steven Becker | And this is where the friction comes in. This is where the creation of liquidity comes in from the edge case. If you find that folks don't see the value of a proposal, they may change from the MKR to Dai. They're basically gonna start voting with their feet. As soon as you don't see the value in the future proposals that we will have in the construct, you will be able to make a decision on the basis of that and say, "Yes, I want to keep my MKR in the voting construct." Or, "No, I don't see any further value inherent in it." So the incentive is through the actual voting on a proposal. The value is there. |
| # / 11:08 | Richard Brown | I would be interested in seeing some kind of a larger discussion around potential incentive models because I'm personally deeply, deeply suspicious of that. As far as I'm aware and I'd love to be corrected, there's no prior arch for a successful incentivization model for voting that didn't almost immediately lead to the collapse of whatever that system was, especially in the crypto space. |
| # / 11:38 | David Terry | Sorry, Rich. Do you have some case studies where incentivization actually failed? [crosstalk 00:11:42]. |
| # / 11:42 | Rune Christensen | Let me jump in there, guys, because what's happening here is that people are, essentially, fixating on voter participation without really understanding what the fundamental context around what actually matters, which is security of the system and that it's governed correctly. The thing is that you don't actually have to have ... Voter participation isn't actually, essentially, a linear gauge of, "Okay, this is how well the system it governed." Or, "This is how secure it is." Right? Because even if you had, let's say, 30% people who are voting, you don't really know how many of them are potentially malicious voters who at any point in time could switch over their vote or something like that. In general, you can't rely on that as a security assumption in the first place. You can't just be like, "We'll just assume that we'll have tons of voters and things will always be great, and we'll never have malicious votes. We'll never have stupid votes." |
| # / 12:44 | Rune Christensen | This is how, of course, the system is designed from the bottom up, right? Fundamentally, you cannot assume that governance will actually always vote correctly and always behave correct. There has to be protection mechanisms in place to actually ensure that if something goes wrong, if so there is a majority or malicious voters, we're able to protect against it. This is where we get into the big topic of emergency shutdown. Ultimately, that is actually what drives the system. That is actually the real governance model. It becomes really clear once you think about the actual consequences of having this MKR triggered emergency shut down. So if we put the MKR triggered emergency shut down at 50K for instance. That means that the only threshold of voter participation that really matters is 50K at this point because it doesn't matter if you have someone with, let's say, half of the entire MKR supply who comes and tries to perform some sort of governance attack. As long as there's 50K active voters who can then protect and do an emergency shutdown and redeploy without the attackers MKR stake. Also, on the flip side, it doesn't matter if you have 50K honest voters who are all actively voting with their 50K MKR because you can still have an attacker come and trigger a troll emergency shutdown which is five ... Oh, sorry. You could have honest voters with 500K, so honest voters with half of the entire total supply 'cause an attacker can still come with just 50K and trigger troll emergency shutdown which is the best you'll achieve with any sort of attack, anyway. So in the end, voting, it's almost like an oracle of the governance process. It's how we actually indicate to the system crypto-economically this is what we've decided, but it's not the actual decision making process because that would just be totally useless if all that mattered is who has the most votes, whose got the biggest bags of the shitcoin and who's got the most MKRs to pump the shitcoin. It would be a total disaster. You wouldn't actually have any sort of intelligent governance coming out of that. |
| # / 15:05 | Rune Christensen | On the contrary, you actually need to completely divorce the individual biasses of anonymous people from the direct decisions made by the [inaudible 00:15:16]. This is why we use scientific governance. Obviously, it makes sense to talk about voter incentives just from a theoretical perspective, but it needs to be done in the context of what is it we actually want to achieve. There's really only one thing that matters in MKR governance. That is it always have a minimum amount of active voters ready to trigger an emergency shutdown to defend against some sort of governance attack, some sort of proposal that is intended to be pushed through by a MKR majority outside of the scientific governance framework. As long as we have this minority that's ready to defend against a governance attack, that's all we need in terms of active voters. Obviously, it's nice to go beyond that, but there's no crypto-economic risk here. There's no fundamental flaw. It's really a question of nice-to-haves beyond that. |
| # / 16:19 | Richard Brown | I think that's a decent summary. It ties into something that Steven was talking about earlier, and it's ultimately, a tragedy of the commons. We've seen in the chat that ... As I think those one of the risks that we assumed from the beginning is that it's going to be same risks that are inherent in any voting system, is "Will my vote actually matter? Probably not, so why bother?" Or, "It's already going the direction I like, so there no point." Or, "It's going so far in the direction I don't like. There's no point." The only way that we can solve those problems outside a new and foolproof incentivization mechanism is transparency and communication, and these calls, and getting people that are deeply engaged with the system educated, on board, and interested in participating. I think it's far too early for us to be overly concerned or obsessed with the numbers that we're seeing. I'm actually kind of happy with the numbers we're seeing. As long as they improve over time, we're in a good spot. |
| # / 17:19 | Richard Brown | I would like to keep on moving 'cause we have a bad habit of squeezing the risk team out of time in these calls, and it's ultimately the reasons why we're here, is to hear from the risk team as well. I wanna move onto the next section. I wanna talk briefly about the previous mechanisms that we've put into place for the voting system, talk a little bit about numbers, and then I'm gonna hand it off to Cyrus and the risk team to give us some deeper insights. Last Friday, we had an executive vote. That executive vote was exciting for a couple reasons and not as good as it could have been for a few other ones. Excuse me. The vote was ratified in 10 hours and 27 minutes, which is excellent. I think that's a good sign for the voting process in general, but Maker as well 'cause it shows that we have some agility here. And that's going to be a recurring theme as we have these calls, velocity and communications are going to be significant issues that we need to optimize. |
| # / 18:22 | Richard Brown | There's a lifecycle of the governance process that starts in a call, moves to Reddit for debate, moves into various chats or social medias, then it hits an official proposal process, then it goes into, potentially, a governance vote. Then it goes into an executive vote, and then there's a cooling off period. Then there's another call to discuss those issues. So we have this lifecycle that we need to optimize as much as possible so we can maintain enough agility to react to market pressures as they arise. I think that we have seen, already, some improvement there. I'm excited about it. |
| # / 19:00 | Richard Brown | There's a post in MKRGov, which is a sub-reddit that everybody in this room should be familiar with or getting familiar with. It's where we're posting summaries of the debates, summaries of these calls. I'm sorry, live ongoing debate summaries of these calls and metrics around the votes that we've had and various discussions that are directly related to governance and this is the corporate that appears to be interested in governance. Everybody here should, come by and join the discussion. And some of posts though. I got it ... I want to talk too you a bit, a bit about the apathy issue and the concern with the voters, and the turnout that we've seen so far, the reasons why I'm not super concerned about it. So we're on vote number five, I think over the course of the year. So it's hasn't sunk into the general background of unspoken assumptions of what it means to be involved with Maker. |
| # / 19:55 | Richard Brown | And I think that's going to be improving over time. We're a rapidly approaching a cap of number of people that can find room in these calls. And so obviously things are happening. As long as you maintain that momentum, we're in a good spot. But to set some context here, we have a total of 8,000 ish Maker holders. And if we wanted to create a venn diagram of those 8,000 holders, we need to, we have a smaller circle of holders that are interested in Makerdao itself, the ones that are interested in Makerdao and governance. And then after that they have to be actively interested in participating in governance, feel that they have enough insight to contribute to the conversation. Then they need to set up the portal. Then they need to feel that their vote is required and that is that last step, which seems to be a blocker for a lot of our members. |
| # / 20:43 | Richard Brown | And the idea that we just keep on pushing the idea that even if you're not the biggest holder in the world, and I can assure you that I was at the bottom of the list of holders that voted in the last call. That last vote. A lot of small Maker voters can make a large impact on the system as Rune was alluding to previously. So even if you don't think you're going to move the needle but yourself, if it's extremely important that you signal your support with whatever tiny bags you might have or to signal your, your descent with the whatever you have, there's a huge amount of value there. |
| # / 21:22 | Rune Christensen | One the thing I would like to add this to the whole question of gas cost? around voting where. Like in general, I really think that for the executive vote that's more where you have like whales and funds and so on having a big impact. But the thing is that again, like I've just talked about that and it's sort of the fundamental dynamics of the governance is that the executive vote has to vote a certain way and that's the way that's the sense of governance processes already decided, right? Because if you vote in a different way, you're performing your governance attack, right? Unless, I mean unless there is a legitimate governance split, in which case anyway you need some sort of like, there will be an emergency shutdown in either of these cases and they will either be a new system or two new systems following from that, right? And as a result, like they say the vote is really a very technical thing and it's not really a very much in like, it's not so much about decision making. |
| # / 22:13 | Rune Christensen | Where you have decision making is in the things like the governance polls for instance. And what's really cool is that in the future governance polls will not even require gas at all. So as a small holder, even if you just want to cast a vote of one MKR, you'll be possible to do like it will be possible to do that with just like signing a message from your private key so not actually having to like pay gas for unchained transaction. And that's really gonna help with the dynamics of this. But really most importantly, the way you can have the most possible impact as a small voter is to join the governance call. Because in the governance call you'll have just as much of an impact as everyone else. Right. And ultimately you'll even as a part of that process be dictating how the funds and the large MKR holders have to vote in the executive. |
| # / 22:54 | Richard Brown | Yeah, exactly. I want to remain cognizant of time. So I'm going to cut off my speech finally there. One of the things I want to do before I hand it off to risk though is, point out the factor or revisit the idea that we were getting super popular around these parts and these calls are getting some notice in the media and with influencers on the twitter sphere. So it's going to be increasingly important that we treat these as a normal meeting as opposed to in the past where we had the luxury of just chit chats and being as dramatic as we felt we could get away with. So there's last week's call was literally, there were reporters in the call, which we're happy to have with us. And a summary of the call went up less than an hour later. So there's an enormous amounts of visibility on these calls. |
| # / 23:45 | Richard Brown | So we need to be careful about a few different things. One of those things is numbers. So we need to be careful that if we're coming up with a Google doc or a spreadsheet that has best guess numbers in there that were clearly explaining to people that these are ideas and maybe not facts. I want to ensure that we don't fall into the trap of creating too much urgency where that might not actually exist. So it's an environment where we will eventually be competing to have, our ideas heard and our or will met. But yeah, we need to do that in a way that's beholden to the risk team. The risk team will come up with data, they will come up with charts and they will come up with recommendations and then we can debate those. But we need to shy away from falling into the trap of opinionating and speechifying as much as we possibly can in these calls. |
| # / 24:42 | Richard Brown | I think that's it for today. That's it for the gray beard wisdom aspect of the call. So as long as we keep things businesslike while at the same time this means it's continued to maintain some radical transparency because that's, I think the core of our success in these calls is that we're not filtering things 15 times before it actually makes it into the community, we're talking about real things as they happen, straight from the people that are actually executing some of these tasks. So let's maintain that transparency but let's not say things that are going to alarm people without context. |
| # / 25:16 | Richard Brown | Okay. That's it for me. Cyrus, I want to hand it off to you now. So here's some context. In the course of these calls going forward, every time that we have a call that's about demand and supply and balance like this one is. That means that there's some issue in the risk parameters that we need to address stability fees or the inventory is off or something. That means that we need analysis and we need the risk team to provide that analysis for us. Fill us in on what they've been up to and give us some suggestions about what the future should look like based on the data. And that's what Cyrus is going to be bringing to these calls. So Cyrus, I'm going to hand it off to you now. |
| # / 25:55 | Cyrus Younessi | Sure. I just want to add one thing to what you were saying about the governance is that, just want to remind people that this is largely a work in progress. And even though, you know, the media may take something from our calls and run it as a story, for example, that the governance poll that we held two weeks ago where we had, for example, two options, that may not be the way that it continues to be done. We're still working on how to make it as fair and unbiased as possible or other methodologies or statistics that we pull up. You know, this isn't just the final way that we're doing things. We're very much working on how to be transparent, fair, decentralized and so forth. |
| # / 26:47 | Richard Brown | Perfect. Thanks for that. Cyrus that's a great reminder. The governance process that we have in Makerdao is evolving over time. And if you've been following these last 25 calls, you'll have seen that in very slow motion as it happened. We're, balancing two different things. We're very cognizant of the fact that what we're involved in is a very serious thing and that we can't perform experiments on the governance process and we can't, you know, expose every lever there is, but we want to do things, as rapidly as we possibly can and still maintain stability. And that's, it's a good reminder. That the way that governance looks right now is not going to be the way it looks in a month or two because we're constantly improving with feedback. |
| # / 27:28 | Cyrus Younessi | Yeah. Yup. Okay. So I'm going to talk a little bit about Dai price. I'm going to do a quick screen share. |
| # / 27:51 | Cyrus Younessi | All right, so first up we have the Dai price. This is weighted average of several different exchanges that we're following. You know, the governance, the executive vote happened around a week ago, March 8th, and for the first few days we didn't really see much change, but, eventually we saw some volume hit and we saw a bit of an uptick from about 96 cents up to about 98 and a half cents. I don't think these, these dollar prices really traded for much size. The Dai supply also continued to grow but at a much slower pace than we've seen in recent weeks and months. So you can see that even though we're still growing, it's still tapering off a bit. We've also seen an uptick, a small uptick in CDP wipes, that is represented by the pink bars here. So you can see in the few days leading up to the executive vote. And you know, the couple days leading up to today's call, we saw some maybe a small uptick in people wiping down their CDPs. |
| # / 29:15 | Cyrus Younessi | We can also, you know, we can surmise that this is either, due to expected future increases in the stability fee or just people kind of the current rate has pushed them to close right now. We've also seen some OTC activity that suggests speculators and traders are entering the market buying up Dai, also potentially trying to get ahead of expected closures of CDPs. |
| # / 29:53 | Cyrus Younessi | We are also working on creating more numerical representations of this data, not just charts. That is definitely a work in progress. I'm looking into how we might collect and utilize market maker inventory data, start controlling for other variables such as the ETH price. And there's a lot of talk about, Dai demand in relation to the market price of ETH so we will be getting that out shortly. |
| # / 30:29 | Cyrus Younessi | In terms of statistical data analysis that, that's about it. I think that it's clear that Dai price hasn't quite hit $1. It's been hovering around 98 and a half cents for the past three, four days. |
| # / 30:45 | Cyrus Younessi | Our thinking internally was to just observe the Dai price seems to be going, trending and the general right direction. We're seeing some good metrics in terms of CDPs, Dai supply growth, market makers. I think just waiting and seeing how it evolves could be the right decision. But depending on how the next week progresses, we could explore options of holding another governance poll or maybe just more community discussion on social media. It's about it for the Dai price. Kind of stop for some questions. |
| # / 31:35 | Barron Gati | Quick question, can you talk a little bit about the, you mentioned the statistical data analysis. Can you talk a little bit about the, degree of confidence you use in observing the linkages. If you want to go back to that other chart with the pink bars representing the collateral wipes? |
| # / 31:53 | Cyrus Younessi | Yeah. |
| # / 31:56 | Barron Gati | You mentioned that there was like, you can see leading up to, the stability fee increase X, Y, and Z, but at what point are you considering those random events versus linked and correlated or causal events? |
| # / 32:11 | Richard Brown | Cyrus can you click that ETH price and Dai supply as long as we can see those two things that might help answer the correlation. |
| # / 32:20 | Cyrus Younessi | Yeah, that's a good question. That's, that's part of the reason we kind of take our time and do it right. It does require kind of evaluating past historical data, going back to stability fee increases from last August and trying to figure out which variables exactly we want to control for. We do have a somewhat limited sample size of data. This governance process is fairly new. The data from last year, it's not exactly relevant to today's environment and the two stability fee increases we had in February were not very illustrative in terms of how things react to stability fees. So yeah, the short answer is we don't have statistical confidence today. But it is something that we're actively researching. |
| # / 33:18 | Barron Gati | And just one quick follow up, do you have in place a system to check basically the power of those statistical correlations or you know, having an a priori set up to think through what the data might show and then be thus more confident in the power of what the data are showing and also is that public somewhere like is there, are those data available publicly? Can we take a look at that somewhere or is it just in your system somewhere? |
| # / 33:50 | Cyrus Younessi | Yeah, the data is, is definitely public. Or will be public, whatever isn't, we do have some internal, other ... I'm going to post these in the chat. |
| # / 34:12 | Barron Gati | I actually haven't been to this https://dai.stablecoin.science/, thing. Are the data downloadable from there or is it all in chart form? |
| # / 34:23 | Cyrus Younessi | It's not downloadable from the Dai.stablecoin. |
| # / 34:28 | Barron Gati | Is that a possible change coming or is it going to be permanent? |
| # / 34:32 | Cyrus Younessi | No, no. |
| # / 34:32 | Richard Brown | That's the goal. The short to midterm goal is that we will eventually have a sort of a go to location for risk at Makerdao, which will be public. So the idea is that anyone can dig into the data themselves. Grab the data sets themselves run their own models and attempt to ... we're going to have some debates about which data sets we as a group, feel are most valid and trustworthy. And then we'll be able, we'll be in a position where we can come to these calls ahead of time with our own opinions that are based on the information that's available. |
| # / 35:12 | Vishesh Choudry | So just one, one quick note. As far as lead scraps go, I shared a couple of links in the chat. I also pulled very similar data in terms of the traits that were happening on the centralized exchanges. I mean I think we can sort of confirm a little bit with looking at trading volume, what our level of confidence is, at least in the price that we're getting. So that was one question I had was it seems that recently the trading volume on these decentralized exchanges for ETH/Dai pairs has gone down. And I was kind of curious as to like if anybody had any thoughts on that and what impact that has on our level of confidence in what the price actually is. |
| # / 35:55 | Cyrus Younessi | Yeah, I mean volumes have been generally weak across the markets over the past few days or the past week. I think part of the increase in the Dai prices has been initiated from the OTC side, if I had to guess. I know on our side we were seeing a bit of positive upward activity. Not, I'm not sure how to, how else we can, what else we could do when there's low volume across, centralized and decentralized exchanges. |
| # / 36:33 | Rune Christensen | So the most, the most critical data points, at least one of them. Absolutely the most critical. But I think especially in a situation where now it's the most critical one is to hear from trading desks and market makers what the inventory levels are at. Right. Because my impression is that while we've seen an uptick on the stablecoin.science chart for instance, it's a very superficial change that has not actually have any significant impact on what it actually looks like in terms of, do people still have too much Dai and still trying to get rid of it at all costs and nobody wants to buy it and so on. So I think it would really interesting to hear from, I think we've got someone from the foundation trading desk, but then also if there's anyone else who wants to sort of share that general sentiment in terms of what their inventory looks like. |
| # / 37:25 | Joe Quintillian | This is Joe Quintillian I feel like this is, I'm doing all the foundation trading desks. I feel like the market has moved up, but I'm light volume. It has had a little bit of an effect. But I feel like my feeling is eventually we should, if we don't get further price on, we should try to do one more, increase. Maybe. See see how it goes till Monday, Tuesday, |
| # / 38:00 | Joe Quintillian | But probably the next governance meeting's talking about doing another increase, because I've not seen the push forward, at least to really trades above ... any really good trades above one or even close to really one. So, that's my feeling. Again, I have seen other OTC desks who are for size have come on the bid, so there's size bids below, but we're kind of in this 98.5 to 99 where I see some size going off OTC, but nothing in the 99s yet. |
| # / 38:36 | Rune Christensen | So in terms of the foundation trading desk and sale, have you been able to start selling off DAI and have you seen demand for actually buying DAI from anywhere? |
| # / 38:46 | Joe Quintillian | I really didn't wanna sell DAI below 99, so I have not sold any DAI in ... I've actually [inaudible 00:38:53] a little bit of DAI on the buying at 97s. |
| # / 38:58 | Rune Christensen | Yeah, I think this is a really ... I think this kind of data is super important and I think also of course a critical piece is that whenever we talk about buying and selling DAI, we have to think of it in terms of buying and selling assuming the peg is correct, right? Even if a trading desk is selling DAI below one dollar, that doesn't actually count as them being able to sell DAI, right? Because what that means is now they're taking a loss for essentially having believed in the stability of DAI, right? |
| # / 39:26 | Rune Christensen | So I'm wondering if there's anyone else who had the same impression that they're still unable to sell DAY around one dollar and even though the stability fee has had this superficial effect on the decentralized exchanges, I wonder how it actually feels like in the deeper markets. |
| # / 39:44 | Richard Brown | Louis, are you're comfortable with giving us "things are better", "things are worse" kind of indication? |
| # / 39:47 | Louis Aboud-Hogben | I would say things are definitely the same on our end. You know, the spot prices have moved up a bit, but you know, we're not clearing at any of these prices. Yeah, I'd say it's pretty important we don't sort of get lured into a false sense of security here. You know, the ETH price went down a little bit and I think that has helped DAI out, but our inventories are still the same. We're still overweight and need to clear some. |
| # / 40:17 | Lev Livnev | I'd also like to add here that I think the last couple of people were spot on to bring up volumes, so this run up, even without asking people stuff like this, you can see just from the trading history on dex's for example. The trading volume during the run up was pretty low. There hasn't actually been a lot of DAI traded near one dollar, so if you sum up all the trades that have been near one dollar on the charts you'll see it's like a couple hundred K it's not really meaningful amounts. So I think the fact that there's been volume the past couple days means that the move is less ... the spot prices you're seeing are less meaningful. So I'm not surprised to hear everyone saying this. |
| # / 41:01 | Lev Livnev | Since we're on the topic of discussing inventory, I thought something that would be quite interesting to keep in mind is that we can ... it would actually be surprisingly straightforward to come up with a system where participants in the governance discussion can actually provide proof of their holdings in a zero knowledge way if they weren't comfortable giving to the public details about their accounts or their balances, but they still wanted to be able to substantiate their claims with real data to make sure that people believe them and that they're not just pumping their book. |
| # / 41:46 | Lev Livnev | I think I don't want to derail this technical discussion now, but actually this is not a huge engineering challenge and I think that basically if a couple of people wanted to work together for a little bit to make this happen, then it'll be pretty straightforward and I think it'll give us a really ... it'll formalize this sharing of data, right? So instead of having to kind of awkwardly see what people want to reveal on a call like this, we can just have somewhere where people can post these proofs on a weekly basis if they want. |
| # / 42:16 | Richard Brown | I love that idea and I don't wanna lose that because it was raised in our last discussion as well that there's ways that we can come together, or the OTC desks can come together for the benefit of everybody so. Anybody in this call who has a desk, wants to be involved, please email cyrus@makerDAO.com And he can spearhead that initiative. |
| # / 42:38 | Alex Evans | So it sounds like we came up close to a dollar on low volume, we're still trading below a dollar and market maker inventories are flat, nobody's been able to sell anything off. Those are the kinds of things where historically, we've said would justify an increase in the stability fee and a number of people have brought this up on the chat of if all that is true, why are we waiting here? Why would we wait an extra week? We're still below a dollar, nobody's been able to sell any DAI. ETH could very well hyperventilate from here, and if that happens we could be in trouble. So, why wouldn't we be willing to move faster and give some of these market makers some relief? We're still holding significant amounts of back. |
| # / 43:28 | Rune Christensen | I think this is a really important point. And I totally agree, and I'm of the opinion that if we saw no effect, we should actually do a 4% increase as quickly as possible. I think we've seen some effect and then with that I think we should just go steady with 2% increases, but the thing that is more important than doing a quick increase now and kind of trying to fix the problem as quickly as possible is to ensure that we can continue to actually do intelligent and effective governance in the long run. And so, what we have to also weigh this against is what the kind of community fallout and just the general sentiment around how governance is conducted, right? |
| # / 44:09 | Rune Christensen | And I can think of a great example of that is that one of the main reasons why we are in this situation in the first place is because there were so many complaints about basically doing stability changes that we went back to that, let's do 50 basis points because then people don't get as upset, right? And it turned out that they don't get as upset, but they still don't react at all in the market, right? But I think what we have right now is a system I feel will be effective enough to deal with the current issue we're facing, regardless of what some of the underlying magnitude ultimately is and how significant the change ultimately has to be. |
| # / 44:46 | Rune Christensen | We've definitely seen that 2% changes have an effect, right? So if we keep doing that, like we guarantee that will work out eventually, and there's not an uncertainty that it could all become a big ... that it could all spiral out of control. |
| # / 45:02 | Richard Brown | We're running out of time again. I'd like to recap and make sure that we have action items now, so I still wanna give people a chance to ask some questions. It sounds to me like we have kind of an idea here where risk is because their job is risk, they're taking a cautious stance, so there's an option to wait and see what happens. And so over the next week. And then there's more options out there, potentially two or potentially fore. So we should consider possibly a governance vote to determine do we wait and see, do we raise it by 2% or maybe possibly a third option. I think that that's probably a valid discussion to continue in a subreddit for this call for the debate that's related to this call. |
| # / 45:50 | Rune Christensen | Yeah, well I think it's really important ... I think we should really try to see if we can get consensus on this right now. I think I would say the most straightforward option is yeah, we do a governance poll for no change or 2% increase. So is there anyone who's against that right now? |
| # / 46:09 | David Terry | Last week there was some discussion that it would be nice to have more options on the governance polls so we could get a gauge of what the community sentiment really is, that maybe if we just have one or two options, we're not [inaudible 00:46:23]. |
| # / 46:23 | Richard Brown | Yeah, I agree. I would like those, that range to be less arbitrary and more risk team driven, so have them run some numbers, come up with some evidence that says that from here to here is probably good, but beyond that is risky or something. It's super important that we stay away from opinions here and it needs to be dated if it's scientific governance. |
| # / 46:47 | Rune Christensen | Yeah, I mean it would be super great to have a very elaborate and sort of evidence-based range of options, right? But it's just very important that like you're saying, Rich, it's not like choose-your-own-adventure, what you want is stability [inaudible 00:46:58] MKR holder right? |
| # / 46:58 | Rune Christensen | I think we should try to get that done. I mean, that is a great thing, but I don't think that we should ... I think it would be risky to basically say let's block, try and fix the peg on basically creating some new science that we're not sure if we can do currently, right? So with that in mind- |
| # / 47:18 | David Terry | Could we at least add a ... Sorry, could we at least add a none of the above option? |
| # / 47:21 | Richard Brown | Sure. I think that we're running out of time though, and I think that's [crosstalk 00:47:27]. |
| # / 47:27 | Matthew Rabinowitz | In general. Yeah, when you said zero or two, I mean, I think at least from my perspective, I don't think anybody disputes that we should keep it at zero. I think it's more of do we make it one or two or half or one and a half. I don't think there's any dispute that people think we should change it. |
| # / 47:44 | Richard Brown | Well, it's risky to assume there's no dispute because we're very, very limited group of people here. Not everybody likes [inaudible 00:47:50]. |
| # / 47:50 | Cyrus Younessi | Fair enough. Also there's a question on how long after a stability fee increase happens that we should wait to observe. I mean, how quickly do we expect certain changes to take effect? |
| # / 48:03 | Barron Gati | Really quick on that note, I brought this up to David Utrobin last week after the call which is there's definitely some improvements I think that could probably be made to the reporting of the decentralized exchanges with respect to the bid and ask prices and where the last executions fall between them and where those actually fall with respect to volume being proposed. So if there was a way to like chart that, you might get a more reliable data rather than just the chart that was shown before that shows where the drift is with respect to just the execution prices, because they jump around a little bit. But if you also look at the prices where people are putting out bids and offers with various volumes, you can get a sense for what the market will bare at different rates. And that can help in the form and provide more data and less time maybe necessary to observe between stability fee votes or other votes. |
| # / 49:00 | Rune Christensen | Yeah, so the thing is, I mean, because we don't have time ... there's so many dimensions to this, right? So I think there's certainly no way that we would be able to come up with ... get consensus around the solution right now or in the very short term. So, I think what will happen ... if we created a governance vote for either we wanna do immediately ... like if we want to keep it steady or do we want to do a 2% increase immediatly? You can vote for not making a change if you think we need ... because what you're essentially doing is you're saying let's wait. Like you know, obviously no one thinks that we should just keep ... well, at least I assume the very few people think we should keep interest rates at their current levels, but in terms of do we want to just press ahead with the 2% increase, do we wanna try something more elaborate? Essentially then, voting to not do anything in the short run can from there, what we can then go from there is okay, let's do absolutely nothing or let's try something more elaborate. |
| # / 49:57 | Rune Christensen | Because otherwise, if we decide right now that we want do something more elaborate, essentially what we're deciding is we want to wait. Right? Because we're not going to be able to have something ready to actually do an executive vote next week. And I definitely think there should be an option. Let's try and see if we can get an executive vote through next week because there's still significant imbalance. |
| # / 50:17 | Richard Brown | Okay. We ran out of time and I wanted to actually get into that towards the end of the call, but when I talked earlier about the lifecycle and how we're trying to tighten up this process. So I think that this might be a good opportunity to do that. Let's take ... we're gonna have transcripts and video and audio of this call and I'm going to make a post that's devoted to this call in MKRgov and then we can ... the subject of that post will be what is the next governance pole look like? 0 to 1%, we'll get risk teams to come in with some suggestions and then we'll start thinking about timing. |
| # / 50:56 | Richard Brown | Perhaps that's 24 hours for that post to digest, then we will determine what's the good lifespan of a governance vote. Perhaps it's two days if we get enough activity there, and then we can start tightening up this cycle so that we can ensure that we have an entire iteration of governance executive cooling off and then evaluation between each one of these calls. |
| # / 51:22 | Richard Brown | So let's make sure that this debate doesn't peter out and that we all take a look at the discussion in Reddit so we can continue. |
| # / 51:35 | Richard Brown | So we have six minutes left. There probably isn't a lot of time left for general Q&A I think. If there were some interesting and topical questions in the side that somebody wanted to ask before it's too late we could do that now. |
| # / 51:55 | Rune Christensen | I just would like to talk a little bit more about the actual process, because I feel like we need to really try to de-risk this whole peg process. One thing we did last time was we made this post on the MKRgov asking for arguments for anything other than a 2% change, right? And then because the thing is, I think a governance poll that has a bunch of different options with each option being sort of very obvious, like this option is here because of this data and this is what I'm seeing. Is this the data you agree with? |
| # / 52:27 | Rune Christensen | So we really ask people to actually provide an informed ... at least be kind of like an oracle almost for the kind of data that they think is ... for the kind of reality that they're seeing, right? So it's very important it's not like what do you feel like, but it's something like what do you ... can you help us figure out what's actually going on? And I think we can do that again, and but also again we can maintain that ... Ultimately, there should be an option to do an increase ASAP, so an executive vote next week. And I would say that out of ... there are kind of like a 2% is the most obvious because this is what we already have done. |
| # / 53:07 | Rune Christensen | 50 basis point is also like another obvious one, just because it also has the advantage of being a historical increment we've done before, right? So I think we can- |
| # / 53:18 | Richard Brown | I agree that it's definitely ... it needs to be scientifically based, and I think that we're transitioning right now between ... obviously because we don't have the historical data to give us a very clear signal, but the impact of these things that we're relying on the wisdom of the crowd, but we're slowly transitioning away from that, hoping that a week or two from now we'll have enough signal to not need to ask people what their opinions are or how they feel about things, but in this discussion that we're about to post, we'll get a recap from risk that talks about exactly what the situation was that they're seeing and then we can get more of the analytical people in this call hopefully. They'll be able to chip in with their own grabs and their own thinking, maybe some OTC desks to pitch in. Then from there, hopefully fairly quickly we can iterate into a very valid range for a governance poll. And we can start kicking off as soon as possible. |
| # / 54:08 | Richard Brown | Alright, unfortunately we didn't get to a lot of questions, so we're gonna do the same thing that we did last week where we're goin to take the questions that were asked in the chat and then we're going to format those and add them to the debate for this call as well. It will give people an opportunity to either re-ask or to have other people engage with the questions we didn't get a chance to answer. |
| # / 54:39 | Richard Brown | Okay, that was actually a quite good call. I'm happy with that. We're tightening things up, things got decided and we have next steps and that's always good, so we can iterate on this for next week as well, but I want to thank everybody for joining us once again. Lots of great questions, lots of great data, and I'm looking forward to seeing how the debate shakes up. Thanks everybody. |
| # / 55:00 | Rune Christensen | Thanks everyone. |
| # / 55:06 | David Utrobin: | Thank you Rich, for hosting the call. You're the man. |
| # / 55:09 | Barron Gati | Rune Christensen real quick, do you have an email or a twitter or something? |