Episode 174: January 13th, 2022

--- tags: G&R --- # Episode 174: January 13th, 2022 ## Agenda - [00:02](https://youtu.be/TYjzRyEWaRk?t=2): Introduction with Payton Rose - [02:53](https://youtu.be/TYjzRyEWaRk?t=173): Votes and Polls - [04:46](https://youtu.be/TYjzRyEWaRk?t=286): MIPs Update - [07:55](https://youtu.be/TYjzRyEWaRk?t=475): Forum at a Glance - [11:38](https://youtu.be/TYjzRyEWaRk?t=698): Discussion: Dai's Stablecoin Backing - [31:59](https://youtu.be/TYjzRyEWaRk?t=1919): Discussion: Multidisciplinary Approaches towards Governance Evolution ## Video <https://youtu.be/TYjzRyEWaRk> ## Introduction ### Agenda and Preamble #### Payton Rose [00:00](https://youtu.be/TYjzRyEWaRk) - Hello everyone, and welcome to the 174 Scientific Governance and Risk meeting. My name is Payton, and I go by prose11 online. I am one of the Governance Facilitators here at MakerDAO. I am joined today by incredible people who care about and are interested in the Maker protocol. This is our weekly Governance and Risk meeting, where we discuss current topics, things going on in the Governance realm, and potential risks or opportunities we see for the protocol in the future. - Before we get started, you might have noticed that this call is being recorded. We post the calls in their entirety on YouTube afterward; keep that in mind for when you are speaking. If you have a question, please ask when the time permits. If you are uncomfortable or unable to hop on the mic, feel free to drop the question or comment in the sidebar, and we will cycle to it when conversationally appropriate. - We will go into a presentation and discussion topics. The first topic is about the Dai stablecoin backing. We have been increasing that due to market volatility and a few other reasons I am sure we will highlight during the call. The second topic is more abstract but exciting. It is a multidisciplinary approach towards governance evolution; fancy academic-speak for drawing on different pools of knowledge. It considers what other groups have done and tried in the past and uses that to improve our governance system at MakerDAO. One of our recognized delegates proposed that, so that was also exciting. ## General Updates ### Votes [02:53](https://youtu.be/TYjzRyEWaRk?t=173) #### Payton Rose *Polls:* - 5 weekly polls - Passed - Prioritize Onboarding a Compound DAI Direct Deposit Module (D3M) - Raise the Emergency Shutdown Module (ESM) Threshold - Prioritize DAI Direct Deposit Module (D3M) for Maple Finance - Increase the Dust Parameter for Select LP Token Vaults - Recover DAI Locked in Optimism Escrow - There are currently 11 ratifications polls ending January 14th (very soon!) - 5 ongoing Greenlight Polls, ending on Monday, Jan 24th - OGN, OUSD, RBLD, TUSD, and USDAP *Executive:* - First Executive of 2022 tomorrow! - Expected contents: - Delegate Compensation Payments - LP Vault dust Changes - Recover DAI from Optimism Escrow ### MIPs #### Pablo [04:46](https://youtu.be/TYjzRyEWaRk?t=286) [Weekly MIPs Update #69](https://forum.makerdao.com/t/weekly-mips-update-69/12543) ![](https://i.imgur.com/LVornWL.png) ![](https://i.imgur.com/USDtFGN.png) ![](https://i.imgur.com/2uDavDt.jpg) ![](https://i.imgur.com/obhU6wx.png) ![](https://i.imgur.com/niAwbOA.png) ![](https://i.imgur.com/aL54o6T.png) ![](https://i.imgur.com/L1NfLYe.png) ![](https://i.imgur.com/QGjIpVP.png) ### Forum at a Glance #### Artem Gordon [07:55](https://youtu.be/TYjzRyEWaRk?t=475) Post: [Forum at a Glance: January 7th - 13th](https://forum.makerdao.com/t/forum-at-a-glance-january-7th-13th-2022/12614) Video: [Forum at a Glance](https://youtu.be/NDy_mP4gAHM) ## Presentations ### Dai's Stablecoin Backing [11:38](https://youtu.be/TYjzRyEWaRk?t=698) #### Reducing Stablecoin Dependancy ![](https://i.imgur.com/mjhEGfl.png) - Artem Gordan: We will move on to our first discussion of Dai stablecoin backing and reducing stablecoin dependency. Most discussion is about USDC, the fear of blacklisting of USDC, and Dai being backed by many USDC. Thanks to the help of David, Peyton, and others, I put up a couple of prompts. - We will get started with the first prompt on ideas on reducing stablecoins dependency for Dai, especially in the case of a potential bear market coming along. We would love to be bullish, but we cannot always be bullish. We must expect and prepare for the worst. It is about steering collateral diversification and preparing a bear market. - Nikolaj: I can say that CES and Real World Finance 2022 is about scaling up the Dai supply generated against real-world assets. That is one part of the co-equation here, hopefully. - LongForWisdom: The main issue is the scale; we have a lot of USDC we want to offset. The other main thing we have been pursuing is the institutional vote side right for crypto sets. - Prose11: Thinking about the bear market is essential because unless Dai demand also dries up with a bear market, you have a problem where there is still a ton of demand. The only way to access Dai is through stablecoins through our PSMs. We have been seeing this last week's volatility will increase our backing on stablecoins. Although we have been doing well for a little bit, we are about to push 50% stablecoin backing as soon as the markets get more volatile again. - Unknown speaker: Bear or bull market, it seems Dai demand is going up only. It is an excellent problem to have, but it is a problem. We need to catch up on supply. #### USDC-back Dai PR [15:04](https://youtu.be/TYjzRyEWaRk?t=904) ![](https://i.imgur.com/4WFenXj.jpg) - Artem: The following prompt is a complete 180, bringing us to an exciting discussion on shifting public relations regarding USDC back Dai. The public sentiment was rather negative. Moreover, it seemed like a Dai wrapper. But what about promoting public opinion towards seeing it as a security buffer or as a positive sentiment overall? - LongForWisdom: It is best of both of these things, right? It is hard to flip the perception entirely if we get enough of it. - Unknown: Do you want us to say something nice about USDC? I did come across a funny post for somebody who reinvented Dai. People talked about how expensive transactions got with USDC on-chain and how it got worse because of the blacklist, upgradability, and poor memory management. Transferring USDC is reasonably expensive. If you want to transfer $100 worth of USDC, you are paying nearly half of that in transaction fees in gas. People theorized about making the USBC wrapper that allowed a more gas-efficient transfer. - Dai is one of the most (if not the most) efficient stablecoin implementation on-chain. Maybe there is a positive PR angle where Dai is wrapped USDC. Then, all the people that want to use USDC start using Dai instead because it is cheaper to transfer it between people on-chain. Maybe you can pick up some positive PR that way. - Prose11: One interesting aspect that came to mind was that many risk-weighted assessments have to do with how much non-stable collateral we have. This made sense since the Black Swan scenario. We do not have to worry about the stable backing as we do the volatile ones. In some ways, having more stablecoin backing frees us up a bit on the risk side with the rest of our portfolio. - Artem: If not USDC, what would be the next best stablecoin backing? - LongForWisdom: PAX, GSD. If you mean non-centralized, it is a lot harder to find good backing. - Unknown: PAX, GSD, and USDC are potentially regulated and backed stablecoins. There are also things like USD and more decentralized stables. We may want to take a look at that at some point too. - LongForWisdom: We have discussed a few of these before. In some cases, part of the issue is that market caps are not high enough. We have a lot of USDC that we would need to shift completely to one of these. Also, this questions how stable these will be over the long term. - Unknown: Fundamentally, if these coins are over collateralized, they will have the same supply issues that we have. They will plain explode if they are under collateralized. There is no Get Out of Jail Free card with any of these. - Prose11: The real world angle has been voiced a decent amount in the sidebar. One interesting challenge is that since it is not off-chain, we have to rely on legal structures that add some complication, particularly with the size of our surplus buffer. The liquidation for real-world assets will be much slower than an on-chain liquidation, assuming both work as efficiently as possible. - Unknown: The DAO prefers inaction regarding the USDC issue. There are different solutions, but people do not want to take them because they are not perfect. We can do on-chain DeFi stuff, but that has counterparty risk by investing in other protocols. - We can do off-chain stuff and put USDC in treasuries or other liquid credit assets. People have criticized this from a decentralization standpoint. USDC is flowing in rapidly to the PSM. We have over four and a half billion dollars of exposure sitting there in stables. It is challenging because it is a lot of money, and it is a complex thing to discuss. At some point, we will have to come together as a community and do some deep analysis on it, or maybe hire people to do some deep analysis on it. Structuring even a traditional credit investment for Treasury-type services is much money, so we have to do it correctly. Real-world assets will help, but we are still quite far away from that scaling in a significant way. - Unknown: Until we get the real-world assets online, I see it as we have crypto assets. However, we can only get so much supply from those, and people want to leverage those in bull markets. Having centralized stablecoins is the only thing we can do at this stage. We may get a boost from stake ETH, which could give us a bit of a shot in the arm. However, in the fundamental long term, we need real-world assets. We cannot keep the supply up to match the demand forever. - Unknown: This is also a capital allocation issue derived from being highly decentralized. If we were in a centralized company, you would see a vast investment focused on scaling real-world assets quickly. Since this is so new and novel, the team is hesitant to request more funds and scale this. We are at the stage at Proving Grounds. Once we have proven a base model it is scalable, we need to look at reinvesting deeply in RWA. Given how vital real-world assets are to the protocol long term, we are significantly under-investing under allocating capital. - Brian McMichael: Boosting D3Ms is a risk, but it should be weighed against the risks of having too many stablecoins as well because much of the demand for Dai is coming from places like Aave and Compound. You have to decide where you want to take existential risks. - Unknown Speaker: Were there no competitive issues with Compound that were brought up on the GNR call before that made it less beneficial to partner than Aave? Are they doing something stables themselves? - LongForWisdom: They talked about launching their stablecoin like cross-chain a while back, called Cash. - Unknown: Looking back at shifting PR regarding USCC back Dai, I see some people with vested interests and bags speaking on Twitter. At the same time, we see an all-time high for Dai and Dai being used everywhere. The biggest issue is blacklisting risk, which has been discussed many times. It does not matter if there is 4 billion Dai or 2 billion Dai. If Maker could get blacklisted, it is game over. I am wondering what is the problem here? I am not seeing Dai being used less necessarily. - Unknown: We have a good problem of too much demand. We need the supply to catch up, or we will essentially be wrapped USDC. - Nikolaj: Why can it not be wrapped USDC? Why not 50%? What is the goalpost here? Because it is also good to have some stable collateral to cohesion us in the market and downturns. - Unknown: The goal is that we are not centralized on one particular asset; we can survive the collapse of anyone. We are critically dependent on USDC and probably even USDP as well. What we want is a basket of assets jurisdictionally distinct and diversified. That is the goal. - Brian McMichael: Since it was launched, I have been a pretty vocal critic of the PSM. I will take the other side and say that we have reduced the PSM and increased the supply for the last six months. It may just be the case where we have these shocks where we have to take on the USDC to meet the market demand for Dai. We can solve that later by slowly bringing back the collateral as the market allows. It might be working; we are now looking at a local shock. - Unknown: I have another view on that. From what I have seen over the past couple of years, Dai demand steadily picks up in the bull market side. It seems to tick up from farming yields. That is not the entire thing, or else in the reverse case, the demand would drop, which we are not seeing. There is a bit of a drop when we have a significant market drop, but demand seems to start ticking up again after a while. It seems we get demand from both the farming and people as a flight of safety out of volatile crypto assets. I do not see why this trend would not continue. In the bull market case, the supply grows to roughly match or exceed the demand growth because people can mint more. This past trend of demand ticking up over time only increases Dai demand. We need to prepare for bear markets where we take on stablecoins because we do not have any alternative. #### Governance Evolution: Multidisplinary Approaches [31:59 - Payton Rose](https://youtu.be/TYjzRyEWaRk?t=1919) - I wanted to kick off the introduction to this topic to the recognized delegate whom you all suggested from Acre Invest, Young. [32:12 - Young](https://youtu.be/TYjzRyEWaRk?t=1932) - Hi to everyone. Thank you for the invitation to the recognized delegates to submit a topic for this call. I was intrigued by a post recently about multidisciplinary approaches toward governance. I would love to have a conversation or hear from people on this call about how our system of governance and recognized delegate function ought to move forward in a decentralized way. Also, the topic we are broadly speaking in the community around DAOs, voter participation, engagement, free-rider problems, contributions for work. We wanted to see the thinking and people’s views related to MakerDAO and our future. [33:23 - Artem Gordon](https://youtu.be/TYjzRyEWaRk?t=2003) - I have a couple of prompts that you touched on. I have this one. Also, if we want to speak about Andrew Hall's post regarding incentives and MKR participation directly, we could jump to that and then come back to this. [33:46 - Young](https://youtu.be/TYjzRyEWaRk?t=2026) - Maybe one specific prompt, since I am new to the community, I would be curious what conversations have taken place on this topic up until now. Any background on the transition from the foundation to setting up recognized delegates and any history you think would be an important context for me or others who are new to the community as we think about this broad topic. [34:19 - Unknown](https://youtu.be/TYjzRyEWaRk?t=2059) - I can give a brief technical history of how things have changed. We were not always a protocol that had these delegates. This is a reasonably new development. MKR voting is intended to quantify some measure of skin in the game. In a way, it is a plutocracy. It is intended to be that way because we want MKR holders to make good decisions. They would be, therefore, financially penalized if they made bad ones. Good choices, you get a bigger burn. Bad decisions, you get diluted. That is the basic gist of it. - From the very old days, it was always continuous voting. We stake our MKR on a particular state of the system. In the old days, you could only change a couple of parameters like in single collateral Dai. When we released multi-collateral Dai, the system became infinitely more configurable and completely upgradable by governance, with a couple of caveats. The same thing, in continuous voting, your MKR had to park itself on an executive to guard the system's security. There is already this anti incentive to take your MKR and move elsewhere unless you ruin the security of the governance system, and you get diluted. - After a while, it is too complicated to be voting on executives all the time. We started to have a lot more configuration changes. We went to polling to signal what we would not want to include in our governance executive. Now, you can poll. You send a transaction on-chain, but the transactions are quantified into the polling result. On-chain polling is not to be confused with signal polling in the forums. When you do on-chain polling, it is mostly events that are fired and then counting the results of those events. That allowed us to develop a better set of changes and for people to park their MKR and executive. Then, we took a detour into a design space where we could use Starkware and some other L2 to reduce the amount of gas that participating in governance was costing. That led to many ideas around a governance redesign. There is even a MIP for it, like snapshot voting and much other stuff. - Then, the gas problem became big. There is one step in between the gas problem. We were vulnerable to flash loans for a little bit. We did a governance upgrade that prevented people from taking flash loans and voting executives. After that, to reduce the gas costs on individuals from participating, we effectively created a way for people to aggregate their MKR into a single delegate. That delegate can vote the MKR but not take possession of it. Now we are in this delegate system, which is working to help secure the hat but perhaps has the trappings and benefits of representative democracy. That is the history of how we have gotten here. Are there any questions about that? [38:22 - Young](https://youtu.be/TYjzRyEWaRk?t=2302) - Thank you, that was an excellent overview. Makerman, you commented about other stakeholders, vault holders, Dai holders, and their ability to influence governance. How do people view that dynamic today, and what is an ideal structure to incorporate their views if the current state is not perfect? [38:55 - Deniz DUX](https://youtu.be/TYjzRyEWaRk?t=2335) - I was talking to Colby Anderson. He is now with the SES team. We had a good conversation, and he changed my mind on some topics. I am with a DUX team. We have built a governance portal, for example. The voting tooling we are currently using shows that scalability and the gas costs associated with participating and making governance are huge user problems. We have been breaking our heads over an issue. As Christopher just mentioned, the main challenge I saw initially is that the Maker voting system is designed for security. A lot of it is on-chain and is therefore very difficult to tamper with it, especially as it is a relatively new CU. - I learned from that discussion with Colby and after talking with other people that we should consider the voting system that we use as modular. This continuous voting mechanism to protect the on-chain Maker system makes much sense. We should not mess with that in the short term. We have many tooling out there, like Snapshot. Also, there are alternatives out there. We are already looking into gasless voting mechanisms. There is a lot out there that we can use for multiple things in the DAO. We should probably not use gasless voting for changing on-chain parameters. That is probably very risky. We are eager to explore together with GovAlpha to see which governance primitives or actions we can use these gasless voting mechanisms for. That is super, and it is on the radar from our side. It is one of our priorities for this quarter. We are looking into it, not only Snapshot but also other mechanisms. [41:09 - Young](https://youtu.be/TYjzRyEWaRk?t=2469) - Would the community ever have an appetite for polling Dai holders about their views? Could that be a gateway drug toward getting more involved in the Maker Core Protocol? Before being a recognized delegate, I was most active with the vaults and Dai. That is a relationship that more people understand, like stablecoin, and there may be layers of communication in which we can engage the holders of Dai and then draw them in. [41:57 - LongForWisdom](https://youtu.be/TYjzRyEWaRk?t=2517) - This comes up semi-frequently. Can we extend some governance power to Dai holders or vault holders? In theory, we could do stuff like that. Snapshot could work for Dai. We could have Snapshot systems set up versus Dai. Each Dai is a vote, and you can vote on things. We could probably do something similar for vault holders if we wanted. To what extent do Maker holders allow Dai holders and vault holders to affect the system? As Chris mentioned before, the key, I guess safeguard, in theory, to make all the decisions is that they are on the hook financially for bad decisions. That is true to a limited extent of Dai holders and vault holders. If Dai breaks completely, Dai and vault holders are potentially out of luck. For MKR holders, it is more graduated. - If we do a bad job, then we get screwed over. If we do a little bit of a good job, we get like some good burn. The problem with this is the Dai and vault holders have opposing desires and stakes. The Dai holders want the DSR high, while vault holders want the vault fees below. If you let Dai holders set the DSR to a high number and vault holders set the fees to a low number, MKR holders will end up paying the difference. In extreme cases where you assume everyone is perfectly selfish, that is the system you end up with. Traditionally, MKR holders have the voting power because they balance the two. [44:25 - Young](https://youtu.be/TYjzRyEWaRk?t=2665) - There is a whole realm of conversation and engagement that is bigger than the technical functioning of the protocol and some operational details as to how it should run. We could maybe engage a wider audience with this. This post jumped out at me because I am a humanities person. We could add to the conversation, which is very operational on the protocols as a system. There is great potential and importance of our work as a decentralized stablecoin and an emphasis on clean money. I am thinking about how it could be marketed to gauge views that impact the adoption, scalability, awareness, and attraction of what MakerDAO stands for to people. However, I agree with and hear what you are saying. From a business standpoint, it is not giving those stakeholders a foothold in the important decisions around the operation and structure of the protocol. [45:56 - Someone](https://youtu.be/TYjzRyEWaRk?t=2756) - You bring a good point that would be relevant to the last conversation. We may pull Dai holders by signal beyond some Twitter poll or forum poll. We could ask people with skin in the game as Dai holders how much stablecoin backing are they willing to tolerate? We may be able to get excellent signals from them. We are in the stablecoin backing scenario because Dai holders and integrators effectively told us they were unwilling to tolerate a soft peg. If they were willing to tolerate a soft peg, we would not need PSMs. We would just be floating four cents above or four cents below peg at any time. But the Dai holders and integrators were specifically not using Dai because of the floating peg. You may be able to quantify almost where the sentiment is. At what threshold do they find it completely intolerable? Or would they rather have a floating peg than a tight integration without affecting any protocol? [47:11 - Nadia Alvarez](https://youtu.be/TYjzRyEWaRk?t=2831) - It would be interesting to ask the Dai holders what they think. We, in Growth, all of our interactions are with Dai holders, and they do not care about what is behind Dai. They care about Dai being a hard-pegged to one and to find different or on/off ramps with low spread and low fees in the middle. Of course, they care about having gasless transactions if it is possible. They care about liquidity and the fees in the middle of each transaction. They need Dai to be a stablecoin and keep the peg. That is what they care about. This is what everyone looks for in a stablecoin, not just in Dai. [48:39 - Prose11](https://youtu.be/TYjzRyEWaRk?t=2919) - Good conversations are going on in the chat here. I will encourage anyone if you are comfortable hopping on the mic to go forward. [48:46 - Nicolaj Lollike](https://youtu.be/TYjzRyEWaRk?t=2926) - In theory, I had written that MKR holders are incentivized to keep both Dai holders and vault holders happy. But as Chris notes in the sidebar, MKR holders might have imperfect information about the sentiment of these actors, which I completely agree with. And to Nadia's point, I was working with her in the foundation on integrations. We saw in the integrations team that integrators are not following Maker governance because there is too much to follow. I am sure people on this call also think sometimes there is too much. What integrators care about is a stable and useful product. We would try to make sure that no breaking changes were happening, for example, integrators or major upgrades that would have a major impact on the Dai holders, whether they would have to migrate that Dai or something like that. [50:09 - Prose11](https://youtu.be/TYjzRyEWaRk?t=3009) - Thanks, Nicolaj. Good reflections there. A number of people go from vault holder to Dai user to eventually MKR holder. That is a good path, but several people do not take this path. I appreciate conversations like this. [50:24 - Ultra Schuppi](https://youtu.be/TYjzRyEWaRk?t=3024) - It is hard to keep track and learn everything. It takes a long time. Suppose you complain about MKR holders not having enough information to make good decisions. How will we get better if we let in more people with even less knowledge, less attention span, and less willingness to learn? I do not think we have a problem with non-MKR holders participating. If you are a Dai holder and disagree with stuff, join the forum, complain, and interact. That is the only way to learn about this stuff. - Vault holders wanting low stability fees is just one part of the equation. It is just one parameter of complexity. The system we have right now of signal requests is like the House of Commons. Regardless of whether you hold MKR or not and how much you hold, you have just one vote. It is pretty good. It is a great system that we have in there. - In some cases, it is super slow, and sometimes you do not think that the reasonable results are going on-chain. I suffered from this during the surplus buffer discussions we had in the last 12 months. However, it is still a valid system. Of course, it is a system of compromises. It is healthy to do it this way. That is decentralized, I would say. We have a big issue on that topic. [52:03 - Someone](https://youtu.be/TYjzRyEWaRk?t=3123) - I would honestly argue that we have an issue because the voter participation rate is ridiculously low. This is one of the results of the current design of our voting tooling and our governance system. In my conversations with Andrew, I learned that one of the reasons real-life nation-state governments have periodic elections is that asking people to vote on every topic or issue is not feasible. There are many ways to redesign or improve the voting tools and the governance process to ensure that we do not ask for too much attention from our users and delegates. There is a lot to gain by exploring that. I am happy that we are having this discussion. We should not regard our current system design as the way to go because it is technically decentralized. [53:27 - UltraSchuppi](https://youtu.be/TYjzRyEWaRk?t=3207) - We are working on two things here. One is a governance problem. I advocated for a long time that we should hear from all stakeholders. All stakeholders should not necessarily have levers on Maker to adjust the system. You are running a business; you like to hear from the people using it to have your pulse on the people. I have been around the idea and governance issues for decades with many different organizations. I see the real issue at Maker, other places, and stocks are centralized ownership. If we analyze our Maker holder wallets and what happened in governance history, there probably are 50 wallets at most that have delegated. Probably five of those holding 90% of the major Maker delegated. - If we want to talk about governance, we need a governance analysis to understand what is going on in governance. Then we need to think hard about the problems and what we think the solutions could be. A lot of us are going to disagree and discuss. I want to hear from other stakeholders. I would like polling not to cost an arm and a leg. I would like to see that move somewhere else because it does not do anything on-chain except give us a signal. I see this as two different things. I want to hear from stakeholders and have mechanisms for that. I want to look at an analysis of Maker ownership concerning governance. Then we can start to address these issues and think about if there are problems or not. [55:24 - Someone](https://youtu.be/TYjzRyEWaRk?t=3324) - I am going to give a bit of technical feedback. Today, the polling contract knows nothing about MKR when we do on-chain polling. You send a transaction that fires an event as cheap as possible. Then it is off-chain stuff that checks how much MKR was held. We could easily deploy that contract onto an L2, and people could send a transaction on that L2; it would be wildly inexpensive. We could quantify as many tokens as they were holding. We could ask a Dai holder, or LP token that had Dai in it, or 3pool token, whoever would be quantified as a Dai holder, we could ask them questions about Dai and get feedback. [56:08 - UltraSchuppi](https://youtu.be/TYjzRyEWaRk?t=3368) - That is a great technical solution, but it will not solve the problem. Even if the polling on-chain is for free, I do not expect much change because MKR holders have no incentive to participate in governance. The value of your MKR is not changing, if you are taking part in governance or not. That is the flaw that we need to fix if we want to drive up the participation rate of MKR holders. Most of the MKR holders have this as an investment. They do not want to care, and that is okay. But there should be a strong incentive to put their MKR at work by delegating to someone else, which will also drive more active delegators or delegates into the system. If there is zero incentive, it will not change if we do Snapshot voting or polling on an L2. [57:25 - Someone](https://youtu.be/TYjzRyEWaRk?t=3445) - For some time, this is one of the reasons why I brought up this idea of having some reward for participation. If the rewards in Maker were printed, then in this loose sense, the people who do not participate get diluted away by the ones who do. I worked hard trying to come up with a model. The model was this idea of just paying a reward, like putting a bounty on per poll, and particularly executive participation. And that way, it rewards the people who participate in direct proportion to their participation. You can use that reward if you do not have any of these issues about arbitrary levels of whatever. I had it at the point 2% level of Maker inflation. We could have a 50% pass-through on delegates and compensate delegates. And then the 50% pass-through can go to the rest of the people, and the issue is gas paying them out. - You could do this over time, or you can have it fast. I agree with you. The whole idea is that there is no disincentive for people not to participate. I have seen this in so many communities where everyone had equal shares. There would be small groups of people who would not participate in the community. All these people were doing a huge amount of work, yet the people not doing anything benefited. That ends up not working well in the long term. I agree with you, Schuppi, that some approach to do that would work. Unfortunately, Maker governance has not wanted to step toward some reward that would dilute the rest and not participate. [59:20 - Artem Gordon](https://youtu.be/TYjzRyEWaRk?t=3560) - That perfectly brings up the next prompt between you and Schuppi. We cannot meet everyone's needs, and we have to decide on the best approach going forward. We want to improve incentives, user experience, or gasless voting. Also, we want to control some voters' power. Let us say gasless voting would be bad for risk parameters. We also want realistic steps going forward, like short or medium-term plans over the next six to 12 months. How do we decide the best approach, and what could be the best approach to meeting people's needs and not pissing everyone off? [1:00:11 - Someone](https://youtu.be/TYjzRyEWaRk?t=3611) - Can we start with getting polling off L1 and figure out how to get some of our additional stakeholders in? If we get other stakeholders playing, they might buy Maker and play more, and we will get more participation. I see that as low-hanging fruit that, at least if I hear Chris, is an easy one to tap. [1:00:39 - Christopher Mooney](https://youtu.be/TYjzRyEWaRk?t=3639) - It is more UI work than anything or whatever the off-chain quantifies. I do not know who is responsible for that right now. [1:00:50 - Prose11](https://youtu.be/TYjzRyEWaRk?t=3650) - You have to be a little careful when you change these things because you can have unintended consequences from the governance perspective. For instance, if you go to Snapshot, that kind of changes a lot of the assumptions you are making. It is taking a Snapshot of voting waiting time. People who may have sold their MKR past that time can still have their full voting weight. Let us say they are your proposals losing and you wanted to acquire more MKR, currently in the system, you would be unable to do so. There will always be advantages and disadvantages to making a change like that. It is important that when you have a method in mind, you consider what you are losing by making this change. [1:01:38 - Someone](https://youtu.be/TYjzRyEWaRk?t=3698) - One consideration for Chris Mooney's suggestion from a technical perspective is that it would decrease gas costs. Still, it would require anyone participating in those polls to have whatever L2 gas transaction token is required. They would have to get that before they could participate in those polls. [1:02:09 - Christopher Mooney](https://youtu.be/TYjzRyEWaRk?t=3729) - Alternatively, we could count the Snapshot data on all chains at once. We could deploy it to Optimism, Arbitrum, and Mainnet and then count up anyone that sent a transaction. If you are not already on an L2, you must move funds over to that L2 to make the transaction. [1:02:50 - Someone](https://youtu.be/TYjzRyEWaRk?t=3770) - Would it be possible to get data on other protocol's results using Snapshot? I enjoy working around other protocols like Balancer, FAKE Protocol, or whoever else, and they do not get a big turnout with Snapshot. They get maybe 20, 30 different addresses. The only time I ever see a lot of excitement and a huge voter turnout is when you have a new protocol that gets many people excited. Maybe they got an airdrop. Suddenly you see an onslaught of voters come out to participate on Snapshot. Because obviously, it is free. I wondered if someone could collect data and see these OG protocols and what type of turnout they are getting with Snapshot. I am willing to bet that it is not big. If Maker goes into a Snapshot model, you will get a few more turnouts, but I do not think it will improve 100%. [1:04:07 - Prose11](https://youtu.be/TYjzRyEWaRk?t=3847) - I will mark down the link to Snapshot.org. It is pretty easy to browse through. It would be nice to get some comprehensive data. We have looked into some stuff in an upcoming research report, GovAlpha, but it was not specifically related to delegating. I do not believe those metrics are going to be included. I will see what I can do on that trend. [1:04:36 - LongForWisdom](https://youtu.be/TYjzRyEWaRk?t=3876) - I think investigating some cheap gas or alternating makes sense. It is more or less an obvious progression from what we have. But as people said, I did not expect it to change the amount of participation we are getting drastically. It is worth doing because it makes it cheaper for those participating, which is a benefit in itself. I do not think it will fundamentally change the game. [1:05:32 - Prose11](https://youtu.be/TYjzRyEWaRk?t=3932) - Considering the timing, we have potentially a little bit for open discussion if there are no last remarks on this topic. [1:05:50 - Someone](https://youtu.be/TYjzRyEWaRk?t=3950) - The governance contract redesign was MIP26. It was a technical MIP that included some features. There is a forum post about it. People can add comments if they have more ideas, although that current development path is on the backburner, at least from a political engineering perspective. If you have ideas, it is a good idea to quantify them because, at some point, this is going to come to the forefront. We will want to redesign everything. [1:06:25 - Brian McMichael](https://youtu.be/TYjzRyEWaRk?t=3985) - I should mention that the current delegation contracts have a one-year expiration. I do not remember when we started that, but those will all come to an end at some point. Delegates will have to reinitialize a delegation contract. It might be worth considering changes around that time; otherwise, we will roll into another year of the same system. [1:06:59 - Prose11](https://youtu.be/TYjzRyEWaRk?t=4019) - The reason behind the expiration is that something happened to a delegate, especially one that uses a personal address, or what have you to make the votes ramp, that MKR will be stuck. Anyone who did not realize it or is not paying attention would have their MKR delegated to someone who is not doing anything. So a few different reasons. [1:07:26](https://youtu.be/TYjzRyEWaRk?t=4046) - The other more powerful one for me was in the opposite direction. Let us say, for example, Micro Whale. You have got a couple of 100 MKR, 1000 MKR. You delegate that, and then you get hit by a bus. Without the expiration, the person you delegated to effectively controls your voting power into eternity. This is something I have seen blown up happen in the real world with like HOAs and stuff where people age out, and they delegate their voting to a particular actor, and that person ends up having like a supermajority. It becomes a bit of a dictatorship. It would help if you had an age-out delegation. [1:08:14 - Brian McMichael](https://youtu.be/TYjzRyEWaRk?t=4094) - There is also a perverse incentive for a delegate to ensure that their delegates cannot un-delegate from them. So you can imagine all the ways that that could happen. And then they are dark when you get on that path. [1:08:37 - Prose11](https://youtu.be/TYjzRyEWaRk?t=4117) - Awesome, any topics or open discussion? Now is a good time to circle back to something said earlier and that we have moved on from. [1:08:56 - Someone](https://youtu.be/TYjzRyEWaRk?t=4136) - I want to make a quick comment on Snapshot and compare it to Aave. Our forum activity looks like we have around 6x the number of forum posts a month for topics. Their voting participation on Snapshot is significantly higher than ours if you look at any closed poll on their website. I doubt that is unique to Aave. We can go through many other protocols, but I would assume that higher forum participation should generally translate into higher voting through Snapshot. [1:09:40 - Prose11](https://youtu.be/TYjzRyEWaRk?t=4180) - There are external factors as well too. [1:09:44 - LongForWisdom](https://youtu.be/TYjzRyEWaRk?t=4184) - There are staked Aave and other things like how people get paid and rewarded for being staked in governance. That is the compounding factor. [1:10:00 - Prose11](https://youtu.be/TYjzRyEWaRk?t=4200) - They do not have to vote; they just stake. [1:10:04 - Someone](https://youtu.be/TYjzRyEWaRk?t=4204) - The effects of staking are effectively the same as holding an MKR token. We do not require people to stake it. [1:10:14 - Kirk Hutchison](https://youtu.be/TYjzRyEWaRk?t=4214) - People staked in Aave are the first in line to be recapitalizing the protocol? Is there a trade-off? [1:10:23 - LongForWisdom](https://youtu.be/TYjzRyEWaRk?t=4223) - Yes. It is not quite the same as Maker. They get more than their proportion, it is not split overall token holders, but they are all stacking holders. It is right that they still get it even if they do not vote. But that is not about the point that I made earlier. [1:10:42 - Prose11](https://youtu.be/TYjzRyEWaRk?t=4242) - I was also going to suggest that our system potentially discourages active voters because we have so many votes, polls every week, executives every week, the potential for emergency things, and many different types of polls. That complexity and the continued need to appear new things can be intimidating for some voters. Have them conclude. I can keep up with this protocol because they only vote a few times a month, whereas, in Maker, we had 21 polls go up on Monday, for instance. That was an abnormal week for sure, but it does happen. [1:11:21 - Someone](https://youtu.be/TYjzRyEWaRk?t=4281) - We are overcomplicating it. I just clicked through like ten different DAOs, and they all have tremendous voter participation on Snapshot. I do not see any that have less than us. These are even smaller projects than us. I am sure some data can be pulled somewhere. Maybe Snapshot themselves has a report they could share with us to analyze. But from what I see, participation is significantly higher than ours from an individual's wallet standpoint. [1:11:54 - LongForWisdom](https://youtu.be/TYjzRyEWaRk?t=4314) - I agree. We are (x?) right now. As I said before, I still think it is a good idea to pursue. I will make the point that there is no civil protection at all for Snapshot votes. It is completely free to vote. In some cases, this way is legitimate, but there are some cases where they are not. [1:12:14 - Prose11](https://youtu.be/TYjzRyEWaRk?t=4334) - Lots of interesting discussions. I am happy to close the meeting early for once. We tend to go on quite aways. Consider this your last call. [1:12:43 - Someone](https://youtu.be/TYjzRyEWaRk?t=4363) - One quick thing that may be helpful. LongForWisdom suggested to me in discord that we on PE, write up a description of this, but it might be helpful to talk through it. In this week's executive, we will have the Optimism recovery of a 10 million Dai that got burned. That is a fairly unique governance action. We have several people voting on that on here, and recording and hopefully others are watching. I wanted to open the floor if anyone wanted to talk about that. I am happy to go over at a high level what the plan is, and then we can ask questions, just so everyone is comfortable with what is going to show up in the executive. [1:13:30 - Christopher Mooney](https://youtu.be/TYjzRyEWaRk?t=4410) - Great. This week, we are dealing with the 2 million Dai that was burned by the user who lost some Dai. There was a forum post about it and the poll passed. What happened is this user called the burn function on the Optimism Dai contract. That destroyed 10 million Dai on Optimism. In the executive this week, we will be approving the transfer of that Dai from the Mainnet L1 Escrow contract to the user on Mainnet. The very important thing is that this user destroyed the Dai on Optimism, which means that our Optimism Dai total supply is significantly less than 10 million Dai, less than the Mainnet balance of the Escrow contract. This opens us up to transfer the 10 million Dai on Mainnet, to the user's address on Mainnet, while keeping our balance rules and variants in place, which is that the Mainnet Escrow balance should be greater than or equal to the total Dai supply on Optimism. It is fairly simple governance action. We are making a ERC20 approval on Mainnet Dai. Then we are executing a transfer from the past proxy to move 10 million Dai from the Escrow to the user's account on Mainnet. We are not touching anything on Optimism. We are not passing any governance actions from L1 to L2. I will stop there. I feel like I am probably going to start rambling into technical details. Let us see if there are any questions or discussions people have points of clarification. [1:15:29 - LongForWisdom](https://youtu.be/TYjzRyEWaRk?t=4529) - That makes a lot clearer for me, Chris. Thank you. [1:15:34 - Christopher Mooney](https://youtu.be/TYjzRyEWaRk?t=4534) - I have a draft that I will probably put in the forum post for the signal request. We will link it in the Governance Discord channel to write this up too so anyone can read through that. [1:15:53 - Prose11](https://youtu.be/TYjzRyEWaRk?t=4553) - I see Makerman commenting and questioning the need for a tester or verification for the user. If you are following the forum, we could have the wallet address sign a specific message saying: "okay, this is me. I am the one who lost the Dai." It might be interesting. That s not a simple technical explanation. [1:16:27 - Someone](https://youtu.be/TYjzRyEWaRk?t=4588) - I do not know. Maybe it is just me. Before I send a million, I usually send 100. [1:16:35 - Nicolaj](https://youtu.be/TYjzRyEWaRk?t=4595) - This might be too technical, but I was wondering. Why can you call the burn function? Like why is it not off? It seems like people could take Dai out of circulation, and it would have some repercussions. I was interested from an engineering point of view. [1:17:05 - Someone](https://youtu.be/TYjzRyEWaRk?t=4625) - Yes, the burn is off on the MKR token. Maybe there is a good argument that it should have been off from a Dai token. But, we heavily rely on burning in the DAI JOIN adapter. In theory, you could just give the DAI JOIN adapter privilege to burn. But there is also approval burning. You can approve a certain amount, and then someone else can burn if you have approved them too. Is that true? [1:17:42 - Kirk Hutchison](https://youtu.be/TYjzRyEWaRk?t=4662) - Yes., I think at least the Mainnet tokens work that way and MKR too. If you approve for both to burn and transfer. If you approve someone to transfer away your tokens, they could transfer the tokens to themselves and then burn them. [1:17:59 - Nicolaj Lollike](https://youtu.be/TYjzRyEWaRk?t=4679) - I am interested in how this could happen. They really have to go out of their way to burn their tokens. [1:18:08 - Christopher Mooney](https://youtu.be/TYjzRyEWaRk?t=4688) - None of us asked, as far as I know. I like imagining how they did it more than the reality, maybe. The reality is probably something like they were testing some amount of code and did not realize they were connected up their VPC endpoint, like executed the burn, or something. But that would be my guess. [1:18:36 - Someone](https://youtu.be/TYjzRyEWaRk?t=4716) - Chris, sorry, real quick, before we move on. I want to make sure that I address the comments in the chat about documenting the process. Are you all asking about having a formal process for this that we can apply now and in the future, or are you asking about a write-up for the specific action? [1:18:56 - Someone](https://youtu.be/TYjzRyEWaRk?t=4736) - I will speak a bit for delegates here. A number of us would like to see a global process for a criterion of when we are going to do it, whether there is a fee or not, and the process involved for how to do it. We do not want to deal with this in a one-off event model. It does not feel right. We would like to have this as kind of a general thing. That would make us feel better about it. It should be easier than to deal with, at least in my opinion. Anybody else can speak up. [1:19:28 - Young](https://youtu.be/TYjzRyEWaRk?t=4768) - I agree with that. We want to manage the inflow or signal, that there is a process and what it is, and that everyone has an equal understanding of being treated the same. And also that we manage the workload to the extent that a fee is appropriate for the extra resources and time and to encourage or incentivize people to be super careful and not have a sense that if I mess this up, I can go to the community, and they will fix it and undo it for us. So the moral hazard issues are important to think about. [1:20:18 - Someone](https://youtu.be/TYjzRyEWaRk?t=4818) - Yes, that is the comment I was going to make about the hazard that we create, especially at different thresholds. It makes sense for us to probably have a policy. In my mind, I was thinking for these really large amounts that we are moderately incentivized to try and take any deflationary pressure off the peg. Burning 10 million these days is not that big of a deal, but you could imagine when it was 10% of supply, that would have caused really bad repercussions for Dai's peg. We have narrowly escaped enormous sums being locked into like DeFi Ponzi games during DeFi summer a few times. So we have almost had to do this in the past. A whole class of normal users has made this mistake, that of where they have sent the Dai to the Dai contract. It may be able to have a once-a-year sort of debt forgiveness jubilee, where we create a miracle distributor and a token type called, for example, dead. Then you can mint Dai against dead, and the miracle distributor does all this under the hood and mints you a bunch of new Dai, and then the end contract cleans the accounting up at emergency shutdown so that we do not account for dead as a real collateral type. We can handle smaller cases on one giant aggregate once a year, and then for larger cases, we need to figure out where the threshold or the cutoff is. Clearly, 10 million is the cutoff right now. [1:22:06 - Brian McMichael](https://youtu.be/TYjzRyEWaRk?t=4926) - I want to say that this is also a special case because it is L2 and because we can do this with this sleight of hand that does not mess with the accounting. The burn function on L1 Dai is also on-off. This could theoretically happen on L1, and we would not have the same ability to pull this out of a distributor contract whereas the escrow contract. There is the question of what we would do if this had happened on L1, where 10 million Dai was just burned. The mint function is off, and the only contract that can mint new Dai is the JOIN adapter for Dai. So we have to create solutions like Chris was talking about if we want to perform a similar action on L1. [1:23:15 - Prose11](https://youtu.be/TYjzRyEWaRk?t=4995) - This user manually called one of the steps that you would end up going through if you are going to withdraw your Dai from L2 back to L1. In a way, we are just completing the steps. [1:23:31 - Kirk Hutchison](https://youtu.be/TYjzRyEWaRk?t=5011) - They did the first half of withdrawal. We are doing the second half. [1:23:49 - Someone](https://youtu.be/TYjzRyEWaRk?t=5029) - Nik brought up the question in the sidebar, asking if anyone was concerned from the legal perspective. Once we display the ability to perform the action, it will become an obligation to do so in the future. [1:24:05 - Someone](https://youtu.be/TYjzRyEWaRk?t=5045) - Under what jurisdiction? [1:24:11 - Brian McMichael](https://youtu.be/TYjzRyEWaRk?t=5051) - Even from a customer service perspective, not even the legal. [1:24:21 - Someone](https://youtu.be/TYjzRyEWaRk?t=5061) - I would say Maker holders should probably signal on-chain what they are willing to accept as a policy for this. From a legal perspective, it comes down to whether or not a particular jurisdiction can coerce MKR holders into performing an action. [1:24:44 - Someone](https://youtu.be/TYjzRyEWaRk?t=5084) - We can pat ourselves on the back, but the guy who came in here a couple of months ago and lost a few ETH on L2 was a much smaller amount than this, but it meant a lot to them. As things stand right now, we are just taking the stuff on a case-by-case basis. That is probably not the right way to go. [1:25:04 - Someone](https://youtu.be/TYjzRyEWaRk?t=5104) - From a policy perspective, it is important to clarify that both the case that Brian just mentioned with the ETH and the case with this Dai are very specific circumstances. Just because you lose ETH or Dai does not mean that we will be able to recover it in all cases, and not having a clear explanation of that will put us at some level of at best customer service risk. [1:25:37 - Nikolaj Lollike](https://youtu.be/TYjzRyEWaRk?t=5137) - It opens up for some creative accounting. If someone is speculating and losing the Dai for x reasons, and then having them minted at a later stage by the protocol. [1:25:53 - Young](https://youtu.be/TYjzRyEWaRk?t=5153) - The public relations of the small, the little guy, sort of like "sorry, cannot help you." And then the big whales like, "oh, we can help you." Even if there is a technical difference, it is important to keep in mind, hence the policy. I am not sure what the next steps are. There are already some discussions about that policy and how we can move that forward. Since we are taking care of the burn problem for the 10 million holder, we have got to work expeditiously on this. [1:26:41 - Someone](https://youtu.be/TYjzRyEWaRk?t=5201) - LongForWisdom brought up an interesting question, is anyone working on a policy? We need one. It is going to have this fitting in. And then we have to exclude some. We have to develop a policy because of just PR reasons and optics and technically. [1:27:09 - Someone](https://youtu.be/TYjzRyEWaRk?t=5229) - I fully agree. If anybody wants to own that, please step forward. We will be happy to give you our opinions. [1:27:17 - Joshua](https://youtu.be/TYjzRyEWaRk?t=5237) - I remember looking at a Google Doc someone shared with me. I cannot remember who it is now. The document was an initial draft of a policy. I have to figure it out. Someone is working on it. [1:27:36 - Someone](https://youtu.be/TYjzRyEWaRk?t=5256) - I remember SES was getting some work done. They were discussing the (x?)(end?/N) problem, for a little bit, fixing the accounting there. Maybe SES has got the beginnings of a policy somewhere. [1:27:54 - Nicolaj Lollike](https://youtu.be/TYjzRyEWaRk?t=5274) - I think it was PaperImperium that brought it up. Then SES said that they could give out a grant for someone to work on it. [1:28:10 - Prose11](https://youtu.be/TYjzRyEWaRk?t=5290) - That is an incentive problem to a certain extent. People who lose Dai are incentivized to try to get it back. But less often for anyone coming up with a reasonable overall framework. [1:28:23 - Someone](https://youtu.be/TYjzRyEWaRk?t=5303) - Are you saying that this 10 million locked Dai is an opportunity to incentivize somebody to make the policy for us? - Prose11: Perhaps. I do not know. - LongForWisdom: It would have been. If we had not said we would already do it. [1:28:43 - Nicolaj Lollike](https://youtu.be/TYjzRyEWaRk?t=5323) - For me, the complex thing is where to draw the line. When Dai is burnt, that is really nice because it is easy from a technical point of view to say, "okay, we are not worried about it appearing somewhere else." Then there is Dai in the Dai contract, which is another case. What about people who lose their Dai by sending it to some other wrong address, so locking it up in some (x?)(1:29:09)(box?) contract? It needs to be a very hard policy for what cases we can deal with and which ones cannot deal with because we need to do some due diligence on weird third-party contracts to determine whether or not Dai is perpetually locked. That will be a massive workload. [1:29:35 - Someone](https://youtu.be/TYjzRyEWaRk?t=5375) - It sounds like we need PE to issue a technical statement about it first. About what could be and what should not be. I think. We can make stuff up, but I will go to the engineers on this. [1:30:00 - Prose11](https://youtu.be/TYjzRyEWaRk?t=5400) - I want to be mindful of the time here. I appreciate everyone's perspective. Let us keep these discussions going in the forum and possibly again next week when we see you all here. Thank you to everyone. That does it for this meeting. I appreciate you joining us. ## Common Abbreviated Terms `CR`: Collateralization Ratio `DC`: Debt Ceiling `ES`: Emergency Shutdown `SF`: Stability Fee `DSR`: Dai Savings Rate `MIP`: Maker Improvement Proposal `OSM`: Oracle Security Module `LR`: Liquidation Ratio `RWA`: Real-World Asset `RWF`: Real-World Finance `SC`: Smart Contracts `Liq`: Liquidations `CU`: Core Unit ## Credits - Artem Gordon produced this summary. - Kunfu-Po produced this summary. - Andrea Suarez produced this summary. - Larry Wu produced this summary. - Everyone who spoke and presented on the call listed in the headers.