January Governance Cycle Review
- LongForWisdom: This week is the Governance Cycle Review. This is supposed to look back on the cycle, see how it went, and then look forward to the next cycle. We saw seven proposals entered into the January cycle. Six made it past the inclusion polls and were successful. The only proposal that we didn’t continue was the declaration of intent to engage Nexus Mutual to ensure that Maker protocol.
- The successful proposals were significant. We ratified three new mandated actors, and we’ve already covered it, but congratulations to Sam and Juan. We also approved two technical proposals, which are formalizing the PSM and including CropJoin. In terms of voting, the exec passed reasonably quickly this month. Previously, we’ve seen it go down to the wire, but it’s good that it didn’t happen this week. We didn’t have any particular problems last month with the cycle. This is your chance to chime in and say if you disagree with me, but it didn’t seem like any significant controversies or concerns.
- Looking forward, I’m not sure what we’re going to see in February. There’s a lot of MIPs in RFC that could potentially be submitted. If you are planning on submitting things in February, feel free to chime in. I think we’re going to see quite a lot in March. We should be prepared for that, especially if we see budget proposals and the actual Core Unit MIPs themselves.
- Charles: I wanted to have an open conversation regarding the former submission deadline because sometimes people forget if we’re going to keep it from Monday to Wednesday.
- LongForWisdom: We should discuss it. It’s potentially problematic for us to change or start ignoring bits of MIPs. For each of the MIPs that’s formally submitted, we need to do inclusion polls. Therefore, having that open until Friday might be awkward in terms of governance facilitators having to cue up the polls over the weekend and reading through all the MIPs, making sure that they’re valid to be submitted. Perhaps it doesn’t always get done as much as it should do, but we’re supposed to do that. I think Thursday would be the latest. That still leaves the problem of us wanting to amend the MIP or something in some way, which is annoying because it’s a lot of effort.
- I guess we’ll think about it a little bit. Wednesday should be considered the deadline; if anybody is having a lot of trouble finishing anything and asking for an extension, that’s fine. We can consider that as a case-by-case thing. In general, if you haven’t gotten in touch with me and or Charles by Wednesday and you’re planning to submit something, you should tell us that it’s going to be late. Hopefully, that covers any confusion.
- Sébastien Derivaux: I want to talk about medium-term lending loans with MakerDAO. After a post I made on the forum, most people agree in the community to lend some DAI for an extended period. Hence, currently, all vaults are endless. You can borrow as much as you want for as long as you want, except at some point, you will have stability fees that will increase if governance wants you to find a new loan elsewhere. That doesn’t work for the real world because you cannot increase stability fees when you have an asset, for instance, like real estate. People need some stability. The idea was to lend for, let’s say, five years at 2% or 3% or whatever and not changing it afterward. We shouldn’t be stressed about the daily fluctuation. How does it change from what we’re doing already? As we saw, we are lending. We are liquidating when there is a drop in the ETH or BTC price, which works well until it doesn’t. If there’s only some vault to liquidate, we make a lot of money we are all happy. We’ve made 25% of revenues this month from liquidations, but it doesn’t help to liquidate something else when it fails. You can have some fee-generating assets like Uniswap pools in crypto; however, with something like rent where you are getting paid every month a certain amount. If your Maker loan fees are lower, you know that with the rent, you can pay the stability fees and decrease your Maker loan. After some time, you can finish your Maker loan by getting the rent in. That works well when you have a real estate portfolio. If someone doesn’t pay his rent, it’s not a big deal. That’s always the idea with real-world assets, which changed how we see the loans we provide to our customers. This idea may be a bit better in the real world because we are making short-term loans with the real world. Hence, in the case of mature, there is a 1-year notice in New Silver’s case. It’s only three years. They are building something or renovating a real estate property, and they are selling it. That works well if they can sell it. If they cannot sell it, there are two choices, and the same with ETH. If you cannot sell ETH at $1000, the only buyer bids it at $100. Do you want to sell ETH at $100 and take a loss of 90%? It doesn’t make any sense. For ETH, there is no fee revenue accumulating to pay for the loan. However, if you invest in real estate, you have the rent coming every month, which will likely be the same every month. There will be a small variation, but there will be no stress to liquidate if the rent is there because some revenues cover the loan. For some assets, not for all but some, it may be smarter to focus on the borrower’s ability to pay the loan rather than the collateral value on the day to day basis. That was the thought we had, and Lucas was interested as well.
- Lucas: I think you also have to consider how long we will be in a 0% DSR environment. If we start onboarding these long term loans and have 200 million DAI worth of 3% loans, we’re then pushing the DSR up to 5% because no one wants to have DAI. or we can drive the adoption of DAI through the DSR. I’m thinking back to how it was a few months ago. Still, how do those various factors come in? What do we think is going to happen with the DSR? What is the base rate that we should have?
- Sébastien Derivaux: Monet is writing that our asset duration should match our liability durations. It’s not an easy question. There is still some theoretical work to be done. DAI can be very long term because people can only redeem DAI at an emergency shutdown. Maybe, Monet, you want to comment as well?
- Monet Supply: I’ll comment more in the thread, but I know that traditional financial institutions generally put a lot of work to ensure that their asset and liability durations are matched. If market interest rates move up or down, their position doesn’t change too much. However, if we are still committed to continuously maintaining the DAI price at one dollar, we have all of our liabilities be zero duration assets. If we lend out money for long durations while the DAI price goes below one dollar, we have to raise short-term rates, which pushes down the value of our long-term assets. I think it creates a position where MakerDAO loses money when the DAI price goes down, which may be dangerous.
- Primoz: We are talking about duration risk here, and that’s why RWAs would be more suited for short term loans. It’s easier to control the supply and protect the peg. If you’re going to use long term loans, we have this duration risk exactly which Monet has described. It’s more of an issue. I would always prefer to use short term loans, but I know this becomes an issue for borrowers.
- Lucas: Short-term loans have to be insanely cheap for the money. They are insanely cheap on Wall Street.
- Monet Supply: Maybe a solution is to find people who want to hold a long term DAI investment. How banks manage their duration risk is they’ll sell long term bonds or certificate deposits that have a long duration so that if they loan somebody money for five years and have a CD that they owe somebody back in five years, they’re hedged.
- Sébastien Derivaux: Yeah, but a surplus buffer can be seen as a long-term liability even in all cases. Hence, we have at least a long-term buffer, and even above that, there is still some room to work. It’s an open question. I don’t have all the answers, but I think it’s something to work on.
- Matthew Rabinowitz: As an example, in commercial real estate, this is why transactions are a snapshot of the appraised value on closing and also why you don’t do routine appraisals every six months or every year. 95 to 98% of the value is not in the dirt. Instead, it’s in the cash flow stream.
- Sébastien Derivaux: Having long term assets is always good to pay the workforce of MakerDAO. We will have some expenses. If we lend for five years, we know that we will get stability fees for the next five years to cover all of our expenses. All the vaults can be shut down, or the vault can be closed at any time, and that will leave no income from it.
- Matthew Rabinowitz: It’s also an open question concerning what happens when the DSR begins to increase. What happens to the surplus buffer net number that’s caused by real-world assets? It should get compressed because, as you’ve outlined, the market external to Maker is efficient. If it starts pushing real-world assets, they’ll refinance. When they refinance externally, the probability of coming back isn’t so certain.
- LongForWisdom: If I remember your forum posts correctly, Seb, we were trying to figure out what proportion of long term assets would we want in comparison to short term assets. Ideally, we’d like some of both because there’s always going to be some base level demand for DAI. It’s not certain, but you can have maybe 10% of the portfolio on long-term assets that we never think we need to recover 90% of the DAI, right?
- Sébastien: The post’s aim was primarily to know if it’s something we should work on. Whether the governance and community have already decided that Maker will only make short-term loans, that wasn’t clear. The community answered that it makes sense to work on this subject, and it doesn’t make any sense to make it if it’s not a good idea. There are some issues with the DSR, but Monet and Sam are talking about it. Even in crypto, we have the yield protocol. There is a MIP to integrate YFI-DAI into Maker. That’s already some medium-term and most likely long term assets that would be difficult to sell. The final price is always the same, but if the interest rates are moving, the bond price, YFI-DAI, will change this time. We would not want to sell something and have a loss if you know at the end we will make a profit. It’s a work in progress.
- Matthew Rabinowitz: It’s going to evolve into making sure the amortization schedules of whatever the ending borrowers are will imply that the principal gets returned during the term of the cash flow such that that concern you’ve got at the very end doesn’t matter.
- LongForWisdom: Okay, that exhausts this topic. Does anyone else have any other topics they want to discuss before we close up?
- Chris: Did the PSM put Vishesh out of work? Are we never going to see him again?
- LongForWisdom: Vishesh mentioned hes going to take some of January off. Hopefully, we’ll see him back in February.
Links from Chat
Common Abbreviated Terms
MCD: The Multi-Collateral Dai system
CR: Collateralization Ratio
DC: Debt Ceiling
ES: Emergency Shutdown
EV: Executive Vote
GF: Governance Facilitator
GP: Governance Poll
SF: Stability Fee
DSR: Dai Savings Rate
MIP: Maker Improvement Proposal
OSM: Oracle Security Module
LR: Liquidation Ratio
RP: Risk Premium
RWA: Real-World Asset
- Artem Gordon produced this summary.
- David Utrobin produced this summary.
- Denis Mitchell produced this summary.
- Jose Ferrari produced this summary.
- Everyone who spoke and presented on the call, listed in the headers.