 C-FI and D-FI will actually merge over time. You don't want to rely only on ETH because that means you're too exposed to ETH's movement in the market. You have this massive portfolio of assets that is spread across the world, that has no critical reliance on any single entity. Hi everyone, my name is Giovanni. Today we are going to interview one of the biggest names in decentralized finance. Make it our founder, Rune Christensen. Rune, thank you for joining us. Thanks so much for having me. On March 12th, the price of Ethereum crashed by over 30% and MakerDAO experienced a protocol failure which left the platform with over $4 million worth of bad debts. Also, around $8 million worth of investors' funds got lost in the incident. Rune, what are the main lessons you learned from the MakerDAO meltdown on Black Thursday? One thing is, the protocol actually worked as intended and even went ahead to recapitalize itself completely like it was designed to do. But I think the other thing to look at is how the governance responded. So the MKI holders and the Maker governance community responded to this because what became clear was that with basically the amount of keepers that were in the ecosystem, which wasn't as many as I think most people would have expected. And then with the possibility of this kind of extreme congestion and crash and liquidity issues and die happening all at the same time, the first reaction that the governance community had was to increase what's called the bid duration of the auctions. So the auctions that had initially been running relatively fast in order to be as efficient as possible and that way potentially attract more keepers to the ecosystem. Instead, Maker governance changed that so that they lasted six hours and really became very slow auctions so that there's plenty of time to react and even with congestion and technical issues or anything like that, there should be a lot more time for the keepers to actually bid. But one of the other major actions that the Maker governance community took was to onboard a new collateral type, so the USDC stablecoin, because the other issue in addition to the congestion was the lack of access to die liquidity by the keepers. So they simply couldn't access die to bid on the auctions with because nobody wanted to generate die, because nobody wanted to hold the ETH during this crash. So by adding USDC, which is an asset that's completely uncorrelated with crypto, right? So when crypto is falling, when the ETH is falling and so on, USDC is $1, right? So it actually means that even if people are not willing to generate die with Ethereum during a crash like this, they'll still be willing to generate die with USDC because it's not exposed to the same correlated risk, basically. And what that meant is that that created a source of liquidity for the keepers that was reliable even during a crash like this and then also helped to stabilize the situation and prevent the zero bids from happening. You adopted the stablecoin USDC as an additional collateral in order to stabilize the price of die after the accident. Don't you think that relying on a centralized stablecoin contradicts the idea of decentralization at the foundation of MakerDAO? Yeah, so I think first it's important to point out that the entire point of the Maker protocol and the die stablecoin has been to evolve upon the original quite simple decentralized stablecoin model of a single collateralized stablecoin, right? So stablecoins backed by only a single asset and moved from that and then into a type of stablecoin that has many different types of collateral that are all uncorrelated with each other. Because the problem that we actually saw like that everyone, you know, the community in the whole world saw unfold on March the 12th was that there are some serious limitations to having only one type of collateral or even just having only one, you know, sort of a correlated collateral that all behaves the same during a crash, right? So it's always been the plan to add other assets. And it's always been clear that if you want stuff that is uncorrelated from crypto and doesn't all move the same during a crash, you have to move on to real-world assets. You can't just rely on decentralized assets that all have essentially the same business models which is to run a blockchain transaction network. Like the best way to deal with risk is to diversify, right? So, and so the way you would apply that to a decentralized stablecoin like die is that you don't want to just rely on one type of collateral basically, right? So you don't want to rely only on ETH because that means you're too exposed to ETH's movement in the markets and so on, right? And even potentially you're too exposed to the fact that ETH is maybe too reliant on centralized exchanges in the future restrictions. Yeah, basically you're saying that the more diversify is the collateral die depending on the better for the old protocol for the old community. Yeah, exactly. And the way you extend that is so, so if you add USDC, then okay, well now there is one more point. Now there's kind of like an extra point of failure, right? So it's not just Ethereum, now there's also USDC. And given, I mean, and of course, USDC is still it's reliant on custodian, it's exposed to the US legal system. So actually an even better next step would be to then add even more stablecoins, right? So why not also add, you know, three or four other centralized stablecoins so that you don't just add one more point of failure but actually add an extra five points of failure, right? So you even further spread the risk. And then from there it's just going further in that direction, right? So the next, I mean, so the endpoint is thousands of assets, right? It's thousands of real assets that all have different custodians with different security models and that are based in different jurisdictions so that you're not too exposed to one particular political or legal system. I think that the endpoint is very much a form of decentralization, right? We have this massive portfolio of assets that is spread across the world that has no critical reliance on any single entity at all, right? In fact, it can handle a lot of failure. It can handle custodians running away, you know, jurisdictions cracking down and all of these things can be handled actually in the same way that the protocol was able to handle the March 12 crash, right? Which is basically the protocol simply accepts that there is risk, there's always going to be risk and when that risk sets in and you have a loss, the protocol has a way to basically have solved the last recapitalized system and then earned back what was lost during the good times where there isn't a crash. The Black Thursday meltdown was partly caused by the fact that some maker users participating in the liquidation auction managed to buy Ethereum, giving zero die-in returns. These were maker users who preferred to pursue their self-interest against the interest of the community. Don't you think that this is still a vulnerability in the maker system? Well, I think you could say it's a failure of the keeper ecosystem in that it simply wasn't mature enough. There wasn't enough actors there because they're supposed to pursue their self-interest but they're supposed to do so competitively and as a result, you should approach market prices, right? But the problem was that there was no ability for them to compete because most of them were completely went offline basically from just the extreme conditions during the event. And like I said earlier, one of the solutions to this that maker governance immediately put in place was to simply change the duration that an auction has so that there's just more time for more people to participate in a given auction, even though there's always trade-offs with these kind of changes. And I would also expect that in the long run an even more sort of future-proofing solution would be if the community takes a look at the entire auction design and basically redesigns it based on the understanding that these kind of extreme events are going to happen. According to a research from a company called CredMark, the amount of loans that have been given out by centralized platforms like BlockFi, Salesforce, these kind of centralized lenders is still far, far bigger than the amount of loans that are being given out through the centralized platforms like MakerDAO. So the market for centralized lenders is still much bigger than the one occupied by MakerDAO. So why do you think people still prefer to trust in a way centralized entities instead of going for the decentralized path? Well, I think on one hand it's certainly familiarity, right? It's that that's what many people are used to. I would actually expect that over time, once DeFi fully matures and the protocols and platforms fully matures, you would see that what's currently known like the C-Fi and DeFi will actually merge over time. And what's currently known as C-Fi will more become the front-ends and sort of the access points to the various protocols that then allows, you know, that gives people the best possible rates and basically all the different financial options that they want to have access to, right? And that's another of the key aspects of DeFi, right? The composability and the ability to build on top of the protocols that lie underneath and then add whatever features you want, including, you know, making it more user-friendly or adding, I don't know, insurance or what they're just like combining them together in various ways so that it appeals to specific types of end users. And I think overall, the way the ecosystem is shaped right now, it's very well positioned to go, you know, to move in that direction over time and ultimately bridge that gap between the people that prefer the use of friendliness, even if they have to take more custodial risk and use C-Fi right now and then the people that prefer to use decentralized protocols but then obviously the long run would love to have even more user-friendly interfaces as well. The DeFi space is still at its infancy. What are the main milestones that Macadal and DeFi need to achieve in order to mature and onboard the mainstream public? It has basically three core pillars that the foundation has, the Maca Foundation has laid out. So the first one is what we call elected paid contributors. So that is the protocol hiring experts kind of, you know, similar to how the foundation right now hires developers and so on to work with the protocol. In the future, the protocol will itself hire experts to work directly for the protocol and can pay directly for the protocol. And then these experts, these elected paid contributors can cover all the, like all the different day to day operations that are needed. And then the second thing is what's called Maca Improvement Proposals. So Maca Improvement Proposals is a way to codify and structure agreements in the community around how governance should run and how technical upgrades to the protocols should happen. And then finally, there's vote delegates. So a system that allows MPI holders to delegate their MPI voting power to others that may be more active in the governance process and may be more deep into the technicalities. And this will make it a lot easier for regular MPI holders to participate in governance without having to fully understand every single aspect and, you know, consider every single decision instead they can just decide which delegate do they want to attach their voting power to in order to sort of get the change that they would like to see in the system over time. So I think if the community is able to fully integrate these three core pillars into running the Maca protocol for the long run, then the only thing that's in the need from there is just scale and specifically onboarding a lot of tokenized real-world assets into the Maca protocol so that the protocol can reach the scale that it needs to reach and become very diversified, very stable so that it becomes possible for regular people around the world and institutions and small businesses and so on to really trust the stability of the die stable coin at the same level that the fiat currencies are seen today. And maybe even beyond that because you will have that additional advantage of total transparency into exactly what assets are backing the die stable coin and where they're located and what risks they're exposed to. We just pointed out that people are not used to rely on decentralized entity because they kind of, they are more used to and trust their funds to centralized actors. So probably they need some time also to adapt to this new way of thinking. Otherwise it's gonna be very difficult to onboard a massive amount of users. I agree and I also, like I said, I think that a model that will prove to be very popular is this hybrid model where you have centralized entry points into the decentralized finance protocols. So maybe even you will have a custodian that you trust as a user and then that custodian interacts with the DeFi protocols for you so that you still don't have, let's say, you don't have the entire protocol is in custody. It's only your specific custodian that you're exposed to and that allows for these, I guess you can say, kind of like a hybrid approach where you can get some of the advantages of DeFi and you can get some of the advantages of centralized options and maybe when you combine it together then you end up with something that is preferable to a lot of users. That was Rune Christensen, founder at Make a Dow. If you enjoyed the interview, don't forget to like the video and subscribe to our channel. See you next time. Cointelegraph, like, subscribe and hodl.