The crypto market was back in the headlines in early May as the cryptocurrency stablecoin TerraUSD (UST) broke its peg to the U.S. dollar. UST was an algorithmic stablecoin of the Terra blockchain, which used arbitrage with its sister cryptocurrency, Luna, to maintain UST’s value. The collapse of both led to a loss of nearly $60 billion in market capitalization from the two coins.
In this Kenan Insight, we recap a recent crypto conversation on this colossal collapse that Kenan Institute Chief Economist Gerald Cohen had with University of California-Berkeley Haas School of Business Professor Christine Parlour, UNC Kenan-Flagler Business School Professor Eric Ghysels and Paxos Chief Revenue Officer Michael Coscetta. These experts in decentralized finance explain the purpose of stablecoins in cryptocurrency as well as how one like TerraUSD, created to maintain its value equal to the U.S. dollar, could have failed so dramatically. The discussion has been edited for length and clarity.
Gerald Cohen: What happened to TerraUSD?
Michael Coscetta: What you saw from a very high level is a stablecoin that was based on an algorithm that was pegged to a basket of underlying investments, and these assets were not stable. Our CEO [at Paxos] was quoted in in The Wall Street Journal two weeks ago [prior to the collapse] calling out TerraUSD, saying that that does not look like a stablecoin but rather looks like an unstable coin. It’s a clear example of how there are three worlds of stablecoins, depending on how you draw the line: some that are backed by true cash, some that are backed by collateral that could include crypto or other variable value assets and some that are a little bit of alchemy – which, to me, is what the algorithmic stablecoin world looks like. These coins work really well when things are going well and when the algorithm is based on a set of underlying assumptions that hold. But when those assumptions start to shift, very few algorithms are able to keep pace. Needless to say, it was just a pure example of a stablecoin not actually being a stablecoin.
Gerald Cohen: You said there are three types: the algorithmic, the cash-backed and then some combination. So at Paxos, when you say you’re cash-backed, what does that mean exactly?
Michael Coscetta: So USDP, which is our stablecoin, is backed 100% by cash and cash equivalents, which we just determine as short-term U.S. Treasurys, which the U.S. government calls cash equivalent. There’s no underlying investment-grade securities and there is no commercial paper, there’s no crypto backing. Again, those reserves are cash. So when people put a dollar in, they get a dollar USDP, and if they want a dollar out, they can get that instantly as well. And this is regulated by the New York Department of Financial Services. Our reports are audited and published, and I think you’ll see there’s a big difference in the world between being a regulated, registered stablecoin and having true cash reserves that are audited versus attestations that are signed by just the CEO of that company. The second category of stablecoin is what you can call an overcollateralized or crypto-backed stablecoin, where the reserves are built on this basket of different cryptocurrencies. Typically, these are overcollateralized to allow for some deviation in value. Typical to what you would do [for] a margin account. But you know in this case it doesn’t mean the collateral can’t at some point be insufficient, depending on how wide the swings are in the market. Crypto markets swing a lot. The third category are the algorithmic coins, where the value is pegged to a basket of variable quantities, variable ratios of different underlying assets. And when the value of something goes down, generally the quantity of something else comes in to replace it. And that’s where you saw this massive devaluation of Luna because you saw that that basket flooded with additional coins to balance out the drop in the underlying asset price.
Eric Ghysels: This is reminiscent of the day after Lehman went bankrupt, when there was a run on money markets because they were holding commercial paper issued by Lehman Brothers. And it was anybody’s guess what the value of that was. And so, the mechanics are different, the technology is different, but we’re basically talking about bank runs. And we’ve seen bank runs in all sorts of variations throughout history. So I just gave the example of the day after Lehman as de facto a bank run on money market funds. If you look at the U.S. monetary history, there’s this fabulous book by Friedman and Schwartz, “A Monetary History of the United States,” that goes through, in detail, everything that happened. Particularly interesting is the period after the end of the Civil War and before the creation of the Fed, which is roughly between 1863, when the Office of the Comptroller of the Currency was created, as a sort of light touch regulation, to 1913 when the Fed was created, and then the FDIC later in 1933. So in the decades between the end of the Civil War and the creation of the Fed, we had pretty much every decade a major financial crisis with, physically, people running to the bank. We were on the gold standard. Banks were supposed to hold gold. That was what they were supposed to have in their vaults. And they could print their own dollars. But, of course, the underlying assets were in some cases dubious. And so you had the physical bank runs, and we have the same phenomenon. We’ve seen it many times. It’s just under a new technological form.
Christine Parlour: We have stablecoins because we want to pay for stuff online or basically in the digital universe. And why do people accept means of payment? People accept means of payment because they think that they can turn around and then use that means of payment to get other stuff that they want. So basically the ability to use something as a means of payment is really, really important for its long-term value. And if people’s beliefs about whether or not the thing that you’re using to pay ever get shifted, then essentially nobody will accept that thing and they won’t accept it in like a two-second interval where their beliefs shift and suddenly it’s worth nothing. That’s why the ability to redeem for something else that we accept, like U.S. dollars, is just so important. And I think what the UST situation told us was that you can have a very large and, up to the point of failure, successful-looking payment rail [payment platform], but the market is bigger than any particular company and any particular enterprise. Unless there’s a redemption that stops people’s doubts, you’re going to have problems with any kind of payment rail that isn’t robust to changes in beliefs.
Eric Ghysels: I think stablecoins exist because we want to have a currency that lives on a blockchain in order to basically trade with other assets in this sphere. I think transparency is the big issue. It has been the big issue for any of the bank runs, and I think that is where we have to think about what are the most solid foundations for a stablecoin. Stablecoins will have to be in existence. And I don’t think necessarily central banks should be the providers of stablecoins. There’s a lot of discussion about that in terms of central bank digital currencies. I don’t necessarily want to go into that direction. I think where I would like to see things is, yes, we do need a blockchain-based money, basically, that keeps its value. If you define what money is, it’s a store of value, it’s a unit of account and it’s a medium of exchange. Those are the three features that we have for money, essentially.
Michael Coscetta: I think that transparency is really one of the first steps forward to getting mainstream adoption and then also some stability into the crypto world broadly. You nailed it about what is money supposed to be. The definitions have been clear for a long time, and I think stablecoins are trying to be that but to be it in a more efficient, faster, cheaper way, because the current banking system, the current monetary rails are atrociously old, they’re archaic, they’re slow, they’re ripe for error, for failure. You’ve got banking systems running on batch processes, on code that was written closer to the bicentennial than it was the millennium. And blockchain is an attempt to re-platform the movement of money globally to do what global economies are meant to be, which is to transact instantly, securely and for free. But to do that, a stablecoin has to act like a stable asset in a stable store of value and a stable means of transfer. And what you’re seeing now is there’s a proliferation of a lot of different assets that are clearly not stablecoins. There are also cryptocurrencies that are not stable. I think to lump all crypto together is really dangerous, just like to lump all stablecoins together is dangerous until someone draws a very clean line. And I’m hoping the regulators do that. I’m hoping the regulators come in with very clean set of circumstances and rules and guidelines. It’s very hard to play a game when you don’t know the boundaries and you don’t know the rules of the game. When you do, it’s a much easier place to play, and it’s a much cleaner place to build the strategy and to bring more participants in and I think yields both that transparency and the stability that the industry is craving right now. And I don’t see it coming fast. Congress can’t agree on what day it is, let alone write clean legislation that becomes enacted and signed. I don’t think the central banks are ready to issue a digital currency.
Eric Ghysels: I wrote a paper that was published two years ago with the title “Back to the Future.” The exercise was backtesting systemic risk measures. So after the financial crisis, after the Great Recession, we had all these conversations about systemic risk and a lot of measures were proposed, and they’re being used by central banks to actually monitor the banking system. And not all these measures can be computed based on historical data, because we don’t really have that rich historical data, but what we did in this paper — and this is with historians Ben Shabot from the Chicago Fed and Chris Kurz from the Federal Reserve — we went back in time and we had historical data collected and we did the exercise: What if a central bank had access to systemic risk data? Would it have been able to flag the kind of bank runs that we saw at the time, the collapse of the railroads in 1873 and all the financial consequences of that? The Barings crisis, J.P. Morgan bailing out the U.S. economy, etc.? The answer to that was yes. So I think the fact here is that there are tools to basically target keeping a diligent eye on these stablecoins if we want to, and look at what the underlying assets are very much like we do in in other parts of the financial system.
Gerald Cohen: Given the need for transparency, what’s the appropriate policy response?
Eric Ghysels: Right now, actually, the exchanges are kind of the gatekeepers. They are the ones who list the stablecoins, and listing a stablecoin creates its traction. So currently in some sense, the Binance, the Coinbase, etc., decide on which are the credible quote-unquote stablecoins and which are not, and they do the due diligence process. I’m not trying to argue that they’re not doing that, but it has failed. TerraUSD is a good example of that. So I think it requires what we would call tier-one assets and tier two. We have these concepts of what are safe assets. They’re well defined in the financial sector. Something along those lines has to be done in terms of the backing of stablecoins.
Michael Coscetta: I agree with Eric. I don’t want to speak for our legal team and our compliance team, but you generally want a few specific things. Number one is you need to have a definition, a set of clean definitions that everyone agrees on. You have similar concepts in the liquid money supply today. There should be similar concepts applied to the digital money supply because eventually you have two different money supplies if you’re not careful. The second is a means of reserve, a means of auditing those reserves and a means of understanding and having transparent access to those reserves 24/7, although that’s even better today than what the Fed has. So there’s a technology improvement that also needs to happen here for the Federal Reserve. The third is it would be great if there weren’t three or four or five different federal agencies all fighting to be the one that regulates stablecoins, because I don’t think that’s going to give us any movement anytime soon. It’s just going to be a really strong tug of war with zero displacement and progress made. The fourth is if a company wants to operate a coin that’s not stable, that’s OK. It just can’t be called a stablecoin.
Gerald Cohen: Doesn’t government regulation defeat the original libertarian intent of crypto?
Michael Coscetta: If you want a regulated business to be in the world of crypto, well, they’re not going to be able to play the game of an unregulated product. They’re only going to be able to work with regulated partners. So there is a paradox here of playing in the existing world while simultaneously building technology and building something that scales into the future for what could be a very, very different world. So is it antithetical? I don’t necessarily think so. But again, I think a stablecoin is meant to look and act like a dollar today, not to look like Bitcoin. And Bitcoin is meant to not look and act like a dollar today.
Christine Parlour: I think it’s also useful just to think about how the digital asset space has grown. If it’s true believers who understand the technology and are willing to play in the space, that’s one thing. But if it is a large swath of retail investors for whom these assets represent a significant portion of their income, let’s just say, the government’s mandate is to look after these people along some dimension. And so immediately you get calls for regulation. And I think if the crypto world wants to grow, it wants to get those retail investors in. But at that point, basically, the door is open for regulation.
Eric Ghysels: I completely agree with Christine, but it actually goes beyond that. I think a lot of the traditional actors in the financial sector are staying on the sidelines because they are nervous about the fact that there is no clarity about the regulatory environment for this digital crypto, blockchain-based financial sector. So it’s not only the retail investors — that’s important — but there is a large swath of financial institutions that are adjusting out of it for the moment. And the crypto industry would benefit greatly from having a more solid regulatory environment that would create a more stable universe, not only stable coins.
Christine Parlour: The idea behind the Terra blockchain was absolutely great. It was a low-cost payment rail that could be widely adopted that would basically allow people to get around the sort of inefficiencies in the correspondent banking system. Obviously, there were implementation issues. But I think that as a society, this is something that we need and something that we’ve got to go to. Exactly how we’re going to get there, I don’t know. But definitely the movement is towards innovation in this area. And it’s time.