Chasing stable windmills, Part 3
Over-collateralization with cryptocurrency reserves
How does that work?
Let’s call this coin STBO. Here, there is no fiat collateral. Instead, there are collaterals in some other asset, for example a cryptocurrency like Ethereum. You could probably describe this as the system using ETH as a kind of buffer between STBO and USD. However, ETH has its own life and price in USD, which is not static. How can the price of STBO remain stable in USD, if there are no USD to back it up?
Let’s see an example. A long time ago, in a galaxy far far away, the price of 1 ETH was 1 USD. So, in order to buy 1 STBO, I would buy 1 ETH with 1 USD, give it to the STBO issuer and get 1 STBO back. Remember that, although I now use ETH to buy STBO, the stability of the STBO price is still considered in USD.
A month passes, there are some strange hacks in some Ethereum smart contracts, and now the price of ETH is down to 0.5 USD. I now want some USD, so I decide to sell my STBO for ETH and then exchange that amount of ETH for USD. How much ETH should I get back if STBO is to keep its promise? That’s easy, 2 ETH. But wait, if I originally gave 1 ETH for 1 STBO, and now I get back 2 ETH for that same 1 STBO, who provided the extra ETH? The issuer? I don’t think so. Well, you guessed it — I did. When I initially bought 1 STBO, I actually deposited 2 ETH, effectively getting back 1 STBO and locking 1 ETH as an extra collateral, that would absorb the price fluctuations of Ethereum. And what happens if the price of ETH drops even more? Then the collaterals are not enough, so STBO is automatically liquidated, returning its owner the 2 ETH — whether he likes it or not.
Why would I want to do that though, why would I get into a system that makes me lock extra ETH like that?
Let’s roll back the clock. As before, I have bought 1 STBO and locked 2 ETH, but I now see that Ethereum is about to introduce novel scalability updates that nobody else has noticed yet. I immediately use my 1 STBO to buy 1 ETH. A month passes, Ethereum delivers their updates, its price skyrockets to 2 USD, so I decide to get out of the STBO ecosystem. Remember, I only have 1 ETH at hand. So, I sell 0.5 of my ETH, buy that 1 STBO back, give it to the STBO issuer and get my 2 ETH in return, resulting in a total of 2.5 ETH. Hurray, I just made 0.5 ETH.
Is it useful?
Again with the pressing questions. Who would issue such cryptocurrency, why and who would want to use it?
Compared to fiat-collateralized stablecoins, we have now managed to achieve some level of decentralization. Specifically, the issuance of the coins is decentralized, as well as the parameters of the system, even though the price oracle is still centralized, as we will see next.
Is that enough for a regular user to use it? Remember, she has to buy ETH and lock a larger amount than what she gets (double in the example above). Even more so, she has to accept the risk that, if the price of ETH drops a lot, her STBO may be liquidated and she will be left with only ETH at hand. Would a business accept it instead of USD? Considering the risk associated with it, probably not, even with the (partially) decentralized features.
Would somebody use it as leverage though, to buy even more ETH, like described above? Sure, but as we see next, this is not sustainable in the long term.
Is it technically feasible?
The interesting aspect of this type of stablecoins is that the issuer does not (necessarily) need to be a central authority. Instead, it can simply be a smart contract running on the network of the cryptocurrency that secures it — in the example above, Ethereum. That way, we avoid all of the problems that stem from centralization.
Even in a case of a smart contract though, can the system be fully decentralized? A core parameter is the ability of the smart contract to know the price of ETH in USD at each moment. This is achieved by asking a so-called “oracle”. This oracle is usually either an exchange or a service that calculates the aggregated price of ETH from various exchanges, which of course means that it is centralized. Given that the whole system is dependent on this oracle though, the centralization problems come back with a vengeance.
Could the oracle be decentralized? At the moment, I can’t see how. Given that the price is considered in USD and USD cannot exist in a decentralized setting, there must always exist a bridge between the worlds of STBO and USD. Whether this bridge consists of a single entity or multiple organizations that spread across the world is another question, but a level of centralization will always be present. Metrics like CPI are even worse in that aspect, since they are even more dependent of decisions such as which items to include in the index and how to access their market prices.
The fate of STBO is also inherently interlocked with that of ETH. If Ethereum collapses, so does STBO. In the example above, we assume that the price of ETH will not drop below 50%. If it did, then our 1 extra ETH of collateral would obviously not be enough to cover for that price drop. So, we must assume that the price of ETH itself will remain within a margin. The bigger the margin, the more stable STBO will be, but also the more ETH need to be locked as collaterals.
Is a 50% margin enough? A quick look at the price chart of Ethereum, or in fact any cryptocurrency, quickly answers that question. And what would happen in the above example, if the price of ETH plummets? Well, the 1 STBO that I had bought would be liquidated the moment that the price of ETH drops by 50%. So, in the scenario when I reinvested it, I keep the 1 ETH and the investor that bought the STBO gets the 2 ETH of collateral. In other words, at that moment, I lose 1 ETH, the investor breaks even, and STBO ceases to exist. So much for stability.
We can always create an STBO with an adaptive margin though, right? Sure, after all the margin is a parameter of the STBO smart contract. All we have to do is decide when that parameter changes; and who chooses the new value of it; and how fast can these updates be applied; and how flexible the system is when the rainy days come. Again, a quick look at the history of governance in cryptocurrencies probably answers the question that is on your mind right now.
What if we let every user choose the margin they want? Well, in that case each STBO takes a life of its own, since each is collateralized by a different amount of ETH. Now the buyer of an STBO will look at how much collateral the STBO has and possibly reject it, if it has too little and is too risky. In the end, we will have effectively created a single market for each margin value, which is equivalent to creating multiple STBOs with different margin values, even though they would all be handled by the same smart contract.
Is it economically sustainable?
As we saw, the main reason that somebody would use STBO is as leverage, in order to buy even more ETH. And somebody would want to buy more ETH if she believes that the price of ETH will rise in the future, without dropping under the margin in the meantime. Can this work for long?
Let’s see again how the system works. I have 1 USD, buy 2 ETH, use them to buy 1 STBO and then sell that for 1 ETH. So, as long as there is a belief that the price of ETH will rise, more people follow my lead, more STBOs are issued, and all is good in the world. Could the price of STBO fall below 1 USD, because there are not enough buyers for all of us? Yes, that is the first bad scenario, which is entirely plausible and breaks the price stability. But, as we will see by the end of this article, there are even more sinister forces at work.
A new day comes when I have 1 STBO and the price of ETH is 2 USD. For some reason though, I now believe that the price of ETH is about to drop hard, so I decide to sell my STBO for USD. Notice that I don’t want to redeem it, because in that case I would get back ETH, not USD, and I believe that ETH is about to crash. What happens if many people share my belief? We all try to get rid of STBOs and get USD as quickly as possible, minimizing the loss. So, STBO starts trading at a price less than 1 USD, breaking the peg. Without the peg, things get worse as more people start to believe that something is wrong, so they also try to sell their STBO and thus the system goes up in flames.
But wait, if the price of ETH doesn’t actually crash, wouldn’t there be people that buy STBO, as they start believing that the slump will pass? Yes, but in the meantime panic has sent the STBO price below 1 USD at best, making “price stability” a distant dream, or to plain 0 at worst. Wouldn’t the market absorb the panic though? It could, as long as it never gets widespread, or there is an investor with very deep pockets that wants STBO to succeed really, really bad and buys our positions, trying to absorb the market panic himself.
History, cruel as it is though, has shown that again and again there come days when panic overcomes even the most fearless. It can start by something as simple as a sudden drop of, say, 10% on the price of the underlying asset (ETH) and soon it spreads like wildfire. A system that does not have a guardian angel with deep enough pockets would then simply evaporate. On the other hand, a system with such a guardian angel cannot maintain its “open” and “decentralized” facade for long either.
We jump to a different day now, when the price of ETH is again 2 USD. However, autumn has come or, to use its true name, pumpkin spice latte season. Happy with my earnings, I decide to cash out by selling my 1 ETH for 2 STBO, in order to buy that latte that we have come to know and love (yes, this is a world where the latte costs 2 USD and I skip leg day, don’t judge). It seems though that many people share my taste. No problem, there is always somebody willing to sell right? Well, it seems I forgot that we live in a market where manipulation is allowed, if not endorsed. What a heartless, cunning manipulator has done is buy a very large quantity of STBO when I, and others like me, were selling them off for 1 USD. Now that we want them back, he decides to keep them as “ransom”. You probably guess the result. Craving for that latte, I agree to buy the STBO back for a little more than 1 USD, let’s say 1.05 USD, which is actually good enough since I still make 0.95 USD on my initial investment. And thus, subtly, the peg has once again been broken, making me pay a “fee” of 5% for exiting the whole scheme. Could this have been avoided because the market is large enough and no whale can play such games? I will leave this as an exercise.
So, all in all, a system like that is expected to trade around 1 USD, during the good days, or simply evaporate on its first bad day.