Risk II: The Game of Stablecoin Domination
Lesson 12: Thinking about Stablecoins & Conceptual Design Risk
This is part two of an article considering stablecoins and risk. You can read part one on counterpary risk here. And if you’re interested in a primer on stablecoins, please go here.
No two stablecoins are created equal; and pressure tests can expose the inherent conceptual design flaws in either the coin itself or its wider, supporting ecosystem.
So let’s consider the key features of design risk in order to suss out whether or not you want to trust any particular stablecoin.
What Is Conceptual Design Risk?
Conceptual design risk proves complex, with many errors only becoming obvious in hindsight. So let’s look at two ways in which we might get a better handle on conceptual design risk when trying to determine the merit of a stablecoin. We can breakdown conceptual design risk into intrinsic risk and extrinsic risk.
Both involve risks built into the design of protocols, coins, ecosystems, or tokenomics structures. The difference lies in how intrinsic risks are actively present by virtue of how the system or item operates with users and smart contracts. Extrinsic risks are those where a conceptual flaw or oversight in the design allows for external manipulation in order to take advantage of the system.
Intrinsic risk
An example of an intrinsic risk can involve liquidity pools which fail to maintain adequate collateralization of the coins being lent. Consider the case of one of ICHI’s liquidity pools:
Between April 12 and 13, 2022 the ICHI token plunged 90% due to a problem with one of its yield pools — Fuse Pool #136. The pool was operative on Rari Capital but owned by the Ichi Foundation.
The problem involved an automatic liquidation of the ICHI token within the pool that was triggered by an over-collateralization/over-leveraging issue which allowed users to borrow an unlimited amount of ICHI at an 85% Loan-to-Value (LVT) ratio.
Fuse pools work by allowing users to create bespoke pools for lending and borrowing with custom parameters — e.g. interest rates, how oracles draw on real-world data, and collateral requirements. Each pool is governed by a LVT ratio; if a user’s supplied assets to the pool results in either an increase of collateral above the LVT or a decrease of collateral below the LVT, the assets are automatically liquidated.
Because the LVT ratio for Pool 136 was 85%, this meant for every 100 ICHI, users could borrow 85 ICHI. This high LVT went unnoticed and caused a liquidity issue.
The over-collateralization/over-leveraging triggered the automated process of liquidation in order to support the pool’s value. Low liquidity of ICHI on DEXes meant the token dropped to $0.
Another example is the collapse of Terra LUNA. Arguably, a conceptual risk can be baked into the idea of a coin — such as, an algorithmic stable coin. Such coins are not collateralized and instead rely on maintaining a peg in relation to another coin or asset. The collapse of Terra LUNA’s UST stablecoin is seen by many as proving how algorithmic stablecoins are intrinsically flawed.
But there’s another view on intrinsic risk in this instance: one can say that the design of UST was not solely to blame. The narrow framing of its conceptual design failed to recognize the higher order effects of a stablecoin within the ecosystem of its base layer protocol. With LUNA, the resulting interaction of flawed “parts” in the LUNA ecosystem thus led to its collapse. In that sense, the coin design itself was not flawed. Rather, the protocol as a whole involved an intrinsic conceptual design risk. The bread and butter of the LUNA ecosystem was the lending protocol Anchor. In addition to offering lenders a high APY on their staking (up to 20%), it also rewarded borrowers, and held as much as 72% of UST. Any problems with UST would therefore have significant effects across the ecosystem.
Such an effect occurred in May 2022, when a significant amount of UST was sold on the CURVE exchange. This caused UST to de-peg, with its value dropping to $0.98. LUNA tried to compensate, but to no avail.
Extrinsic risk
When a coin or ecosystem invites external manipulation, that potential equates to an external risk. An example of an extrinsic risk can involve lax security measures that can be exploited by malicious actors (though arguably this also involves an intrinsic risk relating to the design of an ecosystem’s security). Cream Finance lost over $100 million in October of 2021 when a bad actor was able to exploit the flash loan system, which allows for uncapped loan amounts.
Another kind of extrinsic risk is one in which a separately owned platform or system one uses or relies on is vulnerable. Babylon Finance operated a liquidity pool at Rari, which was hacked in April 2022. The total loss in the Rari hack was $80 million, $3.4 million of which was Babylon’s. According to Decrypt, this loss
cratered the project’s [Babyon’s] total value locked (TVL) and cost the [Babylon] team three months of operating costs.
The hack also meant that Babylon could no longer use Rari’s Fuse pools, which let users create lending and borrowing pools for just about any type of asset. As such, BABL, Babylon’s native token, could no longer be used as collateral to borrow funds.
This external event thus led to the demise of Babylon Finance, which will be shut down as of November 2022.
An Easy Way to Test for Conceptual Design Risk
One way to test for conceptual design risk is to address the way that feedback loops affect the stability of the stablecoin, both on their own and within the ecosystem. Surprisingly, there is a relatively simple test to employ in assessing this risk. Ask only:
What would happen if every holder of the stable coin were to wake one morning and decide to redeem their currencies? Could the coin survive?
Think “bank run”!
A narrow framing of conceptual design might miss the higher order effects of a stablecoin within the ecosystem of its base layer protocol. So thinking through the thought experiment means following how the economic behaviors of fear and panic affect the various platforms on which a stablecoin resides and what mechanisms are in place to ensure solvency and perhaps mitigate negative sentiment. Tether survived its first significant stress test (July 2022) after the LUNA collapse by virtue of being solvent.
To see an historical example of failure due to design, let’s compare LUNA-UST and IRON-TITAN.
Some have likened the LUNA-UST collapse to the IRON and TITAN collapse, but the cases are significantly distinct. Unlike IRON, UST was partially collateralized by Bitcoin and Avalanche at about 20%. It was also supported through liquidity exchange pools on decentralized protocols which served as “gyroscopes” to maintain a peg to the US dollar. The coin nevertheless collapsed because the liquidity exchange pools were still too new, and so too shallow, to function properly. Moreover, as we saw above, the native blockchain (LUNA) supported a lending protocol (Anchor), which acted as an accelerant in the bank-run feedback loop. When investors initially lost confidence in LUNA, the stabilization pools proved feckless and the collateral on Anchor incentivized participants to remove capital even faster than they would otherwise. The resulting interaction of flawed “parts” in the LUNA ecosystem thus led to the collapse of the coin, not the coin design itself.
What You Can Use to Measure Conceptual Design Risk
The main factor is solvency due to adequate collateralization, which involves both the nature of assets backing the coin and the time it takes a customer to exchange the coin for the assets.
Let’s take the temporal aspect first. At the present moment, no stablecoin allows for the instantaneous redemption of fiat currencies (e.g., US dollars). Most stablecoins will take some time to unwind, should everyone so desire to exchange their stablecoins for dollars. Holding that point in mind, a good measure of design risk with respect to time (with the best insurance against risk at the top):
- Everyone could redeem in seven days or fewer.
- Everyone could redeem in one month.
- Everyone could redeem in six months.
- Everyone could redeem in a year.
- There is no scenario in which everyone can redeem, which would make the stablecoin a form of Ponzi scheme.
With respect to asset backing, we can look at collateralization levels. The higher the collateralization levels of a given coin, the safer it is. Some coins, after all, have more collateral on hand than stablecoins issued; thus qualifying them as overcollateralized — and thus safer coins, all else remaining equal. It is important, in this regard, to know what “equal” means. What kinds of assets are backing a coin?
Not all assets are equal. For example, Tether has changed its game after criticism. It was originally backed by a combination of US dollars and debt instruments. The debt instruments were commercial papers, which can involve default risks (they are also not registered with the SEC since they mature in 270 days or less). In other words, Tether was backing USDT with a form of asset that itself might have problems being “called in” when needing the money.
And then, there are those stablecoins that are backed by other cryptocurrencies. Because the crypto market is itself volatile, one has to exercise caution when trusting coins backed by other cryptos. RAI, perhaps has the best method for doing this since it is backed by ETH on a reflexive (non-pegged) basis.
A Note about Empirical Risk
Empirical analysis examines the actual historical performance of the coin.
- How well has it in fact maintained its peg to the US dollar over some length of time?
Most coins have been stressed in 2022, both by the UST collapse and by the broader market decline. Because empirical risk, approached in this way, is evidenced through directly measurable market performance, it serves to counterbalance many of the largely qualitative assessments of the other measures.
Empirical data about coins can be easily obtain via CoinMarketCap, Coin Gecko, as well as other crypto blog and analytics sites.
How This Can Be Applied
It’s rather straightforward, but time-consuming, if you want to do your own due diligence on stablecoins. I’d like to say that if the coin is widely used, you’ll be ok. But the rapid demise of UST proves otherwise. USDT and USDC remain the top traded stablecoins with commitments to asset backing. Another point not discussed in this article but certainly worth considering is whether future SEC regulation and investigation might negatively affect a specific stablecoin.
This article is a part of the Crypto Industry Essentials educational program presented by The Art of the Bubble.
Though this article is credited to me, it contains some written material by Sebastian Purcell, PhD from his The Art of the Bubble education series on cryptocurrencies.
If you found this helpful, Subscribe to The Art of the Bubble’s free newsletter.
Join us on Discord for live chat and daily updates.
