Algorithms Are Unreliable; Fiat-Backed is the Answer ~Part1
"You’ve got to sort of think, if you’re an investor, if it’s too good to be true, maybe it is."
There’s only one true test of a stablecoin’s legitimacy: how well does it maintain its peg?
From the start, the goal of stablecoins was to import some of the reliability of fiat currencies into the crypto universe. For many corporate and institutional investors, a key obstacle to embracing crypto has long been its volatility; the value of Bitcoin, for instance, has over the years shown a nasty habit of shifting 20–30 percent in a matter of days.
But if a crypto token’s worth could be tied to the value of, say, greenbacks, its stability could be somewhat ensured, easing investor concerns and potentially opening the floodgates. And so it has come to pass: investment in stablecoins leaped five-fold over a recent 18-month stretch, from $27B in December 2020 to over $140B in August 2022, months after the collapse of Luna kickstarted the so-called crypto winter.
Despite that sizable speed bump, stablecoins are still well positioned to be the main crypto entry point for major investors, lending real urgency to the question of which of the three main types of stablecoin — algorithmic, crypto-collateralized, or fiat-backed — is the most stable. The available evidence increasingly suggests the primary cause of Luna’s fall was not the fragility of the crypto marketplace or the Terra ecosystem, as many have argued, but rather the inherent weakness of the algorithmic foundation.
Let’s set aside all the early algorithmic stablecoins that failed, like Basis Cash, D-Dollar and Iron Finance, and go straight to early 2022, when the top three stablecoins in terms of market cap were Tether’s USDT, Circle’s USDC and Terra’s UST, in that order. Only the last of them was algorithmic, and that, of course, is the one that tumbled in May.
UST had maintained its dollar peg via a link to Luna, Terra’s more established token, and a network of arbitrageurs who bought and sold the coin to profit from price differences and balance liquidity pools. Apparently, after a number of whales exchanged their sizable holdings in UST for other stablecoins, the market lost confidence in UST’s algorithm, sparking the death spiral that shook crypto.
But the details of what triggered the collapse are likely less important than a weakness at the root of the concept. In their purest form, algorithmic stablecoins are wholly uncollateralized. Their value is not backed by any other asset, but by an algorithm, a set of digital instructions that mostly aim to incentivize user involvement and ensure a balance in liquidity pools. Some mint or burn coins to correct a proportional deviation from its peg. Another approach, used by UST, is to tie the ASC to a more established crypto token to facilitate stability.
Most of today’s algorithmic stablecoins are not purely so, but instead incorporate some form of collateralization, which might involve derivatives, more established crypto tokens, or some other form of collateral. But the market is coming to grips with the notion that the underlying foundation is still little more than thin air. Reeve Collins, co-founder of NFT platform BLOCKv, thinks UST’s collapse suggests algorithmic stablecoins are unsustainable. “That crash kind of had a cascade effect,” he told CNBC in June, “and it will probably be the end of most algo stablecoins.”
The market’s now so sensitive that the cascade is playing out in real-time. Within days of Luna’s tumble, at least two other algorithmic stablecoins depegged. First, Waves’ USDN, which has a design much like that of UST, lost its peg for the second time in just over a month and fell into the mid-70-cents range. As of early September, it had lost a third of its market cap, which had peaked above $1 billion.
Also in May, collateralized algorithmic stablecoin DEI lost its peg and was soon trading under 60 cents. DEI has since regained its peg, but its market capitalization is less than a third of what it was at its peak. The transactional reasons for these two depegs are different — after flash loan attacks, DEI saw its collateral ratio fall so far as to make redemptions almost impossible, while Waves had used leverage to artificially boost the value of USDN — but in both cases, as with UST, lost investor confidence spurred widespread departures.
These were, in essence, bank runs, as outlined in a recent post. So it was no surprise when investors began shorting algorithmic stablecoins. In June this led to a depeg for Tron’s USDD, which fell to about 93 cents just 40 days after its much-hyped launch. Next up was Acala USD, or aUSD, a decentralized, multi-collateral stablecoin that’s widely considered an algorithmic stablecoin, though Acala seems to reject that label.
Either way, aUSD lost its peg in mid-August and fell to one cent after hackers exploited an error in its code to mint more than $3 billion in coins. The Acala team recovered and then burned most of the $3 billion, pushing aUSD back up over 90 cents. But as of early September aUSD was trading at about 65 cents and the damage had been done. “Once again, it shows that the algorithmic stablecoin,” says Chris Terry, BPSAA board member and VP of enterprise solutions at SmartFi, “is vulnerable to attack.”
Last but surely not least is Tribe’s Fei USD, or just FEI. After weeks of volatility in FEI’s price, Tribe announced in late August a plan to dissolve the DAO that oversees the algorithmic stablecoin and burn the remaining protocol-owned coins. FEI lost its peg two days later, falling to just over 97 cents, and as of September 1st, its market cap of $142M, much like HUSD, is about one-fourth of its peak. This appears to be the first time the founders of an algorithmic stablecoin that had yet to lose its peg took a look at the market reality and chose, essentially, to destroy their creation. And the fact that the value of Tribe’s main token increased 65 percent following the move suggests it won’t be the last.
While basing the value of a currency on bits of computer code is bold, it’s also an idea about which the market has become terribly dubious. Crypto analyst Blocmates has embraced a dim view of algorithmic stablecoins. “In general, they will all attempt to maintain their peg through code as opposed to being fully/over-collateralized,” the site explained in July. “Yes, it is as stupid as it sounds and has resulted in death 100% of the time.”
More and more observers appreciate that most algorithmic stablecoins are issued with informal collateralization that has little underlying value, which is precisely why a little bit of fear fast becomes a raging pandemic. This also explains why algorithmic stablecoin APY’s go up to 20 percent — greater risk necessitates greater incentive to invest. “You’ve got to sort of think, if you’re an investor, if it’s too good to be true, maybe it is,” SEC chief Gary Gensler said in July.
Do you like this article? There are more to come!
Follow us and Subscribe to stay updated on what’s new.
A few weeks prior, Waves founder Sasha Ivanov said that moving past the depeg would require a refining of USDN’s algorithm. Waves announced a complicated series of steps to strengthen the stability of USDN, including liquidating large accounts, launching a new recapitalization token, and capping trades in USDC and USDT. These moves appear to have fallen short: in the last week of August, USDN again lost its peg, falling to 91 cents as investors fled.
Tron’s USDD has been one of the few to escape the death spiral of market doubt. USDD recovered its dollar peg within about two weeks of its June depeg and has since slightly increased its market cap. This is likely due to USDD’s massive over-collateralization: in late August Tron said it held $2.29B in collateral in TRX, BTC, USDT, and USDC to support $730M in USDD. This would be about 315% collateralization, but it’s not entirely accurate, as Tron seems to be fudging the numbers by including $725M worth of TRX set to be burnt.
Without this, the true total collateral is about $1.62B, putting USDD’s over-collateralization at around 222%. Tron would say this provides a cushion against market volatility, which is not untrue. But at the same time, it also underscores the cost and complexity of the algorithmic balancing act.
It’s getting hard to find a living, breathing algorithmic stablecoin that’s not pushing the envelope or straining the truth in one way or another. Take MakerDao’s DAI, which is backed by $10.6B in mostly crypto assets, led by USDC, and overcollateralized at 141%. As of early September its collateralization rate is even higher, at 170%. This means you’d have to hand over $170,000 worth of Ethereum to borrow $100,000 in DAI, enabling you to access the higher yields of certain lending platforms.
MakerDao and its backers like to promote DAI, the world’s fourth-biggest stablecoin in terms of market cap, as decentralized and controlled by the people. But this could be seen as false advertising, as more than half its assets are in USDC. “It is problematic when the foundation of a so-called decentralized financial revolution is so reliant on collateral that’s the liability of a central issuer,” Bitcoin Magazine asserted. In part because of its links to fiat-backed USDC, MakerDao’s founder recently suggested that DAI might soon depeg from the dollar.
Another algorithmic stablecoin often described as more reliable is UXD, mainly because it’s fully collateralized and relies on derivatives. But even this supposed exemplar of the concept briefly lost its peg in June, falling to just above 98 cents, and has seen its market cap decline by more than half since May.
In the next article, we will move on to discuss the challenges that fiat-backed stablecoins face and look into why fiat-backed is the answer. If you like our article, don’t forget to Follow us and Subscribe to stay updated with our latest blog post. Stay tuned!
About GMO-Z.com Trust Company
Building Financial-Grade Digital Assets. The World’s First Regulated JPY-Pegged Stablecoin Issuer. Visit our website to learn more.
DisclaimerThis content is not financial advice and it is not a recommendation to buy or sell any financial instruments, FX trading, cryptocurrency or engage in any trading or other activities. You must not rely on this content for any financial decisions. Acquiring, trading, and otherwise transacting with financial instruments or cryptocurrency involves significant risks.We strongly advise our readers to conduct their own independent research before engaging in any such activities.GMO Trust does not guarantee or imply that any cryptocurrency or activity described in this content is available or legal in any specific reader’s location. It is the reader’s responsibility to know the applicable laws in their country.