USD crashes the crypto party
Hello Everyone,
Welcome back! In this post we will talk about stablecoins, which aspire to be the most boring assets in crypto [spoiler alert, they are anything but].
Their purpose is to provide and island of stability among the highly volatile ocean that is the rest of the crypto space. And yet their name has proven to be deceptive, and the failure of one algorithmic stablecoin in particular wiped out 18 billion of crypto assets.
Their favourite asset to track? Well, it is USD, of course. Not only does the mighty US dollar dominate the stablecoin space, but it may grow to dominate the entire also the crypto industry itself. The [warning] signs are all here.
Three of the top ten crypto coins selected by market cap are USD stablecoins. Notably, these are also the only coins in the top ten to not represent a native blockchain token [ahem, with the notable exception for DOGECOIN, but we don’t talk about DOGECOIN].
Source: https://coinmarketcap.com (Sep 2022)
What would Satoshi say? The libertarian argument for Bitcoin was to give humanity a store-of-value currency that is outside of the reach of nation states. This was to protect us all from government tendencies to erode our purchasing power by excessive money printing [if you have lived through hyperinflation, this is an appealing argument].
And yet, here we are with the world’s most influential fiat [government-issued] currency being extremely popular on blockchains. How should we interpret this?
Many countries face inflation rates worse than those in the US. Some of these countries coincidently also use capital controls to prevent their citizens from purchasing dollars in their quest to avoid local inflation. Sure, those people can purchase native blockchain assets but these are more volatile that the USD, and are prone to large drawdowns relative to it. Therefore, flocking to the safety at the US dollar, is quite rational.
Others simply enjoy the near-instantaneous transaction settlement which blockchains offer, making transfers trivial to execute. That is unless you mess up a transfer or get scammed, in which case you will quickly discover the drawbacks of not having a trusted intermediary to reverse transactions.
Third, it doesn't hurt that governments cannot block crypto wallets as they can bank accounts. Sure, such drastic actions don’t happen often in democratic countries, but they do happen [insert Canadian truckers example here]. The more digital our world becomes, the higher the pain inflicted on individuals who end up being un-banked, and the more valuable permission-less alternatives to store wealth become.
Naturally, the rise of the DeFi ecosystem has led to the need for stable assets for use in those protocols. Plus it also hasn’t hurt that some protocols decided to attract users by offering high stablecoin interest rates while traditional bank accounts were offering 0.1% [yes, it was a trap].
So, there are plenty of good reasons to hold USD on the blockchain, whether crypto purist like the idea or not. Together with NFTs stablecoins have become one of the “killer apps” of blockchains.
But can these stablecoins meet existing demand and live up to their core promise of being, well, stable? Bitter experience has proven that this may not always be straightforward.
Let’s dive in and find out how hard it can be build a stablecoin.
What are stablecoins?
Stablecoins are cryptocurrencies designed to track the price of another asset, usually in the “real” non-blockchain world. The majority have chosen to peg their value to the US dollar.
Creating a stabelcoin sounds simple. Just buy real dollars to back up the ones you issue on the blockchain, right? So you can imagine how frustrated people can be when a stablecoin fails to live up to this name and de-pegs.
In their defense, this is crypto, and a stable peg is not truly the only goal. An important reason to build on the blockchain is to run permission-less applications. When you hold assets off-chain to back up an on-chain cryptocurrency, your assets may be seized by a government, making the on-chain assets worthless. Then there is also the chance of fraud by the stablecoin issuer.
So, for many crypto fans, there is another, higher, requirement of stablecoins. They aspire to build an asset which is not only stable but which no government could seize, block, or tamper with. This dual optimisation problem has resulted in the different types of stablecoins we have today.
Let’s check them out.
Types of Stablecoins
There are three types of stablecoins:
Fiat-collateralised - USDT, USDC, GUSD, BUSD, USDP, TUSD (centralized, collateralized)
Over-collateralised - DAI, MIM, sUSD
(decentralized, over-collateraliized)
Algorithmic - UST (now USTC, peg is lost), USDN, FRAX, USDD (decentralized, under-collateralized)
Let’s describe the three standard stablecoins models.
Stablecoin models
Source: The Generalist
Fiat-backed stablecoins
This is the most straightforward model. A centralised party has dollars in a bank account. For each dollar in their bank account they can issue a digital dollar coin on the blockchain. These Dollar Coins are then backed one-to-one to USD.
Pros: This model is very easy to understand. Coins backed by real world assets should not lose their peg over longer periods of time since the coin providers offer to redeem them at their face value (1 USD) at any time.
Cons: The USD collateral is stored off-chain so you need to trust the provider and their auditors that the coin actually has backing. Tether has repeatedly faced accusations of being under-collateralised but it has so far kept its peg.
Moreover, fiat-backed USD stablecoins need to comply with US regulations. Therefore, they have included provisions in the smart contracts that govern their stablecoins that allow them to “freeze” issued coins. This can be done for “positive” reasons such as when their stablecoin ends up in the hands of crypto thieves.
But the censorship it can also be applied to anyone who the US government decides to consider an adversary.
By holding fiat-backed USD stablecoins you are effectively still dealing with the US government, just with a few extra intermediaries in between.
Over-collateralised stablecoins
These stablecoins are backed by crypto reserves with higher value than the issued stablecoins. In MakerDAO, the DAI stablecoin is currently backed by 140% reserves.
Pros: The collateral is visible on the blockchain so you can see in real time if the coin has enough backing to support its peg.
Cons: In times of high volatility the collateral value can fall below the value of the stablecoin, making in under-collateralised (see Black Thursday).
Another drawback is that new coins are created by users wanting to borrow the stablecoin (e.g. DAI) against their crypto holdings. So stablecoin supply is limited by borrowing demand.
To reduce these risks MakerDAO introduced one-to-one USDC backing into its stablecoin design (Peg Stability Module, introduced in 2020). Now more than half of the DAI collateral is USDC, reducing the chance DAI becomes under-collateralised due to crypto price volatility and allowing DAI supply to increase beyond traditional borrowing demand. Sounds great?
However, in practice this means that now DAI is exposed to US government intervention in a similar way as USDC. If Circle (USDC’s issuer) was asked by the US government to block the DAI contract addresses, then DAI would suddenly become under-collateralsied. The precedent of the US Treasury sanctioning a smart contract address was set with Tornado Cash. Following this unprecedented event, there are now serious discussions about weaning DAI off this dependency and perhaps ending DAI’s peg to the USD.
As you can see, the race is still on to find a trully permission-less stablecoin. Algorithmic stablecoins have stepped in in an attempt to fill this void [not successfully, but they have stepped in].
Algorithmic stablecoins
Broad definition
An algorithmic stablecoin is one which uses an algorithm to maintain its peg. Stablecoins which are nominally “over-collateralized” by assets issued by the balancing algorithm or stablecoin provider (“endogenously” collateralized) are also considered algorithmic.
Paired-coin algostables
A popular stabecoin algorithm is to pair two assets - a stablecoin (S) and a floating cryptocoin (C) .
When S’s value is above peg ( 1 S > 1 USD), the algorithm lets users “burn” (exchange) C coins and receive S coins at par (at 1 S = 1 USD)
The supply of S increases putting downward pressure on its price
Arbitragers can also use this mechanism to buy S coins at peg and then sell them in the open market at a price which is above peg ( 1 S > 1 USD) for profit
The increased supply and sell pressure from arbitragers causes S’s price to go down towards its peg
Alternatively, when S is trading below beg (1 S < 1 USD), users can “burn” S coins for C coins, also as if S was trading at peg (1 S = 1 USD).
This reduces the supply of S so its price rises towards its peg
Arbitragers can also buy S in the open market below its peg and exchange it with the algorithm for C coins at profit
This buying pressure further pushes up the price of S
This all sounds nice, but can you see what could go wrong? If you have followed the Terra Luna collapse you probably have an idea. Unfortunately, such paired algostables are prone to the now-infamous “death spirals”.
The problem occurs when S is below peg.
Burning S coins in exchange for C coins causes a fall in the price of C
C holders anticipate that further expansion in the supply of C may be needed to stabilize the S peg
They realize this will reduce C’s price so they start selling their C coins
This puts further downwards pressure on the price of C
S holders see that the price of the coin backing their S coins is failing so larger amounts of C coin need to be created per burnt S coin
S coin holders become concerned about the algorithm’s ability to stabilize the peg via the burning mechanism.
More S coins are burnt for C coins, which are then sold for USD or other assets
Arbitragers see where the trend is going and also jump in to sell both assets, often with leverage
Repeat 1 to 8 until eventually the price of C approaches zero (see Luna [Classic] for an illustrative example).
[Yes, this does sound like a bank run…]
Whenever you have an under-collateralized asset you are ultimately playing an elaborate confidence game. Once the community loses “faith” in the ability of a stablecoin’s algorithm to maintain its peg, a death spiral ensues and the peg unravels.
So, to summarize the characteristics of algorithmic stablecoins:
Pros: They are decentralized and permission-less so they fall outside the control of governments. Plus it is easy expand their supply.
Cons: In practice they tend to be under-collateralized and are prone to death spirals.
One can argue that real world currencies have also been under-collateralized following the end of the gold standard. True, they have their governments legitimacy and economies [and armies] as backing. Crypto has to rely on computer code [and adoption].
Is this the end of algorithmic stablecoins?
After the Terra debacle, it seems that algostables may be about to get banned by the US government (labelled "endogenously collateralized" stablecoins).
So are all algostables doomed?
I refuse to completely count algostables out just yet. I will settle for being deeply suspicious of any which rely on the paired-coin model. For all we know, some bright day in the future an algorithmic stablecoin creator may invent a model that works. And if that happens, I promise to write a blog post about it and let you all know.
The state of stablecoins
All these three stablecoin models were coexisting [mostly] peacefully until two events occurred:
The most prominent algorithmic stablecoin collapsed (UST, Terra), leaving the future of algorithmic stablecoins in doubt.
The contagion from the US Treasury banning a smart contract (Tornado Cash) led to Circle locking the USDC assets in the contract
Both of these will leave deep scars on the stablecoin space.
We already discussed why UST was unstable and prone to collapse. Many saw its inherent instability but stayed invested because they trusted its charismatic founder or hoped they would get a last-minute “bailout” from a generous benefactor. Lessons awere learnt and all algostables are now looked at with deep suspicion.
So the future should belong to fiat-backed and over-collateralized stablecoins, right? But the second event has put their business models in peril too by exposing how their vulnerabilities.
The TornadoCash sanctions: Before and After
Let’s go back to June 2022 when life was simple and we only had the bear market to worry about.
Terra had just collapsed in a death spiral it had been unable to contain with its Bitcoin and Luna reserves.
Traders were once again betting agains the Tether (USDT) peg due to skepticism over its fiat reserves [and the similarity of its ticker to the of UST]. To boost market confidence, Tether promised to get an audit from a “Top 12" auditing firm which provided a lot of comedic relief
Circle (USDC) was seen as the reputable stablecoin provider with clear reserve backing
DAI was the coin with visible on-chain collateral.
We all knew who to trust, right?
Wrong! You never know who to trust [yes, I may be watching too much GOT].
Just a few months later, in August 2022, the US Treasury sanctioned the Tornado Cash smart contract addresses.
Chaos ensued in the broader crypto space.
Many DeFi protocols blocked users who had interacted with the Tornado Cash contracts from interacting with their application front ends. It was not clear that this was required by the sanctions but few wanted to risk 30 years in jail.
In the Netherlands, a Tornado Cash developer was arrested and is still being held in prison without any charges.
It would have been nice for stablecoins to emerge unscathed by the drama. But naturally this was impossible. You see, the Tornado Cash contracts allowed people to anonymously transfer not just ETH, but all USDC and USDT.
Following the sanctions, Circle chose to be cautious and froze the USDC (over 70k USD) in the Tornado Cash smart contracts. The amount was small but the signal unmistakable.
USDT, in contrast, positioned themselves as crypto-natives, and refused to freeze any USDT contract assets until they received an explicit request from US authorities. So far none has arrived.
All of a sudden, crypto fans had to reevaluate their beliefs of USDT and USDC. MakerDAO’s founder Rune started floating the idea of reducing DAI’s reliance on USDC and even depegging it from USD to protect its long term future.
The fear that the US government may soon turn against other smart contracts was palpable. The vulnerabilities of centralized fiat-backed stablecoins had finally come into focus.
The future of USD stablecoin dominance
Which led to my twitter thread on the future of USD stablecoin dominance.
In the thread I argue that while USD is currently dominating the crypto stablecoin market, the US Treasury sanctions of TornadoCash have have put this top position in grave danger. Consequently, now is the time for new challengers to USD stablecoins to emerge. This would also make the crypto space less vulnerable to sudden changes of heart by US regulators, and become more resilient in the long term.
One alternative would be a stablecoin tracking a basket of currencies reduce the reliance on any particular country. Meta’s DIEM (previously known as Faebook’s Libra) would have done this but regulators stopped them. Another option is to launch a stablecoin pegged to a currency in a crypto friendly jurisdiction. Potential targets could be the UK or Singapore.
And then, there is the wildcard option, one which could make stablecoins truly crypto native again.
Could RAI rise to the challenge?
RAI is an over-collateralized stablecoin backed only by ETH. Crucially, it is not pegged to USD or any other fiat currency. It was designed from a fork of the DAI code and operates a managed float regime against USD (and by proxy other currencies). You may guess that RAI’s price is a smother version of ETH’s price and that in % terms it moves with the price of ETH but over longer time periods. While this would make intuitive sense, it is not true.
The more precise definition is that RAI’s price moves with the demand for lending and borrowing in the RAI protocol. So its primary driver ends up being the demand for ETH leverage - the desire to go long or short ETH.
You can see this illustrated graphically in the chart below where RAIUSD has been quite stable in a period of high ETHUSD volatility and consequently has long periods of rising and falling versus ETH (RAIETH price). Moreover, it tends to rise against ETH when options activity is high (see options open interest and Put/Call Ratio)
RAIUSD vs RAIETH, Source: Coingecko
Fun facts about RAI:
It launched at the price of Pi (3.14).
Ethereum’s creator Vitalik is a fan - “ RAI better exemplifies the pure "ideal type" of a collateralized automated stablecoin".
But will the border community prefer this crypto-native stablecoin or stick to USD stablecoins? My guess is that in times of high global censorship, users will flock to RAI. Otherwise, until the crypto space is more mature, fiat-pegged stablecoins will likely dominate.
Even if it doesn’t end up “winning” in terms of market share, by providing us with a viable permission-less stablecoin option, RAI is doing an enormous service to the crypto community.
The ever-present threat of CBDCs
It’s time to talk about CBDCs (Central Bank digital currencies). So, why shouldn’t governments adopt the best parts of fiat-backed stablecoins and just run them internally?
Well, because a CBDC with smart contracts can easily be abused by a bad actor to un-bank anyone, including political adversaries. Even if we trust our current governments not to overstep their duties, can be trust all future governments to be equally noble?
Some have argued that Satoshi foresaw that the proliferation of the internet would eventually lead to CBDCs and the unprecedented power these would provide to nation states. This looming threat could have been one of Satoshi’s motivations to provide us with Bitcoin as an alternative.
At this point I must confess that I am not completely against CBDCs. However, I am against them being our sole options for digital stablecoins. We deserve a world where there’s strong competition in the sablecoin markets, with easy ways for users to exit to permission-less decentralized stalecoins. In such a world, the threat of dystopian financial control exercised ruthlessly by a hostile government regime is severely diminished.
Conclusion: Now what?
Despite all their troubles, stablecoins are one of the most popular and useful crypto inventions. However, in light of recent events, we will need to focus more on holding them to the highest crypto standards and promote decentralized permission-less options.
These are the developments I anticipate happening in the next five years:
Crypto lobbyists convince US regulators to allow non-CBDC issued US stablecoins to operate in parallel with CBDCs
A stablecoin backed by a basket of currencies is launched by a private firm and gains some traction
RAI increases its market share but is still not dominant [ < 15% of stabecoin market share]
As interest rates rise, yield-bearing-fiat stablecoins emerge and put pressure on the existing fiat-backed stablecoins to also offer yield
A major developed country launches a CBDCs [I’d bet on the UK or Singapore doing this before the US and Europe]
If we are really lucky, we may also succeed in developing an algorithmic stablecoin that actually works [< 20% chance].
Who will win the stabecoin race?
Once CBDCs are launched they will likely make up at least 50% of the total stablecoin market. But they will also grow the market drastically as they facilitate broader adoption of stablecoins by non-crypto natives so they will be a net positive for existing stablecoins. Moreover, I imagine that for a long period of time CBDCs will run on government-managed blockchains and be separated from the broader crypto space.
What will happen on public blockchains?
I foresee a binomial outcome space. If the US does not over-regulate the space, USD stablecoins will likely make up over 90% of stablecoin market share by value. If they do over-regulate, their dominance may slip to below 50%.
Any rash actions would open the door for other fiat-backed stablecoins of crypto-friendly jurisdictions to gain market share. The more hostile the US acts, the more traction decentralized over-collateralized stablecoins like DAI and RAI will gain too.
Long-term, lower US dominance in stablecoin markets would be the better outcome for the crypto industry [and other fiat currencies], but it would require for the US to massively “fumble the bag”.
For the hobbyists
Have a look at this cool Dune stablecoin dashboard
Read about the rise and fall of UST
Consider Haseeb’s post arguing that Ethereum is now un-forkable due to DeFi and sablecoins in particular
Watch the case for RAI being made at the ETH Denver conference, and described in a video by the Defiant
Check out Vitalik’s post on evaluating automated stable-coins, which provides a clear explanation to the Terra and RAI mechanisms