Stablecoins are tokens whose values are meant to stay completely fixed or at least fluctuate with low volatility. Every stablecoin you should know about today aims to be pegged to the US Dollar.
The purpose: Think of stablecoins as the bridge between a stable off-chain world and the currently-volatile crypto world. Without stablecoins, we’d have a really tough time understanding how much something is worth. What is 0.3 ETH when the price of ETH can jump 10% in a day?
Highly conceptual economic terms (“store of value”, “unit of account”) aside, a successful stablecoin needs to be:
- Worth what it intends to be worth – If it should be worth $1 and it’s worth $0.97, then it’s no different than any other volatile crypto asset. When we use stablecoins, we have to be able to safely assume it’s worth $1.
- As decentralized as possible – In keeping with the spirit of web3, the right stablecoin should reduce its dependence on any state/government/institution/bank/etc.
- Not constrained by capital – Having to put up $1 USD to back 1 stable is safe, but it limits how many stablecoins can be created. In this model, putting the entire crypto ecosystem (~$2 trillion) up as collateral could only mint a fraction of the US GDP (~$20 trillion) in stablecoins.
- Widely accepted – If it’s not usable, then it’s not useful.
Cut to the chase: To-date, there are zero stablecoins that have achieved all four dimensions. The most prominent and well-functioning stablecoins are pretty centralized and/or capital-constrained.
Stablecoins you should know
A basic sorting of stablecoins is based on the type of collateral used for the stablecoin:
- Fiat collateral: There is some amount of fiat currency (e.g., US Dollars) in a bank for each stablecoin distributed
- Crypto collateral: The stablecoin issuer requires you to post some crypto assets as collateral to create more of the stablecoin
- Algorithmic: No collateral posted. When needed, the issuer expands/contracts the stablecoins supply to move price.
There are 6 stablecoins worth paying attention to. 4 are at least partially collateralized by fiat and 2 are at least partially collateralized by crypto.
USDC: Issued by Center, a joint venture between Circle and Coinbase. The second largest stablecoin, though seen as the safest – USDC has much tighter transparency and reporting, which is a boon when held against doubt in Tether’s operations. Holds US Dollars as collateral.
BUSD: Issued by Binance. BUSD is common where Binance is popular. Supported mostly on the Binance Smart Chain (BSC) and less popular elsewhere. Holds US Dollars in custody as collateral.
Note: Binance is banned in the US. Binance tends to ask for forgiveness rather than permission, leading to a particularly scammy ethos around Binance-affiliated contexts. For example, Binance Smart Chain is Binance’s L1 blockchain that is known more for low-rate scams than any serious usage.
FRAX: Issued by Frax Finance. Partially collateralized by USDC and partially algorithmic. An emerging stablecoin.
Defending the peg
The most obvious way to ensure a stablecoin’s price stays at $1 is to offer to buy it from anyone at $1. This is obvious for fiat-collateralized stablecoins, but even FRAX has this mechanism, which has helped it perform quite well.
One way we know this mechanism works is because of Fei Protocol, a largely irrelevant stablecoin issuer (FEI). FEI is an algorithmic stablecoin that initially relied on a penalty to enforce the peg.
The idea: When FEI is below $1, penalize people for selling it and reward people for buying it. Vice versa when FEI is above $1. The intention was to deter the harmful actions and reward the healthy actions.
The result: FEI never held peg. FEI crashed to a low of ~$0.13 when factoring in the penalty.
Another way to defend the peg is by using stableswaps, which are exchanges designed around assets that should be priced the same. Stableswaps let you exchange millions of stablecoins without materially affecting the price.
Curve is the most well-known stableswap. Curve “pools” certain assets together to allow trading among them. USDT+USDC+DAI, as the most stable of stables, are pooled to reinforce their pegs. An emerging stablecoin like UST could then get pooled with this core, effectively linking the UST peg to the immense strength of USDT+USDC+DAI.
A last way to defend the peg is by introducing new collateral. Maker brought in USDC (read: a proxy US Dollar) after demand for DAI led to protracted periods of >$1 DAI. Though not encoded as collateral, Terra brought in Bitcoin after a volatility scare threatened the no-collateral UST.
The algorithmic opportunity
Algorithmic stablecoins haven’t made much headway. Terra was the most successful example to-date, and its purchase of Bitcoin is a halfway admission that algorithmic models can’t survive on their own.
Yet, whoever can crack the algorithmic model could solve one of the largest problems in DeFi. The current state of stablecoins is hard-limited by US Dollars – the amount of stablecoins has to be a function of the amount of US Dollars, based on how fiat-collateral stables work. Building a truly global currency used at global scale needs to break from this limit.
Indicative of the larger puzzle: Many parts of web3 sounds like the state of the stablecoin problem. Despite plenty of high-profile attempts, the space still needs to solve the core primitives that will allow it to achieve its final goals.
What does the path to a successful algorithmic stablecoin look like? A lot of work on incentives. Algorithmic coins rely a little bit on certain levers provided by the core protocol that help push the price to peg and a lot of arbitrage and incentives that drive the market to use those levers.
Protocols like FEI demonstrate how hard it is to simulate and predict these behaviors. Protocols like Terra demonstrate that even algorithmic success in the initial phases does not mean success in the later phases – when adoption is high, there could simply be too much on the line to risk trusting your incentive model at-scale.
The bottom line: The opportunity is clear, but examples from history may suggest we don’t see a fully algorithmic stablecoin for a while.