Types of Stablecoins: Fiat-Backed, Crypto-Backed, Algorithmic

Not every stablecoin works the same way underneath, even though most aim for the same visible result: a price that stays close to one unit of a fiat currency. Three broad designs dominate the space, fiat-backed, crypto-backed, and algorithmic, and the differences between them matter far more than their shared, roughly stable price suggests. Two stablecoins can trade at nearly identical prices while carrying very different risk profiles, because the mechanism behind the peg is what actually determines how the coin behaves under stress.

Fiat-Backed Stablecoins

Fiat-backed stablecoins, the largest category by usage, are issued against reserves of cash and short-term, low-risk instruments held by a central issuer. Tether and USD Coin are the two most widely traded examples. The model is conceptually simple: for every token in circulation, the issuer is meant to hold equivalent value in reserve, and large holders can typically redeem tokens directly for the underlying asset. That simplicity is also the trade-off. Holders are relying on a centralised issuer to manage reserves honestly, keep them accessible, and publish credible, regular attestations. The mechanism is easy to understand but depends on trusting an off-chain custodian rather than verifiable code.

Not all reserves are equal, even within the fiat-backed category. Some issuers lean heavily towards cash and very short-term government instruments, which are easier to value and liquidate quickly. Others have historically held a broader mix that includes other short-term corporate debt or investments, which can be harder to price precisely and slower to convert back into cash during a period of stress. The headline claim of being fully backed doesn’t say much on its own; the composition, maturity, and liquidity of what actually sits behind that claim is where the real difference between issuers shows up.

Crypto-Backed (Collateralised) Stablecoins

Crypto-backed stablecoins flip the custody model. Instead of a company holding fiat reserves, users lock other cryptocurrencies into smart contracts as collateral and generate the stablecoin against that collateral. Dai is the best-known example of this approach. Because the collateral itself is volatile, these systems require overcollateralisation, locking in more value than the stablecoin issued against it, and use automatic liquidations to protect the system if collateral value falls too far. The advantage is transparency: anyone can inspect the collateral on-chain, with no need to trust a bank statement. The trade-off is capital inefficiency, locking up more value than you receive, and exposure to smart-contract and liquidation risk instead of custodial risk.

Collateral for these systems isn’t always a single asset either. Many crypto-backed designs accept a basket of different cryptocurrencies as collateral, each with its own risk characteristics, and some have expanded to accept tokenised real-world assets as collateral too. A broader collateral base can reduce reliance on any one asset’s price behaviour, but it also adds complexity: more collateral types generally means more parameters to manage, more price feeds that need to stay accurate, and more edge cases for the system’s liquidation logic to handle correctly under stress.

Algorithmic Stablecoins

Algorithmic designs attempt to hold a peg with little or no hard collateral, instead using code-driven incentives to expand or contract supply in response to price. When the token trades above target, the protocol is designed to increase supply to push the price back down; when it trades below target, mechanisms are meant to reduce supply or create buying pressure to push it back up, often by routing the imbalance through a second, more volatile token. This design is the most capital-efficient of the three, since it doesn’t require locking up large reserves or collateral. It has also proven, as a category, to be the most fragile, because the stabilising mechanism itself depends on confidence and market demand rather than on a hard asset sitting in reserve. A sharp loss of confidence can undermine the very mechanism meant to restore the peg, rather than being absorbed by it.

The mechanics vary between algorithmic designs, but a common thread is that the system relies on an incentive, such as a profit opportunity for arbitrageurs or a bond-like instrument promising future value, to persuade participants to help restore the peg rather than simply matching it with a hard reserve. That incentive only works for as long as participants believe it will be honoured. If confidence in the broader system drops, the same participants who would normally help restore the peg have a reason to exit instead, which is a core reason this category has struggled to hold up during periods of serious market stress.

Comparing Transparency and Reserve Quality

The most useful comparison between stablecoins isn’t price; nearly all of them look similar on that measure most of the time. It’s how verifiable and resilient the backing actually is. For fiat-backed coins, look at how frequently reserves are attested, by whom, and whether the disclosed composition leans towards cash and short-term instruments or longer-dated, less liquid assets. For crypto-backed coins, the relevant question is how the collateral ratio has held up historically during sharp market moves, since that is when liquidation systems are tested. For algorithmic designs, ask what actually stands behind the peg when demand for the secondary token dries up. None of these questions can be answered by the price chart alone.

A handful of patterns are worth treating as caution signs regardless of category: reserve reports that are infrequent, unaudited, or produced solely by the issuer itself with no independent involvement; collateral ratios that sit close to the minimum required level with little buffer; and secondary-token incentive schemes with no clear limit on how much new supply can be created to defend a peg. None of these guarantees a problem is coming, but each one reduces the margin for error if market conditions turn stressful.

Why the Category Matters More Than the Price

Because all three designs are built to trade close to their target under normal conditions, price alone is a poor way to judge a stablecoin’s risk. It only tells you how the mechanism is performing today, not how resilient it would be under stress. Tools that track live reserve and peg data across issuers, such as our reserve watch desk, are more informative than any single price snapshot, precisely because they surface the underlying data rather than just the headline number.

This is not financial advice, and no stablecoin should be treated as risk-free simply because its price is usually steady. The category and the specific mechanism behind an individual coin both matter. For a deeper look at exactly how arbitrage and redemptions keep a peg in line day to day, see our companion piece on how stablecoins hold their peg.