Three fundamentally different designs carry the stablecoin label. Fiat-backed coins hold actual dollars (or equivalent liquid assets) in reserve. Crypto-backed coins hold on-chain assets in over-collateralised vaults. Algorithmic coins attempt to hold a peg through supply-adjustment incentives rather than reserves. The three designs share the goal of price stability but differ in where that stability comes from — and where it can break. As of mid-2026, the total stablecoin market has crossed $320 billion, with fiat-backed designs representing the overwhelming majority.
Fiat-backed stablecoins
A fiat-backed stablecoin is exactly what it says: one token, one dollar (or euro, yen, or other currency) held in reserve by a regulated issuer. To mint tokens, a user or institutional partner deposits cash with the issuer. To redeem, they return tokens and receive dollars back. The reserves are custodied at banks or in government securities — assets that can be liquidated quickly to meet redemptions.
The largest examples are USDC (Circle, ~$78 billion in circulation as of mid-2026), USDT (Tether, ~$183–190 billion), PYUSD (Paxos/PayPal, ~$4 billion), and RLUSD (Ripple, ~$1.7 billion). Together they make up the dominant share of the stablecoin market.
Pros
- Peg is anchored by real, external assets — the mechanism does not depend on market confidence in a second token.
- Redemption at par is contractually available, enabling reliable arbitrage correction.
- Regulatory frameworks (US GENIUS Act, EU MiCA) now provide legal certainty for the category.
- Low volatility even under market stress, assuming reserves are liquid and transparent.
Cons
- Issuer counterparty risk: if the issuer is insolvent or reserves are illiquid, redemptions may be impaired.
- Centrally controlled: issuers can freeze, blacklist, or seize tokens — a material concern for some use cases.
- Users do not earn the yield on reserves; that accrues to the issuer.
- Reserve transparency varies: Circle publishes monthly Deloitte attestations; Tether publishes quarterly BDO attestations.
Risk profile: Low-to-moderate, principally issuer credit and reserve quality risk. The USDC brief de-peg in March 2023 (resolved within hours when deposit insurance confirmed reserves) is the most visible recent stress event.
Crypto-backed (over-collateralised) stablecoins
Crypto-backed stablecoins use on-chain assets — ETH, BTC, or other tokens — as collateral, issued at an over-collateralised ratio. A user who wants $100 of DAI (MakerDAO's stablecoin) must lock up significantly more than $100 in collateral; historical ratios have been 150% or higher.
The peg holds via automated liquidation: if the collateral value falls toward the value of the outstanding stablecoin, smart contracts sell the collateral to buy back and burn the stablecoin, reducing supply and pushing the price back toward $1. The system is transparent — all positions and collateral are visible on-chain.
Pros
- Decentralised and permissionless: no single issuer controls minting or redemption.
- Fully on-chain: reserves are verifiable without trusting any institution's attestation.
- No counterparty credit risk in the traditional sense.
Cons
- Capital-inefficient: every $1 of stablecoin requires >$1 (often $1.50 or more) locked in collateral.
- Collateral cascade risk: if crypto markets crash sharply, collateral can fall in value faster than liquidation bots can sell it, leaving the system under-collateralised.
- Typically limited to crypto-native users with on-chain collateral to pledge.
- Smaller scale than fiat-backed designs.
Risk profile: Moderate. Works well under orderly market conditions; the failure mode is a rapid, correlated collapse in collateral prices that outpaces liquidation.
Algorithmic stablecoins
Algorithmic stablecoins attempt to hold a peg through incentive structures and supply adjustments rather than by holding reserves. The canonical design mints or burns the stablecoin based on demand signals, often using a linked "seigniorage" token to absorb volatility.
The definitive cautionary example is TerraUSD (UST). UST's peg relied on an arbitrage loop: burn UST to mint LUNA when UST traded below $1; burn LUNA to mint UST when it traded above. The mechanism worked while LUNA had independent market value. When a large-scale sell-off began in May 2022, LUNA's price fell, the loop became reflexive, and the peg collapsed within days. Estimated direct losses were around $40 billion in UST/LUNA market capitalisation. The collapse also destabilised other parts of the crypto market, contributing to losses far beyond the Terra ecosystem.
Pros (theoretical)
- Capital-efficient: does not require locking up more than $1 in assets per dollar of stablecoin issued.
- Can achieve large scale without large reserve balances.
Cons (in practice)
- The incentive mechanism is circular: it depends on confidence that the mechanism will work, which evaporates fastest in the scenario it most needs to hold.
- Regulators in the US and EU now explicitly exclude or restrict algorithmic stablecoins from the payment stablecoin category. The GENIUS Act (July 2025) requires 1:1 liquid backing for payment stablecoins.
- No surviving large-scale algorithmic stablecoin has held up through a major market stress event.
Risk profile: High. Most market participants and regulators now treat the algorithmic design as unsuitable for payments or savings.
A comparison at a glance
| Feature | Fiat-backed | Crypto-backed | Algorithmic |
|---|---|---|---|
| Reserve | Cash + Treasuries at regulated custodian | On-chain crypto collateral (150%+) | None (or minimal) |
| Peg mechanism | Redemption arbitrage | Liquidation engines | Incentive loops |
| Capital efficiency | 1:1 (efficient) | >1:1 (capital-intensive) | <1:1 (efficient in theory) |
| Transparency | Periodic attestations | Fully on-chain | Fully on-chain |
| Issuer control | Centralised; can freeze/blacklist | Decentralised governance | Usually decentralised |
| Main failure mode | Issuer insolvency or illiquid reserves | Collateral crash faster than liquidation | Reflexive confidence collapse |
| Post-2022 regulatory status | Clear legal framework (GENIUS Act, MiCA) | Not specifically excluded | Restricted or prohibited for payments |
| Examples | USDC, USDT, PYUSD, RLUSD | DAI | (failed) TerraUSD |
Payment stablecoins as an asset class
Regulators and payment infrastructure providers have converged on a specific sub-category: the payment stablecoin. This is a fiat-backed coin that: holds 1:1 liquid reserves; is issued by a licensed entity; publishes regular independent attestations; and supports fast, on-demand redemption. USDC, USDT, PYUSD, and RLUSD are the established examples.
Payment stablecoins are the native asset class for stablecoin-optimised payment chains. Tempo, which launched on mainnet on March 18, 2026, was purpose-built to carry exactly this type of asset: its TIP-20 token standard is designed for regulated stablecoins, its Fee AMM converts between any USD-denominated TIP-20 so users never need a separate gas token, and its payment lanes guarantee blockspace for stablecoin transfers. All of the major payment stablecoins can operate on Tempo.
What to read next
The taxonomy above tells you what each type is. For how the peg actually holds — and what breaks it — see How do stablecoins keep their peg?. For a comparison of the two largest specific coins, see USDC vs USDT.