Stablecoin is a cryptocurrency designed to maintain a fixed value, almost always pegged to the US dollar. The category covers fundamentally different designs — fiat-backed, crypto-collateralized, algorithmic — and the differences matter more than the marketing suggests. Three failures (TerraUSD in May 2022, USDC in March 2023, Tether's quarterly attestation gaps) make stablecoin selection a real custody question, not a "they're all the same" question.

The three major designs

Fiat-backed (USDC, USDT, PYUSD). The issuer holds USD or USD-equivalent assets (Treasury bills, bank deposits) in a 1:1 ratio with circulating tokens. Redemption is through the issuer. The peg works as long as the reserves are real, accessible, and adequately diversified across banks. This is the dominant design — USDC alone has roughly $40 billion in circulation as of 2026.

Crypto-collateralized (DAI, LUSD, crvUSD). Users lock cryptocurrency (ETH, wstETH, RWAs) into smart contracts at over-collateralization ratios (typically 150-200%), mint stablecoins against the collateral. The peg holds through liquidation mechanics if the collateral value drops. DAI is the canonical example; it has held its peg through multiple crypto-market crashes.

Algorithmic (failed: TerraUSD; surviving: minor experiments). No collateral, peg maintained through contraction/expansion of a paired governance token. TerraUSD lost its peg in May 2022, triggering a $40B collapse over five days. The category is essentially dead post-2022; remaining experiments (Ethena's USDe is closer to a delta-neutral structured product than a pure algorithmic stablecoin) take different shapes.

The 2023 USDC depeg episode

On March 10, 2023, Silicon Valley Bank collapsed. Circle held about 8% of USDC reserves at SVB ($3.3 billion of $40B). Over the weekend, USDC traded as low as $0.88 before the Fed announced it would backstop SVB deposits. By Monday morning, USDC had restored its peg.

For a US-resident holder, the lesson is precise: fiat-backed stablecoins carry banking-system risk on top of issuer risk. Circle has since diversified reserves across multiple banks, but the underlying exposure remains.

How to think about stablecoin choice in 2026

For day-to-day on-chain activity, USDC is the conservative default for US holders: regulated issuer (Circle), monthly attestations from Deloitte, MiCA-compliant for EU operations, broad cross-chain support. USDT carries higher liquidity globally but more opacity around reserves; for short-term operational use this is fine, for multi-month holding it adds risk.

For long-term "cash position" within a crypto portfolio, do not concentrate. A 50% USDC + 30% USDT + 20% DAI mix is a defensible diversification across issuer risk, banking-system risk, and on-chain peg-mechanism risk. Or — better still — for amounts above $50K, move out of stablecoins entirely into Treasury Bills via a US bank account.

The yield product warning

Stablecoin "savings" products at exchanges and DeFi protocols are not zero-risk. Anchor Protocol promised 19.5% APY on TerraUSD; depositors lost essentially everything. Lower-risk versions (Aave's USDC supply at 3-5%, Maple Finance's institutional pools at 8-10%) carry real but smaller risks — smart-contract risk, depeg risk during stress, illiquidity during market events.

Further reading: Proof of Reserves, MiCA.