What is an Algorithmic Stablecoin? Full Guide (2026)

An algorithmic stablecoin is a cryptocurrency designed to maintain a stable value, typically $1, not by holding dollar reserves in a bank but by using smart contracts and on-chain algorithms to automatically expand or contract token supply in response to price deviations.

The category once reached $22 billion in supply before the $40 billion destruction of TerraUSD in May 2022 reduced it to less than 2% of the $312 billion stablecoin market, and it has since been quietly rebuilt under new architectures that trade pure decentralization for practical resilience.

As covered in our stablecoin FAQ guide, the main stablecoin types are fiat-backed, crypto-collateralized, and algorithmic or hybrid, with algorithmic representing the highest-risk and highest-controversy category in the entire market.

This guide covers how algorithmic stablecoins work mechanically, the three core model types, the history from Basis to Terra to Ethena, the advantages and risks, how they compare to fiat-backed alternatives, and the current state of the category in 2026.

Key Takeaways

• An algorithmic stablecoin maintains its peg through smart contract-driven supply adjustments rather than fiat reserves, using one of three core models: rebasing (adjusting every holder's token count automatically), seigniorage (a mint-and-burn arbitrage loop with a companion token, the model that destroyed TerraUSD), or fractional-algorithmic (partial collateral plus algorithmic stability modules as in Frax v1).
• Algorithmic stablecoins make up less than 2% of the $312 billion stablecoin market in 2026, down from $22 billion before the UST collapse, with the only post-UST design to achieve meaningful scale being Ethena's USDe at approximately $4.5 billion, which uses a delta-neutral derivatives position rather than companion token reflexivity.
• The GENIUS Act explicitly prohibits endogenously collateralized stablecoins from qualifying as permitted payment stablecoins in the US, directly banning the UST model, while synthetic dollar models like USDe remain in a regulatory gray zone pending OCC final rulemaking on reserve asset definitions.

How Algorithmic Stablecoins Work

An algorithmic stablecoin replaces a bank vault with a rulebook written in code. When the token trades above $1, the protocol mints new supply to bring the price down.

When it trades below $1, the protocol burns tokens or incentivizes redemption to reduce supply and push the price back up. The whole process runs automatically on-chain with no custodian, no reserve audit, and no counterparty that can freeze accounts.

There are three core model types. As covered in our super-stablecoins guide, understanding the distinction between these models is the most important analytical step before evaluating any yield-bearing or non-fiat-backed stablecoin.

Rebasing stablecoins adjust the number of tokens in every holder's wallet proportionally to maintain the peg.

If supply needs to expand by 10%, every holder receives 10% more tokens.

If supply needs to contract, every holder loses tokens. The price stays near $1 but the number of tokens fluctuates.

Ampleforth (AMPL) is the canonical example, operating since 2019 without a major collapse because its niche scale limits bank run risk. AMPL's rebasing applies to all wallets daily based on a 30-day moving average of the exchange rate versus the CPI-adjusted 2019 dollar.

Seigniorage stablecoins pair the stablecoin with a companion governance or seniority token. Users can always burn one dollar of the companion token to mint one unit of the stablecoin, and burn one unit of the stablecoin to receive one dollar of the companion token.

The arbitrage loop is supposed to maintain the peg. When demand for the stablecoin grows, the companion token gains value because burning it is the only way to mint new stablecoins.

The fatal flaw is reflexive tokenomics: when confidence breaks, both assets can enter a death spiral where each redemption drives the other's price lower, accelerating the next redemption. TerraUSD and LUNA are the definitive example.

Fractional-algorithmic stablecoins are partially backed by collateral and partially stabilized algorithmically. The collateral ratio adjusts dynamically based on market conditions.

When the peg is stable and confidence is high, the algorithmic component carries more weight. When confidence falls, the collateral ratio increases.

Frax v1 pioneered this model, starting at 100% collateralized and gradually reducing the algorithmic component as peg stability was maintained.

Frax v2 subsequently reversed course and moved back toward full over-collateralization after the post-UST regulatory environment.

The synthetic dollar model (Ethena USDe) is often grouped with algorithmic stablecoins but is technically fully collateralized. For every dollar of USDe minted, Ethena holds an equivalent long spot position in ETH, BTC, or liquid staking tokens and an offsetting short perpetual futures position on a centralized exchange.

The long and short cancel each other's price exposure, leaving a delta-neutral position. The net revenue stream is the funding rate on the short perpetual position, which is how staked USDe (sUSDe) generates its APY. USDe's risks concentrate in derivatives market liquidity, exchange counterparty exposure, and funding rate sustainability rather than in the reflexive tokenomics that destroyed UST.

History and Evolution of Algorithmic Stablecoins

Basis (2017 to 2018) was one of the earliest algorithmic stablecoin designs, using a three-token model of bonds, shares, and stablecoins. It shut down in December 2018 after raising $133 million when the team concluded that basis bonds might qualify as securities under SEC regulations.

The concept survived as inspiration for nearly every subsequent seigniorage model.

Empty Set Dollar, Dynamic Set Dollar, and Basis Cash (2020 to 2021) attracted capital during early DeFi summer before collapsing as yield incentives dried up. All three demonstrated that pure rebasing and seigniorage models were structurally unstable without collateral backing at any meaningful scale.

TerraUSD (UST) and LUNA collapse (May 2022) is the defining event of the category. UST reached approximately $18.7 billion in supply at its peak, making it the largest purely algorithmic stablecoin ever created.

The Anchor Protocol artificially inflated demand by offering approximately 20% APY on UST deposits, which was unsustainable but kept capital flowing in. When large-scale withdrawals began in May 2022, users flooded the arbitrage loop, burning UST to mint LUNA and selling it immediately.

The selling pressure drove LUNA's price lower, making the next round of redemptions require more LUNA to compensate, driving LUNA lower still. The death spiral destroyed approximately $40 billion in combined value in one week. Do Kwon, co-founder, was subsequently arrested and convicted.

Frax v1 and v2 (2020 to present) represent the most sustained fractional-algorithmic experiment. Frax launched at 100% USDC collateralized and progressively reduced the collateral ratio as peg stability was maintained.

Frax v2 subsequently reversed course and moved toward full over-collateralization following the post-UST regulatory environment. Frax v3 introduced RWA-backed stability and the frxUSD product.

Ethena USDe (2024 to present) is the first major post-UST algorithmic-adjacent stablecoin to achieve real scale.

As covered in our Ethena USDe review, it launched in early 2024 and grew to approximately $4.5 billion in supply by 2026. The delta-neutral basis trade backing represents a fundamentally different risk architecture from UST because it uses real collateral rather than an unbacked companion token.

Advantages and Risks of Algorithmic Stablecoins

Advantages

Capital efficiency is the primary commercial argument for algorithmic stablecoins. They can in theory expand supply without requiring equivalent dollar reserves, meaning growth is not constrained by the amount of fiat capital willing to enter a custody account.

Decentralization is the second core advantage. Purely algorithmic stablecoins have no issuer holding reserves at a bank and no counterparty that can freeze redemptions or comply with government seizure orders. Yield generation is the third advantage specific to synthetic dollar models. USDe generates yield from its own stability mechanism rather than requiring an issuer to share reserve income.

Risks

Death spiral risk is the structural failure mode of seigniorage designs. When redemptions accelerate faster than the protocol can absorb them, the peg mechanism becomes the accelerant of collapse rather than the solution. UST demonstrated this at $40 billion scale.

As covered in our stablecoin risks guide, death spiral risk is rated as the single most severe risk in the stablecoin category and the primary reason the GENIUS Act explicitly prohibits endogenously collateralized stablecoin designs.

Reflexive tokenomics is the root cause behind death spirals. Any stablecoin whose peg depends on a companion governance token is inherently reflexive: the companion token's value depends on confidence in the stablecoin, and confidence in the stablecoin depends on the companion token's value. This creates a self-reinforcing loop that amplifies both bubbles and collapses.

Funding rate risk applies specifically to synthetic dollar models like USDe. If perpetual futures markets flip into persistently negative funding rates, the protocol must pay to maintain its short positions, depleting the insurance fund and reducing or eliminating sUSDe yield. Sustained negative funding rates are the primary systemic risk for USDe that the UST comparison does not address.

Regulatory exclusion is the commercial ceiling risk. The GENIUS Act explicitly prohibits endogenously collateralized stablecoins from qualifying as permitted payment stablecoins, structurally excluding the UST model from US regulated payment infrastructure. USDe's status is pending OCC final rulemaking.

Comparison to Other Types of Stablecoins

As covered in our top institutional stablecoins in June 2026 guide, Ethena USDe is classified as an institutional yield-bearing stablecoin rather than a payment stablecoin. It sits in a distinct regulatory category from both algorithmic stablecoins and GENIUS Act-compliant payment stablecoins.

The most important commercial distinction for any business evaluating stablecoins in 2026 is the GENIUS Act compliance column. Fiat-backed stablecoins from compliant issuers qualify as permitted payment stablecoins. Pure algorithmic stablecoins are explicitly excluded. USDe is in a regulatory gray zone pending OCC final interpretation.

Current State and Outlook in 2026

Algorithmic stablecoins in the pure seigniorage sense represent less than 2% of the $312 billion stablecoin market in 2026. The surviving semi-algorithmic and synthetic dollar products collectively represent approximately $6 to $8 billion. AMPL operates as a niche rebasing product.

FRAX and USDD have both moved toward higher collateralization ratios. USDe at approximately $4.5 billion is the largest non-fiat-backed stablecoin by a significant margin.

The GENIUS Act's prohibition on endogenously collateralized stablecoins is the most commercially consequential regulatory decision affecting the category. Section 3 defines a prohibited stablecoin as any digital asset whose primary reserve backing consists of the value of the issuer's own affiliated token, directly prohibiting the UST model.

The OCC's final implementing rules will determine where the line falls for synthetic dollar models using third-party crypto collateral with exchange-hedged short positions.

USDe faces three key risks heading into H2 2026. Counterparty concentration on centralized exchanges including Binance, OKX, and Bybit creates custody and operational risk that UST did not carry.

Funding rate sustainability as the bull market cycle matures could reduce or eliminate sUSDe yield if rates turn persistently negative. And the OCC determination on whether delta-neutral collateral qualifies as non-endogenous will set USDe's US regulatory ceiling.

As covered in our highest-paying yield-bearing stablecoins guide, sUSDe has delivered between 8% and 11% APY through delta-hedging but that range is directly dependent on derivatives market conditions that have no guaranteed floor.

The category's trajectory is clear. Purely seigniorage-based algorithmic stablecoins are unsalvageable at any meaningful scale.

The category's future belongs to hybrid models with real collateral floors and algorithmic stability modules operating within those floors, and to synthetic dollar models that achieve capital efficiency through derivatives-based hedging rather than companion token reflexivity.

The question for the next two years is whether USDe's basis trade model holds under sustained negative funding rates.

Conclusion

Algorithmic stablecoins in 2026 are not dead but they are permanently smaller.

The category that once reached $22 billion in supply now accounts for less than 2% of a $312 billion market, and the only post-UST design to achieve real scale is USDe's synthetic dollar model, which achieves stability through real collateral and derivatives hedging rather than the reflexive tokenomics that destroyed TerraUSD in a week.

The three core model types (rebasing, seigniorage, and fractional-algorithmic) each carry structurally different risk profiles, with seigniorage designs uniquely vulnerable to death spirals that no protocol parameter can prevent once confidence breaks.

The GENIUS Act's explicit prohibition on endogenously collateralized stablecoins closes the US regulated payment market to the UST model permanently.

For stablecoin users, investors, and institutions in 2026, the practical guidance is straightforward: use fiat-backed GENIUS Act-compliant stablecoins for payment and treasury operations, consider USDe or sUSDe for yield-seeking exposure within a risk-managed allocation that accounts for funding rate and exchange counterparty risk, and treat any purely algorithmic design without real collateral backing as structurally unsuitable for any commercial purpose.

Read Next

• The Rise of Super-Stablecoins: USDe, USDD, and the 20% Yield That Broke DeFi
• Ethena's Staked USDe Stablecoin: Complete Review
• Stablecoin Risks in 2026

FAQ:

1. What is an algorithmic stablecoin?

An algorithmic stablecoin is a cryptocurrency that maintains a $1 peg through smart contract-driven supply adjustments rather than dollar reserves, automatically minting new tokens when the price rises above $1 and burning tokens when the price falls below $1.

2. What is the difference between an algorithmic stablecoin and a fiat-backed stablecoin?

The difference between an algorithmic stablecoin and a fiat-backed stablecoin is that an algorithmic stablecoin maintains its peg through code-driven supply rules with no dollar reserves, while a fiat-backed stablecoin like USDC holds one dollar of cash or Treasuries in custody for each token, making it redeemable at par regardless of market confidence.

3. Why did TerraUSD (UST) collapse?

TerraUSD collapsed in May 2022 because its mint-and-burn arbitrage loop with LUNA became reflexively destructive when large-scale withdrawals began, with each redemption minting more LUNA and driving its price lower, accelerating the next redemption in a death spiral that destroyed approximately $40 billion in one week.

4. What is the difference between an algorithmic stablecoin and Ethena's USDe?

The difference between an algorithmic stablecoin and Ethena's USDe is that a pure algorithmic stablecoin has no real collateral, while USDe is a synthetic dollar fully collateralized by a delta-neutral position combining long spot ETH and BTC with offsetting short perpetual futures, making it backed by real assets rather than algorithmic mechanisms alone.

5. Are algorithmic stablecoins legal under the GENIUS Act?

Algorithmic stablecoins using endogenous collateral, meaning stablecoins backed primarily by the issuer's own token, are explicitly prohibited from qualifying as permitted payment stablecoins under the GENIUS Act, while synthetic dollar models like USDe are in a regulatory gray zone pending OCC final rulemaking.

6. What are the three types of algorithmic stablecoins?

The three types of algorithmic stablecoins are rebasing stablecoins which adjust every holder's token count automatically (Ampleforth), seigniorage stablecoins which maintain the peg through a mint-and-burn loop with a companion token (the UST model), and fractional-algorithmic stablecoins which combine partial collateral with algorithmic stability modules (Frax v1).

7. What is the death spiral risk in algorithmic stablecoins?

The death spiral risk in algorithmic stablecoins is the structural failure mode where accelerating redemptions trigger more companion token minting, the selling drives the companion token price lower, requiring even more tokens to compensate the next redemption, creating a self-reinforcing loop where the stability mechanism becomes the accelerant of collapse.

8. What algorithmic stablecoins still exist in 2026?

The algorithmic stablecoins that still exist in 2026 are Ampleforth (AMPL) as a niche rebasing product, Frax which has moved toward full over-collateralization in v2, USDD on Tron as a partially collateralized hybrid, and Ethena USDe at approximately $4.5 billion as the largest non-fiat-backed stablecoin using a delta-neutral basis trade model.

Disclaimer:
This content is provided for informational and educational purposes only and does not constitute financial, investment, legal, or tax advice; no material herein should be interpreted as a recommendation, endorsement, or solicitation to buy or sell any financial instrument, and readers should conduct their own independent research or consult a qualified professional.