Stablecoins are supposed to be boring. One dollar, one token, no drama. But the history of stablecoins is punctuated by spectacular failures — events where the peg broke, value evaporated, and holders learned the hard way that “stable” is a design goal, not a guarantee.
Six major depegging events stand out. Each followed a recognizable pattern. Each produced warning signs that were visible before the collapse. And each teaches a specific lesson about where stablecoin risk actually lives.
This is not a history lesson for its own sake. The patterns repeat. The next depeg will rhyme with one of these. Knowing the playbook is the difference between watching from the sidelines and being the exit liquidity.
Pattern 1: The Algorithmic Death Spiral
Iron Finance (IRON/TITAN) — June 2021
What happened. Iron Finance operated a partially collateralized algorithmic stablecoin. IRON was backed ~75% by USDC and ~25% by TITAN, a governance token. To mint IRON, you deposited USDC and TITAN. To redeem IRON, you received USDC and TITAN. The peg was maintained by arbitrage: if IRON traded below $1, arbitrageurs could redeem for $1 of underlying assets, profiting the difference and pushing the price back up.
How it broke. Several large holders (including, reportedly, Mark Cuban’s position) began redeeming IRON simultaneously. Each redemption created new TITAN tokens (the 25% portion). The selling pressure cratered TITAN’s price. As TITAN’s price fell, each subsequent IRON redemption required more TITAN tokens to make up the $1 value — which created more selling pressure on TITAN — which cratered the price further — which required even more TITAN.
The death spiral math. TITAN fell from $65 to effectively $0 in approximately 24 hours. IRON’s peg broke to $0.90, then $0.75, then $0.69, where it stabilized as a claim on the USDC collateral only. The TITAN component of the redemption was worthless. Total value destroyed: approximately $2 billion in TITAN market cap.
The warning signs:
- Extremely high yields (Iron Finance was offering 500%+ APY in some pools — unsustainable by any measure)
- Reflexive collateral structure (backing token’s value depends on demand for the stablecoin it backs)
- Concentrated holder base (a small number of large positions could trigger cascade)
- No circuit breaker mechanism
UST/LUNA — May 2022
What happened. Terra’s UST was the largest algorithmic stablecoin in history at approximately $18 billion in circulation. The mechanism: UST was minted by burning $1 of LUNA, and UST could be redeemed by burning it for $1 of newly minted LUNA. Anchor Protocol offered 20% APY on UST deposits, attracting massive capital inflows.
How it broke. The exact trigger is debated — likely a combination of large coordinated sells on Curve pools and Anchor withdrawal activity. On May 7, 2022, approximately $2 billion in UST was unstaked from Anchor and sold. The selling pressure pushed UST below $0.98. Arbitrageurs began redeeming UST for LUNA, minting new LUNA and selling it. LUNA’s price dropped, requiring more LUNA per redemption, creating more selling pressure. The Iron Finance spiral, at 100x scale.
The numbers. UST fell from $1.00 to $0.10 in six days. LUNA fell from $80 to $0.0001. Combined value destroyed: approximately $60 billion. Anchor’s 20% yield — which had attracted the majority of UST deposits — was revealed as a subsidy funded by the Luna Foundation Guard’s Bitcoin reserves, not sustainable economics.
The warning signs:
- 20% “risk-free” yield on Anchor (no legitimate mechanism generates 20% on a dollar stablecoin without subsidy or hidden leverage)
- Reflexive collateral structure identical to Iron Finance, just larger
- Luna Foundation Guard accumulating Bitcoin as a “reserve” — implicitly acknowledging the algorithmic mechanism alone was insufficient
- Curve pool imbalances visible days before the collapse (the UST/3CRV pool skewed heavily toward UST)
- Concentrated Anchor deposits (~72% of all UST was in Anchor, creating a single point of withdrawal pressure)
The lesson. Algorithmic stablecoins have a 100% failure rate at scale. Every mechanism that uses a reflexive token to maintain the peg contains the same death spiral dynamics. The scale of UST did not make it safer — it made the collapse more violent. The spiral doesn’t care about market cap. It cares about the ratio of sellers to new demand, and in a panic, that ratio goes to infinity.
Pattern 2: Banking Contagion
USDC and the SVB Crisis — March 2023
What happened. On March 10, 2023, Silicon Valley Bank was seized by the FDIC. Circle disclosed that approximately $3.3 billion of USDC’s reserves — about 8% of the total — were held at SVB. Over the weekend, with banks closed and no clarity on whether depositors above the $250K FDIC limit would be made whole, USDC depegged.
The depeg. USDC fell to $0.877 on Saturday, March 11 — a 12.3% discount. The Curve 3pool (USDC/USDT/DAI) skewed violently: USDC comprised over 60% of the pool as holders dumped it for USDT and DAI. Coinbase suspended USDC-to-USD conversions for the weekend, removing the primary redemption mechanism.
The resolution. On Sunday, March 12, the Federal Reserve, FDIC, and Treasury announced that all SVB depositors would be made whole — including uninsured deposits above $250K. USDC re-pegged to $1.00 by Monday morning. Circle confirmed full access to SVB funds. The entire episode lasted approximately 48 hours.
What it revealed:
- Even the “safest” stablecoin (at the time, USDC was considered the institutional gold standard) carries banking system counterparty risk
- Weekend and holiday periods create redemption blackout windows where the peg cannot be defended
- 8% exposure to a single bank was enough to trigger a 12% depeg — the market priced in worst-case loss, not expected loss
- The resolution required intervention from three federal agencies — if the government hadn’t backstopped SVB depositors, the depeg could have persisted or worsened
The DAI Cascade
The downstream effect. DAI was approximately 50% backed by USDC at the time of the SVB crisis. When USDC depegged, DAI mechanically followed — falling to approximately $0.90. This was not a failure of MakerDAO’s mechanism. It was functioning exactly as designed: DAI’s price reflected the value of its collateral, and its collateral had lost value.
The cascade demonstrated a critical and underappreciated risk: stablecoin-on-stablecoin dependency. DAI’s use of USDC as collateral meant that USDC’s banking risk was DAI’s banking risk. The decentralization of the MakerDAO protocol did not protect DAI from the centralized banking exposure embedded in its collateral.
MakerDAO responded by diversifying collateral away from USDC concentration and accelerating RWA (real-world asset) allocations. The dependency has decreased but not disappeared.
The warning signs:
- Concentration of reserves at a single banking institution (Circle’s SVB exposure was public information before the crisis)
- Weekend/holiday redemption gaps with no mechanism to defend the peg
- Stablecoin-on-stablecoin collateral chains creating hidden correlation
- Bank run dynamics at SVB were visible for days before the seizure (stock price collapse, deposit outflow reports)
Pattern 3: Chronic Instability
USDD — Ongoing
What happened. USDD, issued by the Tron DAO Reserve (backed by Justin Sun’s ecosystem), launched in 2022 as an “algorithmic stablecoin” that later added collateral backing. It has never experienced a single catastrophic depeg. Instead, it has experienced something arguably worse: chronic, persistent instability that erodes trust through a thousand small cuts.
The pattern. USDD regularly trades between $0.97 and $1.02. It drifts. It wobbles. It spends days or weeks below $0.99 without any specific catalyst or crisis. The peg isn’t broken — it’s loose. Like a door that doesn’t quite close all the way.
Why it persists:
- Thin liquidity. USDD’s trading volume and DEX liquidity are a fraction of USDC or USDT. In thin markets, small trades move the price. The arbitrage mechanism that keeps major stablecoins pegged requires deep liquidity to function efficiently. Without it, the peg drifts.
- Opaque reserve management. The Tron DAO Reserve claims overcollateralization, but the reserve composition is dominated by TRX (Tron’s native token) and BTC — volatile assets that can swing 20% in a week. The “overcollateralization” is a moving target.
- Ecosystem dependency. USDD’s utility is concentrated within the Tron ecosystem. It doesn’t have the cross-chain, cross-exchange liquidity depth that anchors USDC and USDT. When it drifts, there aren’t enough arbitrageurs with enough capital deployed to pull it back quickly.
- Credibility deficit. Justin Sun’s track record of promotional claims and the Tron ecosystem’s reputation create a baseline skepticism that makes every wobble feel like a potential precursor to a larger failure. Trust, once questioned, creates its own selling pressure.
The lesson. Not all depeg risk is catastrophic. Chronic instability — a stablecoin that’s “mostly” pegged — can be more corrosive to adoption than a single dramatic event. Institutions and protocols that integrate USDD carry the operational burden of accounting for a token that might be worth $0.97 or $1.02 at any given moment. That margin of error is unacceptable for treasury management, settlement, or lending protocol collateral.
Pattern 4: Rumor-Driven Panic
FDUSD — 2025
What happened. First Digital USD (FDUSD), a stablecoin issued by First Digital Trust in Hong Kong, experienced a sharp depeg driven not by reserve failure or algorithmic collapse but by rumors and concentrated exchange dependency.
The trigger. Justin Sun publicly accused First Digital Trust of insolvency (accusations related to a separate custodial dispute, not directly to FDUSD reserves). The accusations were unproven at the time, but the combination of a high-profile attacker, FDUSD’s relatively thin liquidity outside Binance, and the crypto market’s hair-trigger sensitivity to stablecoin risk created a self-fulfilling panic.
The mechanism. FDUSD’s liquidity was concentrated overwhelmingly on Binance, where it served as a primary trading pair. When the rumors hit, holders began selling FDUSD for USDT and USDC on the limited DEX liquidity available. The thin order books amplified the price impact. FDUSD dropped to approximately $0.95-0.97.
The resolution. First Digital Trust published reserve attestations, Binance maintained FDUSD trading pairs, and the peg gradually recovered as the market concluded that the insolvency accusations lacked substance. But the damage to confidence was lasting — FDUSD’s market cap declined significantly in the aftermath as holders migrated to stablecoins with deeper liquidity and less concentrated exchange dependency.
The warning signs:
- Single-exchange dependency. When the majority of a stablecoin’s liquidity exists on one exchange, the stablecoin’s fate is tied to that exchange’s willingness and ability to maintain it. Exchange delistings, suspensions, or even maintenance windows become existential risks.
- Thin DEX liquidity. The peg defense mechanism requires arbitrageurs to buy at discount and redeem at par. If DEX liquidity is too thin to absorb selling pressure, the peg breaks regardless of reserve quality.
- Reputation attack surface. A stablecoin with limited brand recognition and concentrated distribution is vulnerable to reputation attacks that would bounce off USDC or USDT. The less established the issuer, the more credible any accusation sounds.
- Geographic/regulatory concentration. A Hong Kong-based trust company issuing a stablecoin primarily used on a non-US exchange creates jurisdictional uncertainty that amplifies panic when questions arise.
The Warning Signs: A Unified Framework
Across all six events, the warning signs cluster into five categories:
1. Yield That Defies Explanation
If a stablecoin or stablecoin protocol offers yield significantly above the risk-free rate without a clear, sustainable mechanism, the excess yield IS the risk. Anchor’s 20% was a subsidy. Iron Finance’s 500% APY was reflexive token emissions. When the yield seems too good, ask: who is the counterparty, and do they know they’re the counterparty?
2. Reflexive Collateral
Any structure where the backing asset’s value depends on demand for the stablecoin it backs contains a death spiral. This includes algorithmic mint/burn mechanisms (UST/LUNA, IRON/TITAN) and stablecoins backed primarily by their own ecosystem’s governance token (USDD backed by TRX). The collateral must have value independent of the stablecoin.
3. Liquidity Concentration
When a stablecoin’s liquidity exists primarily on one exchange, one chain, or one trading pair, the peg is fragile. Deep, distributed liquidity across multiple venues is the structural defense against depeg. Check Curve pool balances, DEX depth, and exchange order books. Skewed pools and thin books are the canary.
4. Reserve Opacity
If you cannot determine, with reasonable specificity, what backs a stablecoin and where those assets are custodied, you are trusting the issuer’s word. That trust has failed repeatedly. The market has learned this lesson multiple times and still hasn’t fully internalized it.
5. Correlated Collateral Chains
If Stablecoin A is backed by Stablecoin B, a failure in B cascades to A. The DAI/USDC/SVB episode demonstrated this explicitly. Evaluate not just what backs your stablecoin, but what backs the things that back your stablecoin. One level of indirection does not equal diversification.
The Depeg Checklist: What To Do When Your Stablecoin Breaks the Peg
A practical guide for the moment you see your stablecoin trading at $0.95:
Step 1: Identify the pattern. Is this an algorithmic spiral (reflexive selling creating more supply), a banking crisis (reserve access disrupted), a liquidity crunch (thin markets amplifying small sells), or a rumor-driven panic (no fundamental reserve issue)? The pattern determines the response.
Step 2: Check the reserves. Go to the issuer’s attestation page. Check on-chain reserves if available. Check Chainlink Proof of Reserve feeds. If reserves are intact and the issue is liquidity or panic, the depeg is likely temporary. If reserves are impaired, the depeg may be structural.
Step 3: Assess your exposure. What percentage of your portfolio is in this stablecoin? What’s your cost basis? Can you absorb a 10% loss? A 50% loss? A total loss? The answer determines your urgency.
Step 4: Check redemption pathways. Can you redeem directly with the issuer at par? What are the minimum redemption amounts and processing times? If primary redemption is available and you meet the minimums, you may be better off redeeming than selling at a discount.
Step 5: Don’t panic-swap into a correlated asset. Swapping USDC for DAI during the SVB crisis was not diversification — DAI was 50% backed by USDC. Swapping into USDT was. Understand the collateral chains before you move.
Step 6: Consider the asymmetry. If reserves are intact and the depeg is driven by panic, buying at $0.95 for redemption at $1.00 is a 5% arbitrage. This is how depegs resolve — arbitrageurs profit from the panic. If you have conviction in the reserves and the capital to deploy, the depeg is an opportunity, not just a risk. But only if you’ve done Step 2 rigorously.
Step 7: Document and learn. After the event resolves, review what warning signs you missed and update your stablecoin allocation model accordingly.
The Algorithmic Stablecoin Track Record
A blunt assessment: algorithmic stablecoins have a 100% failure rate at scale. Every project that has attempted to maintain a dollar peg through purely algorithmic mechanisms — without full fiat or crypto collateral backing — has either failed catastrophically or survived only at negligible scale with persistent instability.
| Project | Peak Market Cap | Outcome |
|---|---|---|
| Iron Finance (IRON) | ~$2B ecosystem | Collapsed to ~$0.69, TITAN to $0 |
| TerraUSD (UST) | ~$18.7B | Collapsed to ~$0.02 |
| Basis Cash | ~$200M | Abandoned |
| Empty Set Dollar | ~$100M | Abandoned |
| Dynamic Set Dollar | ~$50M | Abandoned |
| Neutrino USD (USDN) | ~$900M | Chronic depeg, effectively abandoned |
| USDD | ~$750M | Chronic instability, added collateral backing |
The pattern is consistent enough to constitute a law: reflexive peg mechanisms fail under stress because the very mechanism designed to restore the peg accelerates its destruction. The larger the stablecoin, the more violent the failure, because the volume of tokens seeking exit exceeds the market’s capacity to absorb newly minted governance tokens.
Newer designs (Frax v2, Ethena’s USDe) have moved away from pure algorithmic mechanisms toward hybrid models with real collateral. This is an implicit acknowledgment of the track record. The question is whether these hybrid models carry their own novel risks — and in the case of Ethena’s delta-neutral strategy, the answer is clearly yes.
Mark’s Take: The six depeg events map to four patterns, and every stablecoin you hold is exposed to at least one of them. The only defense is diversification across issuers, chains, and collateral types — and the discipline to evaluate warning signs before they become headlines. Algorithmic stablecoins are a solved question: they don’t work. The interesting risk is in the stablecoins that “work” — until their specific failure mode activates.
Explore the complete depeg history with timeline data on MarketCrystal’s Depeg History page, and monitor live peg deviations on the Peg Monitor.
The Bottom Line
Every stablecoin depeg followed a recognizable pattern. Algorithmic death spirals. Banking contagion. Chronic instability. Rumor-driven panic. The mechanisms differ, but the warning signs cluster: unsustainable yield, reflexive collateral, concentrated liquidity, reserve opacity, and correlated collateral chains.
The depeg playbook is not about predicting which stablecoin fails next. It’s about recognizing the pattern when it starts, assessing your exposure, and having a plan before the market forces you to improvise.
Stablecoins are infrastructure. Infrastructure fails. The question is not whether another depeg will happen — it will. The question is whether you’ll see the pattern before or after it takes your capital.
We don’t predict. We follow the data. And the data has patterns.
MarketCrystal provides trend analysis and market commentary for informational purposes only. Nothing in this publication constitutes financial advice, investment recommendations, or solicitation to buy or sell any security. Cryptocurrency markets are volatile; you may lose money. Always conduct your own research. Past trends do not guarantee future results.
Follow the plumbing.