Stablecoin Depeg Recovery: What Happens After a Peg Breaks
By Jorge Rodriguez — Stablecoins
Why some stablecoin depegs recover in 48 hours while others collapse permanently
A phase-by-phase framework for reading a live depeg event as it unfolds
What to do, and what not to do, when a stablecoin in your yield position loses its peg
Introduction
When a stablecoin loses its peg, most DeFi yield users face an immediate decision: is this a temporary dislocation that will correct in 48 hours, or the opening phase of a permanent collapse? The difference between those two outcomes determines whether holding through the dislocation is rational patience or a catastrophic mistake. **Stablecoin depeg recovery** is not random. The collateral structure of a stablecoin almost entirely determines whether it recovers. A stablecoin backed 1:1 by T-bills and a stablecoin backed by a reflexive token algorithm are both called stablecoins, but their depeg trajectories have nothing in common. This article breaks down how depeg events unfold, what drives recovery versus permanent failure, and what DeFi yield users should actually do in the hours after a peg breaks. [S&P Global's 2023 stablecoin analysis](https://www.spglobal.com/content/dam/spglobal/corporate/en/images/general/special-editorial/stablecoinsadeepdiveintovaluationanddepegging.pdf) identified over 1,900 depeg events between early 2020 and mid-2023. Most were short-lived and inconsequential. A handful were catastrophic. The distinction between the two groups follows from the underlying mechanics, not from market conditions alone. The [Lince Yield Tracker](https://yields.lince.finance/tracker) surfaces stablecoin yield opportunities across protocols and chains. Understanding peg mechanics is foundational to evaluating whether any yield is worth the peg risk beneath it.
What Actually Causes a Depeg
Not all depegs are the same. The cause determines the recovery trajectory almost entirely. Treating every depeg as equivalent is the most common source of bad decisions during live events. There are four distinct failure modes, and each has a different recovery profile. **Type 1: Collateral shortfall** The reserves backing the stablecoin are insufficient, inaccessible, or at risk. The clearest modern example is USDC in March 2023, when Circle confirmed $3.3 billion of its reserves (approximately 8% of total backing) was held at Silicon Valley Bank at the time of its collapse. USDC dropped to $0.86 within hours. The recovery mechanism was direct: issuer confirmation that the remaining reserves were intact, followed by a regulatory backstop. [Chainalysis documented the on-chain cascade](https://www.chainalysis.com/blog/crypto-market-usdc-silicon-valley-bank/), showing USDC trading at $0.87 by 2am on March 11 as panic spread faster than information could travel. This type of depeg is recoverable because the collateral itself is not destroyed, only temporarily inaccessible or perceived as at risk. **Type 2: Algorithmic death spiral** The stablecoin's peg is maintained by a mint-burn mechanism involving a volatile companion token. When the stablecoin dips below $1, arbitrageurs burn the stablecoin to mint $1 of the companion token and sell it, creating buy pressure on the stablecoin. Under normal conditions, this mechanism restores the peg. Under stress, if confidence collapses, the companion token price drops faster than arbitrage can close the gap, triggering companion token hyperinflation and accelerating the stablecoin's collapse. This is the UST/LUNA failure mode. Recovery is structurally impossible once a **death spiral** begins because the mechanism becomes reflexively self-destructive at scale. The [Forbes analysis of the Terra collapse](https://www.forbes.com/sites/rahulrai/2022/05/17/the-death-spiral-how-terras-algorithmic-stablecoin-came-crashing-down/) documents how LUNA supply inflated from roughly 350 million to 6.5 trillion tokens in under five days. **Type 3: Oracle manipulation or market attack** Price feeds are exploited or spoofed, causing the on-chain reported price to diverge from the true market price. This can trigger false liquidations and secondary depegs across protocols relying on the same oracle. Usually recoverable once the oracle is corrected, but the damage window is immediate and the protocol-level impact can be permanent even after the price normalizes. **Type 4: Regulatory action or bank-run panic** Regulatory uncertainty, a freeze on issuer assets, or coordinated withdrawal pressure driven by sentiment rather than fundamental impairment. USDT experienced this in late 2018 when sustained questions about Tether's reserve composition drove it to approximately $0.85 on some exchanges. Recovery depended on trust re-establishment, not mechanical intervention. A **bank run** in DeFi follows the same logic as a traditional one: it is self-fulfilling even when the underlying asset is fundamentally intact. The key differentiator from Type 1 is that here no actual collateral impairment has occurred, only the perception of one.
The Recovery Lifecycle: What Happens Phase by Phase
Understanding what actually happens during a depeg event, phase by phase, removes the paralysis that causes the worst decisions. Most depeg events follow the same sequence regardless of which chain the stablecoin lives on. **Phase 1: Price dislocation (T+0 to T+6 hours)** The initial drop. Liquidity exits. Panic selling drives the price below peg. Market makers widen spreads and reduce size. This phase consistently overshoots the fundamental impairment. USDC's actual reserve shortfall was $3.3 billion against more than $40 billion in total reserves, representing less than 8% of backing, but the market briefly priced it as if total collapse were probable. Overshoot is a structural feature of panic-driven selling, not a reflection of actual risk. **Phase 2: Arbitrage activation (T+2 to T+24 hours)** For recoverable depegs, **arbitrage** begins as the discount becomes large enough to justify the transaction cost. Arbitrageurs buy the depegged stablecoin at a discount on secondary markets and redeem it at face value with the issuer or via protocol mechanisms. This is the core recovery force. It only works if the issuer's **redemption gateway** remains open, the underlying collateral is genuinely intact, and redemption capacity can absorb the demand volume without queue delays that extend the dislocation. **Phase 3: Issuer or protocol intervention (T+1 to T+72 hours)** For fiat-backed stablecoins, the issuer makes public statements confirming reserve status and, if necessary, arranges emergency liquidity. For algorithmic stablecoins, protocol teams attempt buybacks or parameter changes. The [Federal Reserve's December 2025 analysis](https://www.federalreserve.gov/econres/notes/feds-notes/in-the-shadow-of-bank-run-lessons-from-the-silicon-valley-bank-failure-and-its-impact-on-stablecoins-20251217.html) of the SVB failure documents how Circle's Saturday morning statement and the FDIC's Sunday evening guarantee bookended the recovery window precisely. Issuer communication speed at this stage is a direct input to recovery timeline. **Phase 4: Peg restoration or collapse (T+24 hours to T+7 days)** The outcome becomes visible at this stage. Either arbitrage pressure and issuer intervention combine to restore the peg, or liquidity fragmentation accelerates and the price gap widens. The on-chain redemption queue and issuer communication cadence are the leading indicators here. A long redemption queue that is processing normally is bullish for recovery. A queue halt, even a temporary one, is a severe warning signal. **Phase 5: Trust rebuild (T+1 week to T+3 months)** For stablecoins that recover, the aftermath involves reserve disclosure improvements, banking partner diversification, and protocol parameter adjustments. USDC diversified its banking partners substantially after March 2023. This phase determines whether the stablecoin retains market share or loses it to competitors who experienced no depeg during the same event. 
Historical Case Studies: Recovered vs Failed
The record across depeg events is consistent enough to be predictive. These six cases span multiple chains and stablecoin types, and together they map the full range of outcomes. **USDC, March 2023 (Recovered)** Cause: Circle's $3.3 billion in SVB reserves became inaccessible when SVB collapsed on March 10. Depeg depth: $0.86 at trough. Recovery time: approximately 60 hours, from Friday night to Monday morning. Recovery mechanism: the FDIC's decision to guarantee all SVB deposits, announced Sunday March 12, restored confidence. USDC was back at $1.00 by Monday open. The collateral was actually safe throughout. The depeg was entirely panic-driven against a stablecoin with more than 92% of reserves intact. The **contagion risk** was immediate: DAI (then approximately 40% USDC-backed) dropped to around $0.90, and Aave saw approximately $24 million in liquidations from USDC-collateralized positions, illustrating how [DeFi protocol insolvency risk](/blog/risk-management/defi-protocol-insolvency-risk) spreads across interconnected protocols during a single issuer event. **USDT, October 2018 (Recovered)** Cause: Sustained market rumors about Tether's reserve composition and banking relationships triggered withdrawal pressure across exchanges. Depeg depth: approximately $0.85 on some exchanges. Recovery time: roughly one week. Tether continued honoring redemptions without halting the gateway. The absence of a redemption halt was itself the recovery signal. Market confidence rebuilt gradually as withdrawals were processed without incident and no reserve shortfall was confirmed. **DAI, March 2023: Partial Recovery, Dependent on USDC** DAI dropped to approximately $0.88 alongside USDC due to its approximately 40% USDC backing at the time. Recovery relied on the **Peg Stability Module (PSM)**, a MakerDAO mechanism allowing 1:1 swaps between DAI and USDC, which provided a direct arbitrage floor once USDC itself recovered. DAI's recovery was fully contingent on USDC's recovery. The PSM created a mechanical floor, but only because that floor was built on USDC's own collateral. This is **overcollateralized** design providing a buffer, while still inheriting the concentration risk of its primary collateral. **MIM (Magic Internet Money), January 2022: Partial Recovery Over Weeks** Cause: Exposure to the Wonderland/TIME treasury scandal triggered questions about collateral quality. Depeg depth: approximately $0.96. Recovery: gradual over two weeks as Abracadabra wound down bad positions and reduced MIM supply. MIM was overcollateralized overall, but perception damage from associated projects created a trust gap that mechanics alone could not close quickly. This case illustrates how association risk can cause depegs even in structurally sound stablecoins.  **UST/LUNA, May 2022: Failed, Total Collapse** Cause: Large coordinated withdrawals from Anchor Protocol, which had been offering 20% yield on UST, combined with targeted sell pressure on UST liquidity pools, triggered the initial depeg. The mint-burn mechanism then entered a death spiral. The **Luna Foundation Guard (LFG)** deployed approximately $1.5 billion in BTC reserves in an attempt to defend the peg. The selling pressure still overwhelmed any intervention capacity. LUNA supply inflated from roughly 350 million to 6.5 trillion tokens within five days as arbitrageurs redeemed UST for $1 of LUNA and immediately sold. Depeg depth: $0.00 within five days of the initial break. No hard collateral floor existed, so no mechanical force could restore the peg once the spiral exceeded the foundation's capacity. **IRON Finance, June 2021: Failed, First Large-Scale Algorithmic Collapse** Cause: IRON was a partially collateralized stablecoin: 75% USDC and 25% TITAN token backing. A surge in TITAN's price drove protocol growth, but mass withdrawal triggered TITAN selling, reducing backing below the required 75% ratio instantly. A bank run followed with no mechanical recovery floor. Predating UST by almost a year, IRON Finance on Polygon established the partial-collateral failure pattern that UST would later repeat at far larger scale. Partially collateralized algorithmic designs are inherently fragile because any collateral token price collapse below the required ratio creates an undercollateralization event with no recovery path.
What Predicts Recovery vs Collapse
Five structural variables explain virtually all of the divergence between recoverable and unrecoverable depeg events. These are not post-hoc rationalizations. Each variable can be assessed before a depeg occurs. **Variable 1: Hard collateral floor** The single strongest predictor. If a stablecoin is backed 1:1 or better by real assets (T-bills, cash equivalents, overcollateralized crypto), the depeg cannot rationally represent fundamental value loss. It represents panic only. Recovery is a function of demonstrating collateral access, not rebuilding destroyed value. USDC, USDT, and DAI all had recoverable depegs because their collateral was intact even when prices collapsed. UST and IRON Finance had no hard floor and no recovery. **Variable 2: Redemption gateway status** If the issuer keeps the redemption window open (you can still exchange 1 stablecoin for $1 of underlying collateral), arbitrage forces recovery by arithmetic. If redemptions are halted, the peg-restoring mechanism is gone. Any redemption halt during a depeg event should immediately change your risk assessment. Holding USDC during its depeg was also holding [counterparty risk in DeFi](/blog/risk-management/counterparty-risk-defi) to Circle as the issuer. The SVB event is a textbook case of how that issuer counterparty risk materializes in practice. **Variable 3: Reflexive vs non-reflexive mechanism** Algorithmic stablecoins maintain their peg through mechanisms that become self-destructive under stress. Fiat-backed and overcollateralized stablecoins have non-reflexive backing: the collateral does not depreciate because the stablecoin depegs. This is the core structural difference between recoverable and unrecoverable events. No amount of intervention capital can overcome a reflexive collapse once it reaches critical momentum, as the LFG deployment proved. **Variable 4: Issuer communication speed** In the USDC event, Circle's initial statement and the FDIC announcement bookended a roughly 48-hour recovery window. Silence or delayed communication extends panic and causes prices to overshoot the fundamental impairment by a wider margin than necessary. Communication speed is a quality signal about the issuer's operational readiness for crisis scenarios, and also a direct input to recovery timeline. **Variable 5: Liquidity depth** Thin liquidity allows small sell orders to drive large price dislocations. Deep liquidity absorbs panic without exaggerated depeg. Large-cap stablecoins with extensive CEX market making recover faster because there is more capital available to absorb the dislocation and arbitrage restores the peg more efficiently. Liquidity depth is the reason major stablecoins typically overshoot less than smaller ones during equivalent stress events. 
What DeFi Yield Users Should Do During a Depeg
The worst decision-making during a depeg comes from treating every event identically. A collateral shortfall event and an algorithmic death spiral require opposite responses. The first step is identification, not action. **Step 1: Identify the depeg type before acting** Check the stablecoin's backing type before making any exit decision. Is it fiat-backed with publicly confirmed reserves, overcollateralized by crypto, or an algorithmic mechanism? This tells you whether a hard collateral floor exists. If you are unsure, consult the [stablecoin risk tier framework](/blog/stablecoins/stablecoin-risk-tiers) as a starting classification, then verify directly from the issuer's documentation. **If the stablecoin is fiat-backed (collateral shortfall type):** • Check whether the issuer's redemption window is currently open • Find the issuer's official statement quantifying the shortfall and confirming remaining reserve coverage • Calculate the actual impairment: a 5% shortfall on a 100% backed stablecoin does not justify pricing the stablecoin as if total loss were probable • If redemptions are open and total collateral coverage exceeds 90%, holding through the arbitrage recovery typically produces a better outcome than selling at the panic trough • Watch for the recovery signal: issuer confirmation of reserve coverage, followed by visible resumption of redemption processing **If the stablecoin is algorithmic or partially collateralized:** • Exit immediately if you cannot verify a hard collateral floor. There is no arbitrage force that restores a peg without hard collateral underneath it • Do not wait for a buyback announcement. Foundation buyback programs are consistently insufficient against death spirals at scale • Evaluate contagion exposure across your broader position. What other assets in your portfolio are correlated to the collateral token or connected ecosystem? **Arbitrage opportunities during recoverable depegs:** • Buy the depegged stablecoin at a discount on secondary markets and redeem at $1.00 face value with the issuer (fiat-backed only, with open redemptions) • PSM arbitrage: if DAI depegs, check the PSM swap rate before selling at a discount on secondary markets • Gas cost and redemption queue processing time affect profitability. This trade is viable only above a minimum discount threshold **What not to do:** • Do not sell at the trough of a collateral-intact depeg without first reading the issuer's official statement • Do not assume stablecoin equals safe and ignore the event entirely. Even temporary depegs trigger LP impermanent loss, protocol liquidations, and yield disruption • Do not hold through an algorithmic depeg waiting for recovery. The mechanism will not restore what it is structurally incapable of restoring
How to Assess Depeg Risk Before Holding for Yield
The best time to evaluate depeg risk is before capital deployment, not during the event. Four areas of due diligence map the risk clearly. **Collateral type and transparency** • Fiat-backed with regular attestation (USDC, USDT): lowest structural risk; issuer counterparty risk applies and should be evaluated separately • Overcollateralized crypto-backed (DAI, LUSD): a mechanical floor exists; peg quality depends on collateral diversity and the depth of the liquidation system • Algorithmic or partially backed: highest structural risk; the yield premium on these instruments is compensation for collapse risk, not free incremental return **Backing ratio and disclosure frequency** • Real-time or daily attestation carries meaningfully higher confidence than quarterly audits • 100%+ backing with a disclosed asset breakdown carries lower risk than backing without specific reserve composition disclosure • The USDC March 2023 event was recoverable partly because Circle's reserve composition was publicly known, allowing the $3.3 billion shortfall to be mathematically contextualized against the full reserve base in real time **Historical depeg record** • Has the stablecoin lost its peg before? How far did it fall and how long did recovery take? • Did the issuer maintain redemptions throughout, or were they halted at any point? • How quickly and transparently did the issuer communicate during the event? **Protocol integration risk** • A yield pool heavily weighted to a lower-quality stablecoin carries concentrated peg exposure that may not be reflected in the headline APY • Check lending protocol liquidation triggers: if a stablecoin used as collateral depegs, it can trigger your own liquidation even if you do not hold that stablecoin directly • Apply the [DeFi due diligence checklist](/blog/risk-management/defi-due-diligence-checklist) to any protocol before deploying capital to stablecoin yield positions Use the [stablecoin risk tier framework](/blog/stablecoins/stablecoin-risk-tiers) as a starting classification for any stablecoin representing more than 20% of your yield exposure, then apply the variables above to calibrate the actual peg resilience before committing capital.
FAQs
### What is a stablecoin depeg? A stablecoin depeg occurs when the market price of a stablecoin diverges from its target value, typically $1.00. A depeg can be upward (above $1) or downward (below $1), though downward depegs are far more consequential for holders. The term covers events ranging from brief 1-2% dislocations that correct within hours to permanent collapses like UST in May 2022. The severity and direction of the recovery is determined by the underlying collateral structure, not by the size of the initial price drop. ### Why do stablecoins lose their peg? Stablecoins lose their peg for four main reasons: collateral shortfall (reserves are inaccessible or impaired), algorithmic mechanism failure (the mint-burn mechanism enters a death spiral), oracle manipulation or market attack (price feeds are exploited or spoofed), or bank-run panic (coordinated withdrawals driven by fear rather than fundamental impairment). Each cause has a different recovery trajectory, which is why identifying the type of depeg event is the most important first step for any holder evaluating their options. ### How long does a stablecoin depeg typically last? Recovery timelines vary significantly by depeg type. Fiat-backed stablecoins with intact collateral and open redemptions can recover in 24 to 72 hours, as USDC demonstrated in March 2023. Depegs driven by perception gaps and trust damage can take one to three weeks. Algorithmic stablecoins in a death spiral do not recover. There is no universal timeline because the underlying cause determines whether a recovery mechanism exists at all and how long it takes to operate. ### What is the difference between a stablecoin depeg and a collapse? A depeg is a price divergence from the target peg, which may be temporary or permanent. A collapse is the permanent loss of peg value, typically associated with algorithmic stablecoins that enter a reflexive death spiral. The structural distinction is the presence or absence of a hard collateral floor. If hard collateral backs the stablecoin at or near face value, the depeg reflects panic rather than fundamental value destruction and is mechanically recoverable. If no hard collateral floor exists, there is no mechanical force to restore the peg and the event is a collapse, not a depeg. ### Did USDC fully recover from the 2023 SVB depeg? Yes. USDC recovered to $1.00 by Monday March 13, 2023, approximately 60 hours after the initial depeg began late Friday night. The recovery was driven by the FDIC's decision to guarantee all SVB deposits, announced Sunday March 12 at 6:15pm Eastern. Circle's transparency about the $3.3 billion exposure and continued operation of its redemption gateway meant that once the guarantee was confirmed, arbitrage pressure restored the price within hours. Circle subsequently diversified its banking partners to reduce concentration risk in future stress events. ### Why couldn't UST recover the way USDC did? UST had no hard collateral backing. Its peg relied entirely on a mint-burn mechanism with LUNA as the companion token. When UST fell below $1, arbitrageurs could burn UST to receive $1 of newly minted LUNA and sell it, creating buying pressure on UST. Under stress, the LUNA selling from this arbitrage drove LUNA's price down faster than the mechanism could restore UST's peg, triggering hyperinflation of LUNA supply. The Luna Foundation Guard deployed approximately $1.5 billion in BTC reserves, but the scale of the spiral overwhelmed that intervention. Without a collateral floor, there was no arithmetic basis for recovery. ### What should I do if a stablecoin in my yield position depegs? Identify the backing type first. For fiat-backed stablecoins: check whether redemptions are open, find the issuer's official statement, and calculate the actual collateral coverage ratio before deciding to exit. For algorithmic stablecoins: exit immediately if you cannot verify a hard collateral floor. Also assess contagion exposure across your broader DeFi positions, since a depeg in one stablecoin can trigger liquidations in positions where that stablecoin is held as collateral, even if you do not hold the depegged asset directly. ### Can arbitrage always restore a stablecoin's peg? No. Arbitrage restores a peg only when two conditions hold: the issuer's redemption gateway is open, and the underlying collateral is actually worth approximately $1 per stablecoin. If redemptions are halted or the collateral is structurally impaired, arbitrage has no effective mechanism. For algorithmic stablecoins in a death spiral, the arbitrage mechanism itself accelerates the collapse by minting and immediately selling companion tokens, which is the opposite of peg restoration.
Conclusion
The stablecoin depeg lifecycle follows predictable patterns based on collateral structure. Recoverable events share three features: intact hard collateral, an open redemption gateway, and issuer communication speed. Unrecoverable collapses share the opposite: reflexive mechanisms with no hard floor, making intervention mathematically futile above a certain scale. USDC recovered in 60 hours because the collateral was never actually destroyed. UST could not recover because no collateral existed to create a floor. Understanding whether you are in a recoverable or unrecoverable situation is a skill that can be developed before any specific event occurs. It requires knowing what backs a stablecoin, whether the redemption gateway is open, and whether the mechanism is reflexive under stress. That knowledge determines whether holding through a dislocation is prudent patience or the early stage of a permanent loss. Use the [Lince Yield Tracker](https://yields.lince.finance/tracker) to evaluate stablecoin yield opportunities against their underlying peg risk. The tracker surfaces real yield data across protocols and chains, giving you the information to weigh yield premium against stablecoin risk type before capital is committed, not after a peg breaks.