Terra/Luna fall

Design and ascent of UST/LUNA

The algorithmic dream

Terra sold a seductive idea: a dollar-pegged stablecoin (UST) without dollars. If UST traded above $1, you could burn $1 of LUNA to mint one UST; if it slipped below, you could burn UST to mint $1 of LUNA. Arbitrage was supposed to keep the peg, and as demand for UST grew, so would LUNA’s price—reflexivity as a feature. In a bull market, that feedback loop felt like free energy. Every new UST printed meant more LUNA burned; the chart went up and so did the confidence.

Whitepapers spoke of “seigniorage” and “algorithmic stability,” but the real pitch was simpler: a decentralized dollar that paid you to hold it. The mechanism depended on belief that LUNA would always be valuable enough to absorb redemptions. As long as the music played, it worked. When it stopped, there was no reserve under the stage.

Anchor’s 20%: the marketing budget

The accelerant was Anchor, Terra’s “savings” protocol that paid ~20% APY on UST deposits. That number was not magic; it was a subsidy from the Anchor “yield reserve,” topped up by Terraform Labs and investors. In practice, Anchor was a customer acquisition cost masquerading as passive income. Billboards in Seoul advertised “safe 20%,” and crypto Twitter compared Anchor to a high-yield savings account instead of a venture-funded loss leader.

Deposits poured in. By early 2022, Anchor held over $14B UST. The yield reserve drained faster than it was replenished, but few noticed; dashboards showed a serene 19.5% and the assumption that “someone smart will keep it topped up.” The reflexive loop tightened: more UST printed to meet deposit demand, more LUNA burned, higher LUNA price, and even more confidence that the peg was unbreakable.

Bridges spread the risk

UST didn’t stay on Terra. It flowed through Wormhole and other bridges into Ethereum DeFi, parking in Curve’s 3pool and later the planned 4pool alongside FRAX. Protocol treasuries held UST for “diversification,” DAOs used it as a pseudo-stable reserve, and market makers treated it as interchangeable with USDC. Exposure fanned out quietly; few asked what a bank run on UST would do to their own balance sheet.

Terraform Labs sensed the growing paranoia about “backing” and started buying Bitcoin for the Luna Foundation Guard (LFG) reserve. The plan: if UST wobbled, LFG would deploy BTC to defend the peg. It was a marketing masterstroke—“algorithmic, but also backed by hard money”—but it didn’t change the core math: if redemptions outpaced confidence, newly minted LUNA would have to absorb the shock.

Depeg, death spiral, and collapse

The first cracks

In early May 2022, liquidity in Curve’s UST pools thinned as Terra moved funds to prep the 4pool launch. A few large sellers pushed UST below $1. Traders noticed; some called it an attack, others saw a market test. Arbitrageurs started burning UST for LUNA, increasing LUNA supply to defend the peg. The reflexive loop that looked virtuous on the way up turned vicious: more redemptions meant more LUNA minted, which meant lower LUNA price, which meant less value to absorb future redemptions.

Anchor withdrawals accelerated. Depositors who treated UST as a savings account suddenly saw a stablecoin at $0.99 and thought “I’ll be first in line.” The yield reserve was irrelevant now; survival depended on confidence that LUNA’s market cap could cover the rush to the exits.

Death spiral in real time

As UST slipped to $0.90, LFG deployed billions in BTC to buy UST. It slowed the slide for hours, not days. UST redemptions printed LUNA so fast that the supply chart went vertical—billions of new tokens in 48 hours. Price followed the opposite trajectory: from tens of dollars to pennies, then fractions of a cent. Exchanges halted trading, then resumed, then halted again. Terra validators even paused the chain to patch code and prevent “governance attacks” from the flood of new LUNA.

By mid-May, UST was below $0.20, LUNA was effectively zero, and the arbitrage “backstop” had become a money shredder. The BTC reserve was gone. The famous “decentralized central bank” had burned through its assets and trust in a weekend. Screenshots of Do Kwon’s tweets about being “entertaining” critics aged poorly. Livesavings evaporated. Funds that had stuffed treasuries with UST wrote panicked updates. Retail holders flooded forums with disbelief: “How is this possible? It was a stablecoin.”

“I alone am responsible for this decision.” — Do Kwon, as LFG announced reserve deployment

Shockwaves and contagion

The collapse didn’t stop at Terra. Every pool holding UST imbalanced violently; Curve’s 3pool tilted, wiping out stablecoin yields elsewhere. Market makers ate losses. Funds that had borrowed against LUNA or held UST as dry powder faced margin calls. CeFi lenders with Terra exposure tightened credit, seeding the next crisis. Chains that depended on bridged UST saw liquidity vanish overnight. “Algorithmic stablecoin” became a punchline and a policy talking point; even unrelated projects were asked to explain how their coins wouldn’t “do a Terra.”

Investigations began within days. Korean authorities raided offices; class actions piled up; U.S. agencies sniffed around securities and fraud angles. The collateral damage wasn’t just financial—it was reputational. Retail who thought they’d found a safe 20% felt tricked; institutional desks reevaluated every peg they touched. Risk models suddenly included “reflexive death spiral” as a scenario, not a meme.

Fallout and scars

Collateral matters, narratives break

After Terra, the market repriced every stablecoin by its collateral quality. Overcollateralized models (DAI with more ETH/USDC backing than supply) and fiat-backed models with daily attestations gained mindshare. Reflexive designs looked radioactive. Projects like FRAX added exogenous collateral and caps; new algo-stable proposals were met with eye-rolls. “Show me your reserves” became a reflexive question, not a courtesy.

Even non-stable projects internalized the lesson: reflexivity cuts both ways. Token models that assumed infinite demand now had to include “confidence shock” in their spreadsheets. Bridge risk and counterparty risk felt tame compared to an on-chain bank run that printed its own death sentence.

“You can’t algorithm your way out of a bank run if your collateral is just belief.” — A risk officer, post-Terra

Regulators, lawsuits, and the hunt for executives

Lawmakers grabbed Terra as Exhibit A. U.S. senators cited UST in stablecoin hearings to argue for reserve requirements. South Korea issued travel bans and arrest warrants; Interpol notices followed. Civil suits stacked up in multiple jurisdictions. The message to founders: if your token claims to be a dollar, expect banking-style scrutiny and personal liability when it snaps.

Exchanges tightened listings for anything algorithmic. Auditors, once content with “proof of reserves” for custodial stables, started demanding real-time data and bank-like segregation. Insurance conversations surfaced: who eats the loss when a peg breaks? The legal fog thickened, but so did the expectation that “just code” was no shield against consumer protection laws.

Community scars and reconstruction

Terra tried to reboot with “Terra 2.0,” a new chain without UST, airdropping new LUNA to old holders. It landed with a thud; trust was gone. Builders scattered to other ecosystems; some apps forked to different chains. Meanwhile, forensic blogs documented the blow-by-blow: Curve pool imbalances, BTC reserve movements, frantic governance proposals. Those write-ups became cautionary literature for every risk committee that followed.

For users, the scar was personal. Stories of lost college funds and retirement savings circulated alongside memes of “LUNAtics” who’d mocked skeptics. The dark humor turned to quiet anger. The cultural shift was palpable: less swagger about “number go up,” more emphasis on boring audits and circuit breakers.

Lessons etched in stone

Terra’s fall etched a few rules into the industry’s muscle memory: promises of “risk-free” double-digit yields are marketing, not physics; reflexive pegs without thick, exogenous collateral are brittle; transparency beats charisma; and decentralized doesn’t mean unaccountable. The most durable takeaway was humility. Stablecoins are critical plumbing; if they fail, everything built atop them cracks.