Chapter 24: Stablecoins: The Bridge Between Crypto and the Dollar

24.1 The $150 Billion Question

On any given day, stablecoins settle more transaction volume than PayPal. In 2024, the total value transferred through stablecoins exceeded $10 trillion — a figure that rivals the annual GDP of Japan. Over $150 billion in stablecoins circulate across public blockchains, serving as the connective tissue of the entire cryptocurrency ecosystem. And yet most people outside of crypto have never heard of them.

This is the paradox of stablecoins: they are simultaneously the most boring and the most systemically important product in cryptocurrency. They do not promise 100x returns. They do not generate breathless headlines about price surges. A stablecoin that works perfectly is one whose price chart is a flat line at $1.00. But behind that flat line lies an extraordinary set of engineering challenges, economic assumptions, and regulatory questions that cut to the heart of what money is, who gets to issue it, and what happens when the machinery breaks.

When the machinery does break, the consequences are catastrophic. In May 2022, the algorithmic stablecoin UST lost its dollar peg and triggered a death spiral that destroyed over $40 billion in value in less than a week. That collapse — the Terra/Luna disaster — sent shockwaves through the entire crypto industry, toppling hedge funds, lending platforms, and exchanges like dominoes. It remains the single most instructive failure in cryptocurrency history, and we will dissect it forensically in this chapter.

But the story of stablecoins is not just a story of failure. Stablecoins have become the primary mechanism through which dollars flow into and through the crypto ecosystem. They serve as the base currency for decentralized exchanges. They enable lending and borrowing protocols. They provide dollar access to people in countries with capital controls or hyperinflating currencies. In Argentina, Nigeria, and Turkey, stablecoins are not speculative instruments — they are survival tools.

This chapter examines the three fundamental architectures for stablecoins — fiat-backed, crypto-collateralized, and algorithmic — and evaluates each on its merits and risks. We will trace the full forensic timeline of the Terra/Luna collapse, model the mathematical dynamics of the death spiral, and ask whether USDT, the largest stablecoin by market capitalization, represents a "too big to fail" risk to the broader financial system. By the end, you will understand not just how stablecoins work, but why they matter — and what happens when they don't.

💡 Core Insight: Stablecoins are crypto's attempt to solve the hardest problem in monetary economics — creating a new form of money that is simultaneously stable, decentralized, and capital-efficient. No design has achieved all three.

24.2 Why Stablecoins Matter

To understand the importance of stablecoins, consider what the crypto ecosystem looks like without them.

The on-ramp and off-ramp problem. Traditional financial rails — bank wires, ACH transfers, credit card networks — operate on business hours, settle in days, and charge fees that vary by jurisdiction. When a trader in Seoul wants to buy Bitcoin at 2 AM on a Saturday, she cannot wire dollars to an exchange. But she can swap USDT for BTC instantly on any exchange in the world. Stablecoins function as the always-on, borderless bridge between fiat currency and cryptocurrency. They are the on-ramp that never closes.

DeFi collateral. As we explored in Chapter 21, decentralized finance protocols require collateral to function. You cannot borrow against thin air. The dominant collateral assets in DeFi are ETH and stablecoins. When a user deposits USDC into Aave to earn yield, or provides DAI-ETH liquidity on Uniswap, they are relying on the stability of that dollar peg. If the stablecoin depegs, collateral values collapse, triggering cascading liquidations across interconnected protocols. Stablecoins are the foundation upon which the DeFi edifice rests.

Trading pairs. On centralized exchanges, the most liquid trading pairs are denominated in stablecoins. BTC/USDT is the highest-volume trading pair in cryptocurrency. Without stablecoins, traders would need to convert to fiat for every trade — a slow, regulated, and expensive process. Stablecoins enable 24/7 trading with instant settlement and near-zero fees.

Cross-border payments. Sending $10,000 from the United States to the Philippines through traditional banking channels costs $200-500 in fees and takes 3-5 business days. Sending $10,000 in USDC on the Stellar or Solana network costs less than $0.01 and arrives in seconds. For the $700 billion global remittance market, stablecoins represent a profound efficiency gain — one that is already being captured by services like Circle's cross-border payment infrastructure and Stellar-based remittance corridors.

Dollar access in emerging markets. Perhaps the most consequential use case for stablecoins exists beyond the crypto ecosystem entirely. In countries experiencing currency instability — Argentina (100%+ annual inflation in 2023-2024), Nigeria (naira devaluation and capital controls), Turkey (lira depreciation) — citizens use stablecoins as a store of value. When you cannot trust your local currency and you cannot open a US bank account, a USDC wallet on your phone becomes your savings account. This is not a theoretical use case. Chainalysis data shows that emerging markets consistently rank among the highest for stablecoin adoption relative to GDP.

Settlement volume. In aggregate, stablecoins settled approximately $10.8 trillion in on-chain transaction volume in 2024. To put this in perspective, Visa processed approximately $12.3 trillion in the same period. Stablecoins are no longer a niche crypto instrument — they are a globally significant payment rail.

The stablecoin trilemma. Before we examine specific designs, it is worth naming the fundamental constraint that governs all stablecoin architecture. Every stablecoin designer faces a trilemma among three desirable properties:

1. Price stability — maintaining a tight peg to the reference asset under all market conditions.
2. Decentralization — not depending on a single issuer, a trusted third party, or a permissioned banking relationship.
3. Capital efficiency — not requiring significantly more than $1 of collateral to back each $1 of stablecoin.

No design has achieved all three. Fiat-backed stablecoins excel at stability and capital efficiency but sacrifice decentralization. Crypto-collateralized stablecoins gain decentralization but sacrifice capital efficiency (150%+ collateral required). Algorithmic stablecoins aim for decentralization and capital efficiency but sacrifice stability — catastrophically, as Terra demonstrated. Understanding this trilemma is essential to evaluating any stablecoin, because every design is making a choice about which property to compromise.

📊 By the Numbers: As of early 2025, USDT accounts for approximately 60% of all stablecoin market capitalization ($90B+), USDC holds roughly 25% ($30B+), and DAI represents about 3% ($5B+). The remaining 12% is distributed across dozens of smaller stablecoins, each with its own architecture and risk profile.

24.3 Fiat-Backed Stablecoins: The Centralized Solution

The simplest stablecoin design is also the most controversial: hold dollars in a bank account, issue tokens against them one-to-one, and let authorized participants redeem tokens for dollars. This is the fiat-backed model, and it powers the two largest stablecoins in existence.

24.3.1 USDT (Tether): The $90 Billion Elephant

Tether was launched in 2014 as "Realcoin" before rebranding, making it the oldest surviving stablecoin. It is issued by Tether Limited, a company incorporated in the British Virgin Islands with operational ties to the cryptocurrency exchange Bitfinex. As of early 2025, USDT's market capitalization exceeds $90 billion, making it the third-largest cryptocurrency by market cap after Bitcoin and Ethereum.

The peg mechanism for USDT is straightforward in theory. Authorized participants (large institutions and OTC desks) can create new USDT by depositing US dollars with Tether Limited, and they can redeem USDT for dollars at a 1:1 ratio. This creates an arbitrage loop: if USDT trades below $1.00 on exchanges, authorized participants buy cheap USDT and redeem it for a full dollar, profiting from the spread and pushing the price back up. If USDT trades above $1.00, participants deposit dollars and mint new USDT to sell at a premium, pushing the price back down. This is the same authorized participant / creation-redemption mechanism used by exchange-traded funds (ETFs) in traditional finance.

The reserve controversy. The critical question for any fiat-backed stablecoin is: are the dollars actually there? Tether's history on this question is, to put it charitably, troubled.

For years, Tether claimed that every USDT was "backed 1:1 by US dollars held in reserve." In 2019, the New York Attorney General (NYAG) revealed that this was not true. The NYAG's investigation found that Tether had commingled funds with Bitfinex, loaned $850 million to Bitfinex to cover a shortfall from the payment processor Crypto Capital, and that its reserves were not fully backed by cash.

In 2021, Tether settled with the NYAG for $18.5 million and agreed to provide quarterly reserve reports. These reports — called attestations, not audits (a critical distinction we will examine shortly) — revealed that Tether's reserves were composed not primarily of cash, but of commercial paper (short-term corporate debt), secured loans, corporate bonds, and other investments. At one point, Tether held more commercial paper than many major money market funds.

Attestation vs. audit. Tether publishes quarterly attestation reports prepared by BDO Italia. An attestation is a snapshot: it verifies that on a specific date, the reported assets existed. An audit is a comprehensive examination of financial controls, processes, and ongoing operations over a period. Tether has never completed a full audit by a Big Four accounting firm. Critics argue that an attestation is insufficient because it does not reveal what happens between snapshot dates — whether reserves are temporarily shuffled, whether assets are illiquid, or whether the quality of reserves fluctuates. Tether responds that its attestations are transparent and that it has always honored redemptions.

The commercial paper reduction. Under regulatory pressure and market scrutiny, Tether progressively reduced its commercial paper holdings, claiming to have eliminated them entirely by late 2022. Its reserve composition shifted toward US Treasury bills, which now constitute the majority of its reserves. As of 2024, Tether reports that approximately 85% of its reserves are in US Treasuries and cash equivalents. If accurate, this makes Tether's reserves higher quality than most US money market funds — but the "if accurate" caveat persists because no full audit has been completed.

The Bitfinex connection. Tether Limited and the cryptocurrency exchange Bitfinex share common ownership through the holding company iFinex. This relationship has been a persistent source of concern. Critics allege potential conflicts of interest: an exchange that can effectively print its own trading currency. Research by Griffin and Shams (2020) in the Journal of Finance argued that Tether issuance was used to support Bitcoin's price during the 2017 bull run, though Tether disputes these findings.

Despite all controversies, USDT remains dominant. Traders trust it not because they trust Tether Limited, but because the Lindy effect — the longer something survives, the more likely it is to continue surviving — applies with force. USDT has maintained its peg through multiple crypto crashes, regulatory actions, and sustained criticism. For most market participants, this track record matters more than the opacity of its reserves.

24.3.2 USDC (Circle): The Regulated Alternative

USDC was launched in 2018 by Circle, in partnership with Coinbase, through the Centre Consortium. It was explicitly designed as the "regulated" alternative to USDT — transparent reserves, monthly attestation reports from Grant Thornton (and later Deloitte), and compliance with US money transmission laws.

Reserve composition. USDC's reserves are held in segregated accounts at regulated US financial institutions, invested primarily in short-dated US Treasury bills and cash deposits. Circle publishes monthly attestation reports and has undergone examination by Deloitte, a Big Four accounting firm. This transparency premium is real: institutional DeFi participants and regulated entities generally prefer USDC over USDT.

The Silicon Valley Bank depeg. On March 10, 2023, Silicon Valley Bank (SVB) was seized by the FDIC after a bank run. Circle had approximately $3.3 billion of USDC reserves deposited at SVB — roughly 8% of USDC's total reserves at the time. When this became public, USDC depegged, falling to approximately $0.87 on some exchanges. The irony was acute: a stablecoin designed around transparency and regulated banking was damaged precisely because it held reserves at a regulated US bank.

The depeg lasted approximately 48 hours. On March 12, the Federal Reserve, Treasury Department, and FDIC announced that all SVB depositors — including Circle — would be made whole. USDC recovered to $1.00. But the episode revealed a fundamental tension in fiat-backed stablecoin design: holding reserves in the banking system exposes you to banking system risks. The very safety net that is supposed to back the stablecoin can become a source of instability.

We will examine this episode in detail in Case Study 2.

24.3.3 BUSD: Death by Regulation

Binance USD (BUSD) was issued by Paxos Trust Company and branded by Binance, the world's largest cryptocurrency exchange. At its peak, BUSD held a market capitalization exceeding $23 billion. In February 2023, the New York Department of Financial Services (NYDFS) ordered Paxos to stop minting new BUSD, citing "several unresolved issues related to Paxos's oversight of its relationship with Binance." The SEC also issued Paxos a Wells notice, suggesting it viewed BUSD as an unregistered security.

BUSD's market cap plummeted from $16 billion to near zero within months as users redeemed their tokens. The episode demonstrated a crucial risk for fiat-backed stablecoins: regulatory risk. A stablecoin can have perfect reserves, flawless redemption mechanisms, and a rock-solid peg, and still be killed by a regulatory order. The issuer is a single point of failure — and regulators have the power to turn that single point off.

⚠️ Key Risk: Every fiat-backed stablecoin has a centralized issuer who can freeze tokens, blacklist addresses, and be compelled by regulators to halt operations. This is the fundamental trade-off: you get reserve backing and peg stability, but you sacrifice censorship resistance and decentralization.

24.4 Crypto-Collateralized Stablecoins: Decentralization Through Over-Collateralization

The centralization of fiat-backed stablecoins — the issuers, the bank accounts, the regulatory dependencies — motivated a fundamentally different approach: what if you could back a stablecoin entirely with on-chain crypto assets, eliminating the need for a trusted issuer or a bank?

The catch: crypto assets are volatile. You cannot back a $1 stablecoin with $1 of ETH, because if ETH drops 30%, your stablecoin is suddenly backed by only $0.70 of collateral. The solution is over-collateralization — requiring more collateral than the value of stablecoins issued.

24.4.1 DAI and MakerDAO: The Pioneer

DAI, issued by the MakerDAO protocol (now rebranded as Sky), is the original and most successful crypto-collateralized stablecoin. Launched in 2017, DAI maintains its dollar peg through a system of collateralized debt positions (CDPs), now called Vaults, governed by MKR token holders.

How Vaults work. A user deposits collateral — ETH, WBTC, or other approved assets — into a Maker Vault smart contract. Against this collateral, they can mint (borrow) DAI up to a maximum loan-to-value ratio determined by the collateral type. For ETH, the typical collateralization ratio is 150%, meaning you must deposit $150 of ETH to mint $100 of DAI. The user pays an ongoing stability fee (interest rate) on their minted DAI.

Liquidation. If the value of a user's collateral falls below the minimum collateralization ratio — say ETH drops and the ratio hits 145% against a 150% requirement — the Vault is liquidated. Keeper bots (automated programs) bid on the collateral in a Dutch auction, repaying the DAI debt plus a liquidation penalty. This mechanism ensures that DAI remains over-collateralized even during market crashes.

The Peg Stability Module (PSM). In practice, maintaining a precise $1.00 peg through collateral management alone proved insufficient. MakerDAO introduced the PSM, which allows direct 1:1 swaps between DAI and USDC (and other approved stablecoins). The PSM effectively pegs DAI to USDC, which is itself pegged to the dollar. This dramatically improved DAI's peg stability but introduced a philosophical tension: DAI's decentralization depends partly on the centralized USDC. When USDC depegged during the SVB crisis, DAI also depegged — a dependency that troubled decentralization purists.

Multi-Collateral DAI. The original Single-Collateral DAI (SAI) accepted only ETH as collateral. Multi-Collateral DAI, launched in November 2019, expanded accepted collateral to include WBTC, LINK, UNI, real-world assets (RWA), and USDC (through the PSM). This diversification improved DAI's resilience but also increased its complexity and governance surface area.

Governance. MKR token holders govern the Maker Protocol — setting collateralization ratios, stability fees, liquidation penalties, and approving new collateral types. This governance is consequential: incorrect parameter settings can lead to under-collateralization (too aggressive) or capital inefficiency (too conservative). The protocol navigated "Black Thursday" (March 12, 2020) when ETH's price crashed 43% in a single day, causing $8.3 million in under-collateralized liquidations and requiring an emergency debt auction to recapitalize the system.

24.4.2 LUSD (Liquity): The Immutable Design

Liquity's LUSD represents a more radical approach to crypto-collateralized stablecoins. It accepts only ETH as collateral (no governance tokens, no RWA, no centralized stablecoins), has no governance mechanism (parameters are fixed at deployment), and its smart contracts are immutable (no upgrade path). The minimum collateralization ratio is 110%, significantly lower than Maker's 150%, enabled by a novel stability pool and redistribution mechanism.

LUSD is arguably the most decentralized stablecoin in existence — no admin keys, no governance votes, no off-chain dependencies. The trade-off is inflexibility: the protocol cannot adapt to changing market conditions, add new collateral types, or fix bugs. It is also limited to ETH as collateral, constraining its scalability.

24.4.3 GHO (Aave): The Lending Protocol Stablecoin

Aave, the largest decentralized lending protocol, launched GHO in 2023. GHO is minted by borrowers on Aave using their deposited collateral, with interest rates set by Aave governance. Holders of the Aave safety module's staked AAVE (stkAAVE) receive discounted borrowing rates. GHO represents the trend of DeFi protocols issuing their own stablecoins to capture the seigniorage revenue that would otherwise flow to external stablecoin issuers.

24.4.4 The Capital Efficiency Problem

The fundamental limitation of crypto-collateralized stablecoins is capital inefficiency. If you must lock $150 of ETH to mint $100 of DAI, the capital locked in the system earns no return (beyond whatever the ETH price does). This means the effective "cost" of DAI is the opportunity cost of the locked collateral. For a large institution, locking $150 million in ETH to access $100 million in stablecoins is a poor use of capital compared to simply buying USDC, which requires $100 million and provides $100 million.

This capital efficiency disadvantage explains why DAI's market cap ($5 billion) is a fraction of USDT's ($90 billion). The overcollateralization requirement limits scalability — the system can only grow as fast as users are willing to lock up excess collateral. Various innovations attempt to address this: Liquity's 110% minimum ratio (lower than Maker's 150%), Maker's acceptance of USDC through the PSM (effectively borrowing capital efficiency from the fiat-backed model), and the introduction of real-world assets as collateral (which have lower volatility and thus require less overcollateralization).

But the tension persists: every step toward greater capital efficiency requires either accepting more risk (lower collateral ratios, riskier collateral types) or reintroducing centralization (USDC in the PSM, real-world assets that require off-chain custodians). The crypto-collateralized model is caught between its decentralization ideals and the economic realities of capital efficiency.

🔗 Cross-Reference: The MakerDAO Vault system is a specific implementation of the collateralized lending architecture discussed in Chapter 22. The liquidation mechanisms here directly parallel the Aave and Compound liquidation systems covered there.

24.5 Algorithmic Stablecoins: The Impossible Dream?

The most ambitious — and most dangerous — stablecoin architecture attempts to maintain a dollar peg with no reserves at all. Instead of dollars in a bank or crypto in a smart contract, algorithmic stablecoins use market incentives, arbitrage mechanisms, and token supply manipulation to target a $1.00 price. The appeal is obvious: if it works, you get a stablecoin that is fully decentralized, capital-efficient (no over-collateralization required), and scalable without limit. The history of attempts, however, is a graveyard.

24.5.1 The Basic Mechanism: Seigniorage Shares

Most algorithmic stablecoins are based on a variant of the "seigniorage shares" model, first described by Robert Sams in 2014. The concept involves two tokens:

1. The stablecoin — targets $1.00
2. The share/governance token — absorbs volatility and captures seigniorage (the profit from issuing new currency)

When the stablecoin is above $1.00 (excess demand), the protocol mints new stablecoins, increasing supply to push the price down. The seigniorage from this minting accrues to share token holders.

When the stablecoin is below $1.00 (excess supply), the protocol must contract the supply. This is the hard part. Various mechanisms have been proposed: burning stablecoins, issuing bonds/coupons that promise future stablecoins, or — in Terra's case — allowing holders to burn stablecoins in exchange for newly minted share tokens.

The fundamental challenge is asymmetric: expanding supply (when price is above peg) is easy and profitable, but contracting supply (when price is below peg) requires someone to take a loss. This asymmetry means algorithmic stablecoins are inherently fragile. They work as long as confidence is maintained, but when confidence breaks, the contraction mechanism can amplify rather than dampen the crisis.

24.5.2 The Graveyard

Before Terra, the algorithmic stablecoin graveyard was already full:

• Basis (2018): Raised $133 million, never launched, returned funds to investors after concluding that its "bond" mechanism would be classified as a security.
• Empty Set Dollar (ESD, 2020): Depegged and never recovered.
• Iron Finance/TITAN (2021): Suffered a bank run in which TITAN's price went from $65 to near zero in hours. Mark Cuban, who had been publicly yield farming on the protocol, called for regulation after losing money.
• Fei Protocol (2021): Used a "direct incentive" mechanism that punished sellers during depegs, creating user outrage and ultimately failing to maintain stability.

Each failure shared common features: the peg held during calm markets, broke during stress, and the recovery mechanism failed precisely when it was needed most.

24.6 The Terra/Luna Collapse: A Forensic Analysis

The Terra/Luna collapse of May 2022 is not merely the largest algorithmic stablecoin failure — it is one of the largest financial collapses in cryptocurrency history, and one of the most instructive case studies in mechanism design failure. We examine it in forensic detail because understanding why it failed is essential to evaluating any future stablecoin design.

24.6.1 The Mechanism

Terra was a Layer-1 blockchain built on the Cosmos SDK. Its ecosystem centered on two tokens:

• UST (TerraUSD): The algorithmic stablecoin, targeting $1.00.
• LUNA: The native token of the Terra blockchain, serving as the volatility absorber.

The peg mechanism was elegant in its simplicity: at any time, a user could burn $1 worth of LUNA to mint 1 UST, or burn 1 UST to mint $1 worth of LUNA. This created an arbitrage loop:

• **If UST > $1.00:** Arbitrageurs burn $1 of LUNA to mint 1 UST, sell UST for >$1.00, pocket the difference. This increases UST supply, pushing its price down toward $1.00.
• **If UST < $1.00:** Arbitrageurs buy UST below $1.00, burn it to mint $1 of LUNA, sell LUNA for $1.00, pocket the difference. This decreases UST supply, pushing its price up toward $1.00.

The mechanism relies on a critical assumption: LUNA must have a stable, independent source of value. If LUNA's value depends on the Terra ecosystem's health, and the ecosystem's health depends on UST maintaining its peg, you have a circular dependency — a doom loop waiting to activate.

24.6.2 The Anchor Protocol: The 20% Bait

Anchor Protocol was a lending and borrowing platform on Terra that offered approximately 20% APY on UST deposits. This yield was not generated by productive lending activity — the demand for borrowing was far lower than the supply of deposits. The yield was subsidized by the Luna Foundation Guard (LFG) and Terraform Labs, burning through reserves to maintain an unsustainably high rate.

The 20% yield served as the primary demand driver for UST. At its peak, over 70% of all UST in circulation was deposited in Anchor. This meant that UST's demand was not organic (people using it as money) but artificial (people chasing yield). When the yield subsidy inevitably ran out, the demand floor would collapse.

In March 2022, Anchor governance voted to implement a dynamic rate that would gradually reduce the yield. The rate began declining from 20% toward 18%, and deposits started to outflow. This was the slow leak before the dam break.

24.6.3 The Attack (May 7-8, 2022)

The precise trigger of the UST depeg remains debated, but on-chain analysis reveals the following sequence:

May 7: Approximately $150 million of UST was withdrawn from Anchor Protocol and sold through the Curve Finance UST-3pool (a stablecoin liquidity pool on Ethereum). The pool became imbalanced — heavy on UST, light on other stablecoins — signaling selling pressure. UST dipped to $0.985.

May 8: An additional ~$350 million in UST was dumped into Curve pools. The Luna Foundation Guard (LFG) began deploying its Bitcoin reserves to defend the peg, lending $750 million in BTC to market makers with instructions to buy UST. But the selling continued. UST fell to $0.975.

Whether this initial selling was a coordinated "attack" by a single entity (the "attacker" narrative) or simply panicked selling by large holders who had already decided to exit (the "bank run" narrative) remains debated. On-chain forensics suggest elements of both: concentrated selling by a small number of wallets triggered broader panic.

24.6.4 The Death Spiral (May 9-13, 2022)

May 9 — the first depeg: UST lost its peg decisively, falling to $0.90 and briefly to $0.60 before recovering to $0.90. The arbitrage mechanism was working — people were burning UST for LUNA — but at devastating scale. As UST was burned, massive quantities of new LUNA were minted, flooding the market with LUNA supply and crashing its price.

Here is the mathematical doom loop:

1. UST falls below $1.00.
2. Arbitrageurs burn UST to mint $1 worth of LUNA.
3. The newly minted LUNA is immediately sold (arbitrageurs want dollars, not LUNA exposure).
4. Massive LUNA selling crashes LUNA's price.
5. Because LUNA's price is lower, even more LUNA must be minted per UST burned (burning 1 UST mints $1 of LUNA — at a lower LUNA price, that means more LUNA tokens).
6. More LUNA minting leads to more LUNA selling, which further crashes LUNA's price.
7. The falling LUNA price undermines confidence in the Terra ecosystem, causing more UST holders to panic and sell.
8. More UST selling means more UST below peg, which means more UST-to-LUNA burns.
9. Return to step 3. The spiral accelerates.

This is the reflexive death spiral. Each step in the arbitrage mechanism — which is supposed to restore the peg — instead accelerates the collapse. The faster the arbitrage works, the faster LUNA's price falls, which creates even more UST selling pressure.

The numbers are staggering:

• On May 7, LUNA's circulating supply was approximately 345 million tokens at ~$80 each (market cap ~$28 billion).
• By May 13, LUNA's supply had hyperinflated to 6.53 trillion tokens, each worth fractions of a cent.
• LUNA went from $80 to $0.00001 — a decline of 99.99999%.
• UST went from $1.00 to $0.04.

May 10: The LFG deployed its remaining Bitcoin reserves — approximately $3 billion worth — in a futile attempt to defend the peg. The Bitcoin was sold into the market, contributing to a broader crypto selloff (BTC fell from $36,000 to $29,000 that week). The reserves evaporated without restoring UST's peg.

May 11-12: The Terra blockchain was halted twice to prevent governance attacks, as the cost of acquiring a majority stake in LUNA (now nearly worthless) had become trivially low. Trading in UST and LUNA was halted on major exchanges.

May 13: It was over. UST was at $0.17 and falling. LUNA was effectively worthless. The Luna Foundation Guard reported that its $3.5 billion in reserves — including 80,394 BTC — had been almost entirely depleted in the failed defense. Over $40 billion in combined UST and LUNA market capitalization had been destroyed.

24.6.5 The Human Cost

Behind the numbers were real people. The Terra ecosystem had attracted a particularly broad base of retail participants, drawn by the Anchor Protocol's 20% yield. Many were not sophisticated crypto traders — they were ordinary people attracted to what appeared to be a high-yield savings account. Stories emerged of retirees in South Korea who had placed their life savings in Anchor, of small business owners who had parked operating capital in UST, of individuals in developing countries who had chosen UST over their local currency.

Do Kwon's reaction during the crisis oscillated between defiance and desperation. On May 9, as UST first depegged, he tweeted "Deploying more capital — steady lads." By May 11, as the spiral accelerated beyond all possibility of recovery, his tone shifted. After the collapse, Kwon proposed "Terra 2.0" — a new blockchain that would airdrop tokens to holders of the now-worthless LUNA and UST. This proposal was approved by community governance, though Terra 2.0 has had minimal adoption.

In March 2023, Do Kwon was arrested in Montenegro while attempting to travel on a fraudulent Costa Rican passport. He was extradited to the United States, where the SEC had charged him and Terraform Labs with securities fraud. In January 2025, a US jury found Terraform Labs liable. The SEC alleged that Kwon had misrepresented UST's stability mechanism and had concealed the role of Jump Trading in secretly restoring the peg during a previous UST depeg in May 2021 — an event that Kwon had publicly cited as proof that the mechanism worked autonomously.

24.6.5 The Contagion

The Terra collapse was not contained. It sent dominoes falling across the crypto industry:

• Three Arrows Capital (3AC): The $10 billion crypto hedge fund had significant LUNA exposure and was unable to meet margin calls. It defaulted on loans from multiple lenders and filed for bankruptcy in July 2022. Its founders, Su Zhu and Kyle Davies, fled to avoid creditors.
• Celsius Network: The crypto lending platform, which had lent to 3AC and had exposure to stETH illiquidity, froze withdrawals on June 12, 2022, and filed for bankruptcy in July.
• Voyager Digital: The crypto brokerage had lent $650 million to 3AC. When 3AC defaulted, Voyager froze withdrawals and filed for bankruptcy.
• BlockFi: The crypto lender received an emergency line of credit from FTX (itself later revealed to be insolvent) and eventually filed for bankruptcy.

The total destruction of value from the Terra collapse and its contagion effects is estimated at $60-80 billion. It triggered the crypto "credit crisis" of 2022, which in turn contributed to the conditions that led to FTX's collapse in November of that year.

24.6.6 What the Design Got Wrong

The Terra/Luna mechanism had several fundamental design flaws, all of which were identified by critics before the collapse:

Circular value dependency. LUNA's value depended on the Terra ecosystem's fees, adoption, and growth. The ecosystem's growth depended on UST maintaining its peg. UST's peg depended on LUNA having value. This is a circular dependency with no external anchor. In calm markets, the circle reinforces positively (growth begets confidence begets growth). In stressed markets, it reinforces negatively (fear begets depegging begets destruction of LUNA value begets more fear).

No terminal backing. If all UST holders simultaneously tried to redeem, there was no pool of external assets to satisfy them. The "backing" was LUNA, whose value would be destroyed by the very act of mass redemption. Fiat-backed stablecoins have a terminal backstop: actual dollars. Crypto-collateralized stablecoins have over-collateralization buffers. Algorithmic stablecoins backed by their own ecosystem token have nothing external to fall back on.

Artificial demand through unsustainable yield. The 20% Anchor yield created fragile demand. When the yield declined, demand evaporated, leaving the mechanism to handle massive net selling — exactly the scenario it was least equipped to manage.

Bank run dynamics. Like a fractional reserve bank, Terra/UST was vulnerable to self-fulfilling crises. If enough people believe the peg will break, they sell UST. Their selling breaks the peg. The rational individual response (sell before it crashes further) is collectively destructive. The mechanism had no circuit breaker to halt this reflexive spiral.

The burn/mint was too fast. The protocol allowed unlimited minting of LUNA as UST was burned. This meant the hyperinflationary spiral had no speed limit. Some subsequent algorithmic designs have proposed rate-limiting the burn/mint mechanism to slow spirals, though it is unclear whether this prevents collapse or merely extends the agony.

⚠️ The Mathematical Inevitability: Once the reflexive death spiral began — once LUNA was being minted faster than it could find buyers — the outcome was mathematically determined. No amount of reserve deployment could overcome the exponential growth of LUNA supply. The LFG's $3 billion in reserves was a finite resource fighting an infinite feedback loop. See code/terra_death_spiral.py for a mathematical simulation of this dynamic.

24.7 How Stablecoins Maintain (and Lose) Their Peg

Understanding peg maintenance requires understanding the different mechanisms across stablecoin types and the conditions under which each breaks.

24.7.1 Fiat-Backed: Arbitrage

For USDT and USDC, the primary peg mechanism is authorized participant arbitrage, as described in Section 24.3. This mechanism is robust because the arbitrage is backed by a real asset (dollars in a bank account). The risk is not that the mechanism fails in theory, but that the real asset becomes inaccessible — because the bank fails (SVB), because the issuer is lying about reserves (Tether controversy), or because regulators freeze the accounts (BUSD).

The USDC depeg during SVB illustrates the boundary condition: when 8% of reserves became temporarily inaccessible, the market priced in the risk that Circle might not be able to honor all redemptions. The peg was restored only when the US government guaranteed all depositors.

24.7.2 Crypto-Collateralized: Liquidation

For DAI and LUSD, the peg is maintained through the over-collateralization buffer. As long as collateral value exceeds outstanding stablecoins, the system is solvent. The liquidation mechanism enforces this by automatically selling collateral when it drops below the minimum ratio.

The risk is a "Black Swan" event where collateral value drops so fast that liquidations cannot be processed in time. On Black Thursday (March 12, 2020), network congestion on Ethereum caused DAI liquidation auctions to process at near-zero bids — liquidators were buying collateral for essentially nothing, and the system briefly became under-collateralized. MakerDAO had to conduct an emergency MKR auction to recapitalize. This experience led to significant redesign of the liquidation system (Liquidations 2.0, using Dutch auctions).

24.7.3 Algorithmic: Reflexive Mechanisms

For algorithmic stablecoins, the peg mechanism relies on market confidence and rational arbitrageur behavior. As Terra demonstrated, these mechanisms work in calm conditions but can enter death spirals under stress. The fundamental problem is that the mechanism's strength is proportional to market confidence, and market confidence is what collapses first in a crisis.

The confidence threshold. Every algorithmic stablecoin has an implicit confidence threshold below which the mechanism breaks. Above the threshold, arbitrageurs trust that the peg will be restored and actively buy the underpriced stablecoin. Below the threshold, arbitrageurs conclude that the peg will not hold and either exit or actively short. The difference between survival and death spiral is whether the selling pressure pushes the system past this threshold. The threshold is not a fixed number — it is a function of market sentiment, liquidity depth, and the perceived credibility of the protocol team.

The asymmetry problem. There is a deep asymmetry at the heart of algorithmic stablecoin design that no mechanism has overcome. Expanding supply when price is above the peg is easy: you mint new tokens, and the seigniorage is pure profit for someone. Contracting supply when price is below the peg is hard: someone must absorb a loss. In the Terra model, LUNA holders absorb the loss through dilution. In bond/coupon models, coupon buyers accept the risk that their coupons may never be redeemed. In every case, the contraction mechanism requires a willing counterparty to take the other side of a losing trade — and in a crisis, those counterparties disappear.

This asymmetry means algorithmic stablecoins can grow easily (expansion is profitable) but shrink dangerously (contraction requires sacrifice). They are mechanisms that work beautifully on the way up and break catastrophically on the way down — a pattern familiar from many financial instruments throughout history.

24.7.4 The Stablecoin Hierarchy of Needs

Synthesizing across all three architectures, we can articulate a hierarchy of what makes a stablecoin robust:

1. Terminal backing — Is there an asset of independent value that can satisfy all redemptions? Fiat-backed stablecoins have dollars. Crypto-collateralized have overcollateralized crypto. Algorithmic stablecoins have... the mechanism itself, which is circular.

2. Redemption credibility — Does the market believe that redemptions will be honored? This depends on reserve transparency, issuer track record, and regulatory standing.

3. Liquidity depth — Is there enough trading liquidity to absorb large sell orders without significant price impact? Thin liquidity amplifies shocks.

4. Contagion isolation — Is the stablecoin isolated from risks in other parts of the ecosystem, or are there dependency chains (as DAI depends on USDC)?

A stablecoin that fails at level 1 (no terminal backing) is fundamentally fragile regardless of how well it performs at levels 2-4. This is why the algorithmic model has failed repeatedly: without terminal backing, every other layer of the hierarchy is built on sand.

24.8 Stablecoin Regulation: The Reckoning

The Terra collapse and the broader crypto crisis of 2022 galvanized regulatory action on stablecoins worldwide. Regulators who had been studying stablecoins since 2019 suddenly had a $40 billion catastrophe to point to as justification for intervention.

24.8.1 MiCA (EU)

The European Union's Markets in Crypto-Assets (MiCA) regulation, which came into full effect in December 2024, includes specific provisions for stablecoins (which MiCA calls "electronic money tokens" or EMTs and "asset-referenced tokens" or ARTs):

• Reserve requirements: Stablecoin issuers must maintain reserves equal to 100% of outstanding tokens, held in segregated custody.
• Redemption rights: Holders must be able to redeem stablecoins at face value at any time.
• Licensing: Issuers must be authorized as electronic money institutions or credit institutions.
• Market cap limits: For stablecoins pegged to non-euro currencies that exceed certain daily transaction volumes or market cap thresholds, additional restrictions apply — a provision clearly aimed at limiting the use of USD-pegged stablecoins in Europe.

MiCA has already had market effects. Tether delisted USDT from regulated European exchanges in advance of MiCA compliance deadlines, though it remains available on non-EU platforms. Circle obtained an Electronic Money Institution (EMI) license in France, positioning USDC as the MiCA-compliant dollar stablecoin in Europe.

24.8.2 United States

US stablecoin regulation has been more fragmented, with multiple agencies claiming jurisdiction:

• The SEC has at times argued that certain stablecoins are securities (the BUSD Wells notice to Paxos).
• The CFTC has argued that stablecoins are commodities when used as trading collateral.
• The OCC issued interpretive letters in 2021 allowing national banks to hold stablecoin reserves and participate in blockchain networks.
• Congressional legislation: Multiple stablecoin bills have been introduced, including the Clarity for Payment Stablecoins Act and similar proposals. These generally propose Federal Reserve oversight for large stablecoin issuers, reserve requirements (cash and short-term Treasuries), and a prohibition on algorithmic stablecoins. As of early 2025, comprehensive stablecoin legislation has been advanced but the final regulatory framework continues to evolve.

24.8.3 The Algorithmic Stablecoin Ban Question

Several regulatory proposals, in the US and elsewhere, have included provisions to ban or severely restrict algorithmic stablecoins. The Terra collapse provides powerful ammunition for this position. However, critics argue that banning algorithmic stablecoins is both technically difficult (how do you define the boundary between an algorithmic stablecoin and a crypto-collateralized one?) and philosophically problematic (should regulators prohibit experimentation with novel monetary mechanisms, given that the history of money is a history of experimentation?). The practical enforcement challenge is also significant: algorithmic stablecoins can be deployed on permissionless blockchains by anonymous developers, making a blanket ban difficult to implement. The debate continues, with most regulatory frameworks trending toward strict reserve requirements that effectively prohibit the pure algorithmic model without naming it explicitly.

⚖️ Regulatory Tension: Stablecoin regulation faces a fundamental trilemma. Regulators want stablecoins to be (1) fully backed and safe, (2) subject to bank-like oversight, and (3) open and innovative. Achieving all three simultaneously may be impossible — heavy regulation favors large incumbents and may push activity offshore.

24.9 USDT: Too Big to Fail?

As of early 2025, USDT is embedded in the infrastructure of the crypto ecosystem at a level that makes its failure a systemic risk.

The numbers: USDT accounts for approximately 60-70% of stablecoin market cap, 60-75% of stablecoin trading volume, and serves as the base currency for the majority of trading pairs on major exchanges. It is the most used stablecoin in emerging markets, the dominant stablecoin on Tron (where most non-trading stablecoin transfers occur), and the stablecoin with the deepest liquidity across the most platforms.

What happens if USDT fails? Consider the scenario: Tether announces that its reserves are impaired and it cannot honor all redemptions at $1.00. The immediate effects would be:

1. USDT depegs — price drops below $1.00, potentially to $0.50-0.80 depending on the magnitude of the shortfall.
2. Exchange trading pairs break — BTC/USDT, ETH/USDT, and thousands of other pairs become mispriced. Traders holding USDT cannot exit at face value.
3. DeFi liquidation cascades — USDT is used as collateral across lending protocols. A depeg triggers mass liquidations, crashing the prices of other assets.
4. Contagion to other stablecoins — confidence in stablecoins broadly collapses. USDC and DAI may temporarily depeg as panicked holders rush to redeem.
5. Emerging market impact — millions of people using USDT as a dollar savings vehicle lose a portion of their savings.

The counterargument: Tether bulls argue that the scenario is implausible because (a) Tether's reserves are now primarily in US Treasuries, which are the safest asset class in existence; (b) Tether has consistently honored redemptions through every crisis; (c) Tether generated over $6 billion in profit in 2023-2024, giving it substantial equity buffer; and (d) Tether's demise would harm so many actors in the crypto ecosystem that there are strong incentives for rescue.

The Tron factor. An often overlooked dimension of USDT's systemic importance is its dominance on the Tron blockchain. As of 2024, more USDT circulates on Tron than on any other chain, including Ethereum. Tron's low transaction fees (fractions of a cent) make it the preferred network for peer-to-peer stablecoin transfers, particularly in emerging markets. In Nigeria, Turkey, and across Southeast Asia, USDT on Tron is the de facto digital dollar. This means that USDT's systemic importance extends beyond the crypto trading ecosystem into the daily financial lives of millions of people who may have no involvement with DeFi or cryptocurrency trading.

The money market fund analogy. To understand the systemic risk of USDT, consider the analogy to money market funds in traditional finance. In September 2008, the Reserve Primary Fund "broke the buck" — its net asset value fell below $1.00 per share due to holdings of Lehman Brothers commercial paper. This triggered a panic across the $3.5 trillion money market fund industry, threatening to freeze short-term credit markets and prompting the US Treasury to guarantee money market fund shares. USDT is, in many ways, the crypto ecosystem's equivalent of a giant money market fund — unregulated, uninspected, and embedded in every corner of the financial system it serves.

The honest assessment: Nobody outside of Tether's inner circle knows with certainty what Tether's reserves look like on any given day. The attestation reports are snapshots. No Big Four audit has been completed. Until that changes, USDT's risk is fundamentally unquantifiable — not necessarily high, but unknowable. In risk management, unknowable risks deserve more caution, not less.

24.10 The Future of Stablecoins

The stablecoin market is evolving rapidly along several dimensions:

Yield-bearing stablecoins. A new category of stablecoins distributes the yield from underlying reserves to holders. If Tether earns 5% on $90 billion in Treasuries, that is $4.5 billion per year in revenue that accrues to Tether Limited, not USDT holders. Yield-bearing stablecoins (like Mountain Protocol's USDM, Ondo's USDY, or Ethena's USDe) pass some of this yield through. The regulatory question — whether a yield-bearing stablecoin is a security — remains unresolved but is critical to the category's growth.

Ethena's USDe. Ethena's USDe represents a novel approach: a "synthetic dollar" backed by delta-neutral positions (holding staked ETH while shorting ETH perpetual futures). The yield comes from staking rewards and the funding rate premium on perpetual futures. With over $3 billion in TVL by early 2025, USDe demonstrated rapid product-market fit but also introduced new risk vectors: negative funding rates, exchange counterparty risk, and smart contract risk. Ethena is not technically a stablecoin — it is a structured product — but the market uses it as one.

Real-world asset (RWA) backed stablecoins. Stablecoins backed by tokenized Treasury bills, money market fund shares, or other real-world assets represent a convergence between traditional finance and crypto. These designs attempt to combine the transparency of on-chain assets with the safety of traditional fixed-income instruments.

CBDC competition. Central bank digital currencies (discussed in Chapter 32) could compete with or complement stablecoins. A digital dollar issued by the Federal Reserve would eliminate the reserve risk entirely — but would also eliminate the privacy, programmability, and permissionless access that make stablecoins attractive. Whether CBDCs replace stablecoins or coexist with them is one of the defining questions of digital currency policy.

Multi-chain expansion. Stablecoins are increasingly deployed across multiple blockchains via native issuance or bridging. Circle's Cross-Chain Transfer Protocol (CCTP) enables native USDC burning and minting across chains, eliminating the need for wrapped tokens. This multi-chain future increases stablecoin utility but also introduces cross-chain bridge risks.

The seigniorage revenue question. One of the most commercially significant dynamics in the stablecoin market is the question of who captures seigniorage — the profit from issuing money-like instruments. Tether earns the yield on its $90+ billion in reserves (primarily US Treasury bills yielding approximately 4-5% in 2024-2025). This generates billions in annual revenue — Tether reported $6.2 billion in net profit for 2023 alone, making it more profitable per employee than almost any company in the world. This revenue accrues entirely to Tether Limited's shareholders; USDT holders receive none of it. The same dynamic applies to Circle and USDC, though Circle's profit figures are smaller due to USDC's smaller market cap.

This seigniorage capture is the business model driving the entire stablecoin industry and is the primary motivation behind yield-bearing stablecoins (which share seigniorage with holders), DeFi protocol stablecoins like GHO (which capture seigniorage for the protocol), and bank-issued stablecoins (which allow traditional financial institutions to participate). The question of who deserves the yield generated by stablecoin reserves — the issuer, the holders, or some combination — will likely be resolved through market competition and regulatory mandates in the coming years.

24.11 Summary and Key Takeaways

Stablecoins are the circulatory system of the crypto economy — the mechanism through which value flows between traditional finance and decentralized protocols, between trading pairs on exchanges, and between savers in emerging markets and the stability of the US dollar.

The three stablecoin architectures represent different positions on the trade-off space:

The Terra/Luna collapse demonstrated that algorithmic stablecoins backed by their own ecosystem token contain a fatal reflexive loop: the mechanism designed to restore the peg can instead accelerate its destruction when confidence breaks. The $40 billion loss and the contagion that followed — Three Arrows Capital, Celsius, Voyager, and ultimately contributing to FTX's collapse — make it the most consequential mechanism design failure in crypto history.

Fiat-backed stablecoins dominate the market but carry centralization and regulatory risk. Crypto-collateralized stablecoins offer decentralization but at the cost of capital efficiency. No stablecoin has solved the trilemma of stability, decentralization, and capital efficiency simultaneously.

As stablecoins settle trillions in annual volume and serve as dollar access for millions in emerging markets, the question of USDT's reserve transparency — and the systemic risk it represents — is no longer a crypto-insider debate. It is a question of global financial stability.

🔗 Bridge to Chapter 25: In the next chapter, we examine lending and borrowing protocols — the DeFi applications that depend most critically on stablecoin stability. We will see how Aave, Compound, and MakerDAO manage risk, and how stablecoin failures cascade through the lending stack.

Key Concepts Review

• Stablecoin: A cryptocurrency designed to maintain a stable value relative to a reference asset, typically the US dollar.
• Peg: The target price a stablecoin attempts to maintain (usually $1.00).
• Fiat-backed stablecoin: Backed by reserves of fiat currency or equivalent assets held by a centralized issuer.
• Crypto-collateralized stablecoin: Backed by cryptocurrency assets locked in smart contracts, typically over-collateralized.
• Algorithmic stablecoin: Attempts to maintain peg through supply/demand manipulation without external reserves.
• CDP/Vault: A collateralized debt position in which a user locks collateral to mint stablecoins.
• Death spiral: A reflexive feedback loop in which a stablecoin's peg-restoration mechanism accelerates rather than reverses the depeg.
• Seigniorage: The profit earned from issuing new currency, captured by share/governance token holders in algorithmic designs.
• Attestation: A point-in-time verification of reserve holdings, less comprehensive than a full audit.
• Depeg: When a stablecoin's market price deviates significantly from its target value.
• PSM (Peg Stability Module): MakerDAO's mechanism for direct 1:1 swaps between DAI and approved stablecoins.

Chapter 24 of Blockchain and Cryptocurrency: A Complete Introduction. Part V: DeFi.