rev #undefined (initial)
CHAPTER 5 · PART A
A Stablecoin Dystopia
The trust shift from "I trust my bank" to "I trust this mechanism" is only valid if we're honest
about what mechanisms have failed and can still fail. This chapter is what separates a credible
book from a promotional pamphlet.
What's Already Broken
The Collapse That Proved the Critics Right
May 2022. TerraUST — an algorithmic stablecoin backed not by dollars in a vault, but by a
complex relationship with its sister token LUNA — lost its peg and entered a death spiral. In
days, roughly $40 billion was vaporized: $18 billion in UST market cap and $22 billion in LUNA
value.
The mechanism was supposed to be elegant: if UST dropped below $1, arbitrageurs would burn
UST and mint LUNA, restoring the peg through supply reduction. But when confidence
evaporated, the feedback loop reversed. More UST sold, more LUNA minted, LUNA crashed
99.99%, and the arbitrage couldn't keep up with the panic.
Anchor Protocol — a lending platform built on Terra — had attracted $18 billion in deposits by
offering 20% APY on UST. Twenty percent annual return on what was marketed as a dollar.
When new money stopped flowing in to sustain the yield, the system failed like a bank run
meets a Ponzi scheme.
The contagion rippled outward. Three Arrows Capital became insolvent. Celsius froze
customer withdrawals. The entire crypto credit market crumbled.
The human cost was staggering. A young trader in Colombia lost 100% of his life savings: "The
guilt is unbearable. I've had big drawdowns before, but this time I'm zero, nothing." A disabled
retiree living on $197 per month: "I'm not rich. So this hurts me." The Terra subreddit became
a space of anguish — communities organized suicide prevention outreach for members
expressing despair.
IMF Managing Director Georgieva: "When it's not backed with assets but promises 20%
return, it's a pyramid. And what happens to pyramids? They eventually fall to pieces."
Money carries dreams and fears. When stablecoins fail, the emotional toll is devastating.
The Catalog of Failures
Terra was the largest, but not the only one.
Iron Finance / TITAN (June 2021): A partial-collateral stablecoin — 75% USDC, 25% TITAN
token. TITAN went from roughly $60 to $0 in a single day. Mark Cuban lost around $870,000
providing liquidity. A precursor warning to Terra that the community failed to heed.
USDC (March 2023): Circle disclosed $3.3 billion — 8% of reserves — was stuck at the
collapsing Silicon Valley Bank. USDC fell to $0.87. $4.5 billion was redeemed in days — a
modern digital bank run. It recovered when the FDIC backstopped SVB deposits. The lesson:
even fully-backed stablecoins carry counterparty risk from the banking system they're
supposed to replace.
USDT: Briefly hit roughly $0.95 during the May 2022 panic. Quickly recovered, but showed
even the biggest stablecoin is vulnerable to confidence shocks. USDT also dipped to $0.97 in
2018 during solvency rumors.
Beanstalk (April 2022): A flash loan governance attack. An attacker used a momentary loan to
accumulate 75% of governance votes, passed a proposal to drain roughly $182 million in
collateral, and executed it in a single transaction. $77 million was laundered through Tornado
Cash and never recovered. Bean crashed to $0.12. The lesson: even if the economic model is
sound, operational security can zero you out.
Cashio (March 2022): An infinite mint exploit on Solana. A coding error allowed an attacker to
mint 2 billion CASH tokens without collateral and redeem approximately $52.8 million in real
assets. The attacker called himself a "Robin Hood." The protocol was abandoned.
Cross-chain bridge hacks: Roughly $2.5 billion stolen in bridge hacks between 2021 and 2023.
Wormhole lost $325 million — an attacker minted wrapped ETH on Solana without locking
real ETH on Ethereum. Jump Trading replaced the funds. Nomad lost $190 million in a chaotic
exploit where hundreds of copycat attackers drained the protocol. Bridges are the weakest link
— high-value targets with small multisig security or poorly audited code.
Not all of these failures are "crypto problems" in the way critics frame them. Terra was a bank
run meets Ponzi. USDC was old-fashioned banking counterparty risk. Beanstalk and Cashio
were code exploits. The diversity of failure modes shows that stabilizing value is genuinely
hard. The only stablecoins that haven't broken their peg are fully fiat-backed — and even
they've had moments of stress.
Centralization and Censorship
Circle and Tether CAN and DO freeze wallet addresses. Over 800 combined, often at law
enforcement request.
When Tornado Cash was sanctioned, Circle blacklisted 81+ addresses and froze roughly
$75,000 in USDC — demonstrating that your "self-custodied" digital dollars can be rendered
worthless with one smart contract call.
Tether has frozen over 700 addresses. Your "digital dollars" can be confiscated by the issuer.
For users in sanctioned countries or using privacy tools, this is not a theoretical risk.
The philosophical tension is real: the promise of "money like cash" conflicts with the reality of
centralized freeze capabilities. Rune Christensen, MakerDAO's founder: "The current design is
too dependent on real-world assets like USDC. We need a stablecoin that can survive
blacklisting and censorship — truly decentralized." But truly decentralized stablecoins have
historically been the ones that collapse. The tension doesn't resolve cleanly.
The Tether Problem
Tether is roughly $86 billion — two-thirds of all stablecoin value. "Too big to fail" for crypto
markets, but with no one to bail it out.
$192.9 billion in assets versus $186.5 billion in liabilities, with $6.4 billion in excess reserves.
Reserve composition: $141.6 billion in Treasury bills and overnight repo (safe), $17.4 billion in
gold — quietly one of the world's largest gold holders — $8.4 billion in Bitcoin. But also: $17
billion in secured loans with undisclosed borrowers, up from $5 billion. The borrowers and
collateral are unknown.
No full audit. Only quarterly attestations from a Cayman Islands accountant. Tether's own
report acknowledges: "Our figures are a one-day attestation, not an audit." No Big-4
accounting firm has audited Tether. This keeps many institutions away.
Under DOJ investigation for potential bank fraud — no charges filed. Previous settlements:
$42.5 million CFTC fine for misleading reserve claims, $18.5 million NYAG settlement, barred
from operating in New York. At one point in 2017, Tether was found to have had "virtually no
reserves."
On the other hand: Tether's CEO argues USDT is "the dollar for the last mile, for the unbanked."
Almost the entire user base is in emerging markets. He frames Tether as "advancing US dollar
hegemony across emerging markets."
Profitability: $9-10 billion in interest income in 2025. Net profits exceeding $10 billion.
Revenue of $5.2 billion equaling 42% of all crypto protocol revenue. More profitable per
employee than most global banks.
The stress test it passed: after FTX collapsed, Tether honored over $7 billion in redemptions
within 48 hours without breaking the peg. It froze roughly $835 million in USDT related to
crime.
The stress test it hasn't faced: what happens at scale. If Tether's reserves were found to be
significantly short, or if a regulatory action froze its banking relationships, the forced
liquidation of $141 billion in Treasury bills could spill into traditional markets. The FSB has
warned about exactly this scenario.