Asset: UST · Class: Algorithmic synthetic · Date: 2022-05-09
Worst spread: −7000 bps · Duration: Terminal — never recovered
Detected live: No (pre-Pegana retrospective)
Background
UST was an algorithmic stable backed by LUNA mint/burn. Anchor Protocol offered ~19.5%
yield on UST — most of UST’s supply sat there. By Q1 2022 the system was a reflexive
flywheel: LUNA holders trusted UST because UST holders trusted LUNA.
Trigger
A series of large UST exits from Curve’s 4pool started May 7 2022. The depth dropped
enough that UST printed $0.985 in liquid venues — a 1.5% spread for a stable. That
should have been a strong DRIFT signal.
Cascade
- UST holders rushed to mint LUNA to exit — burn the stable, mint the volatile.
- LUNA supply went from ~340M to >6.5B in 72 hours. Price collapsed >99%.
- The reflexive backstop broke: minting more LUNA to defend UST destroyed both at once.
- Pegana would have stayed in
DEPEG → CRITICAL → BLACK_SWAN states with no recovery
transitions.
Recovery
There is no recovery. UST trades fractions of a cent. The protocol was relaunched as
Terra 2.0 without the algorithmic stable.
What Pegana would have shown
- May 7, ~12:00 UTC — Curve imbalance; smoothed spread crosses 30 bps.
PEGGED → DRIFT.
- May 8, ~18:00 UTC — Spread crosses 100 bps despite continued exits.
DRIFT → DEPEG.
- May 9, ~06:00 UTC — Spread crosses 500 bps.
DEPEG → CRITICAL.
- May 9, ~14:00 UTC — Spread crosses 1000 bps.
CRITICAL → BLACK_SWAN.
- No recovery — BLACK_SWAN never auto-exits. Operator review required.
Lesson
Algorithmic stables print false PEGGED for months at a time. The signal isn’t the
daily noise, it’s the depth at which arbitrage stops working. Pegana now reports
liquidity quality alongside spread (see confidence score)
for exactly this reason.
Sources
