By the end of this lesson you can

  • →Explain why a protocol accepting its own token as collateral is reflexive
  • →Compute the moment at which defaulting becomes the borrower's rational choice
  • →Enumerate the independent parties a single leveraged position depends on
  • →Identify who absorbs bad debt, and what that means for a supplier

Junior · enrolled learners

This lesson opens with Venus Protocol, 18 May 2021.

What happened
Venus, a lending market on Binance Smart Chain (now BNB Chain), accepted its own governance token XVS as collateral. On 18 May 2021 the XVS price shot up nearly 90 percent, from about $76 to about $144, and then crashed, in a market whose trading was concentrated on a single venue where XVS liquidity was thin. At the elevated price, borrowers drew large loans in other tokens, such as bitcoin and ether, against XVS collateral. When the price fell, the loans became undercollateralized, and rather than repay, the borrowers kept the tokens they had borrowed. More than $200M of liquidations followed, and the collateral, even when sold, was worth less than the loans: the protocol was left with more than $95M of bad debt, in the form of about 2,000 BTC and 5,700 ETH. Venus's founder attributed the price move to large market orders against a limited unstaked supply and said the protocol worked as intended.
The decision point
The collateral was the protocol's own token, which makes the exposure circular in a way ordinary collateral is not. XVS's value depended on confidence in Venus, and Venus's solvency depended on the value of XVS, so the asset that was supposed to make the protocol whole was the asset that would be worth least in the state where the protocol needed it. Every risk parameter was set as though the collateral were an independent asset, and it was not.
Recorded loss
$100,000,000

What you will be able to answer

  • →Why is a protocol's own token bad collateral?
  • →When does defaulting become rational?
  • →Who absorbs bad debt?
  • →How many parties does a looped position depend on?

Orientation and Year One are open: anyone can read them without an account. From Year Two onward the lessons are for enrolled learners, because progress through the later years only means anything if it is tracked against a record.

It is free. We do not sell the list and there is nothing to buy at the end of it.

Sources and review

Confidence high·Volatility medium·Reviewed 2026-08-06·Owner unassigned

Contested

Whether the XVS move was deliberate manipulation was disputed. The Block Research's Igor Igamberdiev said thin XVS liquidity on Binance let the price be easily manipulated; Venus's founder attributed it to large market orders against a limited unstaked supply and said the protocol worked as intended; Venus's later incident report, as quoted by the Quadriga Initiative, cited extreme volatility, unexpected external market manipulation and poor risk management, and found no specific group that stole from the protocol. The lesson does not require it resolved: a collateral asset whose price can nearly double and collapse within hours produces the same bad debt either way, and that is a property of the depth rather than of anybody's intent.

Reported figures for the bad debt vary: more than $95M (The Block, at 19 May 2021 prices), $100M (The Defiant), about $77M lost by the protocol (Venus's later report, as quoted by the Quadriga Initiative), and $145M in Quadriga's own headline, partly because the valuations depend on which price is used, which is itself the subject of the lesson. The collateral factor in the worked example is illustrative.

The Maker shortfall cited in the bad-debt section is estimated differently by source: about $4.5M (The Block, at the time of the debt auctions) to 6.65 million DAI (a later community post-mortem reported by Cointelegraph). The auctions sold 20,980 newly minted MKR for about 5.3 million DAI.