Over-collateralisation is the only lending model that survived the cycle
The crypto lending sector ran two models. One of them is largely gone, and the reason is structural rather than accidental.
By Staff, Lend Ledger
Crypto lending ran two broadly different models. It is worth being precise about which one failed, because the sector’s history is frequently described as though the whole idea collapsed, and that is not what the record shows.
Model one took customer deposits, paid an advertised yield, and generated that yield by lending onward — to trading firms, to counterparties, and in some cases through undisclosed related-party exposure. Depositors were unsecured creditors of the platform. Several large firms operating this model entered insolvency proceedings in the last cycle, and those proceedings are matters of public record.
Model two required borrowers to post collateral worth more than they borrowed, and liquidated that collateral automatically when its value fell towards the loan. Lenders were secured by a margin the protocol enforced mechanically.
The second model has continued operating. That is not a claim that it is safe, and this site would not make one.
Why the structural difference matters
The distinction is not sophistication. It is who carries the credit risk and whether it is visible.
In the first model, the depositor’s exposure was to the platform’s judgement about counterparties they could not see. When the disclosure was thin — and it frequently was — the depositor could not evaluate the risk they were taking, no matter how carefully they read.
In the second, the exposure is to a mechanism whose parameters are published: how much collateral is required, at what ratio liquidation triggers, what penalty applies. Those are numbers a reader can inspect before committing anything.
What over-collateralisation does not solve
A great deal, and this is the part the sector’s marketing skips.
It does not solve smart-contract risk: the mechanism is code, and code has been exploited. It does not solve oracle risk: liquidation depends on a price feed, and a wrong or manipulated feed liquidates positions wrongly. It does not solve market-structure risk: in a sharply falling market, liquidations can fail to clear at the assumed price, leaving the system under-collateralised in exactly the conditions it was designed for. And it does not solve governance risk, since parameters can be changed by whoever controls them.
The honest summary
The model that survived did so because its risk is legible, not because its risk is small. Anyone deploying assets into it can lose all of them.
Nothing here is financial advice. This publication holds no positions and recommends no platform, protocol, or strategy.