Thanks @techsqrt for the question, it’s a good one, and it’s worth clarifying this.
ELI5 version (very short)
Think of the vault as a big box of money that is split into many smaller boxes (markets). Some of those boxes are tiny and have strict size limits. Even if one small box goes wrong, it can only lose the money inside that box, since it cannot touch the rest of the vault. On top of that, borrowers must always put in more value than they borrow, and risky positions are closed early. That’s why losses are both unlikely and limited.
Detailed explanation
1. Market allocation caps (hard limits)
Each Morpho market used by a kpk vault has a maximum allocation cap, defined and enforced by kpk’s rebalancing system (see: kpk USDC Prime | kpk). In practice:
- A rebalance cannot be executed if it would cause the vault to exceed a market’s cap.
- This mechanically limits how much of the vault can ever be exposed to a single market at any point in time.
2. “<1%” is a snapshot, not a guarantee
The “<1%” figure refers to the current allocation at the time of measurement, not a fixed promise. What is fixed:
- The maximum allocation caps per market.
- kpk’s ability to actively reduce or exit exposure as conditions change.
So the correct framing is that potential losses are bounded by the allocation at risk at the time, which itself is constrained by caps and active management.
3. What bad debt actually means for lenders
If bad debt were to occur in a market, lenders in that market would absorb the realised shortfall - not the full amount supplied to that market. Bad debt only arises if:
- Liquidations fail to fully cover borrower debt, and
- The remaining uncovered portion must be socialised across lenders in that market.
In practice, most liquidations do not create bad debt at all — they are the mechanism that protects lenders by closing risky positions early (i.e. liquidation bonuses for liquidators). The mentioned BTC markets have been live for over 1.5 years, during which multiple liquidations have occurred, but no bad debt has been realised. That is exactly the expected behaviour of a well-parameterised, over-collateralised lending market.
4. Conservative LLTVs create a large safety buffer
The BTC-backed markets referenced here use conservative parameters:
- ~86% LLTV for LBTC
- ~77% LLTV for tBTC
Lower LLTVs in Morpho come with higher liquidation bonuses, which creates a substantial buffer between:
- the first liquidation events, and
- any potential scenario where bad debt could appear.
This means collateral value would need to move significantly beyond liquidation thresholds before lenders are exposed to losses.
5. Active monitoring and emergency exits
Finally, this is not passive exposure. kpk operates monitoring systems that can automatically pull funds from a market if there are signals of potential catastrophic risk, such as:
- oracle ownership changes,
- exploits or hacks in underlying protocols,
- structural failures in collateral systems.
This gives us high confidence that we can exit before allocation caps are reached or bad debt materialises in such scenarios.
TL;DR: Risk is controlled by allocation caps, over-collateralisation, liquidation buffers, historical performance, and active monitoring. Together, these mechanisms make losses both unlikely and bounded.
