Underwriting system

Technical details of the core underwriting layer is specified in the whitepaper

riskAMM (3)DocSend

And the docs

Q&A

1. What if managers adversarially select bad instruments to supply to?

   Collateral presented by the managers would be the first to be wiped out(akin to being 'liquidated' in margin trading), so they are disincentivized to do so. Also, a manager's borrowing power is a function of his reputation score(which has to be earned), so a manager with a low reputation cannot attack the system in a meaningful capacity.

2. What if I as an LP disagree with the managers on which instrument is worth supplying to?

   LPs can always buy protection via shortZCB. They can also lever up by buying longZCB.

3. How are the prices of longZCB/shortZCB settled?

   It is determined by an attack resilient oracle that tracks how much returns the instrument has generated.

4. What if a manager pretends to be, or colludes with, a borrower and tries to approve a loan while not planning to repay?

   We prevent such sybil attacks by enforcing a KYC process/ identity gate for the managers, along with the reputation system. More anti sybil prevention details are in our whitepaper.

   Also, there is a time delay between the time a manager supplies to a lending pool and when a borrower can borrow from the lending pool, during which any malicious supply activity can be spotted.

5. Where does the shortZCB profit come from?

   longZCB/shortZCB are zero-sum -> It comes from the collateral used to buy longZCB.

6. What if a borrower buys shortZCB and doesn't repay?

   Only vault investors can buy shortZCB as a hedging instrument.