What it is

Before a pool opens, the originator deposits at least 15% of the pool in its own money. That deposit sits at the bottom of the stack and is destroyed first. If the pool loses $400,000 on $10M, the originator absorbs all of it and investors lose nothing. The originator would have to lose its entire $1.5M before the junior note is touched at all.

Why it is a check in code, not a promise

The contracts refuse to open the pool without it. The first-loss floor and the ratio caps are checked on every deposit, so the pool cannot reach a state where investor money is at risk ahead of the originator’s. That is the difference between a structural protection and a covenant. A covenant is something you sue over afterwards.

Why at least 15%

Two reasons, and only one of them is about loss absorption. It absorbs normal credit loss. At an assumed 4% loss rate, 15% is a wide margin — though that 4% has never been measured, and Disclosures says so plainly. It makes the notes sellable. This is the bigger reason. An originator with nothing at stake has no reason to underwrite carefully or collect diligently. Skin in the game is what makes an investor willing to sit above it.
Pool one requires 25%, not 15%. The first pool has no performance history to price against, so it carries a wider floor.

What it makes pALPHA

Because the originator’s 15% absorbs the first losses, the junior note is second loss, not first loss — and is priced accordingly, at roughly 16% rather than the much higher rate true first-loss capital would demand. See why the rates are what they are.

The honest caveat

No on-chain junior tranche has yet absorbed a real credit loss as designed. Across every protocol we reviewed, first-loss capital has either been bypassed, covered off-chain, or never tested. Ours is untested too.