Reason this number caught my eye: last year the Fed's stress tests found "loss rates from [non-bank financial institution] exposures (i.e., the percentage of loans that are uncollectible) were estimated at 7%, under a severe recession in scenario one" [1].

That's the scenario in which unemployment goes to 10%, home prices crash by 33%, the stock market halves and Treasuries trade at zero percent yield [2].

[1] https://www.mfaalts.org/industry-research/2025-fed-stress-te...

[2] https://www.federalreserve.gov/publications/2025-june-dodd-f...

▲

sehansen 12 hours ago | parent | next [-]

The categorization the Fed uses for NBFI is broader than private credit. E.g. if a hedge fund gives a loan to a private company, that's not private credit because hedge funds seem to have their own category. And lending backed by securities is also in a different category, it seems.

So I guess the Fed expects these other kinds of lending to be safer than private credit?

▲

npilk 13 hours ago | parent | prev [-]

What's odd is according to the article, this index estimated an ~8% default rate in 2024. So maybe the stress test was measuring something different? It's weird to think the stress test would find a lower loss rate during a severe recession than in the most recent year with data available.

	▲	smallmancontrov 11 hours ago \| parent \| next [-]
		The regulators were modeling a scenario where private credit was dragged down by a problem elsewhere in the economy, not one where the rest of the economy was dragged down by private credit. Everyone understands that center of a financial implosion is always worse than its effects on the broader economy, but regulators aren't tasked with stopping the explosion at ground zero, they are tasked with stopping contagion dominoes from falling, so that's what they model.
	▲	JumpCrisscross 13 hours ago \| parent \| prev [-]
		> maybe the stress test was measuring something different? The Fed is measuring the loss on bank loans to the private-credit lenders. A 10% portfolio loss shouldn't result in those lenders defaulting to their banks. By my rough estimate, one can halve the portfolio loss rate to get the NBFI-to-bank loss rate. So a 10% portfolio loss means we're around a 5% expected long-run loss to the banks. Which is still weirdly high, so I feel like I must be missing something...