Stress Testing Private Credit: From 4.9% to 3.5% Capital, and What That Leaves
Private credit has never been through a full cycle at its current size. The market reached roughly $3 trillion by early 2025 having grown up almost entirely in the post-2008 era of low rates, benign credit conditions, and abundant liquidity. Its stress behaviour is therefore an inference, not an observation.
Our research makes that inference explicit. Applying stress scenarios calibrated to historical crisis episodes across 13,317 transactions worth $11.8 trillion, severe macroeconomic shocks reduce capital adequacy ratios from 4.9% to between 3.5% and 3.6% — approaching regulatory minima.
Read that carefully
The headline is not that the sector fails. It is that the buffer is thinner than the starting point suggests. A 4.9% ratio sounds like room to manoeuvre. Under a severe scenario, roughly a third of it goes, and what remains sits close to the floor rather than comfortably above it.
That is the difference between a capital position that absorbs a shock and one that survives it. The first lets you keep lending through the downturn. The second forces you to stop — which is precisely when borrowers need you most, and precisely when forced deleveraging transmits stress outward.
Why the compound effect matters
The stress result is not a single channel. It is four interacting:
- Deal-level credit risk rises — and cross-border exposures start from a higher base (6.46% versus 2.56% default rates).
- Flows contract. The frictions that price distance in normal times bite harder under stress; the corridors that close first are the thin ones.
- Network effects amplify. With the top 10 corridors carrying 68% of volume, common exposures correlate losses that look independent on paper.
- Concentration converts losses into capital events. Herfindahl concentration means a shock to a dominant exposure is not diversified away.
Cross-border deals do not merely start riskier. They are more sensitive to the shock and more exposed to the network — the channels compound rather than add.
The practical gap
Most private credit stress testing we see does one of two things:
- Applies a flat PD multiplier across the book. This ignores that cross-border and domestic exposures respond differently, and that the corridor structure concentrates the response.
- Borrows a bank template. Bank frameworks assume deposit funding, run risk, and the ability to renegotiate and extend. Closed-end private credit funds have a maturity wall instead — different incentives, different failure mode.
Neither is wrong so much as insufficiently specific to how this market actually breaks.
What a defensible stress framework needs
- Scenarios calibrated to real episodes, not round-number shocks. Deal flow fell 71% from the 2007 peak to 2009 — that is a calibration anchor, not a hypothetical.
- Differentiated PD response by cross-border status, corridor, and sector.
- Corridor-level concentration carried through to the loss distribution.
- A documented path from scenario to capital ratio that a validator can challenge line by line.
That last one is what separates a stress test that informs a decision from one that fills a chapter.
An honest caveat
These are simulations calibrated to historical analogues, in a market that has not yet experienced a severe downturn at scale. They are an argument about magnitude and mechanism, not a forecast. The value is in knowing which channels dominate and where the buffer actually sits — before the cycle tells you.
If you want this run against your own book rather than the market aggregate, that is what our DFI Economic Capital assessment does.
Findings from "Cross-Border Shock Transmission in Private Credit Markets: Evidence from Global Deal-Level Data" (2025), presented at the Bank of England Agenda for Research (BEAR) Conference 2026.
Need this applied to your portfolio?
We turn research into models and advice you can act on — economic capital, climate ICAAP, and model governance.