The UK financial press spent this week on the motor-finance redress bill and the timing of the next rate cut. The number that actually matters for financial stability was two clicks deeper into the Bank of England’s July Financial Stability Report, and almost nobody has printed it: £173bn.
That is the Financial Policy Committee’s estimate of UK banks’ banking-book exposures to private-market funds and to highly-leveraged corporates backed by financial sponsors. It is equivalent to roughly 8% of UK banks’ total committed limits across their wholesale portfolios. It is the first time the Bank has quantified, at this granularity, the pipe that connects a private-credit or private-equity downturn to the deposit-taking core of the UK banking system.
Everyone in markets knows private credit has grown. The FSR’s contribution is to price the connection — to say how much of a supposedly non-bank story sits, in the end, on bank balance sheets.
The number, precisely
Read the £173bn carefully, because its construction is the whole point.
It is a committed-limit figure, not a drawn figure. It counts the credit banks have promised to private-market borrowers — subscription and capital-call lines to funds, net-asset-value facilities, leveraged loans and warehouse lines to sponsor-owned companies — whether or not those facilities are currently drawn. In calm conditions a large share sits unused. The exposure is contingent.
That is exactly why it belongs in a financial stability report rather than a quarterly results deck. A committed limit is a claim on a bank’s balance sheet in the one state of the world where the bank would least want it: a downturn, when funds draw their lines to meet capital calls and buy time, and sponsor-backed borrowers roll over debt they can no longer refinance in the market. The 8% figure looks modest until you remember it is 8% of wholesale limits, concentrated in a handful of large lenders, and correlated — the facilities tend to be drawn together, in the same stress, for the same reason.
Private credit was supposed to move risk out of the banking system. The FSR’s quiet finding is that it moved the risk one step away and then wired it back in through the funding line.
Why this is a channel, not a footnote
The mechanism the FPC is describing is not exotic. It is plumbing.
A private-credit fund lends to mid-market companies. It funds itself partly with equity from pension funds and insurers, and partly with bank facilities — leverage on the fund, plus subscription lines that bridge the gap between calling investor capital and deploying it. A private-equity sponsor buys a company and loads it with leveraged loans, some of which banks originate and warehouse before distributing. In each case the bank is not the ultimate risk-taker. It is the provider of the funding that lets the structure exist.
In good times that is invisible and profitable. In a bad state, three things happen at once. Funds draw their committed lines rather than sell assets into a falling market. Sponsor-backed corporates, shut out of a frozen leveraged-loan market, fall back on their bank relationships. And the value of the collateral behind all of it — private company equity, illiquid loans — falls precisely when it is being tested. The bank’s contingent exposure becomes a real one at the worst moment.
This is why the FPC keeps returning to the same word: opacity. Private markets are marked infrequently and modelled, not traded. A bank can measure its committed limit to a fund with precision. It cannot see, in real time, what that fund holds, how levered its underlying borrowers are, or how many other lenders are exposed to the same names. £173bn is the size of the door. What is behind it is still being mapped.
The scenario built to test it
That mapping is the job of the Bank’s private-markets system-wide exploratory scenario, whose stress narrative the Bank published alongside the FSR. It is worth understanding, because it is calibrated to hit exactly the channel above.
The exercise runs across 46 participating firms — banks, insurers, pension funds, endowments and the asset managers who run the private-credit and private-equity vehicles themselves. That breadth is deliberate: a system-wide test is designed to capture the interactions between these players, not each one in isolation. The whole risk in private credit is that everyone reaches for liquidity through the same narrow set of bank lines at the same time.
The scenario itself is a five-year global recession, front-loaded and severe:
In year one, asset prices fall sharply and UK CPI inflation peaks at 7%. In year two — the crunch — UK GDP contracts by 4%, Bank Rate rises to 7%, the FTSE All-Share falls 35%, and European leveraged-loan spreads widen by 390 basis points. The recovery is slow: unemployment peaks at 7.5% and GDP grows at just 0.7% a year through years three to five.
The design tells you what the Bank is worried about. A 390-basis-point widening in leveraged-loan spreads with Bank Rate at 7% is a refinancing wall for exactly the sponsor-backed borrowers sitting inside that £173bn. A 35% equity fall marks down the collateral. A prolonged, shallow recovery denies everyone the quick bounce that would let a fund avoid drawing its lines. This is not a scenario about whether banks have enough capital in aggregate. It is a scenario about whether the private-markets channel amplifies a shock on the way through.
What the FPC did — and didn’t — do
For all that, the Committee held its main lever steady. The UK countercyclical capital buffer stays at 2%, its neutral setting. The FPC’s read, recorded in its 27 June meeting, is that the banking system is resilient today and that the private-markets exposure is a risk to monitor and measure, not one that yet warrants more capital.
That is the correct call and also the revealing one. Regulators raise buffers when they can size a risk and judge it rising. The FPC’s decision to quantify the exposure but not act on it is a statement that it does not yet trust its own map. The SWES is how it builds that trust: initial information-gathering in this report, interim Round 1 findings later in 2026, a final report in 2027. The £173bn is the baseline the next two years of work will be measured against.
What to watch
Three things follow.
First, treat £173bn as a floor on the conversation, not a ceiling on the risk. It is committed limits at UK banks. It excludes exposures routed through overseas subsidiaries, through the insurers and pension funds that provide the equity, and through the second-order links the SWES exists to find. The system-wide number is larger than the banking-book number by construction.
Second, watch the drawn-versus-committed gap. The single most useful figure the Bank could publish next is how much of that £173bn is actually drawn today, and its own estimate of how fast it would be drawn in the year-two crunch. That conversion rate is the difference between a manageable exposure and a correlated one.
Third, note what this does to the “private credit is safer because it is not in banks” argument. The FSR does not refute it so much as bound it. Risk did leave the trading book. It did not leave the system, and a measurable slice of it walked back in through the funding line. £173bn is the receipt.
Discussion
Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.