When the Lender Is the Borrower

Private credit managers routinely borrow from banks to fund their own lending. Each is a lender and a borrower at once. Doing so gives them leverage on leverage, but whether it’s safe depends on the quality of the loans it makes and on a fund’s ability to run its borrowings, the maturities it has to meet, and the collateral it has pledged. Banks lend against that collateral, whether it be undrawn investor commitments or loans already on the manager’s books.

From outside the ecosystem, the tension between running a lending book and a financing book barely shows. Commentators argue over whether the structure hides the next credit blowup. But the real complexity is internal to the manager. Often, only treasury or the CFO watches both books. The amount of capital to deploy depends on cash on hand, plus what the fund can still draw on its lines, minus what’s already borrowed and coming due.

Where the leverage comes from

Leverage enters the fund in two ways, a subscription line against investor commitments and a NAV line against existing loans. The choice between them comes down to terms, speed, and drawdown, and those factors vary bank by bank. Given its deposit base, a money-center bank prices differently from a smaller lender. One bank favors NAV loans while another favors subscription lines. Treasury, or the CFO, shops the facilities when a deal team needs a large check by a hard deadline.

The subscription line is the older tool. A fund holds a commitment from a limited partner (LP), a large public pension, for example, and pledges that commitment paper to a bank to draw against it. Because the advance rate reflects the credit quality of the investor behind the paper, a blue-chip pension commitment borrows close to 100 cents on the dollar, while a weaker or unknown name may only see 20 cents. The fund draws on the line rather than waiting the usual 15 to 30 days for called capital. It deploys into the deal now and repays when the investor cash comes in.

Called capital retires a subscription draw within weeks, because the investor is contractually bound to fund it. A NAV loan waits on exits, cash the fund still has to produce, so its paper runs for years. Both lines pledge collateral that a bank can seize on default. What the fund gets for that pledge is a stronger reported IRR, and what it takes on is a set of maturities it has to meet.

What happens to returns and to investors

IRR calculations use the dates when investor cash moves. When a fund bridges a deal with leverage and calls the capital weeks later, the investors’ money spends less time in the fund, so the same profit translates to a higher annual rate. But distributions to paid-in capital (DPI) and the multiple on invested capital (MOIC) can’t be moved that way, because they divide dollars returned by dollars contributed and never look at the calendar.

This means that two funds can make the same loans and produce the same profit, and the one with better access to the lines prints the higher IRR. That’s why investors have come to prefer DPI and MOIC, where the proof is exits and full liquidations.

A fund with $100 million in subscription commitments that adds about $75 million across the two lines pushes out roughly $175 million, roughly 1.75 times its equity. When the loans yield more than the lines cost, the gains on $175 million accrue to the investors’ $100 million. Any levered balance sheet works this way: borrowed dollars sit alongside equity dollars, and when the assets earn more than the borrowing costs, the gain concentrates on the equity holder.

If the portfolio sells for $400 million at exit, the fund repays the $75 million of borrowings and returns roughly $325 million to investors who put in $100 million, a 3.25x DPI. The printed IRR looks better still, because it runs on the investors’ $100 million rather than the $175 million at work.

Because a private credit book reads loan by loan, investors can weigh a firm’s leverage, loss ratio, and the risk against peers. But the NAV line adds a risk. Once the fund pledges its positions to back a loan, the assets collateralize something else. Sophisticated LPs might answer with side letters that bar the manager from levering or pledging their exposure. A Blackstone-scale manager might refuse and send the investor elsewhere, while a mid-market manager will often take the side letter.

The financing book meets the loan book

With separate maturities, pledged collateral, and cash needs, managers have to reconcile the financing book against the loan book in one view, with the same position-level visibility that its investors demand. That onus falls first on the finance function. A controller keeps the books in line, while treasury knows where the cash sits and where the leverage sits in the system. Many CFOs come into their roles with accounting experience, making the treasury half the capability a fund is most likely to lack when it starts to borrow.

Treasury manages the borrowings as a ladder. Subscription paper rolls at 90 or 360 days against a NAV loan’s three-year duration, and the fund tracks each maturity against the exits projected to repay it. Underneath sits a treasury and liquidity system that holds each facility’s terms, whether coupons pay in cash or as payment-in-kind (PIK), and whether the fund borrowed 80 to repay 100. Leverage also multiplies what the controller tracks, down to the accounting for NAV notes that the fund itself issues.

Missing a maturity is an expensive mistake, because the bank’s call on the collateral keeps its date even if an exit slips. Even worse, banks have to watch for collateral that their borrower has also promised to someone else. In 2025, the subprime auto lender Tricolor collapsed after pledging the same collateral to several lenders at once, and prosecutors charged its executives with fraud.ⁱ Banks prevent this by means of a look-through, naming exactly which investor and which portfolio company they lend against, so the same asset can back only one facility.

The risk of contagion

The other half of the safety question is the quality of the loans. A car loan or a mortgage leans on repossession and a down payment, and even then the collateral can fall short. Private credit lends against a thicker cushion. Because most of these companies carry 50 to 70% equity underneath, the lender advances only 30 to 40% of the value and sits at the top of the bankruptcy waterfall. On default, it takes the collateral and becomes the fulcrum owner.

On those mechanics, first-lien lenders have stayed fairly safe, and the loud hidden-risk narrative overstates what a single default costs them. In spring 2026, the technology-focused sponsor Thoma Bravo handed the software maker Medallia to its first-lien lenders after the company couldn’t carry its private credit payments.ⁱⁱ The sponsor’s equity fell to zero, roughly a $5 billion loss, while the lenders took the business and made themselves whole. Workouts like that stay manageable because the lender advanced well below the company’s value, so it can take the business and sell it to another sponsor or a restructuring firm at a discount without taking a loss.

Today’s market is producing single defaults, the kind the workout machinery absorbs. We haven’t seen correlated stress. If defaults cluster as mortgages did in 2008, we could see every lender selling seized collateral into the same market at the same time, without knowing where the floor is.

Either way, if you see your financing lines, pledged collateral, cash, and loan book in one view, you can run the ladder in calm markets and answer for it in stressed ones. It’s why maintaining that picture has become part of the infrastructure of running a levered fund. Because the bank already demands it in its look-through, and the investor already reads it loan by loan, you should know your two books at least as well as the people you owe. It’s a collateral question in your books, but a data question in your operations.