The lender's own word for it is a shadow default rate. It sits on one loan in seventeen.

What is PIK interest?

Interest that is added to the loan balance instead of being paid in cash. The borrower keeps the money, the lender books the income, and the principal grows. At the end of the period the borrower owes the original loan plus every deferred coupon plus interest on those coupons.

It comes in two forms and the difference matters. Contracted PIK is priced in at closing, usually on a junior tranche, where the lender knowingly accepts deferred cash in exchange for a higher rate: junior capital for a borrower below $10 million of EBITDA currently carries an all-in yield of 13 to 16 percent across cash and PIK combined (SPP Capital Partners, Market At A Glance, July 2026). Everyone understood the trade when it was struck.

A PIK toggle is different. It is an option, usually the borrower's, to switch a coupon from cash to accrual when cash is tight. Elected once in a genuinely temporary squeeze it is a sensible piece of flexibility. Elected because the business cannot pay, it is a signal, and the market has learned to read it as one.

How much of the market is using it?

More than at any point since 2020, and the composition is the concerning part. As at the first quarter of 2026, payment-in-kind interest was present on 10.6 percent of all direct lending loans and represented 8.9 percent of total interest income, the highest share since the fourth quarter of 2020, against 8.3 percent of interest income a quarter earlier and 5.4 percent at the end of 2021 (Lincoln International, published 7 May 2026, as of 31 March 2026). Close to one dollar in eleven that direct lenders book as income is not cash.

Inside that total sits the category that matters. Loans carrying no PIK at closing but carrying it today, which Lincoln labels bad PIK, represented 55.7 percent of all loans with PIK, or 5.9 percent of all loans, against 6.4 percent in the fourth quarter of 2025 and 2.5 percent at the end of 2021 (Lincoln International, 7 May 2026). Lincoln itself describes that measure as one that may be viewed as a shadow default rate.

One loan in seventeen has switched from cash to accrual since it was written. That is not a contractual feature being used as designed. It is a population of borrowers who could pay when the loan was made and cannot pay now, and the ratio has more than doubled in under five years.

It is worth knowing what the measure does not capture. Of roughly 538 companies showing signs of credit pressure across the business development company universe at 31 March 2026, half did not use PIK at all in the preceding twelve months (PitchBook LCD analysis of more than 170 business development companies, 21 July 2026). A PIK screen is a useful early warning and it misses half the problem, which cuts both ways: a borrower not using PIK is not thereby cleared.

What does a PIK toggle predict?

Delinquency, at a measurable rate. The Financial Stability Board finds PIK used in roughly twelve percent of private credit loans, with toggles in about half of those cases, and that PIK toggles are associated with a one to two percentage point increase in the likelihood of a loan becoming delinquent in the following quarter, against an unconditional probability of three percent (Financial Stability Board, Report on Vulnerabilities in Private Credit, 6 May 2026).

Read that as a relative rather than an absolute. An increase of one to two points on a base of three is an increase of a third to two thirds in the probability of delinquency next quarter. It is the difference between a normal credit and one the lender's model has reclassified, and it happens on the election rather than on any subsequent event.

The regulator adds an observation that explains a good deal of the middle-market usage: borrowers frequently use PIK toggles as a substitute for revolving credit lines. A business without adequate committed working capital facilities reaches for the instrument it has, which converts a liquidity gap into permanent balance sheet. That substitution is the single most common way a solvable working capital problem becomes a leverage problem.

There is a second-order effect on the numbers a borrower may be relying on. The share of direct lending borrowers with fixed charge coverage below one times fell to 19.5 percent from a peak of 40.9 percent, and coverage across the market improved through 2025, but Lincoln notes that the improvement may have been due in part to the increase in PIK usage itself (Lincoln International, as of 31 March 2026). Deferring the coupon improves the coverage ratio that measures the coupon. The ratio gets better while the position gets worse.

What does it actually cost?

The equity. The clearest published measurement of that is the loan-to-value on the bad-PIK cohort, which has risen 36.7 points from a healthy 39.4 percent at inception to 76.1 percent as at the first quarter of 2026 (Lincoln International, 7 May 2026). The average borrower in that group started with roughly sixty cents of equity cushion behind every dollar of debt and now has roughly twenty-four.

Part of that is the accrual and part of it is the value decline that caused the accrual, and the two compound on each other. Deferred interest grows the numerator while the earnings problem that made the deferral necessary shrinks the denominator. There is no rate at which that arithmetic reverses on its own; it reverses only if earnings recover faster than the balance compounds.

The forward consequence is a refinancing that does not clear. A borrower whose loan-to-value is in the mid-seventies is asking a new lender to advance against a value that leaves almost no cushion, in a market where unitranche loan-to-value is being set at roughly fifty percent below $100 million of EBITDA and fifty-five percent above it (Houlihan Lokey, as of 30 April 2026). The gap between where the loan sits and where a new one would be written is equity that somebody has to contribute.

Lenders have drawn the obvious conclusion about new transactions. Fifty-five percent of surveyed private credit lenders would not provide payment-in-kind flexibility on a new leveraged buyout at all (Houlihan Lokey, Q2 2026 Private Credit Survey). The instrument that looks like a borrower-friendly term in an existing document is being written out of new ones, which tells you how the people who priced it are now valuing it.

When is electing it the right answer?

In three situations, and they share one feature: the cash gap has a named end date that does not depend on a forecast improving. A working capital swing around a known seasonal trough, a build or integration with a completion date and a contracted revenue start, or a bridge to a transaction that is signed rather than contemplated. In each case the borrower can say when the accrual stops.

The discipline that separates those from the bad version is to size and time-limit the election before making it. Elect for a defined number of quarters rather than until further notice, calculate the accreted balance at the end of that window rather than the monthly saving at the start, and test the refinancing at the accreted balance against the loan-to-value a new lender would actually write. If that test fails, the toggle is not solving the problem; it is choosing which year to have it in.

The alternative worth pricing first is the one the regulator identified. If the toggle is being used as a substitute for a revolver, the honest fix is a revolver, or an asset-based facility against receivables and inventory, both of which cost less than accruing coupon on a term loan and neither of which compounds. That conversation is available to a borrower who is current and much harder for one who has already elected.

The last point is about timing rather than structure. The election is easiest to make when nothing forces the borrower to explain it and hardest to reverse afterwards. A borrower considering a toggle is, on the regulator's own numbers, in a cohort with a materially raised probability of delinquency next quarter, and that is the moment to open the wider conversation with the lender rather than the moment to take the one option that requires no conversation at all.