What is actually being asked for?
One of three things, and confusing them is the most common way a first request goes wrong. A waiver forgives a specific breach on a specific test date and leaves the covenant untouched thereafter. A reset changes the covenant levels for future periods, usually stepping back down to the original schedule over several quarters. An amend-and-extend moves the maturity date and reprices the facility, and is a separate transaction that may or may not travel with either of the first two.
The three are not interchangeable, and the lender will price whichever one it thinks you actually need. A borrower who asks for a one-time waiver on a problem that will recur next quarter is asking to have this conversation twice, at a higher price the second time. A borrower who asks for a reset when a single event caused a single miss is inviting the lender to reunderwrite the whole facility. The request should match the diagnosis, and the diagnosis should be in the request.
There is also a fourth thing that is sometimes the right ask and is rarely made: a covenant holiday, suspending the test for a fixed number of quarters in exchange for liquidity conditions, minimum cash levels and restricted payments. Direct lending covenant holidays rose 14% quarter on quarter alongside a 31% rise in sponsor equity infusions (Lincoln International, 11 February 2026, as of the fourth quarter of 2025). Lenders are granting them. They are also charging for them, and pairing them with controls that outlast the holiday.
When should the ask be made?
Before the test date. There is no published market figure for how many days ahead of a test date lenders expect contact, and we will not invent one, but the mechanics settle the question without a survey. Ahead of the test the borrower is bringing a forecast and asking the lender to help shape an outcome. After the test the borrower is reporting a fact and asking the lender to forgive one. The first is a commercial discussion. The second is a credit event with a documented date attached.
The practical trigger is the forecast, not the calendar. If the model shows the covenant tightening to inside its cushion in the next two quarters, that is the moment. A typical leverage covenant carries 25% to 35% headroom to the borrower's own model, so a plan that runs 10% behind for two quarters can consume most of it (Sidley Austin, 24 March 2026). Springing covenants add a second trigger that borrowers often miss entirely: they come into existence once revolver usage passes roughly 40%, which means drawing the revolver to fund a soft quarter can create the covenant that then fails.
Timing also interacts with the calendar of the lender rather than the borrower. Credit committees meet on a schedule, approvals in a syndicate require consents from lenders who have their own quarter ends, and an accommodation requested inside the last two weeks of a quarter competes with everything else on that desk. None of that is in the credit agreement. All of it affects the answer.
What has to arrive with the request?
A current model, a downside case, and an account of what changed. The base case is a negotiating document and every credit committee knows it. The downside case is the version that gets priced, which is why a borrower who has not built one has effectively delegated the job to the lender. That version will be harsher than yours.
Alongside it: a thirteen week cash forecast that reconciles to the model, the liquidity position including undrawn revolver capacity and any springing trigger it would touch, the amortization and interest calendar through the requested period, the covenant calculation as the borrower computes it with the definitions used, and a clear statement of what is being asked for and for how long. Where a sponsor is expected to contribute, that contribution should be committed rather than described. Sponsors are currently pushing maturities and conserving equity, driving 74% of institutional maturity-extension amendments while accounting for only 44% of new-money activity (PitchBook LCD, as of 30 June 2026), so an uncommitted sponsor line carries little weight.
The covenant calculation is the item most likely to be disputed, because coverage and leverage are both ratios of negotiated quantities. The Financial Stability Board notes that stripping out EBITDA adjustments can move private credit leverage from a reported 5x to 6x range to something closer to 7x (Financial Stability Board, 6 May 2026). Reconciling your definition to the lender's before the lender does it is worth more in this conversation than any concession on price.
What does the accommodation cost?
A fee, a margin increase, tighter documentation, and usually a permanent change to how the facility is monitored. No market series publishes waiver fees for middle-market credits, so any number offered as typical is an assertion. What can be read is the direction of lender posture, and it has moved a long way in twelve months.
Lenders describe their own year in one direction. Some 98% of surveyed private credit lenders report that underwriting standards became notably stricter since the start of 2026, the share expecting looser documentation collapsed from 33% to 4%, the share expecting tighter documentation rose from 13% to 56%, and 55% would not provide payment-in-kind flexibility on a new buyout (Houlihan Lokey, Q2 2026 Private Credit Survey). A borrower asking for flexibility is asking in the year lenders decided to stop giving it.
Be careful what is accepted in place of cash. Payment-in-kind interest was present on 10.6% of direct lending loans and represented 8.9% of total interest income in the first quarter of 2026, and loans that carried no payment-in-kind at close but do today, which Lincoln describes as a shadow default rate, reached 5.9% of all loans. Loan to value on that cohort rose 33.5 points to 76.0% over the year (Lincoln International, as of 31 March 2026). Converting cash interest to accrued interest solves a liquidity problem this quarter and converts equity into lender claim every quarter after it.
| Measure | Second quarter 2025 | Second quarter 2026 |
|---|---|---|
| Lenders expecting looser documentation | 33% | 4% |
| Lenders expecting tighter documentation | 13% | 56% |
| Lenders quoting original issue discount at 99 or tighter | 75% | 56% |
| Lenders seeing top-tier sponsor spreads at S+475 or tighter | 86% | 41% |
“A waiver request is a credit paper the borrower writes about itself, and the lender will either adopt it or replace it. Most requests are declined on thinness rather than on merit: no downside case, no committed sponsor position, a covenant calculation that does not tie to the lender's own definitions. The businesses that get the accommodation are rarely the strongest ones in the portfolio. They are the ones whose management the credit committee already believes.”
What makes a lender say no?
A structure that does not work at any covenant level. If the business cannot service the interest calendar in its own downside case, resetting the test changes the date of the problem and nothing else. Median interest coverage across 2,785 middle-market borrowers held at 1.6x for the twelve months ended 30 June 2026, and the share of borrowers with improving coverage plateaued after more than two years of gains while median EBITDA growth posted its largest quarter-on-quarter decline on record (KBRA, published 28 July 2026). Lenders can see that the growth which was doing the deleveraging has stopped.
A second accommodation on the same credit. Lincoln International estimates that 30% to 40% of direct lending deals maturing in the next two years have already extended once, and puts an incremental extension and a restructuring in the same sentence as the two available outcomes (Lincoln International, 11 February 2026, as of the fourth quarter of 2025). The second ask is underwritten as a restructuring whether or not it is called one.
And a lender that cannot fund. Direct lending volume fell 55% quarter on quarter to $33.6 billion in the second quarter of 2026 across 154 deals, the weakest since the second quarter of 2023, with some business development companies holding capital back to support existing stressed borrowers (Preqin and PitchBook LCD via Reuters, and Private Equity Wire, both 10 July 2026). If the accommodation you need includes new money, the identity and liquidity of the fund holding your paper decides the answer before your numbers do.
As of August 2026
Sources: Houlihan Lokey, Q2 2026 Private Credit Survey, for lender underwriting posture, documentation expectations, original issue discount quoting and sponsor spreads, with year-ago comparators from the same series; Lincoln International, 11 February 2026, as of the fourth quarter of 2025, for the amendment mix and the share of borrowers that have already extended once, and Lincoln International as of 31 March 2026 for payment-in-kind usage and loan to value on the affected cohort; Sidley Austin, 24 March 2026, for covenant cushions and springing triggers; PitchBook LCD, as of 30 June 2026, for the sponsor share of extension and new-money activity; Financial Stability Board, Report on Vulnerabilities in Private Credit, 6 May 2026, on reported against adjusted leverage; KBRA, Q2 2026 Middle Market Compendium, published 28 July 2026, for interest coverage and earnings growth across 2,785 borrowers; Preqin and PitchBook LCD via Reuters, and Private Equity Wire, both 10 July 2026, on direct lending deployment. No market series publishes waiver fees or the notice period lenders expect, so neither is quoted here.