LBO Debt Covenants Explained: Maintenance, Incurrence & Cov-Lite

Key takeaways

A covenant is the lenders' control over a borrower they cannot manage directly. The two families behave in opposite ways. A maintenance covenant is tested every period — typically quarterly — and the borrower must keep a ratio inside a limit whether or not it does anything; cross the line and you are in default having missed no payment. An incurrence covenant is tested only when the borrower takes an action — raising more debt, paying a dividend, making an acquisition — and is silent the rest of the time
The ratios that get tested are leverage and coverage. Net Debt / EBITDA caps how levered the business may be; interest or cash-flow cover (EBITDA ÷ interest, or cash flow ÷ debt service) checks it can service what it borrowed. The covenant is set above the sponsor's base case with a deliberate cushion — historically around 30–35% of EBITDA — so ordinary underperformance does not trip it, and it usually steps down over the life of the loan as the model deleverages
The single most important and least understood point: the EBITDA in the covenant is not the EBITDA you modelled. The credit agreement defines its own "Adjusted EBITDA", and in the modern market that definition permits add-backs for run-rate cost savings and synergies — sometimes uncapped and projected years forward. Inflating the denominator manufactures headroom, which is why a covenant can hold on paper while the business deteriorates in fact, and why the quality-of-earnings fight is really a fight over the covenant
The market verdict: cov-lite — loans with no maintenance covenant at all, only a "springing" leverage test on the revolver that activates when it is heavily drawn — went from exotic to standard. By the late 2010s the large majority of US and European institutional leveraged loans were cov-lite. The trade-off is structural: sponsors gained a long runway to fix a struggling business without a lender at the table, and lenders lost the early-warning tripwire, which is widely expected to lower their recoveries when deals do fail

A Covenant Is the Lender's Steering Wheel on a Company It Does Not Run

A lender into a buyout faces a structural problem. It is providing most of the money, five or six times EBITDA of it, into a business it does not own, cannot manage, and will not hear from between reporting dates. The sponsor holds the equity, the board and the keys. The interest rate compensates the lender for risk; it does nothing to control it. Covenants are the control — the contractual terms in the credit agreement that let a lender act before its loan is actually impaired.

That is the frame to carry into every covenant question: a covenant is not a fee or a rate, it is a right to intervene. Its entire value is the moment it lets the lender say "stop" — renegotiate, demand an equity injection, or in the limit accelerate the loan — while there is still enterprise value to protect. Whether that moment ever arrives depends on which of two opposite kinds of covenant the loan carries.

Maintenance vs Incurrence: Tested Every Quarter, or Only When You Act

The most important distinction in leveraged finance is between covenants tested by the calendar and covenants tested by behaviour, and conflating them is the fastest way to mark yourself as a candidate who has read about debt rather than lived with it.

A maintenance covenant is tested on a schedule — almost always each quarter — regardless of what the borrower does. The company must maintain a ratio inside a defined limit at every test date. Do nothing, change nothing, and a maintenance covenant can still be breached because EBITDA fell while the test level held. It is the lender's tripwire: it fires on deterioration, before any payment is missed.

An incurrence covenant is tested only when the borrower takes a specified action — incurring additional debt, paying a dividend, disposing of an asset, making an acquisition. If the company takes no such action, the covenant is never tested no matter how badly it performs. Incurrence covenants are the high-yield bond convention; maintenance covenants are the traditional leveraged-loan convention.

Why the difference decides who controls a struggling deal The practical effect is about timing of control. With a maintenance covenant, the lender gets a seat at the table the quarter performance slips — early, while options exist. With incurrence-only terms, the borrower can underperform for years and the lender has no contractual trigger to intervene until the company either misses a cash payment or chooses to take a tested action. The same business, the same decline, but the date on which the lender can do something about it moves from "soon" to "much later". That single difference is what the cov-lite debate is about, and why interviewers use the maintenance-versus-incurrence question to see whether you understand that a covenant's value is the date it gives the lender, not the words on the page.

Knowing which calendar a covenant runs on is half the picture. The other half is what it actually measures.

The Two Ratios: Leverage Caps the Debt, Coverage Checks You Can Service It

Financial covenants in a buyout almost always test one or both of two things, and they answer two different questions about the same balance sheet.

The leverage covenant — Net Debt / EBITDA — caps how indebted the business may be relative to its earnings. It is the headline covenant in most LBOs because leverage is the variable the sponsor is deliberately pushing. The coverage covenant — interest cover (EBITDA ÷ cash interest), or a fixed-charge or cash-flow cover that nets off mandatory debt service and capex — checks the business throws off enough cash to actually pay for the debt it carries. Leverage asks "is the stack too big?"; coverage asks "can the cash flow feed it?".

Covenant	Roughly	What it polices	Trips when
Leverage	Net Debt ÷ EBITDA	Total indebtedness vs earnings	Debt rises or EBITDA falls past the cap
Interest cover	EBITDA ÷ cash interest	Ability to pay interest	EBITDA falls or interest rises past the floor
Cash-flow / fixed-charge cover	Cash flow ÷ debt service + fixed charges	Ability to meet all fixed obligations	Cash flow falls below the combined call on it
Capex limit	Annual capex cap	Cash not diverted from debt service	Spend exceeds the agreed ceiling

The leverage covenant is not a single number for the life of the loan. It is set above the model's opening leverage and then steps down period by period, tracking the deleveraging the sponsor's own base case promises — a deal closing at, say, 5.5x might carry a covenant near 7x at the start that ratchets down toward 5x and below over three to four years. The gap between the covenant and the base case is the whole game, so it has a name.

Headroom: The Cushion Between the Covenant and the Plan, and the Equity Cure That Resets It

The distance between where the covenant is set and where the sponsor's base case projects the ratio to be is the headroom, or cushion. It exists so that ordinary, expected underperformance — a soft quarter, a delayed contract — does not put the company in default. Historically that cushion has been set at roughly 30–35% on EBITDA: EBITDA can come in about a third below plan before the leverage covenant trips. The figure is a market convention that flexes with how hot the financing market is — more headroom when capital is cheap and lenders are competing, less when it is scarce. Approximate, and negotiated deal by deal.

~30–35% Typical EBITDA headroom built into a maintenance leverage covenant at close — the amount earnings can fall below the sponsor's base case before the covenant trips. It is the single most-negotiated covenant term, and in a borrower-friendly market it widens toward the point where the maintenance test rarely bites at all

When headroom is exhausted, the sponsor is not always out of moves. Most credit agreements grant an equity cure: the sponsor can inject fresh equity that is treated, for the covenant test only, as additional EBITDA or as debt prepayment, curing what would otherwise be a breach. The right is deliberately limited — a typical formulation allows no more than two cures in any four consecutive quarters and a small number over the life of the loan — because a cure right without limits would make the covenant meaningless. The cure converts a covenant breach into a capital call on the fund, which is exactly why it exists.

Insider tip The equity cure is the bridge between the covenant and the debt stack the rest of the structure is built on. When an associate models a downside case and the leverage covenant trips, the next line of work is not "the deal is broken" — it is "how much equity cures it, and does the fund have the appetite to write that cheque to protect the position?" A candidate who, shown a breaching downside, immediately asks about the cure and its limits is demonstrating they understand a covenant is a negotiation with a release valve, not a cliff edge. That is a far stronger answer than reciting the definition of net leverage.

All of this assumes the EBITDA in the test is a fixed, honest number. It is not — and that is where the modern covenant quietly comes apart.

Covenant EBITDA Is Not the EBITDA You Modelled

The most consequential point about covenants is the one least visible from a textbook: the EBITDA the covenant tests is whatever the credit agreement defines it to be, and that definition has drifted a long way from reported earnings. Every leveraged credit agreement specifies "Consolidated EBITDA" or "Adjusted EBITDA" through pages of definitions, and in the borrower-friendly market of the last decade those pages came to permit add-backs for projected, run-rate cost savings and synergies — earnings the company has not yet realised, added to the denominator as if it had.

The mechanism is simple and powerful. Inflate defined EBITDA with add-backs and you simultaneously lower the leverage ratio and raise the coverage ratio, manufacturing headroom out of a definition rather than out of performance. Where the add-backs are uncapped or projected far forward — and in the loosest documents they are both — the gap between covenant EBITDA and cash EBITDA can run to double digits as a percentage. The covenant can be comfortably satisfied while the business that actually services the debt is shrinking.

Common mistake Treating the leverage covenant as if it tested the EBITDA on the income statement. It tests the EBITDA the lawyers defined, and the spread between the two is precisely what a quality-of-earnings review and the covenant negotiation are fighting over. A candidate who says "the company breaches at 7x net leverage" without asking whose EBITDA, with which add-backs, capped or uncapped, is missing where the real risk sits. The headline ratio is only as conservative as the definition feeding it — and in the modern market that definition is the most aggressively negotiated paragraph in the whole agreement.

A generous EBITDA definition is one way to defang a covenant. The market found a cleaner way: remove the maintenance covenant altogether.

Cov-Lite: The Maintenance Covenant Disappears, and With It the Tripwire

A cov-lite loan is a leveraged loan with no maintenance financial covenant. The leverage and coverage tests that would otherwise be checked every quarter simply are not there. What remains is incurrence-style protection — tests that bite only when the borrower acts — plus, usually, a single springing covenant: a leverage test on the revolving credit facility that "springs" into existence only when the revolver is drawn beyond a threshold, commonly around 35–40%. Run the revolver below that line and even the springing test never activates.

Cov-lite began in the US institutional loan market, where loans are held by CLOs and funds rather than relationship banks, and spread to Europe through the 2010s. By the late part of the decade it had gone from a feature of the strongest credits to the default setting: the large majority of new US and European institutional leveraged loans came to market cov-lite. What was once a concession to a blue-chip borrower became the standard term, and a loan with a maintenance covenant became the exception that signalled a weaker credit.

Springing covenant — the one tripwire cov-lite usually keeps The springing covenant is the compromise that lets a deal be called cov-lite while still protecting the one lender who sits closest to the borrower's liquidity. The revolving credit facility is typically held by the arranging banks, not the institutional term-loan investors, and those banks insist on a leverage test that activates once the revolver is meaningfully drawn — because heavy revolver usage is itself the signal that the company is running short of cash. The term-loan B holders, sitting further out, get no maintenance test at all. So "cov-lite" rarely means no covenants anywhere; it means the early-warning test has been narrowed to the moment the borrower is already reaching for its emergency liquidity, which is much later than a quarterly leverage test would have fired.

Whether stripping out the tripwire is a problem depends entirely on which side of the loan you sit, and that is where the topic earns its place in an interview.

The Verdict: Cov-Lite Bought the Sponsor Runway and Cost the Lender Recovery

Cov-lite is not, as it is sometimes framed, a straightforward erosion of standards — it is a transfer of control, and naming who gained and who lost is the difference between describing the term and understanding it. The sponsor gained runway: a struggling portfolio company can underperform for years without a lender entitled to a seat at the table, giving the operating plan time to work and removing the risk that a single soft quarter triggers a value-destroying renegotiation. For a fund confident in its turnaround, that time is genuinely valuable.

The lender lost the early-warning tripwire, and the cost shows up at the end. Without a maintenance covenant, the lender often cannot act until the company either misses a payment or runs its revolver into the springing test — by which point enterprise value has frequently eroded well past where an early intervention could have preserved it. Rating-agency and recovery studies broadly expect cov-lite structures to deliver lower recoveries on default than the covenanted loans of earlier cycles, even on first-lien paper that once recovered the bulk of its principal. The early seat at the table was worth real basis points of recovery, and the market gave it away in exchange for spread and volume.

Common mistake Calling cov-lite "riskier lending" and stopping there. The risk did not appear from nowhere — it moved. Cov-lite lengthened the runway for equity and shortened the lender's window to protect itself, so the same default produces a worse outcome for the creditor and a better one for the sponsor than a covenanted structure would have. A strong answer frames cov-lite as a reallocation of control and recovery between lender and sponsor, set by the balance of the financing market at the time, not as a vague decline in discipline.

However the trade-off is judged, the mechanics of what happens when a covenant does bite are the same — and they are the bridge to the part of the market that exists precisely because covenants fail.

What a Breach Actually Triggers: Waiver, Reset, Cure — or the Keys

A covenant breach is an event of default, but acceleration — the lender demanding the whole loan back at once — is the last resort, not the first response, because a lender that accelerates a going concern usually destroys the value it is trying to recover. What follows a breach is almost always a negotiation, and the outcome runs along a spectrum.

1. Waiver and reset. The most common outcome. Lenders waive the breach in exchange for a fee, an increase in the margin, tighter terms going forward, and often a reset of the covenant levels to a more realistic path. The loan stays in place; the borrower pays for the breathing room.

2. Equity cure. If the agreement allows it and the sponsor is willing, fresh equity is injected and counted toward the covenant test, curing the breach without the lenders needing to grant anything — subject to the cure limits written into the document.

3. Amend and extend. A broader renegotiation that reshapes covenants, maturities and pricing together, typically when the breach reflects a structural rather than a temporary problem.

4. Restructuring. If the business cannot support the debt at all, the negotiation moves to the balance sheet itself — a debt-for-equity swap in which lenders write off part of their claim in exchange for ownership, and the sponsor's equity, sitting last in the queue, is often wiped out. The lenders take the keys.

The order matters: each step is more expensive and more adversarial than the last, and both sides have strong reasons to settle early. A maintenance covenant forces that conversation while the cheaper options are still open; cov-lite tends to delay it until only the expensive ones remain. That escalation is the entire reason the restructuring market exists — and why a covenant breach on one desk is the opening of a file on another.

Anyone can recite that a covenant is a leverage or coverage ratio in a loan document — that is the definition. The judgement is knowing what each covenant is worth: that a maintenance covenant's value is the early date it hands the lender, that the EBITDA it tests is a negotiated definition and not the income statement, that 30–35% headroom and an equity cure turn a breach into a capital call rather than a cliff, and that cov-lite did not remove risk but moved it — buying the sponsor runway at the cost of the lender's recovery. Reading covenants as a map of who controls the company when things go wrong, rather than a list of ratios, is the difference between someone who has negotiated a credit agreement and someone who has seen one.

Careers: In a Buyout an Analyst Models to the Covenant, Not Just the Base Case

On a live financing the covenant is not a clause the analyst notes and moves past — it is a constraint the model is built around. The base case has to deleverage fast enough to clear each stepped-down test with the agreed headroom intact, and the real work is the downside: flex EBITDA down, hold the debt, and find the quarter the leverage covenant trips. That date is the number the deal team and the lenders actually negotiate over.

The judgement shows in what the analyst does with the breach. How many turns of headroom does the structure need to survive a recession-case quarter? How much equity cures the worst year, and does the fund have the appetite to commit it? Which definition of EBITDA — and which add-backs — moves the trip date by a quarter or two? An analyst who models only the base case produces a deck that looks safe and tells the team nothing; one who models to the covenant, downside and cure together, produces the analysis the partners use to decide how much debt the business can actually carry. The first is arithmetic — the second is the reason the seat exists.

Take Your Preparation Further

Covenants sit on top of the debt they govern, so read them alongside the structure they attach to. Start with the LBO debt stack, which is the set of tranches each covenant polices, and the Topco–Midco–Bidco structure that decides which entity borrows and therefore where the covenant bites. For the earnings figure the covenant actually tests, quality of earnings and EBITDA add-backs; for what happens when a covenant fails for good, the restructuring interview questions; and for the model the whole analysis runs through, the LBO model and the paper LBO.

To build the structure yourself — debt by tranche, the covenant tests, and a downside that trips them — see the LBO Model Template, and for how a breach turns into a balance-sheet negotiation, the Restructuring & Distressed Debt Primer.

Ready for personalised feedback? Book a 1-on-1 mentoring session with an experienced IB/PE professional.

Debt Covenants in an LBO: Maintenance vs Incurrence, Cov-Lite, and How a Leverage Ratio Trips a Default Before a Payment Is Ever Missed