Covenants
Covenant headroom
Headroom is the gap between where your covenant is set and where you are trading, and it is the difference between a difficult quarter and a default.
Also called headroom · covenant cushion · covenant slack
Headroom is measured in EBITDA you can afford to lose. At 25%, earnings can fall to £1.5m before a 3.0x covenant breaks.
| Scenario | EBITDA at the covenant limit |
|---|---|
| Plan EBITDA | £2m |
| 10% headroom | £1.8m |
| 25% headroom | £1.5m |
| 30% headroom | £1.4m |
Illustrative: £6m net debt against a 3.0x covenant. Not a representation of any particular facility.
What headroom protects
A covenant set exactly at your forecast is a covenant you breach the first time trading disappoints. Headroom is the deliberate gap built in so that ordinary volatility, a lost customer, a slow quarter, a delayed contract, does not put the facility into default.
At £3-15m the customary cushion is 25 to 30% against the base case. On a business forecasting £2m of EBITDA against a 3.0x covenant, that means the structure tolerates EBITDA falling to roughly £1.5m before the test fails.
The three ways headroom is measured
Borrowers, lenders and models often mean different things by the same word, and the confusion causes real errors.
EBITDA headroom is the one that matters operationally: how much earnings can fall before the test fails. It is the number to give a management team, because it is expressed in the units they run the business in. Ratio headroom is the gap between the covenant multiple and the actual, so 3.5x against 3.0x actual is half a turn. It is what appears in compliance certificates and is easy to compute, but it understates risk at high leverage, since half a turn at 4.5x is a far smaller earnings cushion than half a turn at 2.0x. Time headroom asks how many quarters of the current trajectory remain before a breach, and is the most useful of the three when performance is already declining.
All three matter. A structure can look comfortable on ratio headroom and be two quarters from a breach on time headroom.
Headroom is not fixed. A step-down raises the earnings you need each year, and on flat trading it eventually passes you.
| Facility year | EBITDA required |
|---|---|
| At close (3.5x) | £1.71m |
| Year 2 (3.25x) | £1.85m |
| Year 3 (3.0x) | £2m |
| Year 4 (2.75x) | £2.18m |
Illustrative: £6m net debt against a stepping covenant, flat £2m EBITDA. Derived arithmetic.
Why thin headroom is the real risk
Borrowers tend to negotiate hardest on margin, because margin is a number they pay every quarter and headroom is a number they hope never to use. That is the wrong way round.
A breach is not a fee. It hands the lender the right to reprice, to impose conditions, to require an equity injection, or in principle to accelerate. Even resolved consensually it consumes management time at exactly the moment the business needs it elsewhere, and it usually costs more than the margin saved in getting there. Twenty five basis points of margin on a £6m facility is £15k a year. A covenant reset costs considerably more than that, in fees and in leverage lost across the table.
Sizing it from the downside
Headroom sized off the plan is not headroom. The useful exercise is to build a real downside, the one where the largest customer halves or an input cost rises and cannot be passed on, and then set the covenant so the downside still passes.
If the covenant only works in the base case, the structure is telling you something: either the leverage is too high for the earnings volatility, or the instrument is wrong. That is a more valuable finding at term-sheet stage than at the first failed test. A business with high operational gearing, where a small revenue fall produces a large earnings fall, needs materially more headroom than the convention for the same comfort.
Headroom differs by covenant
The 25 to 30% convention belongs to the leverage test. The other covenants in the suite carry their own cushions, and they are not the same.
Interest cover is usually set a little tighter, because interest is more predictable than earnings. Debt service cover is tightest of all: it counts scheduled amortisation alongside interest and measures against cash rather than EBITDA, so it has the least natural slack and is the test most likely to fail first in a facility with heavy repayments. Capex covenants are typically loosest, often with carry-forward of unused allowance.
The practical consequence is that the headline leverage headroom can be reassuring while the binding constraint sits elsewhere. Every test moves differently, and only one of them governs.
Headroom is not one number. Each covenant carries its own, and the tightest is the one that governs.
| Covenant | Relative cushion |
|---|---|
| Debt service cover | Tightest test |
| Interest cover | Middle |
| Leverage | 25% to 30% |
| Capex limit | Widest variance |
Market convention at £3-15m. Levels are negotiated and vary by sector.
What erodes it over the facility life
Headroom is a stock, and several things draw it down without anyone deciding to spend it.
The step-down schedule is the largest. A covenant tightening a quarter-turn a year raises the earnings you must produce simply to stand still: on £6m of net debt, a covenant moving from 3.5x to 2.75x lifts the required EBITDA from about £1.71m to £2.18m. A business trading flat at £2m passes comfortably at the start and fails by year four with nothing going wrong. Accrued PIK, where it counts as debt, raises the numerator every quarter. A mid-year acquisition adds debt before it adds a full twelve months of earnings, so leverage spikes unless the definition annualises. A capex programme depresses cash and, where the covenant is cash-based, consumes headroom directly.
Each of those is foreseeable at signing, which is the argument for modelling the tightest point of the schedule against the downside case rather than checking the opening level and moving on.
What restores it
Three mechanisms give headroom back, and they are worth negotiating for even if you never use them.
An equity cure lets shareholders inject funds to fix a ratio, usually capped in frequency and sometimes in amount, and the drafting decides whether the injection reduces debt or is treated as EBITDA. The second is more valuable, because it flatters the ratio on both sides. A springing covenant only tests once a trigger is crossed, typically revolving facility utilisation, which is effectively unlimited headroom below the line. And a reset mechanism tied to a defined event, such as a completed acquisition or a disposal, resets the schedule to reflect the new shape of the business.
When headroom is thin
The worst response is to wait for the test date. A lender that learns about a breach from a compliance certificate is in a materially different frame from one told six weeks in advance with a plan attached.
The practical sequence is to re-forecast honestly, quantify the shortfall against the covenant, decide which lever you are asking for, and go early. The levers are a reset of the schedule, a waiver for a defined number of periods, an equity injection, a disposal applied to prepayment, or an amendment converting a maintenance test to a springing one. Each is priced, and the price is lower when the request arrives with a credible forecast rather than after a failure.
This is also the point at which competitive tension matters most, because a borrower with an alternative lender interested in refinancing negotiates a waiver on quite different terms from one without.
Common questions
How much covenant headroom should I have?
At £3-15m the customary cushion on a leverage covenant is 25 to 30% against the base case. On £2m of EBITDA against a 3.0x covenant, that tolerates earnings falling to roughly £1.5m before the test fails. A business with high operational gearing, where a small revenue fall produces a large earnings fall, needs more than the convention for the same comfort.
How do you calculate covenant headroom?
Three ways, and they answer different questions. EBITDA headroom is how far earnings can fall before the test fails, which is the number to give management. Ratio headroom is the gap between the covenant multiple and the actual, which is what appears in compliance certificates. Time headroom is how many quarters of the current trajectory remain before a breach, which is the most useful measure when performance is already declining.
Is headroom the same on every covenant?
No. The 25 to 30% convention belongs to the leverage test. Interest cover is usually set tighter, and debt service cover tighter still because it counts scheduled amortisation as well as interest and measures against cash. Capex covenants are typically loosest. The tightest test governs, so the headline leverage headroom can look reassuring while the real constraint sits elsewhere.
Can I lose headroom without my business getting worse?
Yes, and it is common. A step-down schedule tightens the covenant each year, so a business trading flat can move from comfortable to breached in about three years with nothing going wrong. Accrued PIK counting as debt, a mid-year acquisition before earnings are annualised, and a capex programme against a cash-based test all erode headroom the same way.
Should I trade margin for headroom?
Usually yes. Twenty five basis points on a £6m facility is about £15k a year. A covenant reset costs considerably more than that in fees and in negotiating position, and it arrives at the moment management least needs the distraction. Borrowers systematically over-negotiate the number they pay every quarter and under-negotiate the one they hope never to use.
What restores headroom once it is thin?
An equity cure lets shareholders inject funds to fix the ratio, usually capped in frequency, and the drafting decides whether the injection reduces debt or counts as EBITDA. A springing covenant that only tests above a utilisation trigger gives effectively unlimited headroom below the line. A reset tied to a defined event such as a completed acquisition adjusts the schedule to the new shape of the business.
What should I do if I am heading for a breach?
Go early. A lender told six weeks in advance with a re-forecast and a plan is in a different frame from one that learns from a compliance certificate. The levers are not equivalent: a schedule reset, a waiver for defined periods, an equity injection, a disposal applied to prepayment, or conversion to a springing test. Each is priced, and the price is lower when the request arrives with a credible forecast.
Does headroom matter if my covenants are cov-lite?
The question changes rather than disappears. Incurrence covenants only test when you take an action such as raising debt or paying a dividend, so there is no quarterly failure risk, but the tests still constrain what you can do and when. At £3-15m fully cov-lite structures are rare; springing covenants that test only above a utilisation threshold are the more common middle ground.
The full treatment sits in the guide: loan covenants explained.
This page explains a term as it is used in the UK lower-mid-market. It is general information, not advice on any particular facility. Terms vary between lenders and between deals, and the drafting in your own agreement governs.