Enough that a realistic bad year, not a catastrophe, still passes every test in every quarter, including after each scheduled step-down. Lenders set covenant levels as a cushion against the projections in the model, so the headroom you get depends on which case they use and how aggressive it is. Measure headroom as how far EBITDA can fall before a breach, run your own downside case through every test date, and negotiate levels, step-down timing and the EBITDA definition together, at signing, when you still have the leverage to change them.
- What headroom is
- The distance between projected results and the covenant level
- Best way to measure it
- How far EBITDA can fall before each covenant breaks
- Where it disappears
- Step-downs that assume growth, rising rates, seasonal quarters
- What to negotiate
- Levels, step-down timing, first test date, EBITDA definition, a cure right
- When to negotiate
- Before signing; afterwards every change costs an amendment
How lenders set covenant levels
Most lower-middle-market term loans carry two or three maintenance covenants, tested every quarter: a maximum leverage ratio (debt to EBITDA), a minimum coverage ratio (fixed charge coverage or debt service coverage), and sometimes a minimum EBITDA, a capital spending limit or a minimum liquidity test. Asset-based lines usually carry a single fixed charge test that springs into effect only when availability runs low.
The levels are not taken from a table. The lender takes the financial model, its own version or the borrower's, projects each ratio for each test date, and sets the covenant a cushion away from the projection: the leverage maximum above projected leverage, the coverage minimum below projected coverage. The size of that cushion is a negotiation, and it depends on the lender's view of how volatile the earnings are. A contract business with years of steady margins gets a narrower cushion than a project business with lumpy quarters, because the lender expects the second to move more.
The cushion is measured against the model, so the model decides the covenant. An optimistic forecast produces tight covenants.
Measure headroom in EBITDA, not in ratio points
A term sheet states leverage room in turns of EBITDA and coverage room in tenths of a ratio, and the two cannot be compared as written. Each has to be translated into the same question: how much can EBITDA fall before this test fails? That is the only measure that matters to an owner, and it is often much smaller on one test than the other.
Take a company with debt of 9,000 and EBITDA of 3,000, a leverage maximum set at four times EBITDA, and fixed charges of 1,500 against cash flow available for fixed charges of 2,400, with a coverage minimum of 1.25x.
| Test | Where the company is | Where it breaks | EBITDA can fall by | Cushion as a share of the measure tested |
|---|---|---|---|---|
| Leverage: debt no more than four times EBITDA | 9,000 of debt on 3,000 of EBITDA | EBITDA below 2,250 | 750 | A quarter |
| Fixed charge coverage: at least 1.25x | 2,400 available against 1,500 of charges | Available cash flow below 1,875 | 525 | Just over a fifth |
| Fixed charge coverage, after rates rise and charges reach 1,650 | 2,400 against 1,650 | Available cash flow below about 2,063 | about 337 | About a seventh |
Two things show up. First, the covenant that binds is coverage, not leverage, although a full turn of leverage room is the number an owner notices on the term sheet. Second, on a floating-rate loan, rising rates attack coverage from the other side: interest rises, fixed charges rise, and the cushion shrinks without the business doing anything worse. A rate hedge protects covenant headroom as much as it protects cash; see interest rate hedging requirements.
Step-downs: where headroom quietly disappears
Leverage covenants usually tighten over the life of the loan. The lender's logic is that amortization and earnings growth should bring leverage down, so the maximum steps down with it. The step-down schedule is built from the same projections as the opening level. If the model assumes growth, the schedule assumes growth, and a company that holds steady, which in most years is a good result, watches its cushion shrink every quarter.
| Test date | Debt after amortization | Covenant maximum (times EBITDA) | Plan EBITDA | EBITDA can fall by, on plan | EBITDA can fall by, if it stays at 3,000 |
|---|---|---|---|---|---|
| Closing | 9,000 | 4.00 | 3,000 | 750 | 750 |
| End of year 1 | 8,400 | 3.75 | 3,150 | 910 | 760 |
| End of year 2 | 7,800 | 3.25 | 3,450 | 1,050 | 600 |
| End of year 3 | 7,200 | 2.75 | 3,800 | about 1,182 | about 382 |
| End of year 4 | 6,600 | 2.25 | 4,200 | about 1,267 | about 67 |
Nothing went wrong in the flat case. The company paid every installment and kept its earnings. It simply did not grow at the rate the step-downs assumed, and by year four almost any dip trips the covenant, while the same loan on plan would have room to spare. The fixes are to build the schedule off a flat or modest case rather than the growth plan, delay the first step-down, make the steps smaller, or tie them to actual amortization instead of projected growth.
Stress the forecast before you agree
Run your own downside through every covenant on every test date before accepting the levels. Quarterly tests use trailing twelve-month figures, so a single bad quarter stays in the calculation for a year. Seasonal businesses need the test that falls after the weakest quarter checked with particular care. The stresses worth running are the ones that could plausibly happen to this business, not generic haircuts:
- Revenue: a flat year, then a year like the worst one in the company's own history.
- Margin: input costs or wages rising before prices can follow.
- Customers: the loss or slowdown of the largest customer; see customer concentration and debt.
- Rates: the base rate rising on the unhedged share of a floating loan.
- Timing: a large capital purchase, an acquisition that closes a quarter late, or a slower ramp on a new location. Spending the business cannot defer belongs in the test; see maintenance vs growth capex.
- Definitions: whether the add-backs you expect will actually count under the agreement's EBITDA definition.
The last stress is the one most often missed. Headroom is only as real as the definition it is measured under. If the lender's definition of covenant EBITDA caps add-backs, excludes pro forma savings from an acquisition, or counts revolver draws as funded debt when you expected them to be excluded, the cushion in your model may not exist in the agreement.
What to negotiate, and in what order
| Term | What to ask for | Why it matters |
|---|---|---|
| Base case | Levels set off a case the company expects to beat, not its best plan | Every other number follows from it |
| Opening levels | A cushion that survives your own downside case | The first year after closing is when surprises land |
| First test date | A first test a few quarters after closing | Gives acquisitions and transitions time to settle |
| Step-downs | Later, smaller, and tied to amortization rather than growth | Stops a flat year from becoming a breach |
| EBITDA definition | Add-backs you rely on, stated plainly; pro forma credit for acquisitions | Headroom exists only under the agreed definition |
| Equity cure | The right to fix a shortfall with new equity | Turns a breach into a choice instead of a default |
| Test set | No more covenants than the loan needs | Each extra test is another way to trip |
Lenders will usually give more on timing and definitions than on the headline level, because the level is what their credit committee sees. A wider cushion can also be traded: some borrowers accept a slightly higher spread or more amortization in return for looser covenants, which is often the better deal. An equity cure right belongs in the same negotiation, because it is the backstop for the day the cushion runs out.
Why covenants have to be designed at signing
Before signing, a covenant is a term like any other, and the lender is competing for the loan. After signing, every change is an amendment. The lender has no obligation to agree, and when it does it can charge an amendment fee, raise the spread, add reporting or ask for a personal guarantee it did not get at closing. A breach is worse: it is an event of default that can trigger default interest, block distributions and, through cross-default clauses, reach other loans. What to do when you breach a covenant covers the options, and none of them is as cheap as a covenant that was set properly.
Many covenant breaches are not signs of a failing business. They are signs of a line drawn in the wrong place: a step-down that assumed growth, a coverage test that ignored rising rates, a definition that did not match how the company reports. That is a design problem, and it is solved with arithmetic before closing.
The financing model in the package Transparent builds for every borrower projects each covenant on each test date under a base case and downside cases, so the levels a lender proposes can be checked against the company's own numbers before the term sheet is signed. Once a borrower's documents are in, that package is built in a day. How we underwrite explains the approach.
Common questions
- What is a normal amount of covenant headroom?
- There is no single right number; it depends on how volatile the company's earnings are. The practical test is whether your own realistic downside, applied to every test date, still passes every covenant with room to spare.
- Which covenant usually breaks first?
- Often the coverage test rather than leverage, especially on floating-rate loans, because rising rates raise fixed charges at the same time that weaker earnings reduce cash flow. Translate each covenant into how far EBITDA can fall to see which one binds.
- Can covenants be loosened after closing?
- Only by amendment, which the lender does not have to grant. Expect a fee, a higher spread or tighter terms elsewhere in exchange.
- Do asset-based lines have covenant headroom issues?
- Less often. Most carry a fixed charge covenant that applies only when availability falls below a set level; see springing covenants. The headroom question there is about availability as much as the ratio.
- Should I accept tighter covenants for a lower rate?
- Usually not. A small saving on the spread is worth less than the room to have a bad quarter without being in default.