The covenant that bites first

Bank Rate peaked at 5.25% in August 2023 and sat at 4.75% as this was written, still nearly ten times its average across the 2010s. On the Bank of England’s own measure, 44% of UK corporate debt is now held by firms with interest cover below 2.5×. The binding covenant in this market is not leverage. It is interest cover. A borrower who sizes debt against the spot rate, without testing the through-cycle cash-flow picture, is carrying risk they have not priced.

For most of the 2010s, floating-rate debt was cheap enough that the interest-cover covenant sat comfortably inside whatever leverage package the lender agreed. A business borrowing at SONIA plus 5% on 4× EBITDA, with Bank Rate at 0.10%, was paying around 5.5% all-in — manageable against almost any EBITDA base. The same structure at Bank Rate 4.75% is paying closer to 10%+. The leverage multiple is unchanged; the servicing burden has roughly doubled.

What did Bank Rate actually do?

The path is worth charting precisely, because the speed of the move is as important as the destination. The Bank of England cut to an emergency 0.10% in March 2020. It held there through the whole of 2021. The hiking cycle began in December 2021, accelerated sharply through 2022 with fourteen consecutive increases (the largest a 75-basis-point move in November 2022), and peaked at 5.25% in August 2023. Two cuts to date have brought it to 4.75% as at November 2024.

A borrower who refinanced in 2021 and chose a floating rate locked in at the floor. A borrower who drew down a five-year term loan in early 2023 met the path near its peak and has watched 515 basis points added to their base since liftoff. That range, 0.10% to 5.25%, is the spread of outcomes a single deal structure has had to absorb in five years.

The Bank Rate is the BoE’s own published figure at each MPC decision. The base in a floating-rate facility is SONIA compounded in arrears, which tracks Bank Rate closely but not identically. The chart shows Bank Rate as the floor that anchors the all-in cost; the margin and any SONIA-to-Bank-Rate basis sit on top.

Why interest cover became the binding constraint

Lenders underwrite leverage multiples (net debt to EBITDA) as a measure of quantum. But the covenant that trips first in a rising-rate environment is interest cover: EBITDA divided by cash interest paid. When the rate doubles, EBITDA stays the same but the denominator grows, and the cover ratio compresses without any deterioration in underlying trading. A borrower who was 3× covered at 0.10% Bank Rate may be 1.8× covered at 4.75%, on an unchanged EBITDA. That is inside a 2× covenant floor.

The Bank of England captured the scale of the exposure in its November 2024 Financial Stability Report. By that measure, 44% of UK corporate debt sat with firms where interest cover was below 2.5×, a threshold the Bank uses to flag elevated servicing pressure. That is not a default rate; most of those businesses are still comfortably servicing their debt. But it is a map of where covenant headroom is thinnest, and therefore where a lender’s credit conversation starts before they will move.

What this means for structuring and negotiation

Three things follow for a borrower in the lower-mid-market.

First, the reference rate matters when negotiating a covenant. A cover ratio tested at the current rate of 4.75% will give different headroom than the same ratio tested at a stressed rate of 6%. Most lenders will want to stress-test to a rate above spot. A borrower who knows that number before entering the room, and has modelled how their EBITDA holds against it, is negotiating from fact rather than from optimism.

Second, the leverage multiple itself has not disappeared as a metric: it still anchors how much debt a lender is willing to put on in the first place. The interaction between the two is what matters. A business with variable EBITDA (seasonal, project-based, cyclical) where interest cover can move materially quarter to quarter is more exposed to a point-in-time coverage test than a business with flat, recurring cash flows. The structure should reflect that: either a looser coverage test with tighter leverage, or a cash-sweep mechanism that reduces quantum in proportion to EBITDA performance.

Third, the case for a rate cap or partial fixed leg has strengthened materially. In 2021, with rates near zero and the curve flat, the premium for fixing was hard to justify. At Bank Rate 4.75%, with the direction of travel uncertain, a cap that removes the interest-cover covenant risk for the life of the facility is a structuring choice rather than a speculative one. Whether it is worth the cost depends on the specific covenant package, but it is worth pricing alongside every floating-rate quote.

None of this is a reason to borrow less. Debt at sensible leverage is still the cheapest capital available to a lower-mid-market business, and the field of lenders willing to extend it (banks, credit funds, asset-based lenders) has never been broader. The discipline is in how the structure is built: test coverage against the rate path, not the spot rate; negotiate the test methodology before the level; and make sure the covenant package gives you room to manage through a quarter that undershoots, rather than one that requires a waiver call.