The covenant definitions now set the interest bill as well.
BY MICHAEL SHANG · SIXTEEN YEARS IN COMMERCIAL BANKING · UPDATED · 6 MIN READ
A margin ratchet makes part of a borrower's interest rate depend on its own numbers. The facility agreement carries a grid of ratio bands and margins, and each compliance certificate decides which band applies for the next period. Its purpose is to keep the price in line with the risk without anyone renegotiating. Its catch is that the ratio comes from the covenant definitions, so a disputed add-back or a high cash balance on test day can change the interest paid. It also raises the cost of debt in the quarters when a weakening business can least afford it.
BASIS Practitioner judgment from sixteen years in commercial banking, set out in How Bankers Think (Highbank Press, 2026). Every borrower on this site is a composite constructed for teaching, not a client. The market is mid-market commercial lending as practised in New Zealand and Australia; where a convention differs elsewhere, the page says so.
The first checks a credit desk runs — is the growth real, does EBITDA convert to cash, are suppliers carrying the gap — on twelve numbers from your own accounts, with the senior banker's written read. Free, no signup, no data stored.
PART OF How Banks Price a Business Loan
A bank prices a loan for the risk it sees on the day. Over a five-year facility that risk moves. Without a ratchet, the borrower that deleverages goes on paying a price set for a weaker business and has every reason to refinance, and the lender holding a borrower that has weakened is paid for a stronger one until the next renegotiation. A grid settles both in advance. The borrower has a visible reward for paying debt down, and the lender is compensated, at least in part, as the credit worsens.
Syndicated, club and sponsor-backed loans are where ratchets mostly appear, because leverage is measured every quarter anyway. Many smaller bilateral facilities in New Zealand and Australia carry a single margin that is revisited at the annual review instead. Companies with a public credit rating often have a grid keyed to the rating rather than to leverage.
Five drafting points do most of the work. The ratio: usually net debt to EBITDA, taken from the covenant definitions. The bands: whether a figure exactly on a boundary falls in the higher or lower band. The start: many agreements hold the margin at its opening level for the first year or so, until there is a track record. The timing: the margin changes only after the compliance certificate is delivered, often from the next interest period. And the override: commonly, while an event of default continues, or if a certificate is late, the margin goes to the top of the grid or stops stepping down. Some agreements also recalculate if the certificate proves wrong, and collect the difference. All of these are negotiated deal by deal.
A contract manufacturer of vitamins and supplements, constructed for this page like every figure in it, has a US$50m term loan with no scheduled repayments and quarterly interest periods. Net debt moves around that figure with cash and a small overdraft. The grid belongs to the example and is not offered as a market norm. Each certificate is due 45 days after the quarter ends, and the new margin applies from the next interest period, in US$m.
The June row is the one to look at. The manufacturer's trading did not change because of the add-back, but the quarter-point did. A definition argued over for covenant purposes now has a price of US$125,000 a year for as long as it decides the band, charged on the loan drawn rather than on net debt. In this table that is one interest period, about US$31,500, because September's test would put the margin at 2.50% anyway. Every add-back and every dollar of netted cash has a second effect the document may not have considered when it was drafted.
The timing runs on a lag in both directions. June's improvement is certified by mid-August and paid for from the interest period beginning 30 September, three months after the quarter it reflects. September's slide does not reach the rate until the period beginning 31 December. A borrower that improves waits for its reward, and a lender holding a deteriorating credit waits for its compensation.
A ratchet is automatic and written down in advance. Repricing at review is a decision, taken when the agreement allows the lender to make it, and can move by any amount. A sustainability-linked margin uses the same machinery but tests an emissions or similar target instead of a financial ratio, and its step is usually much smaller. Default interest is a separate charge on overdue amounts or while a default lasts: it is a response to breach, whereas the grid moves inside compliance. A step-up margin, common on bridge loans, rises with the passing of time and has nothing to do with performance.
DEFINED IN ONE PARAGRAPH → · READING THE FACILITY LETTER → · NET DEBT, AS THE DOCUMENT DEFINES IT →
A ratchet is worth stress-testing because it works against the borrower at the wrong moment. Interest goes up as leverage does, so the same fall in earnings that tightens the covenant also raises the interest bill. A forecast built at today's margin flatters the downside case. A lender running a downside should use the margin at the band that the downside lands in, and a borrower should do the same before relying on a covenant reading.
On the lender's side the grid sets an expectation that may not hold. A margin that tops out at a stated level may not pay for a borrower that has moved well beyond the top band, and the margin earned then is what the grid allows, not what the risk would command. The grid is argued at the term sheet alongside the covenant headroom, since the same numbers decide both: the rate the borrower pays, and how close it can come to the point of breach.
COVENANT HEADROOM → · HOW BANKS PRICE A LOAN → · THE ADD-BACK CAP →
The seats that read this term in an ordinary week. Each page says what the seat is for, who arrives in it, and the shape of the round it runs.
THE IN-TRAY · ONE MINUTE
A short file on the ground this page covers. Every option opens with how far it sits from a senior banker’s read.
A US$20m loan. The bank expects to lose 0.5% of it a year on average. It must hold capital equal to 10% of the loan, and its shareholders expect 12% a year on that capital. Funding and running costs are priced separately.
1 FILES · PICK A CALL ON ANY OF THEM
5 MORE ON THE LOAN · PRICE →WHERE DO YOU READ FROM? THE PLACEMENT →
// SUBSCRIBE
One case file and one question, in your inbox every Tuesday morning, New Zealand time. The senior banker's read opens only after you answer.
FREE WITH IT · ANNOTATED PDF
A complete credit paper on a published file, with the approver's notes on what each section has to answer.
No spam. Unsubscribe anytime. Educational content only.
