What are the instruments?
Three, plus the decision not to hedge, which is itself a position. A cap is an option. The borrower pays a premium at the outset and receives payments from the seller whenever the reference rate exceeds a strike. The maximum loss is the premium; the protection is one-directional, so if rates fall the borrower keeps the benefit and has simply spent the premium.
A swap is a bilateral obligation. The borrower exchanges a floating payment for a fixed one, there is no upfront premium, and the rate is certain for the term. What a swap has instead of a premium is a mark-to-market value, which means an early exit at a moment when rates have fallen produces a breakage cost, payable by the borrower.
A collar sits between them. The borrower buys a cap and simultaneously sells a floor, which reduces or eliminates the net premium in exchange for giving up the benefit of rates falling below the floor. It is the instrument that best matches how most borrowers actually feel about the exposure: they want protection against a large move up and are indifferent to a large move down.
The fourth option, remaining unhedged, is defensible and frequently correct. It is a view that the exposure is small relative to the cash flow, or that the debt will be repaid or refinanced before the rate matters, or that the forward curve already prices the outcome the borrower fears. What makes it indefensible is holding it without having calculated the exposure.
| Cap | Collar | Swap | |
|---|---|---|---|
| Upfront cost | A premium, paid at the outset and not recoverable | Reduced or zero net premium, funded by selling a floor | None |
| If rates fall | You keep the benefit and have spent the premium | You keep the benefit down to the floor | You pay the fixed rate regardless |
| If rates rise | Protected above the strike | Protected above the strike | Fully protected |
| Cost of exiting early | None; the option simply lapses or is sold | Possible, because the sold floor carries value | A breakage payment when rates have fallen, typically secured alongside the loan |
| Where it fits | A business that can absorb a moderate move but not a large one | A borrower indifferent to rates falling and wanting low net cost | A borrower who needs certainty for a covenant, a contract or a model |
What does the curve currently price?
Higher rates at the end of 2027 than today. The three-month forward curve puts the rate at 3.90 percent at the end of 2026 and 4.04 percent at the end of both 2027 and 2028, against a one-month term rate of 3.65 percent and a three-month rate of 3.76 percent as at 4 August 2026. The eleven basis point gap between the one-month and three-month rates is itself a signal: the term curve is pricing a move inside the next quarter.
The policy record supports it. The Federal Open Market Committee held its target range at 3.50 to 3.75 percent on 29 July 2026 by a vote of nine to three, with all three dissents in favour of raising by 25 basis points, the first time since September 2016 that three policymakers dissented in the same direction. The June projections put the median 2026 rate at 3.8 percent, 2027 at 3.6 percent and 2028 at 3.4 percent, with the 2027 participant range spanning 2.9 to 4.4 percent and nine participants projecting at least one 2026 hike (FOMC statement, 29 July 2026; Federal Reserve Summary of Economic Projections, 17 June 2026). Market pricing put roughly a 62 percent implied probability on a 25 basis point increase at the September meeting as at 4 August 2026.
Forecasters disagree by more than the size of the move. Named 2027 calls from major research houses span roughly 100 basis points, from two cuts in 2027 at one house to a next move that is a hike at another, with a further house projecting two 2026 hikes and then a prolonged pause. A poll of 104 forecasters in July 2026 produced a median call of a hold through the end of 2027 while 44 of 67 respondents rated the chance of a 2026 hike as high, having said the opposite a month earlier (Reuters, 21 July 2026).
The borrower-facing conclusion is not a rate view. It is that the free option has expired. For eighteen months a floating-rate borrower could defer the hedging decision because the expected path was downward, and deferral was cheap. The curve, the dots and the dissents now all point the other way, and specialist guidance for the second half of 2026 is explicit that borrowers should term out floating-rate debt rather than bet on cuts (ABF Journal, 27 July 2026).
“Hedging is bought at the wrong time and sized against the wrong number more often than any other financing decision I see. Borrowers hedge after the move they were afraid of, on the full facility rather than on the exposure, for the term of the loan rather than the term of the risk. Each of those three is expensive on its own.”
What does a cap cost?
We are not going to print a number, and it is worth explaining why rather than quietly leaving it out. No dealer, exchange, regulator or professional body publishes an indicative premium series for privately negotiated interest rate caps. The figures that come up in a search all originate with pricing calculators built to generate enquiries, none of which discloses a methodology or an as-of date. A premium quoted without the curve, the volatility surface and the date it was struck is not a price.
What can be said precisely is what drives it. Three inputs decide a cap premium. The forward curve, because the expected payoff is the area by which the forward path exceeds the strike, so an upward-sloping curve raises the premium. Implied volatility, because a cap is an option and a wider distribution of possible outcomes is worth more. And the dealer's own credit and capital charge, which is a spread on top and which varies by counterparty and by borrower.
Two structural facts follow from that and they matter more than any quoted level. First, a cap premium rises steeply as the strike falls toward the forward path, which is why a strike close to today's rate is expensive and a strike well above it is cheap: you are buying protection against something the market already thinks unlikely. Second, in an environment where the curve slopes upward, caps are more expensive than they were in an environment where it sloped down, for identical strikes. Both of those are true today.
For a swap the equivalent question has a cleaner answer, because the fixed rate available is readable off the curve. A borrower can see the approximate level of a term swap by reading the forward path above, and the negotiation is about the dealer's spread and the credit terms, not about the level.
Why does the lender require a hedge, and is its requirement the right size?
Because the lender is underwriting coverage, and an unhedged floating-rate borrower has a coverage ratio that moves with something neither party controls. Median interest coverage across the middle market held at 1.6 times over the twelve months to 30 June 2026, and the share of borrowers with improving coverage plateaued after more than two years of gains (KBRA, Q2 2026 Middle Market Compendium, published 28 July 2026). At 1.6 times coverage a sustained one point move in the base rate is not a rounding item, and lenders know it.
The requirement usually arrives as a hedging covenant: a stated percentage of the facility, for a stated minimum term, at a strike no higher than a stated level, documented with a counterparty acceptable to the agent. It is negotiable, and it is negotiated far less often than it should be.
Two questions are worth asking every time. First, is the required notional the right notional? A covenant requiring a hedge on the full facility is hedging the revolver as well as the term loan, and revolver balances swing. The exposure is the drawn term debt over the period the hedge covers, not the commitment. Second, is the required term the right term? A hedge covering five years on a loan the borrower intends to refinance in three is three years of protection and two years of breakage risk, because the swap does not terminate when the loan does.
The third point is a documentation one. Where the hedge is provided by the lender or an affiliate, the hedging obligation is usually secured by the same collateral and sits inside the same intercreditor arrangement, which means a breakage payment ranks alongside the loan. That is normal and it is not free: it converts a market position into a secured claim against the business, and it should be read as part of the credit package rather than as a separate treasury matter.
How should the hedge be sized and timed?
Against the exposure, not the facility. Start with the amortizing schedule of drawn term debt, subtract any tranche that is already fixed, and subtract the portion the business intends to repay from cash within the hedge term. What remains is the notional worth hedging, and it is routinely 40 to 70 percent of the facility rather than 100 percent.
Then match the term to the risk rather than to the loan. If the intention is to refinance in three years, a three-year instrument matches the exposure and leaves the refinancing to be priced on its own terms. If the loan carries a maturity the borrower expects to extend, be aware that between 30 and 40 percent of direct lending deals maturing in the next two years have already extended once, and the lender's own framing puts an incremental extension and a restructuring on the same fork (Lincoln International, 11 February 2026). A hedge sized to an assumed extension is a hedge sized to a negotiation that has not happened.
Then choose the instrument by which risk actually matters to the business. If the fear is a large adverse move and the business can absorb a moderate one, a cap struck above the forward path is the cheap and correct answer. If the business needs certainty because a lender covenant or a customer contract requires it, a swap delivers certainty and the breakage risk is the price of it. If the objective is protection at low or zero net premium and the borrower is genuinely indifferent to rates falling, a collar does that, and the floor is where the cost has gone.
One timing discipline, because it is the one that costs the most when ignored. Do not decide the hedge in the same week the financing closes, when attention is exhausted and the lender's form is in front of you. Model the exposure during the term sheet stage, agree the covenant then, and execute the instrument on its own timetable. A hedging requirement negotiated at term sheet is a paragraph. Negotiated at closing it is whatever the form says.
As of August 2026
Sources: Federal Open Market Committee statement, 29 July 2026, for the hold at 3.50% to 3.75% on a nine to three vote with three dissents in favour of a 25 basis point increase, the first time since September 2016 that three policymakers dissented in the same direction; Federal Reserve Summary of Economic Projections, 17 June 2026, for median projections of 3.8% for 2026, 3.6% for 2027 and 3.4% for 2028, a 2027 participant range of 2.9% to 4.4%, and nine participants projecting at least one 2026 increase; CME term SOFR as at 4 August 2026 for the 3.65% one-month and 3.76% three-month rates and the eleven basis point gap between them; three-month forward curve as at 4 August 2026 for 3.90% at end-2026 and 4.04% at end-2027 and end-2028; CME FedWatch implied probabilities as at 4 August 2026 for roughly 62% on a 25 basis point increase at the September meeting; Reuters, 21 July 2026, for a poll of 104 forecasters producing a median call of a hold through end-2027 while 44 of 67 respondents rated the chance of a 2026 increase as high, reversing the prior month, and for the roughly 100 basis point spread across named 2027 research-house calls; ABF Journal, 27 July 2026, for second-half guidance that borrowers should term out floating-rate debt rather than bet on cuts; KBRA, Q2 2026 Middle Market Compendium, LTM ended 30 June 2026, published 28 July 2026, for median interest coverage of 1.6 times and the plateau in the share of borrowers with improving coverage; Lincoln International, 11 February 2026, for 30% to 40% of direct lending deals maturing in the next two years having already extended once and for the framing that puts an incremental extension and a restructuring on the same fork. No indicative premium series exists for privately negotiated interest rate caps and no premium figure has been substituted; the pricing calculators that publish such figures disclose neither methodology nor as-of date. The sizing, term-matching and timing disciplines and the two questions to ask of a hedging covenant are drawn from our own mandate practice. Nothing here is investment advice. Companion articles on this site cover what the curve does to a 2027 refinancing, how coverage rather than leverage decides a financing, and what a second maturity extension actually is.