A cap fixes the worst rate the loan can pay. Whether that rate passes the covenant is a separate calculation, and the premium belongs in it.
Only if the strike is set against the covenant rather than against a view on rates. On a $66.9M floating loan with $6.4M of NOI and a 2.50 per cent spread, a cap struck at 6.50 per cent on SOFR still lets the all-in rate reach 9.00 per cent, and the DSCR after the cap premium falls to 0.99x, well below a 1.15x cash sweep trigger. The cap limits the loss; it does not keep the loan inside its covenant.
Worked in full in Real Estate Fund Management by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
The case is the rate-shock sheet of the debt workbook that accompanies Real Estate Fund Management. A stabilised asset carries an interest-only floating-rate loan, sized at 1.25x on a 6.50 per cent, 30-year amortising constant (7.66 per cent); on the closing interest-only rate the same loan covers 1.41x. The sponsor bought a three-year cap and amortises its premium at $420k a year. The loan agreement springs cash management at 1.15x and defaults at 1.05x. All figures are illustrative.
| Input | Value |
|---|---|
| Net operating income | $6.4M |
| Loan amount, interest only | $66.86M |
| SOFR at closing | 4.30% |
| Credit spread | 2.50% |
| Cap strike, on the index | 6.50% |
| Cap premium, amortised per year | $420k |
| Cash sweep trigger | 1.15x |
| Default trigger | 1.05x |
A cap is struck on the index, not on the all-in rate. The borrower still pays the spread on top, so the most the loan can ever cost is the strike plus the spread: 6.50 plus 2.50, or 9.00 per cent. That is the rate to test the covenant against, because it is the rate the cap guarantees.
Capped all-in rate = MIN(index, strike) + spread
DSCR before premium = NOI / (loan × capped all-in rate)
DSCR after premium = NOI / (loan × capped all-in rate + annual cap cost)
Excel: =NOI/(Loan*(MIN(Index,Strike)+Spread)+CapCost)
At closing the all-in rate is 6.80 per cent, debt service is $4.55M and the DSCR is 1.41x, or 1.29x once the premium is counted. Then SOFR moves.
| SOFR move | Index paid | All-in rate | Debt service | DSCR before premium | DSCR after premium |
|---|---|---|---|---|---|
| At closing | 4.30% | 6.80% | $4.55M | 1.41x | 1.29x |
| +100bps | 5.30% | 7.80% | $5.22M | 1.23x | 1.14x |
| +200bps | 6.30% | 8.80% | $5.88M | 1.09x | 1.02x |
| +300bps | 6.50% | 9.00% | $6.02M | 1.06x | 0.99x |
| +500bps | 6.50% | 9.00% | $6.02M | 1.06x | 0.99x |
The cap does its job from +300bps: the DSCR stops falling. But it stops at 1.06x, which is inside the cash sweep and only just above default on the covenant line, and at 0.99x on the line the fund actually lives with once the premium is paid. At the capped rate the asset does not cover its own financing: NOI of $6.4M against $6.44M of debt service and premium leaves a deficit of $37k a year.
Solve for the index at which the sweep bites. A 1.15x test on $6.4M of NOI allows $5.57M of debt service. Divided by the loan, that is an all-in rate of 8.32 per cent, or an index of 5.82 per cent. The loan therefore has 152bps of headroom on SOFR, and the cap only starts working 68bps above the point where cash is already being trapped.
Count the premium as debt service, which is what it is economically, and the trigger index falls to 5.20 per cent: 90bps of headroom. To keep 1.15x after the premium with the same $420k cost, the strike would have to sit at 5.20 per cent rather than 6.50.
The debt yield does not move along the whole path: $6.4M over $66.86M is 9.57 per cent at every rate. That is why lenders lean on it in volatile markets, and why a loan can pass its debt yield test while failing its coverage test on the same day.
A lower strike costs more, so the useful question is not which DSCR each strike produces but how much premium each one can afford. The break-even premium is the annual cost at which the worst-case DSCR after premium lands exactly on 1.15x.
| Strike | Worst all-in rate | Worst debt service | DSCR before premium | Break-even premium a year |
|---|---|---|---|---|
| 5.00% | 7.50% | $5.01M | 1.28x | $551k |
| 5.50% | 8.00% | $5.35M | 1.20x | $216k |
| 6.00% | 8.50% | $5.68M | 1.13x | none |
| 6.50% | 9.00% | $6.02M | 1.06x | none |
At 6.00 per cent and above, no premium is low enough: even a free cap leaves the loan in the sweep. A 5.00 per cent strike protects the covenant if the dealer quotes less than $551k a year; a 5.50 per cent strike only if it costs less than $216k. Those two numbers are what to take to the hedge provider. The strike the lender requires is often the one in the term sheet, and it is usually set to protect the lender's default test, not the sponsor's distributions.
The usual error is to read the strike as the protected rate. A 6.50 per cent cap sounds like 6.50 per cent money; it is 9.00 per cent money. The second error is to leave the premium out of the coverage arithmetic because the loan agreement tests DSCR before it. Here the premium is 7.0 per cent of worst-case debt service and takes 0.07 off the ratio, which is the difference between a thin covenant pass and a property that needs equity every year. To restore 1.15x at the capped rate with the premium counted, NOI would have to be $7.40M, 15.7 per cent above today.
The same discipline applies when the loan is sized. A loan sized at 1.25x on a 6.50 per cent amortising constant has no cushion against a cap that only binds at 9.00 per cent. The loan sizing template on this site, DSCR, debt yield and LTV, shows how proceeds move when the coverage test is run at the capped rate instead.
The rate path, the cap and the sweep trigger are live in the debt sizing workbook, one of the free workbooks for this case, so a different strike or spread reprices the whole table. For what happens once a coverage test is breached and the cure is cash, see why prepaying to cure a covenant costs more than a deposit.
On the index. The borrower keeps paying the credit spread on top, so the most the loan can cost is strike plus spread. A 6.50 per cent SOFR cap on a loan at a 2.50 per cent spread caps the all-in rate at 9.00 per cent, not 6.50. Covenant tests should be run at that capped all-in rate.
Most loan agreements test DSCR before the premium, but the premium is a real annual cost. In this case $420k a year takes the worst-case DSCR from 1.06x to 0.99x. Leaving it out tells you whether the lender is protected, not whether the property covers its own financing.
Solve for the index at which the covenant you care about is breached, then strike below it. Here a 1.15x sweep bites at 5.82 per cent SOFR, or 5.20 per cent once the premium is counted. Then compare the dealer's quote with the break-even premium: $551k a year at a 5.00 per cent strike.
Debt yield is NOI divided by the loan amount, so the interest rate does not enter it. On this loan it is 9.57 per cent at every point of the rate path, while the DSCR falls from 1.41x to 1.06x. That is why lenders favour it in volatile rate markets.
The rate shock and the cap premium are worked in the Chapter 10 debt workbook of Real Estate Fund Management. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: real estate investing, finance and fund management → · All 324 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.