Credit spreads and the cost of debt

Credit risk is the risk that a borrower does not pay. Investors will lend into that risk, but only at a higher return than they would accept from a borrower who cannot default — and the difference between the two is the credit spread.

The risk-free rate is the base because a government borrowing in its own currency is taken as certain to repay. Its yield is pure compensation for waiting — for lending money now and consuming later. Anything a company pays above that is compensation for the possibility of not being repaid at all.

So the spread moves inversely with credit quality. A strong issuer pays a small premium; a weak one pays a large premium; and the same company pays different premiums at different maturities, because ten years is more time for things to go wrong than one.

Spreads are published by credit rating and maturity, quoted in basis points. One basis point is 0.01%, so 100 basis points is 1%.

FS Co is rated AA and has ten-year bonds in issue. The ten-year government bond currently returns 4.2%. Find AA against 10 years — 52 basis points, which is 0.52% — and add it.

Two things to be careful about. Read down to the right rating and across to the right maturity, since the table rewards a misread with a plausible-looking wrong answer. And convert the basis points before adding: 52 is 0.52%, not 52%, and not 5.2%.

Look at the size of the steps as you go down the ratings. Between AAA and BBB at ten years the spread roughly quadruples, from 33 to 122 basis points. Between BBB and CCC it multiplies by ten. Credit risk is not priced in a straight line — it accelerates sharply as quality falls, which is why a downgrade of one notch matters far more at the bottom of the table than at the top.

The table jumps from 10 years to 30, so a 15-year bond has no column. Take the two nearest maturities and work out where between them the bond sits.

Folks Co is B-rated with 15-year bonds in issue, and the 15-year government bond yields 5%. B-rated spreads are 350 basis points at ten years and 450 at thirty.

The middle two steps are where marks go. Divide by the width of the gap — twenty years, not fifteen and not twenty-nine — and then multiply by how far into that gap the bond falls.

It is a straight-line estimate of something that is not a straight line, and the table itself shows it: B-rated spreads widen fast in the early years and then flatten. Over a twenty-year gap, straight-line interpolation is an approximation and worth calling one in an answer.

The yield is what the bondholder receives. The company’s cost is that yield after tax relief, which folds the two ideas together into one line.

Four things determine it, and a question will hand you all four: the company’s credit rating, the maturity of the debt, the risk-free rate at that maturity, and the tax rate.

MP Co is BBB-rated with five-year bonds in issue. Equivalent government bonds yield 3.7% and tax is 28%.

3.42% is the figure that goes into MP Co’s weighted average cost of capital. Notice how little of it is credit risk: of the 4.75% the bondholders receive, 3.7 points are simply the price of time and only 1.05 is the premium for lending to MP Co rather than to the government. Tax then removes more than either.

A rating is an opinion on the probability of default, and it is revised. When it moves, the spread moves with it, and the company’s cost of capital moves whether or not any of its existing debt is repriced.

• Gearing rising — more debt against the same earnings leaves less cover for the interest, and the next lender charges for it.
• Earnings becoming more volatile — a customer lost, a commodity price turning, a single contract carrying too much of the revenue.
• The sector or the country deteriorating — an issuer rarely escapes a downgrade of the sovereign it operates in.
• The market’s appetite changing — spreads widen across every rating in a crisis, with nothing having changed at the individual company.

Follow the arithmetic through a downgrade of a ten-year issue from BBB to BB. The spread goes from 122 to 250 basis points, so the yield rises by 1.28% and the post-tax cost of debt at 30% by about 0.90%. On ZMW200 million of borrowing that is ZMW2.56 million of extra interest a year before tax.

The larger cost is not the interest. A higher cost of debt raises the weighted average cost of capital, and the WACC is the rate every project is appraised at — so projects that were marginally worth doing at the old rate now have negative NPVs. A downgrade does not just make borrowing dearer; it shrinks the set of investments the company is able to justify.

Which is why gearing is not only a financing question. Taking on debt lowers the cost of capital while the rating holds, and raises it once the rating gives way — and finding where that turn happens is what the rest of this lesson is about.