Corporate Bonds and Credit Spreads: Investment Grade to Junk

Somewhere in the middle of every high yield bond indenture, in a section nobody reads at the time and everybody reads afterwards, sits a sentence to the effect that the company will not incur additional indebtedness unless, on the date it does so, its consolidated coverage ratio would be at least two to one. That sentence, and the twenty pages of definitions that decide what counts as indebtedness and what counts as coverage, is the actual product being sold, and the rating is a three-letter summary of it, produced by an agency reading the same document with a house view about what usually happens. The coupon is what you are paid for accepting it. A corporate bond is a Treasury plus a promise from a company, and the promise is written down in full somewhere, which is more than can be said for most things bought in financial markets.

Credit investing separates two things. One is the lending of money over time, the other is the risk of the borrower, and doing that cleanly is the whole discipline. The number that separates them is the spread.

The queue, and the documents that describe it

When a company borrows, its lenders line up. The order decides who is paid first if things go badly. Senior secured debt has a claim on specific assets. Senior unsecured debt ranks next, which is where most investment grade corporate bonds sit. Subordinated debt stands behind that, and the shareholders come last, which is why equity can go to zero in a restructuring while bondholders walk away with a substantial share of face value.

Position in that queue matters more than the coupon in the one scenario that decides the outcome, and two bonds from the same issuer, with the same maturity, can trade at very different prices on seniority alone, so a buyer who has compared only the yields has compared nothing.

Two documents settle the question. The indenture sets out the covenants, meaning the promises the company has made about how much more it may borrow, what it may sell, and what tests it must pass before it may pay a dividend. Weak covenants let a company load on debt after you have lent to it, transferring value from bondholders to shareholders without a single missed payment and without anything that would appear in a news story. The offering circular sets out seniority, security and the call schedule. Both are worth more of your time than the rating. The rating is an opinion about them.

The underlying mechanics of coupons, prices and yields are identical to any other bond. They are worked through in how bonds work.

The letters, and what the agencies say they mean

Rating is an old business. John Moody began on railroad bonds in 1909, which makes it older than the Federal Reserve. The scales the three agencies use today line up closely enough to read across.

Moody’s	S&P and Fitch	Category	What it signals
Aaa	AAA	Highest grade	A very small number of issuers worldwide
Aa1 to Aa3	AA+ to AA-	High grade	Very strong capacity to pay
A1 to A3	A+ to A-	Upper medium	Strong, with some sensitivity to conditions
Baa1 to Baa3	BBB+ to BBB-	Lowest investment grade	The line most mandates are written around
Ba1 to Ba3	BB+ to BB-	Speculative	High yield begins here
B and below	B, CCC and below	Substantial risk	Default is a live possibility

It repays reading the agencies’ own definitions. Moody’s describes obligations rated Baa as medium grade, subject to moderate credit risk, and as such possibly possessing certain speculative characteristics, and that last clause sits inside the lowest investment grade rating, the one an enormous quantity of institutional money is required to treat as safe, where it has been sitting in plain sight the entire time.

The line between Baa3 and Ba1 is the one with consequences. Index rules and investment mandates are drawn along it. A downgrade across that line can force selling by holders who have no opinion on the credit whatsoever. The mandate does the selling. A bond that crosses downward is a fallen angel. One that climbs back is a rising star. The scale of what that can mean became clear in May 2005, when two of the largest American industrial borrowers were downgraded to speculative grade within days of each other, and the high yield market had to absorb a volume of paper it had not been built to hold.

Ratings are backward looking by construction. They are opinions formed from published financial statements by committees that move deliberately, which is a virtue in a rating and a limitation in a warning, and Penn Central was rated investment grade shortly before it defaulted on its commercial paper in 1970. The market has repriced many credits well ahead of the agencies in the cycles since, which is the argument for watching spreads, provided you remember that spreads are also frequently wrong, merely wrong sooner.

What 140 basis points has to cover

Suppose a ten-year Treasury yields 4.00% and a ten-year BBB rated corporate bond yields 5.40%. Both figures are illustrative. The spread is 5.40% - 4.00% = 1.40%, which the market would quote as 140 basis points.

Ask what those 140 basis points have to pay for. Assume for the sake of the example that this bond carries a 1% chance of defaulting in any given year, and that holders recover 40 cents on the dollar when it does:

1. Loss given default is 100% - 40% = 60%.
2. Expected annual loss is 1% x 60% = 0.60%, or 60 basis points.
3. That leaves 140 - 60 = 80 basis points.

Those probabilities are assumptions built for a worked example. The structure is what carries over. A spread buys compensation for expected credit loss, plus a residual for everything expected loss does not cover, and that residual is paying for illiquidity, for the risk of a downgrade well short of default, and for the plain uncertainty in the default estimate itself, which in most cases is the largest of the three.

Run the same arithmetic on a high yield bond at a 450 basis point spread, with an assumed 4% default rate and the same 40% recovery. Expected loss is 4% x 60% = 2.40%, leaving 210 basis points. High yield pays a great deal more. It surrenders a great deal more of it to defaults, and the surrender arrives in concentrated bursts, years apart.

Professional quotes usually use option-adjusted spread, written as OAS, which strips out the value of any call or put feature so that bonds with different structures can be compared on credit alone. A callable bond quoted at a nominal spread of 200 basis points might carry an OAS of 150 once the issuer’s right to redeem early has been valued and deducted. When two spread figures on the same bond disagree, look for the embedded option.

Spreads as a signal, and the two sensitivities

Credit spreads move faster than economic data and a good deal earlier than ratings. When lenders grow uneasy they sell corporate risk and buy Treasuries, so corporate yields rise while Treasury yields fall and the spread widens from both ends at once. That gives a continuous reading where most macro data arrive monthly and revised, which is why spreads appear alongside the yield curve, jobless claims and the leading index in recession indicators.

The relationship runs in the other direction too. That is what makes credit awkward as a forecasting tool. Wide spreads raise the actual cost of borrowing for actual companies. That slows investment and hiring, and makes the deterioration more likely. Central bank policy works partly through this channel, which is covered in how the Federal Reserve works, and the point was made explicitly in the spring of 2020, when the Federal Reserve announced facilities to buy corporate bonds and spreads narrowed sharply on the announcement, well before any material quantity of bonds had been purchased. The spread was measuring the expected behaviour of a central bank.

Meanwhile a corporate bond carries two exposures that have to be tracked separately. Interest rate duration measures what happens when Treasury yields move. It is calculated exactly as described in duration and interest rate risk. Spread duration measures what happens when the spread moves while Treasury yields stand still. For a straight corporate bond the two numbers are close. A bond with a spread duration of 7 gives up about 7 x 1% = 7% of its value if its spread widens by 100 basis points, with the Treasury curve entirely unchanged.

Where corporate bonds disappoint

The diversification argument breaks first. High yield credit correlates with equities in exactly the weeks when the correlation matters. Both claims depend on the same companies staying solvent. Holding junk bonds as the defensive portion of a portfolio means owning equity risk with a bond’s ceiling on the upside, which is worth checking against asset allocation.

Liquidity breaks second. Corporate bonds trade over the counter through dealers, and a retail sized lot of $10,000 face value can carry a markup of a percent or more, buried inside the price where no commission line will ever show it. The cost of getting out of a single corporate bond in a stressed week can exceed a full year of the spread you were being paid to hold it, which turns a good credit decision into a poor investment.

Call features break third. A great many corporate bonds can be redeemed early, so a period of falling rates removes the bond you most wanted to keep and hands back cash to be reinvested at the new lower rates. Price that possibility with yield to worst. It is explained in bond yields explained and computed for you by the bond yield calculator.

For most investors the practical route is a fund, where hundreds of issuers dilute any single default and trading costs are absorbed at institutional scale, and the trade-offs are set out in bond ETFs explained. Before choosing between credit and government paper at all, look at what the Treasury curve is paying on the yield curve tool and read the yield curve explained, because in some conditions the spread on offer is a poor reward for giving up a borrower that cannot run out of its own currency.