Corporate bonds look cheapish
Yesterday Unhedged complained that, given the economic backdrop, stocks looked a little expensive. So what is one to own instead? It is hard to recommend fixed income, given the high volatility — mostly upward volatility — of interest rates. For the first time in a long time, even decent-quality bonds are as scary, or scarier, than stocks.
Another leg up in rates, and leg down in bond prices, cannot be ruled out when no one knows for sure where the Federal Reserve will stop. Flows into bond funds have been deeply and persistently negative all year, except for the brief period this summer when the market saw the mirage of a Fed pivot.
Still, there is a case to be made for buying bonds, and specifically corporate bonds. Here is that case, put very crudely, in a single chart:
That is the yield on bonds from the highest tier of junk-rated bonds, compared to the earnings yield on the S&P 500 (that is, the reciprocal of the price/earnings ratio). Right now, BB bonds yield more. This does not happen very often. The yield differential is about what is was in the acutely frightening early days of the coronavirus pandemic. Are things as bad now as they looked then? There are assorted perfectly technical quibbles that could be raised with this comparison, but it gets to the heart of the argument. Bonds look cheapish, relatively.
Why cheapish, not plain old cheap? A lot of what makes corporates look cheap compared to stocks is higher Treasury yields — a higher risk-free rate. The additional compensation investors are getting for taking on BB default risk, about 330 basis points, is only middling by historical standards. Here is a long-term chart of the spread:
In addition, the compensation for trading down from the lowest rung of investment grade (BBB) to the highest rung of junk (BB) is not all that special. The difference between the BBB and the BB spreads (the spread spread, if you will) stands at 1.4 per cent, just a bit above its 10-year average.
On some measures, high-yield bonds are not cheap at all. Marty Fridson, of Lehmann Livian Fridson, uses a model that shows a fair value spread for the entire high-yield index (BBs and everything junkier) at about 7.2 per cent, well ahead of the 4.8 per cent spread the index currently offers. The main reason is that his model demands more yield when the availability of credit is constrained — as it is now, according to the Fed’s loan officer survey, for example.
During the past few years, most companies pushed their debt maturities years into the future. According to Scott DiMaggio, co-head of fixed income at AllianceBernstein, only about 20 per cent of the bonds in the US market mature in the next three years, “a very manageable number”. But there is more to credit risk than refinancing, as Unhedged recently discussed.
That said, many BB rated bonds are issued by companies that seem very unlikely to default outside of a very severe recession: the index includes Sprint, Ford, Occidental, Yum Brands, T-Mobile and so on.
Interest rate volatility, and the accompanying direction risk, hang over the whole discussion. Investors will remember the brutal 17 per cent fall in the JNK — the largest of the junk-bond exchange traded funds — over the past year. Remember, though, that for most investors duration risk must be weighed against the riskiness of equities and the vulnerability of cash to inflation.
The bonds in the JNK have average maturity of between five and six years. The five-year Treasury has risen almost 2 per cent in the past 12 months. If rates rise another two percentage points, take a moment to think about what that implies about inflation, and what it will do to your stock portfolio.
Finally, the flat yield curve presents an opportunity. Quite good yields are available in bonds with maturities of just a year or two. The yield to worst on Ford’s BB+ bonds maturing in September of 2025 is 6.5 per cent, for example. Sprint’s bonds due June 2024 pay 5.5 per cent.
The corporate bond market is by all accounts very illiquid right now. It is not easy to buy bonds. But I think there is a case to be made for doing so, at the short end, and in the higher reaches of the junk spectrum.