Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The writer is a bond portfolio manager at Barksdale Investment Management and co-author of ‘Undiversified: The Big Gender Short in Investment Management’
Anyone who takes Finance 101 learns that a dollar today is worth more than a dollar in the future. In a normal environment, this basic tenet translates into an increase in the yield earned by an investor from the shortest maturities to the longest, reflecting the risk and opportunity cost of holding on to debt for more time — the so-called “term premium”.
But this maths no longer applies since last autumn’s “inversion of the yield curve”. A two-year Treasury bond currently pays more interest than a 30-year one. Since coupons on bonds issued by companies incorporate the so-called “risk-free rate” of a similar-maturity Treasury yield, you might think that corporate bond issuers are seeing the same inversion. But the effect is not as pronounced.
Credit spreads, which represent the risk premium demanded by investors to hold a corporate rather than a Treasury bond, are set by the market’s assessment of an issuer’s perceived level of risk. Like Treasury yields, they vary by maturity. Investment grade-rated companies pay even more to issue longer debt, relative to shorter maturities, than the US government (normally) does.
This second upward-sloping curve, called a “credit curve,” makes intuitive sense; if it’s riskier to lend to the US Treasury for a longer period, it should be riskier still to do so for a company, even one such as, say, Apple, with a double A credit rating, a $3tn market cap and $166bn in cash. Some might argue that Apple is a better credit than the US, given the wrangling over debt ceilings in Congress and the risk of a default. But a corporate bond is not supposed to trade “through”, that is at a negative spread to, its sovereign in the developed world.
Data from JPMorgan shows that at the beginning of 2022 (and the tightening cycle in interest rates), the increased spread for five-year non-financial corporate debt relative to three-year debt was a mere 0.09 percentage points. It was another 0.35 points for 10- versus five-year debt and 0.36 points for 30 versus 10-year debt. This suggested, reasonably, that there was little credit risk involved in buying five-year Apple debt relative to three-year debt, and more risk involved in “reaching for yield” all the way out to a 30-year Apple bond. IPhone disintermediation risk is a lot easier to forecast for three years than 30.
While the longer-dated credit curve hasn’t changed much since the start of the US interest rate tightening cycle, there have been more intriguing shifts in shorter-term debt.
For example, in June, Nasdaq issued a three-year bond at a spread of 0.90 percentage points, 0.45 points lower than its five-year spread of 1.35 points. Ordinarily a credit curve this steep would signal concern about a deterioration in Nasdaq’s credit quality from 2026 to 2028. But the coupons of the bonds tell the opposite story: the three-year bond has a coupon of 5.65 per cent, while the five-year pays 5.35 per cent. A host of other companies that have issued three-year bonds as part of a larger financing in 2023, including Deere, Benz, PNC, Stanley Black & Decker and Walmart, experienced the same phenomenon.
This places fixed-income investors evaluating such issues in a dilemma. They must consider whether the risk of further Treasury curve inversion is high (worse for the three-year), whether the Federal Reserve will pause and the curve will “bull steepen” (better for the three-year), or whether Nasdaq will improve as a credit in the near term (better for the five-year). If an investor, who probably did take Finance 101, decides to buy the five-year, they are ignoring their expensive MBA by accepting a lower yield for longer Nasdaq risk.
The current corporate bond conditions also matter for other types of investors. For example, an equity investor might attempt to value Nasdaq using a discounted cash flow model, which uses the weighted average cost of capital as an input. But does a calculation of Nasdaq’s WACC make sense when the curve inversion might be distorting its true cost of debt?
When the Treasury curve eventually normalises, none of this may matter much. The cost of debt of Apple, Nasdaq and their ilk will probably go back to showing a reassuring increase as maturity lengthens. But for now, as the heavy autumn issuance calendar hits, corporate bond investors, issuers and underwriters will continue to navigate a financing environment that defies their usual playbook.
Barksdale may hold interests in the securities mentioned