Jeff Davis CFA is a veteran bank analyst. The views and opinions expressed in this piece are those of the author and do not necessarily represent the views of S&P Global Market Intelligence; Mercer Capital, where he is the managing director of the financial institutions group; or StillPoint Capital, where he is a registered representative.
I read a great piece from the bank team at Robert W. Baird & Co. earlier this week. The gist of the thesis is that bank valuations (P/TBV basis) are excessively cheap given significant tightening in high-yield credit spreads over the past few months versus bank valuations that have not improved much. Historically, the two have exhibited a high degree of negative correlation.
Widely followed economist David Rosenberg must have seen the piece or had a client ask him about it. He tweeted the same day that banks are trading for a "credit disaster" compared to high-yield bonds that are trading as though "the default rate is about to get cut in half." He went on to note that both cannot be right and that the disparity may be the biggest anomaly in the market presently.
Credit spreads are the equivalent of a mirror for lenders. Ignoring short-term technical factors that may affect bond prices, widening spreads indicate an increasing probability of default and vice versa. The second derivative of widening spreads is more important, however: expected loss in the event of default.
Both the U.S. Treasury and corporate bond markets are credited with having better vision about where the economy is headed than the equity market. For instance, the U.S. yield curve flattened and then inverted in 2019, forcing the Fed to cut short-term rates three times as a harbinger that the economy was likely going to slow in 2020 before the COVID-19 pandemic crushed it. Credit spreads peaking in November 2008 about four months before equity markets bottomed in March 2009 is another of many examples of the bond market front-running the equity markets.
My theory about why this is the case is that all bond investors can hope for other than when buying bonds at a discount is contractual interest and repayment of principal. In other words, not much. Fixed-income investors must be on top of the factors that affect bond returns. Equity investors are slower to recognize changing fundamentals because narratives matter to stocks, too.
Rosenberg is right about the dichotomy between depressed bank valuations and tight high yield spreads. One of the markets is wrong, although bank valuations also reflect depressed net interest margins that afflict the sector. Maybe both also reflect a possibility that rates eventually go negative in the U.S. as has occurred in Europe and Japan.
I agree with Baird that the risk-reward proposition is much more favorable for banks than high-yield bonds. If the economy gets better and losses are not too bad as implied by the high-yield market, then bank stocks should perform very well. Or, if high-yield bonds are wrong and credit losses will increase significantly, then bank stocks are already priced for such an outcome, whereas high yield is not.
Relative valuations aside, what if high-yield bond prices largely reflect liquidity flows rather than fundamentals?
By this I mean pricing is distorted because Fed intervention has caused capital to flood into high yield bonds. Further, some or maybe much of the capital has flowed into passive exchange-traded funds that are indiscriminate buyers. There are no investment committees to pass on specific bonds, just allocations to be made.
Consider this year's move-in spreads compared to past downturns.
The option-adjusted spread, or OAS, on the ICE BofA Single-B US High Yield Index was 363 basis points on Feb. 21, 2020, when the market began to focus on the risk of a pandemic. Spreads were tight compared to the long-term average (552 bps) during 1997-2020. In four weeks, OAS gapped to 1,189 bps on March 23, 2020, when the Fed announced it would purchase investment-grade credits. A few weeks later, the Fed announced it would buy high yield ETFs and fallen angels that were investment-grade rated on March 22. Four months later, the OAS had narrowed over 50% to 532 bps, the long-term average.
A monster rally occurred in the prior financial crisis, too, but not in the blink of an eye like this time.
OAS for B-rated credits peaked Nov. 21, 2008, at 2,084 bps and then narrowed to 1,459 bps on April 9, 2009, the same amount of time as has elapsed this year since the March bottom. At the time, the Fed and U.S. government were clear that they were not going to let large banks fail, but policy did not involve buying corporate bonds.
Corporate defaults were high in 2009 at 11.0% for B-rated credits according to S&P Global Fixed Income Research. The next year was shaping up to be high, too, yet defaults plummeted to only 0.87%. In the shallow recession of 2001, B-rated credit defaults peaked at 11.5%, then required five years to hit 0.82% in 2006.
I once heard a high-yield expert — I believe it was Marty Fridson, who is also a contributing analyst to S&P Global Market Intelligence — comment that the reduction in high yield default rates in 2010 was statistically impossible. The market expected improvement — OAS had fallen to 597 bps by year-end 2009 — but not that much. Investors dived into high yield as they realized rates would be low for a long time and because the economy was improving. As a result, a lot of companies were able to refinance debt that might have defaulted.
The same phenomenon is occurring in 2020. Liquidity trumps fundamentals with Fed support.