ALTHOUGH the stockmarket has rallied strongly in recent weeks, it is striking that the corporate bond market has performed far less well. This is a worrying sign for those who believe in a lasting equity rally. The corporate debt market currently offers spreads that presuppose a high level of defaults; if that many companies are going bust, it is hard to see how equities can perform well.
Richard Cookson is a strategist at HSBC and a former writer of the Buttonwood column. He has looked (with some help from the Economist) at the long-term relationship between corporate bonds and shares. At the moment, for example, the dividend yield on the S&P 500 is around half the spread over government bonds of BAA corporate debt. In the early 1980s, the dividend yield was more than three times the spread.
But if you think that makes corporate debt look cheap, think back to the first half of the 20th century. During that period, it was common for the dividend yield on shares to be higher than the nominal corporate bond yield - never mind the spread. After all, companies had to pay the bondholders before they paid the shareholders. Equities were deemed to be a lot more risky and thus required a higher yield.
All that changed in the 1950s. Why? A look at the figures from the Barclays Capital Equity-Gilt study gives the best clue. If you rebase the real value of UK dividends in 1900 as 100, then by 1952, the index had fallen to 42. Investors were not enjoying any real dividend growth. (For the US, the history is shorter but still instructive; there was virtually no real dividend growth between 1926 and 1957.)
From that point on, and with the exception of a calamitous period in the mid-1970s, investors could count on consistent real dividend growth. In 2007, UK dividends were more than twice as high in real terms than they were in 1952; in nominal terms they rose more than fortyfold. In the US, dividend levels have trebled in real terms since 1957. Since equity investors could count on a steadily rising income, they were prepared to accept a lower initial dividend yield.
Now, of course, dividends are being slashed. Banks, among the biggest payers in the past, are giving up. Standard & Poor's estimates US payouts will be cut by 23% this year, the fastest rate since the late 1930s. So the historic dividend yield on the S&P of 3% translates into a prospective yield of 2.3%.
Should Mr Cookson be even roughly right, and equities have to be repriced in line with corporate bonds, that implies some further significant falls in share prices. Investment grade bonds trade around five percentage points above government bonds; US equities trade on a yield just a fraction of a percentage point above 10-year Treasuries.
Readers' comments
The Economist welcomes your views. Please stay on topic and be respectful of other readers. Review our comments policy.
Sort:
BlairStripOne,
I believe the Baa that is mentioned in the article refers to Moody's rating of corporate bonds. Baa is equivalent to BBB from S&P...
I agree with the main point of the two markets disjointed although there are some technicals here. The implied default rate of bonds is surely a little false because many of the holders such as hedgefunds became forced sellers (as with loans) when their funding from banks prime brokerage arms was cut off. Ditto cds implied levels of default probs when correlation desks had to unwind by buying cds. But something I don't agree with is:
"At the moment, for example, the dividend yield on the S&P 500 is around half the spread over government bonds of BAA corporate debt. In the early 1980s, the dividend yield was more than three times the spread."
Is is really valid to compare a USD index on US stocks to the non USD denominated debt of a European coprporate? And I might be wrong on this but I suspect there's a decent chance I'm right - in the 80s wasn't BAA a quasi-government entity with very low debt levels and now post the Ferrovial takeover it has lots of debt piled onto it to pay for the takeover?
"From that point on, and with the exception of a calamitous period in the mid-1970s, investors could count on *consistemnt* real dividend growth..."
C'mon - click on that spellcheck icon.
Peter Bernstein discussed the correlation between bond and dividend yields recently in terms of inflation and dividend growth. I recommend reading the whole thing, but I will give one excerpt here:
"Dividends are variable over time and bond coupons are fixed from issuance to maturity. Therefore, the correlation in returns should be just the opposite of the relationship prevailing since 1954 - negative during inflationary episodes and positive when the price level is relatively stable."
This is something that Napier points out in "Anatomy of the Bear", that bonds recover before equities.
The argument that dividend growth made up for paltry yields always seemed a bit over done. How many people held those shares from 1957 til 2007 to get that big real return?
I read that yields pre-1950 rivaled (or often beat) bonds because of the much higher price volatility of equities in those days. That was written by someone trying to justify 2% yields.
The last ten years kind of puts that argument where it belongs...