January 31, 2009

Past Predictions

At the inception of Diamond Hill in May 2000, we expected the total return of the S&P 500 to moderate to no more than 5% per annum over the coming decade, considerably lower than what had been experienced in the prior decade. The primary reason for this forecast was a starting valuation level that was simply too high. At the end of 2003, we provided an update. Despite losing more than 5% per annum through the end of 2003, we were not inclined to raise the forecast materially, bumping our forecast to 6% per annum for the decade ending 2013. Through August of 2008, this forecast was faring well, as the compound return had been a bit above 5%. However, after the past four months, the trailing 5- year annualized return has dropped to a -2.2%. The S&P 500, therefore, would need to return approximately 14.9% per annum over the next five years in order to make the end of 2003 prediction of 6% per annum over the next decade come true. Thus far, it is shaping up to be a lost decade in terms of equity returns and even relatively downbeat forecasts have erred on the optimistic side.

A convenient excuse might be that the unfolding credit crisis, which originated in the subprime mortgage sector and can now be fairly characterized as broad-based, could not have been foreseen so long ago. While this is true, it ignores one reality. Easy credit, both in terms of low interest rates (spurred by central bank rate cuts in the wake of the tech and telecom crash and the coincident mild economic recession) and lax underwriting, contributed to the strong economy and stock market in the previous years. Without this easy credit, housing prices would not have risen as much as occurred. The portion of consumer spending that was driven by home equity extraction, made possible by the rise in home prices, would have been muted. Furthermore, the expansion of other types of consumer credit, such as revolving credit cards, has also allowed consumer spending to outpace incomes for some time. Finally, narrow spreads in various credit markets also led to booming environments in commercial real estate and mergers and acquisitions including highly leveraged going private transaction. These were positive influences on stock prices. It would seem intellectually dishonest to accept the favorable early benefits of these factors on the original forecast, while blaming the eventual consequences of some of the irrational behavior as being unforeseeable.

A second possible defense is far simpler. Perhaps, despite the economy facing real challenges, stock prices have overreacted and are now at levels implying high future returns. Many managers with stellar long-term records have publicly called now the best ‘buying opportunity’ of their careers. Warren Buffett, who has seldom made general “market calls,” wrote an op-ed in The New York Times on October 17, 2008, summarizing his belief that while the economic news will be grim for a time, it was now time to buy stocks. The S&P 500 closed at about 940 that day. We will explore this more in a bit.

Why Bother?

Since we are active managers at Diamond Hill, we have never owned the S&P 500 Index through an index fund or ETF and more than likely will not in the future. In constructing our Fund portfolios, we have always been willing to veer far from the benchmark. So, our outlook only tangentially affects what we do. There are a couple reasons for our interest, however. First, although we are seldom involved in client decisions regarding asset allocation among various asset classes (equities, fixed income, and real estate with each often further subdivided into different categories), these forecasts give a sense of our opinion for the general U.S. equity market. Second, it provides information in setting the discount rate to use for an individual company under analysis.

The Difficulty of Normalizing Earnings in Today’s Environment

Corporate earnings and interest rates are the two fundamental underpinnings for the long-term value of stocks. A third component, which we’ll term investor psychology, is also very important to the short-term performance of stocks. However, this is ordinarily a less predictable component of returns that in the long-term tends to recede in importance as investor’s intermittent bouts of fear and greed tend to cancel one another out. Recently, however, earnings have demonstrated great instability as well. After corporate profits (excluding write-offs, a topic for another discussion) retreated in 2001 approximately 20%, S&P 500 earnings grew steadily and briskly once again over the ensuing 5 years. At the onset of 2008, many strategists forecast 2008 S&P earnings to eclipse $90 per share, up from the mid to high $80’s in 2006 and 2007. When companies finish reporting fourth quarter earnings this winter, S&P 500 earnings are likely to be in the mid $60 per share range, which again excludes a heavy amount of write-offs, especially in the financial sector. The year 2009 is likely to be another down year for earnings, perhaps receding to the mid to high $50’s per share, a level last seen in the year 2000. This has been a huge miss, obviously. For energy and basic materials companies, commodity prices have undergone significant downward swings. In the financial sector, banks have had to increase provisions for loan losses while insurers have marked down investment portfolios. And the upheaval in the credit markets spilled into the real economy at an accelerated rate starting in October, leading to a rapid falloff in orders and uncertain capital spending budgets that is wearing on most industrials. Clearly, past income statement data may be less reliable for predicting the future than normal. As long-term investors, we attempt to  estimate the economic earnings a company will generate over an entire economic earnings cycle. Even dividends, which have historically risen at a slow but steady pace, in part because managements tend to be conservative in setting dividend policy (once instituting a dividend at a certain level, they have been reluctant to reduce it later), are likely to fall in 2009 as banks preserve capital needed to shore up capital ratios. Today’s economic environment presents challenges in estimating “average” earnings.

The Future Outlook

Let’s return to the premise that the steep decline in the S&P 500 might leave it unusually cheap. The S&P 500 closed at approximately 826 on January 31. At this level, using trailing dividends and estimated earnings, the S&P 500 is trading at approximately a 3.5% dividend yield (at a time when the 10-year Treasury bond is approximately 2.6%) and less than 13X earnings. Provided that the economic slump does not drag on for a multi-year period and that corporate profits maintain their relative share of GDP, the combination of the resulting earnings growth, the current dividend yield, and a small bit of multiple expansion suggest to us that the expected return for the S&P 500 is now 10.5% – 12.5%. The bad news is that while Treasuries currently offer miniscule returns, many areas of the bond market offer expected returns that approach these same returns. The good news is that we do expect equities will once again allow investors to compound capital commensurate with the risk involved.

Screen Shot 2013-09-09 at 10.33.38 PM

Originally published January 31, 2009

back to top