By all accounts, interest rates are set to rise in the near future. Inevitably, that will have an influence on stock valuations, and investors are correct to consider what the consequences will be, as well as what they can expect from equities. Short-term predictions abound on what will happen in the market, but the utility of short-term predictions is ... dubious. A longer-term horizon offers a much better chance for predictions that might not be completely accurate but are, at least, better informed. Warren Buffett ( Trades , Portfolio ) publicly provided a model for making such a prediction, under circumstances that might be considered similar to today's, and it turned out to be highly useful in framing expectations. It might be again today.
In two Fortune magazine articles, one published in late 1999 and one in late 2001, Buffett asked investors to keep in mind the returns of the Dow Jones Industrial Average over the two preceding 17-year periods:
For comparison's sake, with the next 17-year sequence approximately 97% of the way complete, we see these results:
Buffett pointed out that the divergent results over the 1965-1981 and 1982-1998 periods weren't at all explained by better gains for the economy as a whole over those periods. In fact, it was just the opposite:
What really fueled the growth in stock market returns were two things - a decline in interest rates, and growth in companies' profit margins.
Regarding the history of profit margins, going into the late 1990s, the previous 50 years had corporate profit margins as a percentage of gross domestic profit normally falling between 4% and 6.5%, and in 1998 to 1999, margins were at the high end of the normal range.
All of this good fortune leading up to the end of the 1990s made for some very enthusiastic investors. Fueled by the returns in the rearview mirror, investors with less than five years of experience expected 10-year returns of 22.6%. Investors with 20 years of experience expected returns of 12.9%. Buffett argued that both of these expectations were completely unrealistic.
Looking at these two factors - interest-rate movements and corporate profitability - Buffett stated that one or the other, or both, would have to continue to improve for investors to get "juicy" returns in the upcoming years. Yet both factors did move in beneficial ways, and the returns were still not "juicy." Here's why.
As to interest rates, he offered that it was "always possible" that rates would keep going down. And this possibility materialized.
As to corporate profitability, he expected returns to the historical norm. Yet corporate profitability improved dramatically since his Fortune articles.
On stock market valuations, the multiples that investors would pay 17 years hence on earnings, he did not offer a prediction, but he cautioned that expecting investors to keep paying more and more for earnings wasn't a good bet. At the time, the market was trading at approximately 30 times earnings.
Adding up these variables and how they might move over the 17 years to come, in 1999, Buffett posited essentially normal GNP growth (at the time 3% real, 5% with inflation), plus 2% in dividends, and gave 7% as the likely returns of the equities, not including the costs of investing. Two years later, at the end of 2001, with stock prices 20% lower than in late 1999, he predicted 8% nominal returns.
Inherent in these predictions was that if interest rates, corporate profit margins, dividend payout rates, and valuation held steady, economic growth would control market rewards.
It was, at heart, an extremely simplified prediction meant to show that investor expectations - fueled by the recent past performance of the stock market, and not the underlying fundamental drivers of the market - could not be met. While Buffett pointed out that further improvements in profit margins and interest rates could certainly lead the way to increased returns, he saw improved profit margins as unlikely and a further interest-rate decline as unpredictable.
Fast-rorward (nearly) 17 years, and let's review
Is it remarkable how close Buffett came:
Buffett was mildly generous on this long-term estimation, at least if we cut off the record at the end of last year.
Of course, history didn't play out in precisely the manner Buffett suggested it would. Over the intervening 17 years, the long-term risk-free rate did decline, from roughly 6% to around 3%. This alone (in Buffett's words in 1999) would have the effect of doubling the value of stocks - though valuations went decidedly down during those 17 years.
Profit margins have reached levels previously unattained - moving from the 6.5% that Buffett indicated historically was the top of the normal range, to levels of about 9% over most of the past decade.
Counteracting these extremely powerful forces, valuations came down from a level of 30 to 33 times reported earnings at the time of Buffett's first piece to a more "normal" level of about 21 - this despite the downward move in interest rates.
Profit margins may or may not stay at today's historically high levels. They do not have to revert to pre-2000s levels, but doing so would make profit growth from today's levels that much harder to achieve.
Interest rates seem destined to go up, though the yield curve remains fairly flat. Expectations - today - are not for dramatically higher long-term rates, and there are sufficient demographical reasons to think that long-term rates can remain low for the foreseeable future. Still, that's a bet embedded into future valuation multiples as being within today's range if you want to get solid market returns from here.
If you take Buffett's original premise that future returns are likely to be guided most strongly by the rate of growth in the economy plus likely dividends, an optimistic starting point would be a return to 3% annual real GDP growth, plus 2% dividends. Both inputs could be higher or lower, of course, and continued net buybacks of stocks could fuel per-share growth at a marginally faster rate than GDP growth. But having GDP elevate back to 3% annual real returns as a starting point is not overly cautious by any means, and buybacks at today's prices won't deliver the same kick that buybacks in 2009 to 2011 did.
If you're agnostic on the direction of margins and interest rates, you're likely to come to the conclusion that stock-market returns over the next 17 years - or 10 or 20 or whatever - will tend toward a range of 6% to 8%, assuming 2% inflation.
You can raise or lower estimates from there depending on your biases (or carefully constructed models, if you have those), but consider some of the parting words of Buffett's 2001 Fortune article. They've come acceptably close in their long-term predictions:
"If the percentage relationship [between market valuation and GDP] falls to the 70% or 80% area, buying stocks is likely to work very well for you. If the ratio approaches 200% - as it did in 1999 and a part of 2000 - you are playing with fire. As you can see, the ratio was recently 133%."
Today, that relationship stands at around 125%. That's obviously higher than the range Buffett designated as attractive, yet it's lower than it was at the moment in 2001 that he seemed to find a tolerable time to invest - and predicted 8% future returns. People should put words in somebody else's mouth to arrive at a conclusion they've already determined they want to reach, but Buffett's analysis, given at a different time in different circumstances, offers some guidance on a way to analyze longer-term market rewards that don't center on whether the market is too expensive or too cheap or too exposed to interest-rate fluctuations. At all times, a host of variables will move the market for short periods, but a certain few variables will dominate the most likely drivers of stock market returns over the long term.
For more insights from Motley Fool Asset Management, please visit www.foolfunds.com
 Does that not "seem" normal to you? As measured by end-of-quarter price to earnings, the market has averaged EPS of 24 over the past 26 years. If we exclude the four quarters immediately following the initiation of the Great Recession, when one quarter of very large write-downs skewed the P/E up, the average P/E of the rest of the time period dating back to 1989 is 21.5.
About GuruFocus: GuruFocus.com tracks the stocks picks and portfolio holdings of the world's best investors. This value investing site offers stock screeners and valuation tools. And publishes daily articles tracking the latest moves of the world's best investors. GuruFocus also provides promising stock ideas in 3 monthly newsletters sent to Premium Members .
This article first appeared on GuruFocus .