FREE NEWSLETTER
AFTER THE WILD RIDE of the past two weeks, stock investors are in search of reassurance. Will this movie have a happy ending?
If we’re venturing into the stock market, we should ideally have at least a 15-year time horizon. That gives us 10 years to make money and another five years to look for the exit. Those final five years may prove crucial if the first 10 years don’t turn out so well.
What return can we reasonably expect over the initial 10 years? We might start by estimating the economy’s growth rate, consider what that would mean for corporate earnings, and then ponder what value investors will put on those earnings. Problem is, with this approach, we begin by making reasonable estimates—and end up engaging in wild speculation.
The first step doesn’t seem so tough: estimating the U.S. economy’s 10-year growth rate. If we look back over the past half century, the real (after-inflation) GDP growth rate has averaged 2.8% a year. The worst year was a 2.8% contraction in 2009 and the best was a 7.3% spurt in 1984. Still, 38 of the 50 years were above 0% but below 5%. That’s moderately reassuring: It tells us that, most of the time, GDP growth hasn’t been that far from the 2.8% long-run average. Moreover, nine of the 12 outliers occurred in the first 20 years—and just three in the 30 years since.
The not-so-good news: GDP growth has been slowing over the past 50 years. In the 17 years since 2000, it’s averaged just 1.8% a year. Partly, that reflects the Great Recession. But it also reflects slower growth in the labor force—a trend that will continue as the U.S. population ages. Taking that into account, we might assume the economy expands 2% a year faster than inflation over the next 10 years. If inflation is also 2%, that would put nominal GDP growth at 4%.
Will corporate earnings also grow at 4%? That question triggers three others. First, will earnings per share grow slower than overall corporate earnings, as companies sell shares to finance growth and issue stock options to employees? Historically, earnings per share have lagged behind overall corporate earnings by two percentage points a year. But over the past 10 years, shareholders haven’t suffered any dilution, as companies use spare cash to buy back stock.
Second, will profit margins contract from today’s historically high levels? We can only guess at the answer. Third, will earnings of U.S. multinationals get a boost either because they snag a larger share of foreign markets or because foreign economies grow faster than the U.S.? Both are possibilities, but—once again—we can only guess at the answer.
The upshot: We might assume that earnings per share do indeed grow at 4%, but accept that it could easily be somewhat faster or slower over the next decade. Tack on today’s roughly 2% dividend yield, and we would be looking at an estimated 6% annual total return, or four percentage points a year faster than inflation.
But this assumes that stock prices climb along with earnings per share. Will they? Share prices today are richly valued relative to corporate earnings, with stocks at 23.9 times reported earnings, versus a 50-year average of 19.2. That is worrisome. But stocks have been richly valued for much of the past quarter century, so maybe investors shouldn’t be concerned.
Vanguard Group founder Jack Bogle likes to distinguish between the market’s investment return—corporate earnings growth plus dividends—and its speculative return, which is the change in the value put on earnings, as reflected in the market’s price-earnings ratio. The dilemma: To come up with a prediction for 10-year returns, we need to speculate on how speculative other investors will be.
As we were reminded yet again over the past two weeks, investors flit between exuberance and despair with remarkable speed—and that means short-term returns are anybody’s guess. But don’t despair: As we lengthen our time horizon, changes in P/E ratios become less important and instead the market’s return is increasingly driven by the combination of earnings growth and dividend yield. In other words, if we have a truly long time horizon, we have a reasonable shot at notching something close to 6% a year, and that 6% should be comfortably ahead of what we could have earned with bonds.
Follow Jonathan on Twitter @ClementsMoney and on Facebook. Check out his two earlier articles about 2018’s market decline: The Morning After and Taking Stock.
Want to receive our weekly newsletter? Sign up now. How about our daily alert about the site's latest posts? Join the list.
to speculate a bit further, I have seen this piece
http://knowledge.wharton.upenn.edu/article/siegel-shiller-stock-market/
where Prof Siegel’s estimate of EPS growth is 3.5% real ‘with 2.5% due to stock buybacks and 1% organic growth. Nominal return is 7.5% including 2% inflation.’ So he seems to think there’s going to be a larger effect of buybacks on EPS growth than you appear to imply.
I was wondering what you thought about the effect of buybacks? (as far as I understand, you say buybacks have roughly balanced stock dilution in the past 10 years: would you say they are likely to continue to do so, or boost EPS growth?).
Anyway, to be fair, Prof. Shiller has a much lower 10 yr estimate for market returns, so that yours is roughly half way between the 2 😉
Jeremy, whom I’ve always enjoyed talking to and reading, is perennially bullish. Back in the 1990s, he seemed to suggest that stock investors could count on earning 7 percentage points a year more than inflation — the historical average — so it’s notable he’s now down to 5.5. It’s also notable that he thinks companies will, after inflation, only grow 1% a year organically. That’s surprisingly low — and suggests he’s either quite pessimistic about the U.S. economy or thinks that, before buybacks, earnings per share will grow slowly because companies will be dishing out so many stock options to management. Maybe corporations can add 2.5 percentage points with buybacks. But that seems like an awful lot — and you have to wonder where they’ll find the cash, if their organic growth is only 1% after inflation.