As the 2010s come to a close, it’s safe to say that the decade has been an enchanted one for U.S. stocks. The hard question is what lies in the decade ahead, and the data suggests that investors shouldn’t expect a repeat performance.

When the U.S. narrowly averted an economic meltdown at the end of the last decade, few anticipated that the following one would feature a historic rally for U.S. stocks. The S&P 500 Index has returned 13% a year from 2010 through November, including dividends, easily outpacing its long-term annual return of roughly 9%. It’s also the market’s fifth best decade since 1880. Only the 1920s, 1950s, 1980s and 1990s produced higher returns.

The surprising rally from the rubble of the 2008 financial crisis has touched off a fierce debate about what fueled the resurgence and, more important, whether it can continue. One popular theory is that the Federal Reserve goosed the market by unleashing a mountain of cheap money through low interest rates and quantitative easing. Another theory is that corporate earnings were boosted by a handful of fast-growing companies, such as Facebook Inc. and Amazon.com Inc., and by a wave of share buybacks, all of which lifted the market. The implication is that as long as the Fed keeps the money flowing, and highfliers and stock repurchases continue to supercharge earnings, the party can continue.

But none of those theories stand up to scrutiny. To see why, it’s helpful to break down the market’s return into its component parts: dividends, earnings growth and changes in valuation. Blogger Ben Carlson recently posted numbers compiled by late Vanguard Group founder Jack Bogle for each decade since 1900. I ran the numbers going back to 1880 using data compiled by Yale professor Robert Shiller, and the results closely resemble Bogle’s.

So what do the numbers reveal about the last decade? Of the S&P 500’s 13.3% annual return since 2010, 2.3% came from dividends, 10.2% from earnings growth and 0.8% from the change in the market’s valuation, as measured by the 12-month trailing price-to-earnings ratio. In other words, the vast majority of the gains can be attributed to a spike in earnings rather than investors’ willingness to pay more for stocks. In fact, the decade’s earnings growth was the highest on record.

That’s a problem for those who credit the Fed for the market’s stellar decade because cheap money doesn’t appear to have pushed investors to splurge for stocks. That jibes with the recent experience of much of the developed world, where a flood of monetary stimulus in Europe and Japan over the last decade failed to expand their stock market valuations. It’s also consistent with the longer-term record in the U.S., which shows no correlation between P/E ratios and the level of interest rates since 1871, as measured by 10-year Treasury yields (-0.12), counted monthly. (A correlation of 1 implies that two variables move perfectly in the same direction, whereas a correlation of negative 1 implies that two variables move perfectly in the opposite direction.)

Nor is there any indication that monetary policy is behind the record jump in earnings. Here again, earnings growth has been muted in Europe and Japan in recent years despite aggressive stimulus by central banks. And there’s been no correlation in the U.S. between the level of interest rates and subsequent earnings growth, whether measured over rolling one-year (-0.06), five-year (0.02) or 10-year periods (0.04) since 1871. 

While the focus on earnings is correct for understanding the last decade, the theories about what’s behind their surge are no more appealing. For one, the growth was generated by more than just a handful of companies. Of the roughly 370 companies in the S&P 500 for which earnings growth over the last decade can be calculated, 183 increased profits by more than the average of 10.2% a year for the index, according to Bloomberg data. And the median earnings growth was 9.6%, which closely matches the index’s average and shows that the gains were broadly distributed.

Buybacks don’t account for the growth, either. Yes, all else equal, buybacks reduce the number of outstanding shares and thereby lift earnings per share, the number commonly used to gauge earnings growth and P/E ratios. But total earnings have grown by 9.4% a year since 2010, just shy of earnings-per-share growth of 10.2%, so buybacks don’t appear to have contributed much. It’s not even clear that the recent rate of buybacks is unusually high.

Instead, the better explanation is also a simpler one: Earnings are extraordinarily volatile, even more unstable than stock prices by some measures. The volatility of yearly changes in earnings per share has been nearly three times greater than that of stock prices since 1871, as measured by annualized standard deviation — a whopping 52% for earnings compared with 19% for stock prices. And lest you’re tempted to conclude that much of that earnings volatility is a relic of the ancient past, consider that two of the three most severe earnings recessions on record were the previous two around the dot-com and housing busts. The only one that rivaled them was the 1920-21 recession (no, not even the Great Depression).

First « 1 2 » Next