For every thing there is a season, and a time for every purpose under the heaven.
—Ecclesiastes
Over the last decade there has been a burgeoning interest in the effects that demographic patterns have on the economy and stock market, with many popular books and academic papers finding important connections between the two. In their book The Fourth Turning: What the Cycles of History Tell Us About America’s Next Rendezvous with Destiny, William Strauss and Neil Howe write that the human life cycle is a predictable chronology of 1) Childhood; 2) Young adulthood; 3) Midlife; and 4) Elderhood. Each of these phases coincides with predictable patterns of earning, spending, saving, and investing. According to Strauss, Howe, and other acolytes of demographic destiny, understanding where our economy is headed is a simple matter of analyzing demographics, since where each group is in their life cycle will determine what they will be doing and how they will be investing and spending in the future.
According to Harry Dent Jr., in his book The Next Great Bubble Boom: How to Profit from the Greatest Boom in History; 2005–2009, “The point is that there is a new information-based science built on predictable cause-and-effect impacts of how we change as we age that is just as predictable on average as life insurance actuarial tables for when we will die…. Demographics as a new science is the greatest breakthrough we have seen in economics.”
Dent and others cite simple facts to support their arguments. For example, consumer spending accounts for approximately 60 to 70 percent of gross domestic product, with the biggest spenders between the ages forty-five and fifty-four. You can therefore make very accurate predictions regarding the economy and the stock market by looking at how many people there are at any given time between ages forty-five and fifty-four, and how many there will be at various points in the future.
Demographic Demons
After establishing that the majority of gross domestic product is the result of consumer spending in his book The Great Bust Ahead, Daniel Arnold succinctly concludes:
Now that we know that we are responsible for the GDP, it is only common sense that “who we are” must have a powerful bearing on the GDP. For example, if we were all fifteen years old with virtually nothing to spend, the GDP would be pathetic. If on the other hand we were all forty years old with good incomes (and spending it all as we do) the economy would be going gangbusters. So, at any given time, the more of us with more money to spend there are, the better the economy (GDP) is going to be. [Emphasis in original]
Both Arnold and Dent then plot out how many big spenders there are in the country at any given time. To their eyes this presents a perfect picture of how well the economy—and by extension the stock market—will be performing in the years ahead.
Using these simple facts, the most popular books on demographics come to similar conclusions: the 78 million baby boomers—the largest generation in U.S. history—will drive the stock market to new heights through 2009 or 2010 but then bring it crashing down as they enter retirement and sell their stocks. According to the demographic data, the trend does not turn positive again until the year 2021 when the echo-boom generation starts hitting their peak spending years.
It Works—Until It Doesn’t
The problem with the approach advocated by Dent and Arnold is that the inflation-adjusted Dow Jones Industrial Average does broadly mirror the number of forty-five-to-fifty-four-year-olds in the population until about 1987, when the correlation between demographics and the Dow becomes much fuzzier. Both Dent and Arnold start “adjusting” the data to make the two correlate more highly again—first it’s technology innovation, then it’s the introduction of the birth control pill which allowed baby boomers to delay parenthood and therefore their peak spending years, then it’s that in the 1990s the preferred index was not the Dow but the NASDAQ, and on and on.
Whatever the reason, demographic patterns began diverging from the DJIA in 1987—and they haven’t gotten back in sync since. The impulse to tweak and adjust the data is understandable, but doesn’t always make sense. For example, both authors point out that the NASDAQ got the lion’s share of investors’ money at the height of the tech boom in the late 1990s, arguing that is the reason the Dow didn’t do as well. Yet they conveniently ignore that the same was true for the AMEX stock exchange in the late 1960s: during that bubble the AMEX went up at six times the rate of the Dow.
Remember hindsight bias from the last chapter? If you continually adjust the graphs as “new” data becomes available, the charts will show perfect correlation after the fact but will be of little use to users now. For example, say foreign investors start aggressively buying U.S. securities in the next several years. While that is something that is very difficult to predict now, it might seem obvious after the fact, tempting the Dent school of demographers to once again use hindsight to adjust their charts to reflect these new buyers. Unfortunately, the new trend will have provided zero predictive information in real time. Without these ongoing modifications, the data doesn’t seem to hold up. According to the chart on page 21 in Dent’s The Next Great Bubble Boom, if the Dow truly mirrored the immigration-adjusted births as of 2005, it would currently be trading at nearly 20,000, not the actual 10,500 level it was at in the summer of 2005.
Talking ’Bout My Generation
Part of the problem may be the baby boomers themselves. Who’s to say that they will actually behave like the generations that preceded them when deciding how to invest or behave during retirement? In each of their previous cycles, boomers have acted very differently from those who came before them, from their hippie, antiestablishment youths to their embrace of individualism and entrepreneurship in middle age. Thus far, they have defied prediction. According to a new report called New Retirement Survey conducted by Ken Dychtwald for Merrill Lynch, boomers will likely continue their rebellious ways right through retirement. Dychtwald finds that the long-held assumptions about retirement—that it begins at sixty-two, that it’s spent pursuing leisure activities, and that people accumulate most of their savings for it before age fifty—are all wrong in regard to the baby boomers. It turns out that baby boomers born in 1946 (who are now approaching the age of sixty) did the bulk of their retirement saving and investing after turning fifty, and the majority of them expect to continue working well past the normal retirement age of sixty-five. The baby boomers are turning the accepted notion of what a “typical” retiree looks like on its head. As a result, the era of age-based assumptions may be ending.
Baby boomers may also buck their forebears by continuing to embrace stocks rather than bonds as their preferred investments during retirement. Unlike their parents’ generation, they have a deeply embedded belief that stocks are the best long-term investments, and as they consider their expanding longevity, they may fear the relative penury that a bonds-only portfolio might offer. They are also the Peter Pan generation that continues, well into their fifties and sixties, to redefine what it means to be “old”—in their case, five to ten years from whatever age they are now!
It has also been convenient for demographers to look at baby boomers as one monolithic cohort. There are actually large differences among them, depending upon when they were born. Demographer Jonathan Pontell has gone so far as to separate the baby boom into two groups: the traditional boomers whom he assigns the birth years 1946–1956 and the group he calls “Generation Jones,” or those born between 1956 and 1964. According to Pontell, the two subgroups have very little in common, possess very different attitudes about life and work, and have very different ideas about how to achieve what they want in life. It doesn’t take much imagination to see that these differences might lead to very different investment and retirement strategies.
Academic Debate
The academic debate about the impact of current demographic trends has been fierce, with two camps emerging: the first claims that demographics are destiny; the second that the aging of the baby boomers will have very little effect on the stock market. The stakes in this debate are enormous, since the baby boomers are the largest generation ever in American history and therefore will have a potentially huge impact on the stock market in the years to come.
On the side of “demographics as destiny” are many luminaries such as Jeremy Siegel, author of Stocks for the Long Run, Robert Shiller, author of Irrational Exuberance, and a trio of finance professors who have published a paper called “Demography and the Long-Run Predictability of the Stock Market.” The authors—John Geanakoplos of Yale University, Michael Magill of the University of Southern California, and Martine Quinzii of the University of California at Davis—rely less on the absolute number of big spenders than Dent and his ilk, but nevertheless claim that demographics are the most important factor in determining long-term trends for stock prices. The authors note that “since the turn of the century, the live births in the US have also gone through alternating twenty-year phases of baby booms and baby busts: for example, the low birth rate during the Great Depression was followed by the Baby Boom of the fifties and the subsequent Baby Bust of the seventies. These birth waves have resulted in systematic temporal changes in the age composition of the population in the postwar period, roughly corresponding to the twenty-year phases of the stock market.” [Emphasis added.] In other words, we see the same boom and bust phenomena in the birthrate as we do in the stock market. Perhaps one causes the other.
Unlike Dent, who plots the absolute number of boomers with an expected correlation with absolute returns for equity markets, the academics have identified a ratio that seems fairly predictive of the market’s future: the ratio of middle-aged to young investors. While the professors don’t claim their model will forecast the exact levels of the Dow Jones Industrial Average or the S&P 500, they argue that when the ratio of middle-aged to young investors increases, the market’s PE ratio increases, and when the ratio decreases the market’s PE ratio decreases. While their ratio only generally predicts what price investors will be willing to pay for every dollar of earnings, it did accurately predict the great boom between 1982 and 2000, when both the ratio and PE’s were rising. The ratio is now predicting falling PE ratios as the ratio of middle-aged to young investors declines. Their model also works in markets outside of the United States, predicting the massive decline in Japanese stock prices as the ratio in Japan peaked in the late 1980s and went into decline all through the 1990s and early 2000s. The professors say that the ratio will be declining in the United States until 2018, when it will again turn strongly positive, as the huge echo-boom generation enters the investment fray.
Consistent with Reversion to the Mean
These findings are similar to the Leuthold Group’s in chapter 2, the difference here being that the professors find demographics to be the driving force in the expansion and contraction of the PE ratio. John Y. Campbell and Robert J. Shiller offer similar findings in their March 2001 paper “Valuation Ratios and the Long-Run Stock Market Outlook: An Update.” In it they find that the relative level of PE ratios and dividend yields is useful in predicting future returns, with high PE ratios and low dividend yields leading to below average future performance and low PE ratios and high dividend yields leading to above average returns. This hypothesis is consistent with both reversion to the mean and the demographic argument offered by Geanakoplos, Magill, and Quinzii in their paper.
Demographics Don’t Matter
On the other side of the debate are a group of academics who believe that demographic data is useless when forecasting stock market returns. Robin Brooks, an economist at the International Monetary Fund, claims that it’s not the behavior of all the baby boomers that matters; it’s what the richest among them decide to do. In an extension of the 80/20 rule, where 20 percent of the people are responsible for 80 percent of the results, Brooks argues that only the actions of the largest equity holders will really matter. He believes that even if a majority of baby boomers are forced to sell their stocks and mutual funds during retirement, they will still account for only a modest percentage of the markets’ dollar volume, since few have saved enough to be entirely self-sufficient during retirement. Brooks points out that the majority of baby boomers will be reliant on the government for much of their retirement income.
Data from the Federal Reserve back up Dr. Brooks’s claim. In 2001, the richest 1 percent of the U.S. population owned 53 percent of the stock owned by individuals and the richest 10 percent owned 88 percent. Brooks claims that corporations will likely raise their dividends in order to keep these rich owners happy, leading to an era in which dividends are, once again, an important force behind stock returns.
Demographics Do Matter
In the end, it is difficult to argue that 78 million Americans moving into retirement will not have an impact on both the economy and the stock market. Perhaps one of the reasons the Dow is currently at 10,500 rather than the 20,000 that Dent’s demographics suggest is that the market has already begun factoring the demographic shift into prices. This is, after all, one of the most hotly debated stock market issues of our day, and the market usually factors such data into current share prices. But Dr. Brooks’s forecast could be equally true—if shareholders start clamoring for stocks with higher dividend yields, companies eager to keep their stock prices up will do everything in their power to deliver these yields.
The most important point is that the 78 million boomers will affect the market and which types of stocks are rewarded. Just as surely, over the next twenty years the cost of Medicare and Medicaid will soar, as will shortfalls in the Social Security trust fund after 2018. This will probably lead to higher taxes, reduced payouts, and other Draconian measures. Huge governmental deficits and increasing taxes are rarely stock-friendly environments, so even if boomers hold on to their stocks, the very size of their demographic appears certain to affect the stock and bond markets in numerous ways.
Demographics Affect Our Destiny
Demographics were clearly a strong tailwind to the bull market between 1982 and 2000; they now appear poised to act more like a headwind. Short of dramatic changes in U.S. entitlement programs, the government will face unprecedented challenges as the 78 million boomers move into retirement and old age. This will inevitably lead to tax increases and benefit reductions—hardly an environment conducive to double-digit stock market returns. And even if boomers choose to stick with stocks, they will most likely choose those that pay hefty dividends and offer greater price stability. And while it is unlikely that the doomsayers’ prophecies will come true with the age wave crashing into a great new depression, the age wave is nevertheless unavoidable, and very consistent with what mean reversion and contraction of PE ratios forecast for markets over the next twenty years.
Chapter Seven Highlights