The 7 Percent Delusion

Why are Social Security privatizers so sure that stocks will outperform bonds forever?

Among advocates of Social Security reform, it’s an article of faith that stocks will continue to outperform bonds in the next 75 years, just as they have in the past 75 years. Back in 2001, Social Security Administration Chief Actuary Stephen T. Goss wrote that “the long-term ultimate average annual real [i.e. after inflation] yield assumed for equities is assumed to be 6.5 percent.”

There are several problems with this assumption—click here to read about them—but the greatest one is its certainty. As Paul Krugman has put it, privatizers appear to believe that “it’s a sort of economic law that stocks will always yield a much higher rate of return than bonds. They seem to treat that 7 percent rate of return as if it were a natural constant, like the speed of light.”

In fact, most investors know such guarantees don’t exist, as the market’s own behavior shows. The financial services industry is remarkably flexible and is constantly developing new products for investors. If you were so sure stocks would return 6.5 percent after inflation ad infinitum, why not guarantee customers 5 percent after inflation, lock up their money for 10 or 20 or 50 years, and pocket the difference? But such vehicles that guarantee a return on stocks are rare. Fidelity and Janus certainly won’t guarantee you returns on equities of 6.5 percent after inflation for 25 years, or five years, or any years.

The U.S. government guarantees lesser returns. The U.S. Treasury’s last 30-year bond, issued February 15, 2001, yields 4.73 percent before inflation—so more like sub-2 percent real return. If you’re willing to entrust your capital to mercurial Europeans, you can find a government that will guarantee you a return for a couple more decades. France is issuing a 50-year bond that will likely yield about 4.2 percent before inflation.

A handful of corporations in the 1990s issued bonds that guarantee returns for up to 100 years. Disney issued the first, the so-called “Sleeping Beauty” bond, in 1993. On Fidelity’s Web site, I found 15 bond issues that go out past 2047, mostly from blue-chip companies. The most risky among them, Daimler Chrysler, yields about 6.75 percent. Citigroup’s 100-year bond, which matures in 2098, yields 5.77 percent. Keep in mind that all these returns will be knocked down significantly by inflation. U.S. government Treasury Inflation-Protected Securities, which offer a protection against inflation, today offer just inflation plus about 2 percent for 20 years.

Stocks are perceived as being far more risky than bonds—not risky in the sense that you’re more likely to lose all your money if you buy stocks than if you buy bonds, but risky in the sense that it’s difficult to predict precisely how they’ll perform. And that’s why mutual funds won’t guarantee you a 6 percent return on stocks.

Professional investors don’t believe the market will reliably return a certain amount year after year and haven’t clamored for instruments that would allow them to place such bets.

Investors who believe the S&P 500 will rise or fall by a given amount in a given time period can place their money where their mouth is by buying and selling options. The Chicago Board Options Exchange offers options for the distant future, known as LEAPs, but they only extend as far out as December 2007 for the S&P 500. Investors who have a need or desire to speculate further into the future can ask brokers to create so-called synthetic options. Using the same methodology used to value the listed options that trade on the exchanges, brokers can create and price options on any potential outcome over any time frame—say, the S&P 500 closing at 3,000 in February 2015. And the price of the option created can be translated into a probability of the S&P actually closing at or above 3,000 10 years from now.

I asked the options desk at a major Wall Street firm to construct an option that delivered a 6 percent annual return on the S&P 500, with dividends reinvested, for 10, 15, and 20 years (ignoring inflation for the moment). The prices of the options they constructed indicated that investors today believe there is only an 18.8 percent likelihood that stocks will return 6 percent or more annually for 10 years, a 13.3 percent likelihood that they’ll do so for 15 years, and a 9.6 percent likelihood that they’ll do so for 20 years.

Now, such long-term options on broad stock indexes do have a tendency to be pessimistic. And over the next 20 or 50 years, stocks may well return more than 6.5 percent after inflation. We should all hope that they do, because we have so much—our retirement and our kids’ college savings—riding on it. Everyone who buys stocks is, by nature, an optimist. Any time you buy a stock you’re essentially saying you will earn more than the risk-free alternatives. But the assumption of the privatizers is that they’ll be right in the future, year in year out, no questions asked, in an era in which the economy is growing far more slowly than it has in the recent past. They also assume that if you’re setting out to construct income insurance, which is what Social Security is, relying largely on stocks is the best and not particularly risky way to do it. But if professional investors aren’t willing to use stocks to guarantee certain returns, why should we expect amateurs to do so?