The Social Security Debate

I thought I would stay out of this, because I don't really have any particular expertise. But no one has yet posted (or at least I haven't seen it) some of the basic data which, in this case, are the inflation-adjusted returns to the S&P 500 Index. I used data for the first-day-of-the-year close (except for the end of 2004) and for the January CPI to compute a "real" S&P for each year from 1950 to 2004 (data in terms of "2004" Dollars). For the end of 2004, I used the December 21 S&P and an advance estimate of the CPI for December 2004.

What you get is a 4.04% (compunded) average annual real growth in the index, which sounds not too bad. The distribution of returns, however, doesn't look quite so good. The following table shows the (arithmetic) average returns for 1, 2, 5, to, 20, and 30 years, along with the standard deviation of returns, and the percentage of returns over those periods which were negative.


Term............Mean..........StDev..........%Neg
1-Year..........5.44%.........16.99%..........38.2%
2-Year........10.89%........24.95%..........31.5%
5-Year........29.22%........48.33%.........37.25%
10-Year......63.98%........78.41%.........23.91%
20-Year....110.58%......159.38%........34.29%
30-Year......88.42%.......94.89%...........3.70%

Remarkably, the 30-year return is (on average) worse than the 20-year return. This reflects the poor returns in the high-inflation period of the 1970s, the 1987 crash, and the late-1990s slump. (The first 30-year return is for the 1950 - 1980 period.) (A total return of 88.42% over 30 years is an average annual return of 2.11%, less than half the S&P's overall performance).

Even most daunting, however, is the fact that somewhere around 30% of the returns for each holding period (1-year, 2-year, 5-year, 10-year, and 20-year) , other than the 30-year holding period, are negative. The historical data suggest, in other words, that you have about a 1-in-three chance of earning a negative return to holding the S&P 500, even for long periods. Yes, the average looks good. But the downside risk is not something I'd like to bet a whole lot more of my retirement on.