Magnitude of 12-Month Real Losses

May 7’s post explored how frequently your investments might be expected to experience real (i.e., inflation-adjusted) losses after different lengths of time. The graph in that post showed, historically, how frequently investments in stocks, bonds, and a 50%-50% mixture of stocks and bonds have exhibited real losses over periods ranging from 12 months to 52 years.

Pretty good information. It provides you with a gauge of how frequently you might get scared. But there’s another dimension to consider. Just how scared will you get? Are we talking “Ghostbusters” or “Alien”?

The charts below attempt to give an idea of the magnitude of the real losses (and gains) that the markets have historically provided. Rather than the frequency of bad results, it measures their size.

The bottom chart summarizes the results over short (12-month) periods. As might be expected, stocks have been very scary over the short run, exhibiting the worst potential losses over 12-month periods—as bad as a 64% loss in value. In contrast, bonds’ worst 12-month real loss has been only 16%. A 50%-50% mixture of stocks and bonds resulted in a 27% worst-historical-case loss. The range between worst cases and best cases are shown in 10-percentile increments. (As with May 7’s post, I looked at all 12-month periods between January 1, 1926 and December 31, 2008. There have been 977 of them. For stocks, I used total return on the S&P 500; for bonds, intermediate term Treasury bonds; and for inflation, changes in the Consumer Price Index. In other words, the usual suspects.)

If you lengthen the period between measurements, ignoring interim ups and downs, things start to change. The middle chart shows the same data for 60-month (5-year) periods. Stocks still exhibit the potentially worst result, with a worst-case real loss during 1926-2008 of up to 51%, with bonds’ worst real loss over a 60-month period of 28%. With time, stocks' worst performance gets better, and bonds' worst performance gets worse. And the 50%-50% mixture of stocks and bonds now exhibits the best worst case, with only a 21% loss.

And if you lengthen it further to 120 months (10 years), the pattern continues to hold: A 50%-50% mixture of stocks and bonds has provided the least bad worst case, as shown in the top chart below.

By the way, those of us who suffered through 2008 will not be surprised to learn that between 1926 and 2008, the worst 120-month period for stocks, in terms of inflation-adjusted losses, was the 120 months from December 1, 1998 to November 30, 2008. Just thought you might want to know.