We're all market-timers
Posted 02:18 p.m., April 28, 2008
One of the things the brokerage houses like to say is that “market timing” is a bad idea, that no one can consistently beat the overall stock market’s performance, so the only sensible thing for investors to do is “buy and hold” stocks, which provide better returns over the long run than bonds, commodities or other investment options.
The “you can’t beat the markets” concept derives from the “random walk” theory of market behavior, which holds that the stock market by nature follows a generally upward linear path and that the gains and losses in the indexes that cause so much excitement or worry are just random “fluctuations” around the long-term linear uptrend. In support of this theory, economists advance the notion that “the stock market always goes up over the long term,” based on calculations of average changes in the market over extended periods. But as the following chart shows, “always” isn’t quite accurate:
This chart shows the Dow Jones industrial average from its creation in May 1896 through Friday’s close. As the red and blue lines show, there are extended periods when an investor could “buy and hold” and end up either with no gain at all or even a loss.
For example, the first blue line shows that if you “bought the Dow” on Nov. 5, 1896, and held it for 36 years until July 8, 1932, you wouldn’t have earned a penny on your investment. Obviously, there were plenty of times in between when you could have sold your investment and made out quite well, but the point is that “buy and hold” isn’t a guaranteed way to make money.
The first red line shows an even worse scenario: If you had “bought the Dow” at the top of the 1920s bull market, on Sept. 3, 1929, you would have had to hold on to that investment for 25 years, until Nov. 23, 1954, just to break even.
Similarly, the other blue line shows that you could have bought and held the Dow for 16 years from December 1958 until December 1974 without showing a gain, and the other red line shows how, if you had bought the Dow in January 1966, you would have had to hold it for 16 years until October 1982 to break even.
Now, those scenarios are a bit unrealistic – you’d have to have been monumentally unlucky to have bought and sold on the exact dates listed above – but they do remind us that there’s an unavoidable element of timing in every investment decision, and sometimes the timing can work against us.
One thing that makes these scenarios unrealistic is the idea that someone would put all of their money into the stock market on a given date and then pull it all out again on some arbitrarily chosen date years later. In the real world, people put some money in one day, more another day, shift it around, take some out, put more in, etc. What investment advisers recommend is “dollar-cost averaging” – putting some money into stocks on a regular, ongoing basis – which they say helps guard us against the market’s “random fluctuations.”
To test how well that concept works in reality, I’ve used the Dow industrials data to construct a series of hypothetical investment portfolios. Each portfolio represents an investor who begins buying stocks in a particular month, starting with May 1896, and continues to invest the same dollar amount every month for 30 years. After the 360th monthly contribution, the investor then sells the whole investment the following month, presumably to put it into bank CDs or some other more or less risk-free, short-term instrument suitable for funding one’s retirement. There’s a new portfolio starting each month through March 1978, which would be the one cashing out now. (If that's you, congratulations on your retirement.)
The next chart shows the results of this calculation in terms of average annual percentage return as of the cash-out date:
As you can see, this dollar-cost averaging-based strategy can produce highly variable results: negative returns in 1932 and 1942, a paltry 0.79 percent in 1982, but a huge 24.73 percent in December 1999. Notably, anyone who held on past that 1999 peak until September 2002 would have earned an overall return only about half that received by the luckier person who sold at the peak. Once again, the timing element is unavoidable; people who begin their 30-year investment plan near market lows and cash out near market highs get the best returns, which ultimately serves only to confirm the long-established advice to “buy low and sell high.” Duh.
As of this month, using Friday’s close on the Dow, the average annual return over 30 years is 15.28 percent. That’s not too shabby, but it’s not a rate anyone who’s still putting money into stocks can expect to receive at their own cash-out in five, 10 or 20 years. If you look back at the first chart and see the kind of steep advance in the Dow that produced the 1999 peak in returns, that’s what the market will need to look like over the years ahead to produce the same kind of percentage gains in the future.
Comments
Post a comment