Range of Stock Market Percentage Returns
In May, I posted a graph showing the historical range of DJIA (Dow Jones Industrial Average) returns for a variety of holding periods. That graph showed the results for a theoretical investor who bought and held the Dow for holding periods ranging from one to one hundred years, reinvesting dividends for the whole time. It seemed to show that the worst case got better with each passing year, and that historically the more years the investor held onto his investment the less likely he was to lose money.
Graphs like that one are often used to show investors that, while the stock market is very risky in the short term, the long-term investor faces much less risk. And that's true -- in a sense. It's especially true if the risk you are most concerned about is the risk of losing money (measured at the end of the holding period). Let's look again at exactly the same data, but from a different point of view.
Best and Worst Stock Market DOLLAR Returns for 1-10 Years
The graph above shows the results for the same theoretical investor, starting with an initial investment of $10,000. However, instead of showing the maximum, average and minimum annual return for each holding period, it shows the maximum, average and minimum value of the portfolios at the end of the holding periods. It confirms some things we already knew from the earlier graph (you may find it helpful to look at that earlier graph for comparison), namely:
- The market is volatile; you can lose substantial amounts of money -- especially in the short term (the investor lost money if the ending portfolio value was less than the $10,000 that he started with). This supports the adage that you shouldn't put money in the market that you're going to need in less than 3-5 years.
- For 10-year holding periods, the risk of loss was small.
The Worst Loss Was in Year Four, Not Year One
However, the careful reader will also note that the graph sends another message that is different from the original graph. In the original graph, the worst case seemed to be at one year; the worst case now appears to be at four years. While the worst one-year dollar loss only reduced the portfolio to about $5200, the worst four-year loss reduced the portfolio to less than $2700. (Regular readers will not be surprised to know that the four-year period in question was 1928-32. For more on this period, see The 1929-1932 Stock Market Crash Revisited.) In fact, in dollars, the worst case in years two and three is also worse than year one. The earlier graph seemed to show the worst case getting better each year; now the worst case is getting worse every year -- at least for the first four years. How can that be?
The answer is that the original graph, the one brokers and financial planners often show you, showed annualized returns -- that is, percentage returns per year. The ending portfolio dollar value is a function of both annual return and time. The worst one-year loss was 48%, the worst two-year loss 39%/year, and the worst four-year loss 28%/year; so, on an annualized percentage basis, the situation appeared to be improving. However, losing 39% two years in a row does more damage to your portfolio than losing 48% for one year. And, losing 28% four years in a row does even more damage.
The 10-100 Year Results are Even More Surprising -- and More Important
I was a little surprised at the results for 1-10 years; "astounded" is a better description of my reaction to The Range of Dollar Returns for 10-100 Years. Those results have critical implications on retirement planning.
Note: Dow dividends prior to 1929 have been estimated based on another index.
Related Posts:Range of Stock Market Returns for 1-100 Years
The Best & Worst 10-Year Returns in Stock Market History (rolling 10-year percentage returns)
For lists of other popular posts, and a complete index of all stock market posts by subject area, see the sidebar on the left.
Share This Article
Bookmark this on Delicious
To share via Facebook, Twitter, etc. see below.
This work is licensed under a Creative Commons Attribution 3.0 unported license.
Last modified: 3/9/2011