Evaluating past performance? Don't overlook these caveats
April 09, 2014
It is a common investment-industry disclaimer: "Past performance does not guarantee future results." It simply means that whatever numbers a financial adviser, mutual fund or money manager is showing you, take them with a grain of salt.
This is a valuable warning, but of course investment firms hope you will ignore the warning and focus on the numbers. There are several reasons you should not.
The trouble with history
Whether you are looking at something as conservative as savings account interest rates or as exotic as emerging market returns, never expect the past to repeat itself reliably. Here are five reasons why:
- Limited observation periods. Investment literature likes to deal in neat one, three and five-year time periods. From a market history standpoint, these are very short periods of time, and usually dominated by one set of conditions. If an investment approach does not show you its history over both bull and bear markets for its particular asset class, then you are not seeing the whole story.
- Uneven cycles. Even though a longer-term history such as a 10-year period may span more than a full market cycle, those nice, round, calendar-based time periods rarely coincide with a representative range of market conditions. For example, a 10-year period might comprise two bull market periods and only one bear market, or vice versa. Focus your evaluation on complete market cycles.
- Different interest rate environments. The past 30 years have seen 10-year bond yields drop from around 12 percent to less than 3 percent. Meanwhile, savings account rates and CD rates have dropped to nearly nothing. Not only does that mean history is of little use in evaluating bonds and bank deposit products, but it should also be noted that falling interest rates have been a wind at the stock market's back over most of the past 30 years. It will be a different story when rates start to rise.
- Valuation differences. By definition, backward-looking stock returns tend to peak just as a bear market is about to begin. You would do better to look at valuation measures such as price-to-earnings or price-to-sales to judge whether the market is a good buy now.
- New market uncertainty. When a new market gains popularity, the number of issues in that market increases dramatically, effectively rendering the limited history meaningless. This was demonstrated with emerging market debt in the 1980s: When there were relatively few issues and they had a fairly short history, default rates were negligible so returns were impressive. This attracted more investment money, so more emerging market debt was issued, and before long the default rate started to rise. This pattern has been repeated in many other markets since.
So what are past returns good for? With respect to different types of markets, history can give you a feel for what those investments are capable of -- especially if you look not just at long-term averages, but also how widely returns have deviated from those averages at times.
For specific investment managers, past returns can only tell you something about how successful the approach was if you fully understand the conditions under which those returns were earned. Few managers have succeeded under a broad range of conditions, but unless a strategy has proven itself in that way, the track record is incomplete.