I’m currently reading Fund Spy–a book by Russel Kinnel, Director of Mutual Fund Research at Morningstar.
One of the chapters of the book discusses how to intelligently use past performance data. Kinnel begins the chapter by explaining that past performance isn’t a good indicator of future performance. However, he then argues that:
“Fund returns tell you a lot about how a fund behaves in different types of markets and can help you to set realistic expectations.”
For example, Kinnel argues that you could examine an international stock fund’s returns to see whether the fund hedges its currency risk and to see how it responds to fluctuations in the value of the dollar relative to other currencies.
Seems like a neat idea.
But why bother?
Why not just read the fund’s prospectus to see whether it hedges currency risk? It seems to me that looking at fund returns is a roundabout way of assessing the situation.
I’d suspect that it introduces a great deal of noise as well. For example, imagine a year in which the following occurs:
- The currency of the country in which the fund invests declines in value relative to the dollar (this would ordinarily have a negative impact on the fund’s returns), yet
- The fund performs quite nicely.
It would be easy to conclude that the fund hedges away its currency risk and that, therefore, you wouldn’t have to worry about such risk when investing in the fund. But that’s not necessarily the case. Perhaps the fund’s performance was primarily driven by a few big, lucky stock picks.
Examining Losses
Kinnel also suggests the following use for a fund’s past performance data:
“Look at its biggest losses. Obviously the fund could do even worse than its biggest loss, but at least you’ll have some idea of what you’ll need to be able to tolerate.”
I like this idea. There’s no guarantee that a fund’s worst year so far will be its worst year ever, but if its worst year so far is worse than you can handle, you’d better stay away (or only invest a small portion of your portfolio).
Putting Past Performance in Perspective
Kinell concludes the chapter with the following reminder:
“Still, don’t lose sight of the fact that trading costs and expense ratios are better predictors than past performance.”
Sounds like good advice to me.
Want to learn more about investing?
Enter your email address to receive free updates from this blog. (You won't receive any emails other than blog posts, and you can unsubscribe at any time.)Like Cliffs Notes...for Investing
| If you're looking for a brief, plain-English introduction to investing, I'd encourage you to pick up a copy of my book: Investing Made Simple: Investing in Index Funds Explained in 100 Pages or Less. |  |
{ 8 comments… read them below or add one }
It’s always important to remember that old adage “past performance doesn’t guarantee future results.” But it can give you something of an idea. But it’s not great to completely live in the past, expecting the returns of the best years.
Mike, maybe you’re talking about specific funds rather than entire asset classes. But just yesterday, Frank Curmudgeon wrote: “for the 42 years ending 2009 the stock market was up an average of 8.76% annually.” Surely that can be read as an encouragement to buy and hold stocks over the long term—precisely as a result of past performance. And many of us use similar benchmarks when thinking of bonds, cash, and other major asset classes. Past performance may not be a guarantee, but it indicates at least a trend and at best a probability. Obviously it stands to reason to pick funds with the lowest possible fees too, and to diversify and rebalance in order to spread the risk.
Larry: Yes, I’m speaking specifically to Kinnel’s application of past performance figures of a specific fund to predict (aspects of) that fund’s future performance.
I’d agree that in a broader sense, past performance can be useful to learn things like “stock funds exhibit more volatility on a monthly basis than short-term government bond funds.”
One problem with looking at past performance is that most funds haven’t been around long enough. Even if a fund has been around ten years, that’s still not long enough to see what it will do in each type of market.
Mike – Does Kinnel give any advice for researching funds that haven’t had a long time frame?
RJ: He suggests (as with funds with a long time frame) to focus primarily on expense ratio and transaction costs. (More on this–and the book–next week.)
I’m using long run past history of stock markets (not funds as I now only buy index linked and save on the fees) to try and understand when markets are either overvalued or undervalued based on long run historical data. My method is based on the PE10 ratio developed by Professor Shiller. I’m then weighting my stock allocations based on this valuation method to try and squeeze some extra performance.
I have had good luck with funds that have had good managers. The things I look for are the managers age (I am keeping an eye on my Gabelli funds) and fund bloat. If a fund gets too large I am out.
The trouble is a “few big lucky picks” may be lucky, or they may be skillful — they may be Buffett buying American Express or Coca-Cola at just the right time, or they may be a me-too manager punting his way out of burnout.
I mean, people still argue Buffett was lucky.
All we can really be pretty sure of is in 20 years time we’ll look at these posts knowing there are a handful of young fund managers out there who will have delivered 15%-20% returns annualised between now and 2030.
Does any reader know how to identify them from the *thousands* out there? I don’t. I doubt their employers do.
For that reason, I’d rather bet on myself in the main if I go off-index. At least I get some fun out of the extra charges and potential under-performance!