Does past performance matter?

(MoneyWatch) Despite the SEC's warning that for actively managed funds past performance isn't an indicator of future performance, most individual investors continue to base their investment decisions on how the funds have fared in recent years. Are investors right to ignore the warning? Does past performance matter? To answer that question, twice each year the S&P produces its Persistence Scorecard, which, importantly, is free of survivorship bias. The following is a summary of the July 2013 report:

• For the three years ending March 2013, 16.6 percent of large-cap funds, 14.2 percent of midcap funds and 23.1 percent of small-cap funds maintained a top-half ranking over three consecutive 12-month periods -- all below the random expectation of 25 percent.
• For the five years ending March 2013, only 2.4 percent of large-cap funds, 3.2 percent of midcap funds, and 4.7 percent of small-cap funds maintained a top-half performance over five consecutive 12-month periods -- all below the random expectation of 6.25 percent.
• There was also no persistence beyond the randomly expected among top quartile funds. For the three years ending March 2013, just 3.4 percent of the large-cap funds and 6.1 percent of the small-cap funds remained in the top quartile. Randomly, we would expect 6.25 percent to remain. More impressively, there wasn't a single midcap fund that managed to remain in the top quartile.
• There is consistency in the death rate of bottom-quartile funds. Across all market cap categories and all periods studied, fourth-quartile funds had a much higher rate of being merged and liquidated.

The evidence demonstrates that the one thing that is persistent is the lack of persistence in mutual fund performance -- with the exception being that the worst performers tend to repeat, mostly because they have the highest expenses. The efficiency of the market in setting prices protects the dumb managers as much as it creates high hurdles for the smart ones. The bottom line is that very few active managers generate risk-adjusted alpha (returns above their benchmark) on a persistent basis, no more than would be randomly expected. That makes it extremely difficult to separate skill from luck in fund performance; at least that's what several studies found.

• What George Costanza can teach us about investing
• Beware bad financial journalism 
• Hedge funds continue to disappoint

Thus investors relying on past performance are playing a loser's game. They believe they can somehow do what others persistently fail to do. For the misguided, hope, hype, marketing and, perhaps, overconfidence triumph over wisdom and experience.

Image courtesy of Flickr user Eric Skiff.