Survivorship Bias is:
"the tendency for failed companies to be excluded from performance studies due to the fact that they no longer exist. It often causes the results of studies to skew higher because only companies which were successful enough to survive until the end of the period are included."
An example of it comes from this Wall Street Journal article posted on NYU Professor Damodaran's web site:
"Consider 1986: A decade ago, Lipper reported that 568 diversified U.S. stock funds had delivered an average 1986 return of 13.39%. Today, Lipper, which supplied the data for this section, puts the average 1986 return at 14.65%. Why the improvement? The new number is based on the performance of only the 434 funds from the 1986 group that are still in business today."
In the June 2006 edition of the Journal of Financial Planning, a short blurb notes that Morningstar's data may include suvivorship bias of up to 1.6% between 1996 and 2004:
"A study by the Zero Alpha Group and ZAG member Savant Capital Management claims that because Morningstar doesn't eliminate "survivorship bias" in its mutual Rind data it is "systematically and significantly" overstating the performance of actively managed mutual funds.
Morningstar concedes that it doesn't include defunct funds in its performance rankingneither do other major fund data providers-but it argues that not only is it no big deal, the elimination of survivorship bias creates its own problems.
Savant and ZAG (composed of nine independent investment advisory firms, all proponents of indexing) studied Morningstar data from 1995 to 2004 of actively managed funds relative to their related indexes. The survivorship bias, claims the study, inflated managed fund returns relative to their index, on average, by 1.6 percent a year."
Playing with Morningstar's Fund Screener, after having checked "No load funds" I got the following results (with S&P results in parenthesis):
YTD: 10.87 (11.89)
1 Year: 11.12 (10.44)
3 Year: 10.11 (9.28)
5 Year: 7.74 (4.16)
10 Year: 6.78 (6.35)
Please note this isn't the entire database, there were only 200 results. As you can see, if you deduct 1.6% from the 10 year results of 6.78%, the Mutal funds only returned 5.18% - well below the return of the S&P 500 6.35%.
Apparently, there is a similar issue with Hedge Fund returns, although available data is much less reliable as hedge funds are very conscientious about the information they disclose.
The bottom line is "BE CAREFUL" when looking at historical results and aggregate averages of funds. They are probably skewed.
Survivorship Bias is: