Selection & Survivor Bias

One of the stories that was often told to newbie’s in the stock market was how investing just a small sum of 14,500 in 1993 public issue of Infosys would have been worth a few million rupees as on date. There is no falsehood in the statement either since Infosys had for long been a darling stock of the market and has provided unprecedented returns to any investor who invested when it IPO’ed and held on to the stock till date.

But what is not immediately seen is how this return is due to pure Selection and Survivor Bias. Selection Bias is defined in Wikipedia as

“Selection bias is the selection of individuals, groups or data for analysis in such a way that proper randomization is not achieved, thereby ensuring that the sample obtained is not representative of the population intended to be analyzed. It is sometimes referred to as the selection effect.”

When we select Infosys, we are selecting one among maybe hundreds of IPO’s that was seen in the same year (1993). Compounding the error, we fall prey to another common bias which goes by the name “Survivor bias”. Once again, the definition of the bias from Wikipedia

“Survivorship bias, or survival bias, is the logical error of concentrating on the people or things that “survived” some process and inadvertently overlooking those that did not because of their lack of visibility. This can lead to false conclusions in several different ways.”

The reason we selected Infosys was that it was a survivor among the hundreds if not thousands of companies that traded in those days. If you remember a company (which was a hot stock in those days) by name, Orkay Silk Mills, you will know what I am leading you towards.

When analyzing financial databases, its very important that both these factors are taken into consideration as otherwise your results will be very biased and in all probability under estimates the risk and over estimates the reward.

When Mutual Funds are analyzed, you need to keep a eye open to these biases since over time, its a list of those funds / fund houses that could survive. Bad funds generally get merged with a better fund, small AMC’s get taken over and their funds merged with other schemes.

In other words, if you were to get a list of say the best funds of 2000, its very much probable that a lot many funds don’t even exist today. Bad funds don’t survive for long since it showcases failure of the fund manager / AMC and faster it gets deleted, the better it is.

Using ValueResearchOnline, I found 169 funds which have been merged with other schemes and are no longer quoted. Even this list (post 2003) misses fund houses such as IL&FS and hence not complete. Add to that, I could also find another 124 funds that were “Redeemed”.

The question is, what would your return be if you were invested in any of those funds that got merged over time and how does that compare to what the fund that now in existence (into which it was merged) performed.

The art of investing has to begin by starting to ask the right questions. Do remember, there is no Free Money out there. A large cap fund cannot outperform its benchmark by a yard unless it did something different. Now, whether this out performance came in due to additional risk (investing in a large mid cap stock for instance) or cutting of risk (going into cash) comes with its own Pro’s and Con’s. Understanding that aspect enables you to understand performance better and stick with it for a much longer period that you otherwise would have been comfortable with.

If you are serious about learning the biases and fallacies that affect our judgement, do read Thinking, Fast and Slow by Daniel Kahneman, am sure it will provide you with a perspective that is easily missed by majority of investors out there.