Do Index Funds Contribute to the Mispricing of Securities?

In my post on The Myth of the Rational Market, I mention one of the principle authors of the efficient market hypothesis, Eugene Fama. If people are interested in his research (conducted with his colleague Kenneth French), they have a blog which is not frequently updated and has the unique (and appealing) form of a Q&A. The have a straightforward way of answering the questions, with Fama usually being very blunt and French occasionally filling in the details. I mention it because today they have an interesting question with a slightly misleading answer.

Index funds buy stocks "blind" without regard to company fundamentals. Do their activities contribute to mispricing of securities?

EFF: Index funds typically buy cap-weighted portfolios so they do not contribute to mispricing.

KRF: We analyze a general version of this question in "Disagreement, Tastes, and Asset Pricing" (Journal of Financial Economics, 2007). Suppose index fund investors hold a passive market portfolio. Then from a pricing perspective they are sitting on the sideline. They are not overweighting or underweighting any securities, so they do not affect (relative) prices. As a result, it is hard to argue that they contribute to mispricing.

You can see the problem here. Fama and French are operating in a totally theoretical world where index fund investors invest in the "market index." The market index is a market-cap-weighted fund of all tradable securities (or at least all tradable equities). But no fund like that exists. Instead, you have funds for the S&P 500, the DJI, the Wilshire 5000, the FTSE 100, the Russell 2000, etc. So index fund holders are deciding (passively) to own one group of stocks instead of all other stocks based on an somewhat arbitrary reason (the market cap, the location of the exchange, etc.). This seems likely to cause some mispricing--at least in the real world.