Most sensibly designed securities indexes can serve as the basis for an ETF portfolio, sometimes with a few minor changes in the index's structure or composition change policy. At the risk of oversimplification, several rules of thumb seem appropriate. Major benchmark indexes like the S&P 500 and the Russell 2000 are not efficient fund templates, largely because they are too popular. Because they are usually the most actively traded ETFs, some major benchmark index funds can serve as appropriate temporary vehicles for an advisor who wants to take a temporary position in a particular market segment at low cost while developing and implementing a longer-term plan for a client. ETFs based on these indexes are not appropriate long-term holdings, simply because of the market impact cost of their portfolio composition changes. See Equity Index Funds Have Lost Their Way," The Journal of Portfolio Management, Winter 2002, pp. 55 – 64. Winner of a Jacobs Levy/Bernstein Fabozzi Award as an Outstanding Paper. For long-term performance, index popularity is a curse that investors and advisors should attempt to avoid. Some ETFs are based on indexes that appear designed more for their intuitive appeal or marquee value. Some low profile indexes are reasonable fund templates, but indexes should be evaluated on the basis of investment merit. Index investment merit is usually inversely proportional to index popularity.
2. Does it matter who manages or publishes an index?
The rules for publishing information about most indexes are sufficiently clear and the process is sufficiently transparent to eliminate opportunities for gamesmanship by the index publisher. While some indexes seem almost "silly" from an investment perspective, as long as you evaluate the components of the index on the basis of investment merits, there seems to be little reason for concern about the qualifications of the people who develop and publish indexes. As in any other field, it pays to check references and ask questions.
3. Are fundamental indexes or other non-cap or non-float weighted indexes a good idea?
Until the introduction of what has come to be called "Smart Beta" (Essentially a decision to use a non-traditional index weighting model) cap- or float-weighting was almost a religious issue for many long time believers in the virtues of indexing. I am far more concerned about the cost to investors associated with high profile composition changes in the most popular benchmark indexes than I am about the purity of an investment process or the consistency of the index weighting with modern portfolio theory. In my view, it is much more important for a long-term investor to avoid the costs associated with composition changes in overly popular cap- and float-weighted benchmark indexes than it is to assure a particular kind of diversification. Don't hesitate to look at fundamental indexes or any of a wide variety of less widely used index weightings that will reduce the total transaction costs associated with trading in the portfolio of a fund you own and may give you better performance under some circumstances.
4. Should index composition turnover be an issue for a fund investor?
Turnover in an index means turnover in an ETF portfolio based on that index. It is not possible to avoid all portfolio turnover for a host of reasons, but any turnover in an ETF or mutual fund imposes a cost on the owners of that fund. Turnover is generally a smaller problem with a less popular index and with large-cap rather than small-cap stocks or Treasury debt rather than non-investment grade bonds. Other things equal, less turnover is better than more turnover. High turnover in an index and in a fund is a topic for analysis and evaluation by investors and their advisors. A common sense examination of specific transactions is far more useful than any rule of thumb.
5. How important is transparency in an index fund?
Revealing a fund's transactions as soon as practicable after the fund manager has finished trading is a great idea. An investor or another portfolio manager may like the trade and follow the example, making the first manager look good. Trading transparency (revealing the intention to trade before the order is complete) is the greatest curse ever inflicted on most index investors. "Transparency" has become almost a mantra among many users of indexes and particularly of index ETFs. It is certainly appropriate that an investor know what is in the fund - at frequent intervals - but knowing what transactions the manager of the fund will be making on your behalf before the fund can trade is certainly not in your best interest as an investor. This answer does not reflect a currently popular, almost religious, viewpoint that investors are best served if everything the manager of an ETF does is disclosed in advance. In extremely rare circumstances, it might be in a fund shareholder's interest to announce a transaction in advance of completing the trade. I can't name one of those circumstances. I can only assume that the transparency fanatics have made a virtue of necessity because traditional (non-transparent) active management in the ETF structure was not possible until the introduction of NAV-based trading. If you have heard what you think may be a good argument for pre-trade transparency, send me an e-mail and I will be glad to point out the error in the argument.