Let's face it. Index fund managers don't get much respect. In contrast to the 'exciting' world of active management, index fund managers are perceived as a pretty boring lot who require little talent or information to dispatch their duties. Vanguard 500 Index Fund manager Gus Sauter famously listened to a prospective investor say over the phone that 'a monkey could run that fund.' In fact, it takes a special kind of talent to run an index fund. But the strategy and goals of index managers are radically different from those of active managers.
Managing an index fund is a basis-point game. Unlike active equity managers, who hope their fund's performance will consistently rank high relative to their benchmark and peers, index fund managers are usually indifferent about the absolute performance of the index itself. That is not to imply that index fund managers aren't interested in a particular benchmark's performance, or that they don't quietly hope that the equity markets will move higher. But whether the benchmark index appreciates or depreciates is of little consequence.
Of significantly more importance is by how many basis points (one basis point equals 0.01%) their fund underperforms or outperforms its benchmark. This is known as tracking error. Each basis point is critical. Each one will ultimately reduce the fund's correlation with its benchmark. Index fund managers routinely scrutinize their tracking error and take corrective action as often as necessary, to mirror the benchmark as closely as possible. Because the success of an index fund is measured by its correlation to the performance of the benchmark, successful index fund management requires a detail-oriented, micromanaging, perfectionist personality.
Most index fund managers use a 'full replication' strategy, owning each security in the benchmark in the exact weight as the benchmark. This is the most conservative strategy and is generally used, if possible. However, some funds employ a 'sampling' strategy. There is a slightly greater risk of higher tracking error with funds that use sampling, which involves the selection from the chosen benchmark of specific constituent securities with characteristics that are very similar to those of the full benchmark. For example, a manager may own 300 of the S&P 500 names, or 1000 of the Russell 2000. The benefit of this strategy is that the manager may produce results that are comparable to the benchmark, without the added expense of owning each security in it. Intuitively, the downside of the strategy is that the fund ultimately will produce more tracking error over the longer term. However, correlation is often remarkably accurate when employing the sampling method.
Sampling is a common strategy used by fund families attempting to replicate indexes that include a very large number of securities where actual ownership of all may be impractical. Examples of such indices are the Wilshire 5000 or the Lehman Brothers Aggregate Index, which contains approximately 7000 issues.
Assuming quality analytics and diligent portfolio management, an index fund theoretically should provide the benchmark return, less the expense ratio of the fund. To illustrate, if the index is up 10% and the fund's expense ratio is 0.50%, then 'perfect' tracking would imply an index fund return of 9.50%. In reality, very few funds have been capable of performing 'perfectly' and tracking varies. The principal reasons for tracking differences are the timing and size of cash flows, transaction charges, and other expenses. Interestingly, what is often over-looked in the passive index business is that if a fund beats its benchmark, this also is tracking error. In our example, if the fund returns 10.01%, or 10.25%, this is tracking error-positive tracking error, but tracking error nonetheless.
