Yet another WOW paper!
Miller decomposes mutual fund returns into an active portion and a passive portion. He then shows that the fees are for the active portion are much higher than most investors would suspect.
It is well known that actively managed funds are highly correlated with market indices. What Miller does in this paper is to break funds down into a passive index plus an actively managed portions. He then looks at expenses for the funds and then using a sort of weighted average tries to allocate them to the active and the passive portions of the fund.
When this is done, the reported fees for actively managed funds (which are reported for the whole fund but theoretically stem largely from from largely from the active portion, are much higher than a combination of indices and leveraged "market neutral" investments would suggest.
In Miller's words:
"Mutual funds appear to provide investment services for relatively low fees because they bundle passive and active funds management together in a way that understates the true cost of active management. In particular, funds engaging in "closet" or "shadow" indexing charge their investors for active management while providing them with little more than an indexed investment. Even the average mutual fund, which ostensibly provides only active management, will have over 90% of the variance in its returns explained by its benchmark index."Of course that active funds are highly correlated with market indicies is not new. We've known that for a long time (at least since Sharpe's 1992 Portfolio Management paper). What Miller now does however is to attempt to uncover how this impacts the true benfits of active management as well as what investors are paying for this actmanagementment.
Probably the best way to understand the idea is with an example from the paper:
"Consider, for purposes of illustration, the Fidelity Magellan Fund at the end of 2004. Based on monthly data from the preceding three years, an investor could have replicated the risk and return characteristics of the fund (including its R2 of 99%) by placing 90.87% of his or her assets in an index fund that tracks the S&P 500 and the remaining 9.13% in an appropriately chosen marketneutral investment. In this new portfolio, 99% of the variance of this portfolio is explained by the index and we can leverage it in a way that Magellan's beta and variance are also replicated. If we then take 18 basis points as the expense ratio for the passive component of Magellan (the same ratio as the version of Vanguard's S&P 500 index fund marketed to individual investors), Magellan might be seen as "overcharging" investors by 52 basis points on the passive component of its portfolio. If we were to assess those 52 basis points against the 9.13% of the portfolio that is actively managed, we would find that annual expenses account for 5.87% of those funds."See, I told you it deserved a WOW!
Miller, Ross M., "Measuring the True Cost of Active Management by Mutual Funds" (June 2005). http://ssrn.com/abstract=746926