Mutual funds were once hailed as a revolutionary financial instrument able to provide investors with instant access to well-diversified market indices and achieve economies of scale by pooling resources with fellow investors to develop investment portfolios that previously would have been prohibitively expensive. Over the past several decades, as fund sponsors have promoted their superstar managers and market-beating funds, countless funds have been launched, and assets under management swelled to as much as $26 trillion by some estimates.
The mutual fund industry has experienced this rapid growth despite some glaring flaws, which have been highlighted in a few academic studies and empirical data disparaging active management as inefficient and expensive. Although the cost-benefit analysis of active management has never made a truly compelling case for mutual funds, investors have never been presented with a plausible alternative. But now there’s a new kid in town. ETFs have established a legitimate threat to the existence of actively-managed mutual funds, and with good reasons. These passively managed investment vehicles have become tremendously popular primarily because they don’t even attempt what mutual funds have repeatedly failed to do: consistently outperform the market. Given the new investing landscape, it may be useful to look back (in some cases way back) at a few well-known cases against active management.
Over the past 25 years, a number of in depth studies have been performed attempting to quantify the costs and benefits of active management. Here’s a quick overview of a couple of the best known academic studies.
Overview: This groundbreaking study, first published in the Financial Analysts Journal in 1995, analyzed historical data from U.S. corporate pension plans to determine which investment decisions had the greatest impact on the magnitude of total return and the variability of that return. Relevant “decisions” include timing (strategic over / under weighting of an asset class) and security selection (selection of individual securities within an asset class). For each decision, this study calculated the returns attributable to employing both an active and a passive strategy.
Conclusions: The conclusions reached by Brinson et al.are summarized in the adjacent table. These results imply that there is no value added by active weighting and individual security selection. In fact, as presented in the table, the portfolios studied actually lost returns in active weighting and security selection. Rather, it is normal asset class weightings and the passive asset classes themselves that provide the bulk of portfolio return. Specifically, this study estimated that 93.6% of portfolio returns are attributable to asset allocation policy.
Critiques: There are several critiques of the Brinson study, the most thorough of which is attributable to William Jahnke. Among other faults, Jahnke claims that Brinson et. al overvalue the importance of asset allocation policy because they attribute all the market returns to policy. It is also obvious that this study is based on dated results that are now several decades old, although it’s not clear how this diminishes the validity of the results in any way.
Overview: In this 2000 paper, Roger Ibbotson and Paul Kaplan noted that many investors have regularly misinterpreted the Brinson studies. The authors examined 10 years of monthly returns for 94 mutual funds and five years of data for 58 pension funds to address three questions:
- How much of the variability of returns across time is explained by policy?
- How much of the variation in returns among funds is explained by differences in policy?
- What portion of the return level is explained by policy return?
Conclusions: In response to the three questions posed above, Ibbotson and Kaplan came up with the following conclusions:
- By regressing 120 monthly returns of mutual funds against the returns of the relevant benchmark, it was estimated that 90% of the variability in monthly returns of a fund can be explained by the variability of the fund’s policy benchmark.
- By comparing funds with each other through the use of cross-sectional analysis, the authors conclude that approximately 40% of the return difference between funds is explained by policy. Ibbotson and Kaplan note that many have mistakenly thought that this is the question answered by the Brinson study.
- By computing the ratio of compound annual policy return to compound annual total return, Ibbotson and Kaplan estimated that policy accounted for 100% of returns, implying that pension and mutual funds are not adding value above their policy benchmarks through active timing and security selection.
Much More to Debate
I’ve given a very high level overview of only two studies which sought to estimate the value of active management (there are many more). There are numerous complexities involved in each, and each has attracted a fair amount of well-reasoned criticisms (some of which I’ve tried to provide). But the underlying theme is difficult to attack. While some managers are able to generate alpha for their clients on a regular basis (Warren Buffett and Peter Lynch come to mind), the vast majority of managers do not add value by tilting assetallocations or selecting individual securities. In fact, many destroy value by overcharging relative to the value they add.
I’m sure many of you have your own thoughts on the merits of these studies (it can be a sensitive area for some), so it is my hope that this summary will spark some lively debate!
For more on the asset allocation debate, read these fine articles: