Hosting an industry conference? Ask us about including it in this ticker?
What do you think of our site upgrade?

New Research Reveals When Active Beats Passive

November 30
15:23 2009

A new study by William R. Thatcher adds more fuel to the increasingly warm fire burning the fortress Bogle. While “When Indexing Works and When It Doesn’t in U.S. Equities: The Purity Hypothesis,” (The Journal of Investing, Fall 2009)

812643_38612409_Yin_Yang_Rocks_royalty_free_stock_xchng_300does leave enough questions to spur criticism, it successfully articulates the theoretical foundation against a passive-only approach to investing. As such, it continues the trend towards a mixed active-passive management strategy. Moreso, Thatcher uses the language of Modern Portfolio Theory (one of the aforementioned questions) to slay MPT’s own progeny.

In the course of his thesis, Thatcher uses both logic and empirical data to disprove the Efficiency Hypothesis. He defines the Efficiency Hypothesis as the belief indexes outperform active managers in more efficient markets (e.g., large cap stocks) while they tend to underperform active managers in less efficient markets (e.g., small cap stocks). In doing so, he relies heavily on William Bernstein’s April 1999 paper “When Indexing Fails.” Thatcher, a senior consultant at Hammond Associates, proposes, rather than using efficiency as a guide, investors should instead draw on relative performance. In other words, with top performing indexes, passive outperforms active. On the other hand, with bottom performing indexes, active outperforms passive. He calls this the “Purity Hypothesis.”

The idea appeals to common sense. By “purity,” Thatcher means how close the portfolio comes to resembling the actual index the manager uses as his benchmark. A purer portfolio will more closely resemble the index. A less pure portfolio will see returns regress to the mean versus the index; hence, performing better when the index is far below the mean and worse when the index is far above the mean. Thatcher isn’t the first to propose the possibility of this correlation (see “The Emperor Exposed,” Journal of Financial Planning, September 2005), but he does give it a memorable name.

He also provides empirical evidence to defend his hypothesis. Although the science is a bit iffy (see below), the performance data for the ten year period ending 2007 clearly supports the Purity Hypothesis. To his credit, Thatcher does admit – and show – the hypothesis fails to work consistently when viewing 12 month calendar return data. Of course, this may merely suggest the Purity Hypothesis is only one factor in determining when active beats passive and vice-versa.

As mentioned, the paper presents plenty of questions. First and foremost, as the author explains in his work, his reliance on Morningstar data presents a survivor bias in the active manager performance. He attempts to merely append Carhart’s solution (“Mutual Fund Survivorship,” Review of Financial Studies, Vol. 15 No. 5 (Winter 2002), pp. 1439-1463) – i.e., by subtracting 66 basis points from the active performance. A better solution might have been to use Lipper return data, which includes inactive funds as well as a more complete universe of mutual funds. In addition, because Morningstar can sometimes misplace funds, one can easily question the data’s reliability of matching the fund to the correct index. For example, Morningstar, unlike Lipper, does not have a “Multi-Cap” category and has at least on some occasions placed those funds in the “Mid-Cap” category, skewing the overall results as to favor the index rather than the active pool.

One complaint that might seem apparent but Thatcher actually addresses revolves around the selection of the comparison index. His paper only shows the S&P comparison, but he also attaches a (much larger) appendix showing similar results for the MSCI, Russell and Morningstar indices. The results in the appendix might offer the most interesting data of the entire report. Russell, a true passive – or “pure-ranking” – index, performed substantially poorer versus the S&P actively managed – or “rules-based” – index. This very discrepancy has been the source of much debate within the indexing industry.

Finally, and this perhaps might present the most glaring issue with Thatcher’s research, the author looks only at a static moment in time. As first empirically identified in “The Emperor Exposed,” the “Snapshot-in-Time Anomaly” sometimes leads to seemingly conflicting conclusions, depending upon which date the relevant period ends. To avoid this anomaly, performance based research might offer more should it look at rolling periods as opposed to one static period.

In an interview with, Thatcher declared he intends to conduct further studies, including possibly looking at rolling performance periods. Still, his paper remains consistent with a growing trend away from index-only investing and towards a passive-active hybrid. In particular, it reveals the need for both strategies in portfolios, like those found in various ERISA plans, requiring investment in all segments of the market, not merely the one best performing index.

About Author

Christopher Carosa, CTFA

Christopher Carosa, CTFA


No Comments Yet!

There are no comments at the moment, do you want to add one?

Write a comment

Only registered users can comment. Login is sponsored by…

Vote in our Poll


The materials at this web site are maintained for the sole purpose of providing general information about fiduciary law, tax accounting and investments and do not under any circumstances constitute legal, accounting or investment advice. You should not act or refrain from acting based on these materials without first obtaining the advice of an appropriate professional. Please carefully read the terms and conditions for using this site. This website contains links to third-party websites. We are not responsible for, and make no representations or endorsements with respect to, third-party websites, or with respect to any information, products or services that may be provided by or through such websites.