tjbresearch.com
Monday, March 4th, 2013
from ideas to performance

It’s time to return to the topic of analyst-run mutual funds.  As detailed below, I have written about them before, and a few months ago a new working paper on the subject was published by Gjergji Cici and Claire Rosenfeld of the Mason School of Business at the College of William & Mary,SSRN | The title is “The Investment Abilities of Mutual Fund Buy-Side Analysts.” providing additional insight into these unique vehicles.

My previous postings included a background piece about research departments at asset management firms and the rationale for analyst-run funds;the research puzzle | It was called “the analysts get their turn.” the complications and considerations that come from having such a fund (including the possibility of “the research-fund tail wagging the research-function dog”);the research puzzle | The post is titled, “wagging the dog.” and a chart with commentary regarding some of the leading research-based funds.research puzzle pix | One of the five outperformed the S&P 500 during the time shown.

The paper by Cici and Rosenfeld begins with an introduction that says that despite the importance of analysts, “we know very little about their abilities and the role that their research plays in their respective fund families.”  The authors, of course, mean that there hasn’t been much work by academics to understand what analysts at a buy-side shop do and how good they are at it.  (One nice feature of the paper is that it references other studies of interest, but in comparison to the huge number of papers on sell-side research,the research puzzle | Here is a summary of the “decades of work” and the misunderstandings embedded in it. there haven’t been many at all that focused on the buy-siders.)

Those of us that study the research process might feel that we understand it in a qualitative way, but some of the most important skills of analysts defy measurement and it is common for the leaders of investment organizations to have trouble articulating how their analysts’ work is different or better than anyone else’s.

In considering the working paper, my standard caveats apply:  I don’t question the statistical methods or check the math.  I think about the ideas that are put forth and the conclusions that are drawn — and look for analogies and anomalies between them and my own observations.

The basic approach of the professors involved evaluating the performance of the analyst-run funds in comparison to other funds (that are not managed by analysts), both at the same fund family and elsewhere.  The bottom line was that the research funds did well in comparison.  That finding went against the expectations of “common wisdom” and the Human Capital Theory, which would argue that portfolio managers, being more experienced than analysts, should outperform them.  (The list of possible explanations offered by the authors is interesting.They appear in the paragraph starting at the bottom of page two of the PDF.)

To further study the organizational dynamics involved, the professors also gauged how extensively portfolio managers used the ideas of analysts that were in research funds — what they called the “analyst idea utilization ratio.”  There was a great variability in that utilization, with evidence that higher-skilled managers (as defined by past performance) use the specific ideas less than do other managers, sourcing more positions in other ways, although their performance was only on par with the analyst-run funds.  Those that use the analyst ideas to a greater extent actually lag, perhaps because “they either pick the worst analyst ideas or poorly weight those ideas in their portfolios.”

Their analysis also led Cici and Rosenfeld to “believe that the quality of investment decisions undertaken by analysts in their management of analyst-run funds can be viewed as a reflection of the research capability and strength of the investment process within each corresponding mutual fund family.”

Unfortunately, as someone wrote recently, “We see performance and infer process.”I can’t provide the attribution for that statement, which I read recently and now use frequently.  If you know who said it, please send me a note, so that I can credit the author appropriately.  So, that conclusion is too much of a reach for me.  I also have other questions, wanting more evidence on the source of the outperformance and the nature and impact of the structuring choices (many of which have changed over time) that convert the raw material of analyst recommendations into a portfolio.  Also, we need to keep in mind that the filings of these funds are infrequent, so that there is a lot of information not available.

All that said, the paper is a great example of how academic research should trigger further important work by practitioners.  Quite apart from the question of analyst-run funds, the ideas that the authors offer should inspire leaders of investment organizations to evaluate in new ways the interactions of analysts and portfolio managers — and to assess and explain the real impact of research output on investment performance.

This is the tenth posting in an ongoing series on equity research.the research puzzle | This index of the postings is updated online as they are published.