Abridged Version
Full version of this working paper is available at www.athenainvest.com
October 2008

Advisor Perspectives welcomes guest contributions.  The views presented here do not necessarily represent those of Advisor Perspectives.

In order to make sense of the bewildering array of thousands of equity mutual funds, market participants label funds and form small fund groups. Ideally, funds would be labeled by the way managers analyze, buy and sell stocks  - their investment strategy. Implementing this framework requires asking the manager about her or his strategy and labeling the fund accordingly.

But market participants are often skeptical of such an approach and instead look for supposedly objective but indirect labeling. For this and other reasons, Style Box labeling, which is based on the characteristics of the stocks held by the fund, is currently popular. It is widely used by advisors, consultants, managers, and investors for describing, analyzing, and choosing among fund offerings.

It is critical to determine if a fund labeling approach is effective, that it is grouping funds together that are pursuing the same strategy, while avoiding the undesirable consequences of a poorly thought out scheme. I compare Strategy labeling, which labels based on the fund’s stated strategy, to Style Box labeling, which labels based on the stocks held by the fund. I find strong evidence that Strategy labeling is superior to Style Box labeling in terms of forming homogeneous strategy clusters and for identifying top performing managers within each strategy. 

Poorly Conceived Labeling

To understand the problems created by ill-conceived labeling, consider a professional football team analogy. Imagine players, rather than being asked the position they play and labeled accordingly, are labeled based on their weight and 40 yard sprint time. The group of players weighing between 200 and 230 pounds with sprint times between 4.5 and 5.0 seconds are labeled “midsized, midspeed” or mid-mid for short. The mid-mid players include quarterbacks, punters, receivers, and safeties, among others. Such a labeling approach has the advantage of being objective, but in spite of its objectivity, it turns out to be detrimental to team performance.

First, knowing that a player is mid-mid says nothing about what he does on the team. So the label has to include a modifier, such as the player is a “mid-mid quarterback” or he is a “mid-mid receiver”. In order to identify all receivers on the team, it is necessary to look across all weight, speed groups.

Second, the mid-mid label gets in the way of player performance, because mid-mid players are compared to other mid-mid players as a group. So a quarterback has to be concerned about receiving since he is being compared to mid-mid receivers, while the receiver has to be concerned about passing since he is being compared to the mid-mid quarterbacks. As a result, each wastes time on developing skills that help little with their specialty. This undesirable consequence is the insidious result of labeling players based on weight and speed rather than on the position played.

Display article as PDF for printing.

Would you like to send this article to a friend?