Philosophy

Style analysis is a widely used technique for scrutinizing investment managers. It is most powerful and informative when customized to the individual manager. Some managers are index huggers so a single style index works fine for them, but liberated managers are best viewed as custom blends of styles. Style classification provides a convenient description of managers and leads to better benchmarking, so we can assess success or failure. The question "Is performance good?" requires an answer to a second question: "Relative to what?" Style analysis provides this answer in the form of a custom benchmark.

The original concept of style analysis began about 35 years ago, with research conducted by James Farrell who at the time was with TIAA-CREF. Jim suspected that the price movements of certain stocks clustered together. Groups of stocks move in tandem, independent of the movement of other stocks. So he experimented with cluster analyses and discovered "homogeneous stock groupings" which set the stage for today's style indexes. Upon inspecting the constituents of these clusters, Jim assigned them natural names like value and growth, and large and small. Russell, Wilshire, PPCA and others seized upon the idea, and style indexes were born in the 1980s.

This set the stage in 1988 for Dr. William F. Sharpe, Nobel laureate, to invent "effective asset mix", which subsequently became known as "returns-based style analysis." Dr. Sharpe introduced mathematics to solve for the blend of style indexes that most closely matches the return behavior of an investment portfolio. The resulting mix could have been deployed at a low passive management fee in lieu of the manager, so it makes a good benchmark and a good classification profile. And it has really caught on. Returns-based style analysis is used by most due diligence teams today. Some also complement returns-based with holdings-based analysis. For a discussion of the distinctions between these two approaches, see [Surz, 2003].

Now in this announcement we introduce a 21st century quantum leap forward in style analysis that we call "Style Scan." In keeping with the word "quantum", Style Scan is to investing what particle physics is to science; it's particle finance. You could call Style Scan "particle-based style analysis." Importantly this next generation of style analysis is actionable. Security selections and allocations are managed much more efficiently and effectively with Style Scan than with previous approaches.

Style Scan brings style analysis back to its individual security roots using visual displays that have been created for returns-based style analysis, so the framework should be familiar to the reader. Every stock in a portfolio is plotted in Cartesian style space, better known as "X-Y plots", as shown in the exhibit on the right. The Y coordinate, or size, is determined by the company's outstanding capitalization. The X coordinate, or value-growth, can be determined in either of two ways - a three factor model employed by the Surz Style Pure® Indexes, or a single factor price/book factor employed by some popular indexes. The user can choose between these two approaches. This is an excellent way to experience the differences that are caused by the choice of style "lens." Our choice of a style index family significantly effects our perceptions.

The size of each dot in the exhibit is a reflection of the allocation to that stock. Hover above any dot, and the name and allocation are revealed to you. Click anywhere in the plot space and a list of stocks and ETFs with that coordinate are presented. In this way outlier stocks can be quickly identified and voids can be easily filled. For example, we've found that many portfolios, especially "managed accounts" of multiple managers like UMAs, are making an unintended bet against large core because the industry has fallen into a 4-corner style allocation trap. We see this unintended bet as a void in the top center region of the graph. This underweight to large core is a mistake that was costly in 2008 because core significantly outperformed both value and growth. This mistake can be easily remedied with a large core ETF (note this does NOT mean an S&P or Russell index) or a few stocks in this market segment. Style Scan is a powerful and fun portfolio management control.

Also, click on any economic sector in the pie chart that is provided, and stocks in that sector are highlighted in the graph, creating a sector-style mosaic within the style signature.

Consultants and investors use Style Scan to visualize the details of the manager's current approach and to compare managers to one another. Each manager has a unique "style signature" that should be an integral part of all due diligence.

Conclusion

Significant breakthroughs like Style Scan are rare and important. This new 21st century actionable application of style analysis provides exciting insights and controls for investors and their managers.

Interested parties are invited to experience Style Scan firsthand, inputting their own portfolios, or simply some of their favorite stocks, at an on-line live application available on our DEMONSTRATION web page. We've also provided sample indexes for those who simply want to see how Style Scan works.

References

Farrell, James L. "Homogeneous Stock Groupings: Implications for Portfolio Management." Financial Analysts Journal, May-June 1975.

Sharpe, William F. "Determining a Fund's Effective Asset Mix." Investment Management Review, December 1988, pp 59-69.

Surz, Ronald J. "Style Analysis: How and Why." Senior Consultant, February 2003.