Callan to apply MSCI indexes as guide in manager searches

Callan Associates, San Francisco, adopted domestic indexes from Morgan Stanley Capital International Inc., New York, and plans to begin using the indexes — and the methods used to create them — as a tool in portfolio manager searches.

Ivan Cliff, Callan's director of operations, said the consulting firm, which has clients with a total of $908 billion in U.S. institutional, tax-exempt assets, recently completed integrating MSCI data into its own systems and will also use it to help Callan clients keep track of their existing managers and avoid investment style drift.

"Going forward, it's going to start showing up in more and more client reports," Mr. Cliff said. "The other thing we're talking about is utilizing this more in the external manager search process."

Until now, Callan had used information from its own database of investment managers.

"We have a huge database," Mr. Cliff said. "We do everything the hard way, but it gives us quality control."

But he said the MSCI methodology, which uses three variables to identify growth and eight for value, helps the firm save time on the quantitative portion of managers searches, which can be anywhere from one-third to one-half the process.

Typically, if consultants do not use their own data, most work with one of the two largest index providers, New York-based Standard & Poor's Corp. and Russell Investment Group, Tacoma, Wash. In addition, most money managers use Russell or S&P indexes internally for benchmarks.

Russell and Wilshire Associates Inc., Santa Monica, Calif., another index provider, both have their own consulting divisions.

The three fundamental factors that MSCI uses for defining value are: book value to price ratio; 12-month forward earnings-to-price ratio; and dividend yield. The five it uses for defining growth are: long-term forward earnings-per-share growth rate; short-term forward earnings-per-share growth rate; current internal growth rate; long-term historical earnings-per-share growth trend; and long-term historical sales-per-share growth trend.

MSCI crunches these numbers to come up with what it calls "Z scores" for every factor and then determines an overall value Z score and an overall growth Z score for every security in its large-cap, midcap and small-cap indexes. The Z scores, a statistical method to make different data comparable, can then be used to gauge the level of growth or value in a portfolio.

"They really thought about what fundamental financial characteristics at the individual stock level are important to portfolio managers," Mr. Cliff said. "In my opinion, it is a good way to track the relative value-ness or growth-ness of active managers."

The other major index providers — Russell, S&P and Dow Jones-Wilshire, a partnership between New York-based Dow Jones & Co. and Wilshire Associates — have been attempting to improve their own methodologies to provide more definition to their styles.

Dow Jones-Wilshire is planning to introduce U.S. equity indexes in early 2005 covering five investment styles: large-cap growth, large-cap value, small-cap growth, small-cap value and microcap. The companies use six factors for style classification: five-year average earnings growth and sales growth; price/book value ratio; dividend yield; forecasts of price/earnings ratios; and long-term earnings growth.

Consultants agree that improved style definition is important.

"When you're analyzing managers as part of a potential placement, every search is customized," Callan's Mr. Cliff explained. "You have to get into varying flavors — what's right for this particular client, how does this manager fit into the roster of others, do I need deep value or deep growth, who are you trying to offset and who are you trying to complement."

He stressed, however, that the quantitative part of the search is simply a framework to start qualitative discussions about managers.

At Ennis, Knupp & Associates, Chicago, consultants don't spend a huge amount of time crunching manager numbers, according to associate P.J. Kelly.

"We do some screening, but most of the work involves getting to know managers and organizations and their investment process," he said.

"Once you start slicing and dicing into style and capitalization, a lot of managers don't fit tightly within the classifications" the industry has created, he added.

Another problem with defining managers too strictly is that their own internal benchmarks might differ from the benchmarks used by consultants.

For example, if most midcap managers use the S&P 400 index as their benchmark but the consultant is using the Russell midcap index as a screen, "you can throw all the Russell midcap statistics out the window" because those two indexes have diverged over the last few years, said one consultant, who asked not to be named.

"I don't think anyone has come up with the perfect answer," Mr. Kelly said. "Some managers fit very well into how an index provider created it and some don't."

He said his firm tends to hold managers to beating the broad market, but in the case of a manager with a distinct style bias and large dispersion of returns between styles, the firm does compare them to the most appropriate style index.

"We think the ideal situation would be if you gave a manager an index that included the entire opportunity set — for example the Dow Jones-Wilshire 5000 or the Russell 3000 — which are representative of the entire market," he said. "As a plan sponsor your end goal is to outperform the broad market. Overclassification and restriction of skillful managers may divert resources from that goal."