The money management industry hardly suffers from a shortage of benchmarks to measure portfolio performance. Evolving from the one-size-fits-all Standard & Poor’s 500 index in the 1970s, benchmarks proliferated in the 1980s and 1990s with scores of style-, sector- and geographically based indexes, such as the Russell midcap index and Morgan Stanley’s MSCI EAFE index of European and Asian stocks.
The proliferation hasn’t stopped. Increasingly, consultants and investors are looking to improve upon these benchmarks, and create new ones, to better match -- and thus evaluate -- a manager’s investing style.
“As markets have grown more complex, and as institutional investors have become more sophisticated, the S&P, the Russell 2000 and other broad-based indexes have become less relevant,” says Andrew Lo, a professor at Massachusetts Institute of Technology’s Sloan School of Management and director of its Laboratory for Financial Engineering.
Benchmarking is crucial to evaluating investment managers and constructing a portfolio of funds whose overall asset allocation is tailored to an investor’s goals -- whether it is diversity or exposure to specific industries or stock sectors. For a consultant or investor, the first challenge is to determine a manager’s true investment style, which may differ significantly from what’s advertised. Once the appropriate benchmark is found, a manager’s skill can be assessed and a broader portfolio of top managers can be assembled whose aggregated investment styles fit the investor’s overall objectives.
Callan Associates, Mercer and other major consulting firms commonly use custom benchmarks that blend available indexes to create a better match with the money manager’s style. A smaller group of consulting firms choose to analyze the specific holdings of a manager’s portfolio to determine the best yardsticks.
By using benchmarks that conform more closely to a manager’s style, consulting firms can more easily identify a stock picker’s skill. They also help the firms to identify style drift among managers.
“I think these new benchmarks really cut through the clutter and are the best way to evaluate any manager’s performance,” says Louis Navellier, found-er and chief investment officer of Reno, Nevadabased Navellier & Associates, a fund manager with more than $3 billion under management.
One variety of the new benchmarks, so-called holdings-based style analysis, examines the stocks in a portfolio and categorizes them by style class or quantitative characteristics such as size or value-growth orientation. This gives an accurate view of a portfolio at the time the holdings are analyzed (usually on a monthly or quarterly basis). The traditional purpose of holdings-based analysis is to get a clear picture of what’s in a portfolio and how risky the package is. Evaluators compare how the portfolio or its subcomponents are performing against various style indexes.
Returns-based style analysis is a cheaper alternative to holdings-based analysis. It breaks down a fund’s historical returns and finds the public index, or combination of indexes, that best correlates with the returns. It’s less expensive because it doesn’t take into account the actual stocks in the portfolio, just the portfolio’s performance.
Designer benchmarks are a different animal. Instead of analyzing past performance or actual holdings, they are based on a universe of securities whose characteristics, such as market cap or industry sector, fit an investor’s goals and a portfolio manager’s expertise.
“Custom benchmarks are helpful in clarifying and understanding what we have in a portfolio in terms of the kinds of managers and the risks they bring,” says Michael Rosen, a principal at Angeles Investment Advisors in Santa Monica, California. “It’s quite powerful stuff that allows us to control our overall risk.”
As benchmarking becomes more sophisticated, MIT’s Lo remarks, it will intensify pressure on money managers to justify their fees by delivering real alpha -- returns better than beta, or the market return.
Because it’s the least expensive, the most popular approach is returns-based style analysis. Most often this entails creating a custom benchmark by blending two or more popular style indexes, such as the Russell value index or the S&P midcap index. The components of a blended benchmark are investable indexes, so this method of portfolio manager evaluation helps to answer a basic question: Should an investor choose a low-cost passive approach or opt for active management with higher fees? If the active manager isn’t adding value -- real alpha -- to the market return as measured by the benchmark, there’s no reason to pay the higher fee.
A host of companies, including Barra, Ibbotson Associates and a small firm, Zephyr Associates of Zephyr Cove, Ne-vada, sell software to build these new benchmarks. The holdings-based approach to creating a custom benchmark is more expensive, but proponents argue that it often yields a more accurate evaluation because it is based on stocks actually held in a manager’s portfolio. A series of portfolio snapshots from a holdings-based analysis can identify style drift. These snapshots can also provide a more accurate assessment of a manager’s skill and the overall risk in the portfolio.
“It allowed us to take a look to see whether the manager really was adding value or just happened to be in the right place at the right time,” says Thomas Quinlin, president of San Jose, Californiabased Lifestyle Design Group, a consulting firm that also manages about $130 million in assets.
In September, Ronald Surz, founder and president of PPCA, a pension consulting firm based in San Clemente, California, began selling a new kind of software, which he calls Popular Index PODs (short for portfolio opportunity distributions). PODs measure performance against subsets of managers investing in similar stocks and with similar goals, drawn from a computer-generated random group of some 10,000 managers. Plotting a given return against the distribution of outcomes, which crowd the middle of a bell curve, reveals how much alpha a manager is adding, or losing, and to what extent the result is statistically insignificant. If the manager consistently scores at an extreme, it’s highly likely that it’s the result of skill -- or lack thereof.
“PODs give a clear view of alpha,” says M. Barton Waring, a managing director at Barclays Global Investors in San Francisco who helps large pension plans set investment strategies.
A handful of consulting firms are going a step beyond these yardsticks, which are based on a manager’s historical behavior, to designer benchmarks that are set up in advance, consistent with an investor’s goals and a manager’s strategy. These benchmarks have been around for several decades but have been largely sidelined because of the cost and time required to create them. Now they’re gaining a higher profile. “We’re coming back to these because we’re more sophisticated on investment philosophy and strategy and also on the technology side,” says Lo.
Creating a designer benchmark requires a collaborative process between manager and investor, and that can be an advantage. “We like to build a communications bridge from the manager to the fund sponsor and help both parties establish an accountability standard for performance evaluation and, more importantly, for risk management,” says Thomas Richards, co-founder and principal of Richards & Tierney, a Chicago-based consulting firm.
Electronic Data Systems Corp. in Plano, Texas, has used designer benchmarks for the $750 million domestic equity portion of its $2.5 billion pension portfolio since 1995. “We hired some managers who are pretty eclectic, but they weren’t managing to any index,” says David Nixon, director of investments at EDS. “To measure them, we felt we needed something that reflects the pond they’re fishing in.” The result, he says, has been a collection of managers who have added “significant” value.
Consulting firm Hewitt Bacon & Woodrow recently introduced SimIAn, a product that combines designer benchmarks and PODs. Kerrin Rosenberg, an adviser who chairs the London-based firm’s investment policy group, expects a couple of major U.K. pension funds to soon begin using the new product. There is a particularly strong need for these benchmarks in Europe, he says, because the London Stock Exchange and other smaller bourses are often dominated by a handful of giant companies. Oil giant BP, for instance, makes up about 10 percent of the market-cap-weighted FTSE 100 index.
“We want someone building portfolios based on their conviction in companies and not putting companies into the portfolio because they’re a large part of the index and can’t avoid them,” Rosenberg says.
MIT’s Lo, among others, thinks the trend toward custom benchmarks has only just begun. He predicts that in ten or 15 years, individual investors will routinely design their own bogeys. “It will no longer be seen as so revolutionary, in the same way that we have software that helps us keep track of appointments,” he says.
