Home > Library > Stable Times > Volume 10, Issue 2  

Evaluating Manager Performance

After years of debate, the stable value industry continues to search for a uniform way to evaluate the performance of stable value managers. Despite the slow progress, Victoria Paradis, a leading advocate of enhancing manager evaluation within the industry, hasn't given up hope that it will get done. Speaking at the Stable Value Investment Association's 2006 Spring Seminar, Paradis, a former SVIA Chair and a Managing Director with JPMorgan Asset Management, said she still believes the industry can develop a framework for evaluating manager results that will be meaningful not only to its plan sponsor clients but also to the community of pension consultants who help those clients select investment managers for their retirement savings plans.

"Manager comparability is and will remain a challenge for the industry," Paradis said. "But we have made progress. Today, market value reporting of fund performance is far more common than it was years ago. In fact, it's become very common because market value returns are the most useful measure of investment decision-making success. However, it's not a complete solution because there is such a wide range of underlying management strategies. I do believe we can provide a framework for evaluating a stable value manager that recognizes that fund design, not just investment decisionmaking, is critical to determining a fund's success." Paradis cited three key components of performance: fund structure, investment policy, and a manager's own investment performance.

Historically, the performance data reported by managers was the book value returns generated by their portfolios-the actual returns produced for retirement plan participants. But while the industry generally agrees that book value returns are the sole measure of performance for participant reporting, many concede they are a poor way to compare one stable value manager to the next. That's because book value returns can be influenced by a raft of factors outside the managers' control, from investment constraints imposed by the client to the trading patterns of plan participants and the resulting impact on cash flows.

Brad Bennett, a Senior Portfolio Manager with Standish Mellon Asset Management, stated that cash flows can have significant impact on stable value fund performance. He illustrated with an example. He compared the performance of two theoretical stable value funds identical in nearly all respects. Each had a starting yield of 6.5 percent and reinvestment rates of 5.5 percent, although one reinvested $20 each year and the other reinvested nothing. After two years, the fund without any reinvestment had an annualized return of 6.5 percent, while the other earned just 6.31 percent. The entire difference was attributable to the differing cash flows.

While marking a stable value fund's underlying portfolio to market can eliminate the distortions caused by external factors, market-value reporting still doesn't solve the measurement conundrum because there remains such a wide variety of management styles. Managers may have a preferred way of packaging underlying fund components to manage withdrawal risk. Different managers labor under distinct investment policies with differing duration and quality characteristics, permitted investment sectors, and market benchmarks. As a result, when evaluating a manager, it is useful to analyze the success of the fund structure and investment policy, in addition to the underlying investment performance.

Finally, there is no single market benchmark appropriate for the entire stable value industry. In fact, there is no consensus on whether an individual manager should be held accountable to a single, standard-market benchmark or a custom benchmark. "The value of a single benchmark can be lost," observed Ben Allison, a Senior Stable Value Portfolio Manager for Invesco Institutional, "when unique-plan issues force a change to the portfolio that is not based on the manager's investment outlook."

Allison suggested the industry may need to rely on a combination of benchmarks, with the possibility that a given fund's allocation to each could be adjusted over time as appropriate. Paradis, however, warned that a customized benchmark precludes manager comparability because plan sponsors and consultants cannot evaluate managers against any peer group. One compromise solution, she suggested, might be to use a single benchmark, but allow a manager to adjust it to reflect the duration of their portfolios.

Paradis suggested that coming up with a performance measurement framework that adequately accounts for these issues would be valuable for the entire stable value industry. It would, she said, help plan sponsors assess whether the book value performance delivered to participants was attributable to fund design or their manager's investment prowess. By answering that question, sponsors could better decide where to direct their remedial efforts. If, for example, a manager's portfolio was outperforming her fund's market benchmark but the book value return was underperforming her peer group, it would suggest that the problem might lie in the fund's design.

"We should not expect consensus on these issues," Paradis concluded. "This is why hiring a stable value manager is more complex than traditional investment evaluation."

Read Next: Value Managers Find Prudent Ways to Boost Returns