Most pension funds and endowments design and use biased benchmarks that make their comparative performance appear better than it is, when actually they often underperform more appropriate benchmarks, according to a study published in The Journal of Investing.
"The benchmarks are biased downwardly, meaning their returns tend to be less than a fair return for the market exposures and risk exhibited by the institutions' portfolios. Significant samples of both fund types exhibit benchmark bias in the range of 1.4 to 1.7 percentage points per year," the study "Lies, Damn Lies, and Benchmarks: An Injunction for Trustees" said.
"This bias enables a sizable majority of both types of funds to report outperforming their chosen benchmarks when, in fact, most underperform an appropriate passive-management benchmark by a wide margin," says Richard Ennis, former Financial Analysts Journal editor and co-founder of the professional services firm EnnisKnupp.
Earlier research by other authors found a similar bias in mutual fund benchmarks, the study noted.
Benchmark bias masks serious problems in institutional fund management, Ennis wrote. "Fund staff and consultants have strong incentives to justify complex, costly, multi-asset-class portfolios, for which they themselves are the benchmarkers," the study said.
"Trustees may feel they have no choice but to accept the benchmarking and reporting by staff and consultants, but this only perpetuates the problem," Ennis said in the study. "At the very least, investment trustees should step up and take control of benchmarking and performance reporting. For they are the ones charged with watching the watchmen."
He cited other research showing evidence of similar benchmark bias in mutual fund performance reporting, with 67% of the funds studied exhibiting "a meaningful benchmark discrepancy." Those authors concluded that, among mutual funds with a benchmark discrepancy, "the prospectus benchmark significantly overstates" performance.
