How Is This Research Useful to Practitioners?
The authors highlight the influences on decision making by plan sponsors—namely, conflicts and apparent irrational behavior in asset manager selection.
Plan sponsors’ expectations of asset managers’ future performance are driven by past performance, nonperformance factors (e.g., the business processes, quality of personnel, and service delivery) and, to a lesser extent, by investment consultants’ recommendations. Fund flows are driven significantly by past performance and by investment consultants’ recommendations and are only marginally a function of expectations. The authors conclude that plan sponsors respond to past performance and consultants’ recommendations because of the associated personal and professional benefits, not because of any advantage they offer in extrapolating future performance. Plan sponsors’ expectations, past performance, and asset managers’ soft factors are shown not to reliably predict the performance of those asset managers.
The authors interpret investors not fully acting on their own expectations and instead following past performance and consultants’ recommendations when making investment decisions as being consistent with agency effects. Furthermore, the authors believe that plan sponsors rely on past performance and consultants’ recommendations because they feel that these indicators are more defensible to their bosses and stakeholders than their own expectations, especially if their recommendations disappoint. Because past performance and consultants’ recommendations are widely followed measures, plan sponsors recognize that when they deliver subpar performance, they can potentially deflect responsibility. The authors argue that plan sponsors implement herding and scapegoating strategies and that sponsors’ decisions are affected by their strong interest in avoiding blame for poor performance.
How Did the Authors Conduct This Research?
There are two sources for the authors’ data. The first source consists of two series of surveys conducted by Greenwich Associates, a global provider of market intelligence and advisory services to the financial services industry, between 1999 and 2011. In one series, plan sponsors rate the asset managers they have appointed; in the other series, investment consultants rate asset managers using the same criteria. The second source consists of a dataset provided by eVestment, a third-party provider of analytic services for the institutional fund management industry, on the returns of institutional US equity asset managers and their assets under management.
The authors implement several regressions, including regressions estimating expectations of future performance and actual future performance based on past performance, soft investment factors, service factors, number of recommendations, and so forth. They also implement a regression that measures how flows respond to lagged past and expected future investment performance, changes in investment consultants’ recommendations, and variables measuring soft-service quality characteristics, among other factors.
Expected future performance is a poor predictor of actual future performance. So, the tendency of plan sponsors not to act on their expectations may actually not have adverse implications for investors.
The presence of behavioral and agency effects in decision making by plan sponsors should be recognized. The tendency for plan sponsors to extrapolate from past performance and rely on investment consultants’ recommendations when neither of these is informative about future performance is irrational. In addition, as the authors explain, the fact that plan sponsors do not act on their own expectations when making investment decisions is consistent with agency conflicts. Plan directors, when making asset allocation decisions, will likely consider not only the interests of plan participants but also their own personal and professional interests. Because expected future performance is not a good predictor of actual future performance, it would be informative to quantify what type of impact plan sponsors’ tendency not to act on their expectations has on investors.