Some of the managers supposed to be among the sharpest have cost their clients $170 billion dollars over the last two decades. I’m referring to “plan sponsors” who handle pension funds, endowments, and foundations, and Scott Stewart, a former money manager who now teaches finance at Boston University, has documented their value destruction in a study, Absence of Value: An Analysis of Investment Allocation Decisions by Institutional Plan Sponsors, which appeared in the Financial Analysts Journal late last year.
When we initially wrote about Stewart’s research two years ago, he had examined data only through 2001. Since then, he has expanded his analysis to incorporate data from a longer time period, from 1984 to 2007, and across a wider range of asset classes, including equities and fixed income for US and non-US markets.
Plan sponsors destroy value through their selection of active managers – most problematic is their proclivity to fire managers because of poor short-term performance, Stewart says.
Stewart used the PSN database from Informa Investment Solutions, which consisted primarily of separately managed accounts. As of December 2006, PSN held approximately $13.5 trillion in assets – a very large sample, roughly twice the size of the assets held in the top 1,000 US pension plans at that time, according to Pensions & Investments. He ruled out any effects that might reflect survivorship bias.
Stewart focused his analysis on fund flows representing the decisions of plan sponsors to hire or fire a manager. He measured performance against what it would have been under a buy-and-hold scenario – if the manager had not been hired or fired.
Using the most conservative approach for interpreting his results, Stewart concluded that plan sponsors had collectively squandered $170 billion in value over the two-plus decades he studied. That translates to approximately 112 basis points per year, based on fund flows. That is, a plan sponsor with a 100% annual turnover of managers will lose 112 basis points annually on those assets.
Stewart’s results changed little if he included index funds in his analysis or if he did not. The value destruction came from the active side, cycling between active managers.
Value destruction persists for three years following a decision to change managers; only in years four and five does the value destruction decay to zero.
“Plan sponsors never make their money back,” Stewart told me. “If they simply went on vacation, they could save their clients $170 billion – and that doesn’t count transaction costs.”
More volatile asset classes experienced greater value destruction. Domestic growth and international equities were among those generating the greatest loss. Domestic fixed income had the smallest loss, which is not surprising, given that it does not offer the same opportunities for excess returns as do more volatile asset classes.
