Frank Armstrong

The Consulting Game

The one thing Wall Street tells you that you can always take to the bank is: "Past performance is no guarantee of future performance." Of course, they really hate telling you that, but they are required to. The rest of Wall Street's message is designed to make light of the disclaimer, and not for one second do they want you to believe it. They certainly act like they don't believe it. If you ever begin to take the message seriously, almost everybody on Wall Street is going to have to find another job.

The relationship between superior past performance and mediocre to sub-par future performance is as well documented as the one between cigarettes and cancer. Yet Wall Street continues to push its own brands of poison.

As an example of Wall Street's ability to mix messages while aggressively marketing a fatally flawed strategy, let's look at the pension consulting business.

As far back as 1965, Michael Jensen's groundbreaking study of mutual funds documented the consistent failure of active managers to match index performance. In 1968, the AG Becker and Company found similar results in the institutional market.

AG Becker and Company then went on to establish the pension consulting business with the creation of "Green Book" tables comparing results to benchmarks. Having convincingly demonstrated that most managers can't match an index, they went on to identify those few hardy souls that had. By extension, it was implied that this past performance indicated superior skill and cunning that would be duplicated in the future.

AG Becker's consultants would perform manager evaluations for large pensions with an eye to weeding out lagging performers. Then they would suggest replacement of the laggards with a proven past performer selected from their extensive database. Of course, all of the suggested firms had generated significant alpha (see glossary).

Once the replacements were made, a funny thing happened. These impressive past performers generated remarkably mediocre results. So, one or two years later, AG Becker would perform another study and suggest a new round of replacements. The explanation for the failure of their previous recommendations usually boiled down to a change in personnel or process at the manager's firm over which Becker presumably had no control. But, thank goodness they were monitoring the situation before things got completely out of hand!

Unfortunately, the new manager would usually insist on a complete portfolio revamp. Otherwise, why hire a new manager if not to change the investments? This, of course, generates transaction fees, commissions, and miscellaneous expenses that must be borne by the pension fund as a necessary cost of plan maintenance.

The pension consultants made themselves available - for an additional fee, of course - to repeat the cycle endlessly.

The operative philosophy, and fatal flaw, of this approach is that past performance is related to future performance. As literally thousands of studies have shown, this is demonstrably not true. If it were true, a simple Morningstar or similar database search would guarantee all of us consistent superior performance. The inescapable fact is that few pensions match the performance of a simple index strategy of 60% S&P 500 and 40% Lehman Brothers Long Bond. During the period from 1987 to 1996, only 10 of the 145 largest pension plans in the U.S. exceeded this basic strategy.

The failure of active management to add value is undisputable. The further failure to add value through manager selection based on past performance is also undisputable. Yet, AG Becker, Russell, SEI and a host of others persist in beating this dead horse. Worse yet, pensions continue to support the failed "manager search" system - voodoo finance at its worst.

The system is bad enough where true independent consultants are employed. But, it takes a bizarre turn for the worse when employees of a financial institution masquerade as objective consultants. Then the process looses any objectivity at all, and the recommendations are restricted to a small group of second class managers willing to "pay to play," or rebate some of their management fees for participation in the system.

Simply put, investment returns are not generated by the skill of managers, the size of the institution, the brand name of the sponsor, or the diligence of the manager search. None of that matters. They certainly cannot be predicted from past returns.

Returns are best explained by the systematic exposure to risk factors of market, size, and value. Today, advanced economic theory and available modeling tools allow institutions to reliably, effectively, and economically capture these elements through portfolio engineering using passive strategies.

There are never guarantees of success in the investment business. Especially in the short term, returns may look little like projected or "expected" returns. But, chasing past performance is a clear losing strategy. Structuring a portfolio to capture persistent risk premiums on a global basis is a enormous improvement that offers the highest probability of success at the lowest cost and lowest possible risk point.

Frank Armstrong, CFP, is the author of Investment Strategies for the 21st Century as well as the investment guide The Informed Investor. He is the President of Investor Solutions, Inc. a fee-only Registered Investment Advisor.