No `accounting trick,' just a long-term view of plan
Responding to "Companies under fire for incentive calculations" (April 15, page 4):
In your article, the headline refers to an "accounting trick" regarding pension income. The only "trick" to understanding pension accounting, however, is in remembering to look at the bigger picture. Pension plans are very long-term investment structures that require similarly long-term valuation processes. They also almost always are exposed to the public equity markets, which of course go both up and down on a short-term basis. However, over the time frame used to fund a pension benefit, these markets have almost always gone up - thus, the corresponding use of longer-term data in pension accounting. While your article notes a short-term, market-related loss for the Verizon Communications Inc. plan in 2001, it fails to point out that Verizon and other U.S. corporate pension funds have also had large market-related gains over the longer-term period. If, as the article argues, corporate earnings (and executive compensation) were overstated in 2001, then these same results have been understated for many years prior. Since the point of a well-run pension system is to realize long-term gains, then it is not out of line for pension accounting standards to take a long-term view of plan results.
T. Britton Harris IV
Verizon Investment Management Corp.
Reality and MPT
Kudos to Joel Chernoff for his excellent articles on the history, evolution and current thinking regarding Modern Portfolio Theory (Pensions & Investments, April 29).
The contributions of Harry Markowitz, Bill Sharpe, James Tobin and others have been well documented over the years, as have the difficulties associated with "normative" theories about market and portfolio behavior that are far removed from observed reality.
In contrast, "descriptive" theories are based on observable characteristics of real markets and have proven invaluable in solving many of the difficulties associated with the application of MPT. This approach is the basis for the recent monograph, "Stochastic Portfolio Theory," by Robert Fernholz of INTECH (Springer-Verlag, New York, 2002) wherein is presented an illuminating hypothesis about portfolio behavior that is applicable over a wide range of assumptions. Since prehistory, the natural sciences have been descriptive, and it is doubtful we would have progressed to fire or the wheel had they been normative.
One reason Mr. Markowitz's theory continues to be valuable after 50 years is that it is not dependent on normative assumptions about the behavior of market participants. As he stated in Mr. Chernoff's article, "There is no assumption `that anyone else is acting rationally."'
However, market efficiency is a different matter. The argument for market efficiency, at least in the form that the market is "unbeatable," rests on Mr. Sharpe's capital asset pricing theory, and it has long been understood that the normative assumptions of this model are nowhere near satisfied in reality. Mr. Sharpe argues the market is "almost" efficient, but isn't that about the same as "almost" winning a baseball game? Convincing mathematical evidence that the market is, in fact, not efficient is presented in Mr. Fernholz's monograph.
For more than 14 years, INTECH has been providing above-market returns with below-market risk using newer forms of mathematical finance. Perhaps it is time for investment philosophy to move on to more realistic models that are not based on assumptions that we all know are false.
senior vice president
Palm Beach Gardens, Fla.
Letters and other submissions may be sent by mail to the attention of Barry B. Burr, Pensions & Investments, 360 N. Michigan Ave., Chicago, IL 60601; by fax, to (312) 649-5228; or by e-mail to [email protected]