(The following piece appeared in the November 26, 2001 issue of Barron's and is reprinted with the kind permission of editor Thomas G. Donlan and Dow Jones, Inc. ---WB)

Other Voices

Riding for a Fall

The 401(k) is likely to turn out to be a defined-chaos
retirement plan

By William Bernstein

The past 12 months or so brought a revelation to many 401(k) participants: Stocks can actually lose money. This knowledge throws into sharp relief the serious flaws in the burgeoning defined-contribution retirement system. The average 401(k) account held just $41,919 -- woefully inadequate to meet the needs of the typical retiree, even allowing for further contributions and investment growth.

Some commentators, including Barron's Editorial Page Editor Thomas G. Donlan, have praised a bill from Rep. John Boehner, an Ohio Republican, that would allow plan administrators to contract with finance professionals to provide advice to participants running their 401(k) and similar accounts.

Too little, too late, and too expensive: Although the current system seems robust, the exodus from traditional defined-benefit plans to the employee-managed defined-contribution paradigm is a social and economic time bomb primed to explode sometime within the next few decades. Here are some of the reasons:

  • Employees are not saving enough. A worker who earns a constant real salary from age 20 to 65 and saves 10% of it requires a 4% real investment return to sustain a 20-year retirement at the same inflation-adjusted salary level. But most younger workers have relatively low incomes and no savings at all; starting later, at age 30 or 40, raises the required real investment return to 6% to 8%.

  • Future returns will not be nearly this high. The long-term price increase of stocks must track that of earnings and dividends. Over the past century, this has been 2% per year adjusted for inflation. New Paradigmistas point out that reinvested earnings, stock buybacks, and technology-driven improvements in productivity will result in increased earnings growth. But from 1950 to 1975, annualized real per-share earnings growth was 2.2%; from 1975 to 2000, it was 1.9%. Add a 1.45% dividend yield and you get an expected real stock return of just over 3%. The 7% real stock returns seen in the 20th century resulted from a combination of this 2% real earnings growth and dividends averaging 5%; anyone forecasting the same returns going forward should wear a sign that says, "I Can't Add!"

  • The average long-term investor will receive the market return minus plan expenses. The typical 401(k) plan is an absurdly expensive vehicle with fees approaching 3%, according to benefits consultant Brooks Hamilton. Add commissions and other costs from frenetic trading at the funds. The typical fund company services participants in the same way that Baby Face Nelson serviced banks.

  • Poor allocation decisions further degrade performance. At one major company surveyed by Bart Waring of Barclay's Global, almost half the participants owned only one or two funds, incurring unnecessary risk. Worse, many companies encourage purchase of company stock in their retirement plans, exposing employees to the double jeopardy of losing both paycheck and nest egg if the company fails.

  • Watson Wyatt Worldwide examined 252 large companies with both defined-benefit and 401(k) plans for the 1990-1995 period. It found that the defined-benefit plans bested the 401(k) plans by 2.4% per year. (The defined-benefit plans were no great shakes; from 1987 to 1999, the nation's largest pension plans underperformed a 70/30 benchmark of global stocks and bonds by an average of almost 2% per year.)

    Even scarier are the results in the 401(k) plans of the most prestigious financial services corporations: For 1995-1998, the annualized returns of the 401(k) plans at Morningstar, Prudential, and Hewitt Associates were 13.5%, 10.5%, and 11.8%, respectively, versus a 21.2% return for the global 70/30 mix. If employees at the nation's most sophisticated financial companies can't get it right, what chance do folks on the assembly line at Ford have?

    Given low equity returns, high expenses, and poor planning, it is likely that most 401(k) investors will obtain near-zero real returns in the coming decades. Further, a substantial minority will have disastrous results. Only a lucky few will save enough and obtain the 4% to 8% real returns necessary for a comfortable retirement (that is, aside from their bosses, who were smart enough to retain their traditional defined-benefit plans). When the boomers retire between 2010 and 2030, most will find the cupboard bare. The inevitable government bailout will make the savings and loan resolution of the last decade look like lunch at Taco Bell.

    Meanwhile, a real dogfight has erupted between investment advisors and the brokerage industry over the Boehner bill. Little wonder: In the July 23 edition of Investment News, industry sources estimated the size of this market at $18 billion per year. Calculated against the total defined-contribution $3 trillion asset base, that's another 0.6% of annual return flying out of the pockets of employees and into the industry's coffers. The notion that several hours of canned questionnaires and PowerPoint presentations will turn the average corporate employee into a well-informed, disciplined investment manager is absurd to anyone with the remotest sense of financial history and human nature.

    We've seen this movie before, and it doesn't end well. In the 1920s, millions read Edgar Lawrence Smith's Common Stocks as Long Term Investments and convinced themselves that they would never sell their stocks. Unfortunately, they did. The investment bestseller of the next decade was Lawrence Chamberlain's Investment and Speculation, which flatly stated that only bonds should be purchased for investment purposes. The late 1960s saw a renaissance of popular enthusiasm for common share ownership, with more than one-third of households owning stocks. Real returns over the next two decades were nearly zero, and by 1979, when BusinessWeek famously proclaimed "The Death of Equities," just 16% of households held stocks. The average participant is a long-term investor only the way that Tony Soprano is a Catholic.

    The defined-contribution system is broken; we just haven't realized it yet. While it may be possible to fix it by providing participants with much closer attention to expenses -- the exact opposite of what the Boehner bill accomplishes -- it makes more sense to improve the defined-benefit system with equitable vesting, portability and liability protection. These plans operate at a much higher level of efficiency and competence, with correspondingly higher and more uniform returns, than the ever-growing hodgepodge of expensive employee-run accounts.

    Such a paternalistic approach may offend those who make a philosophy out of self-reliance. But most people don't build their own cars or remove their neighbors' kidney stones. We should treat retirement investing the same way. If an employee wants to manage his own retirement account, he should at least be able to show competence in the basic principles of investing, such as the differences between stocks and bonds, the fundamentals of prudent diversification, and the impact of expenses on returns.

    The self-managed defined-contribution concept is fatally flawed. While Wall Street pros may (or may not) be getting it right, the overwhelming majority of employees are floundering, bewildered by a subject they only dimly comprehend. The time has come to throw workers a lifeline, in the form of meaningful pension reform, before we all drown.

    To Efficient Frontier Homepage E-mail to William Bernstein

    Copyright © 2001, Dow Jones, Inc. All rights reserved.