What return should we expect from common stocks?

This is not an academic question.

The answer has visceral importance to all of us. It will shape our expectations. It will change where we invest. It will influence how much we save. And it will change living standards for those near retirement.

So what return should we expect?

If we rely on the conventional wisdom, embodied in "Stocks, Bonds, Bills, and Inflation", the Ibbotson Associates annual yearbook recounting of investment returns since 1926, the answer is reassuring. The long-term return on common stocks is 10.7 percent, a healthy premium over the long-term return on government bonds (5.3 percent) and inflation (3.1 percent).

We like to hear numbers like this, particularly after two dismal years of losses, because it tells us that our money will double in less than 7 years, quadruple in 14 years, and continue doubling into the distant future.   If we just relax, the long-term trend will resume.

Equally important, the conventional wisdom also tells us that equity investors always receive a "risk premium" of about 5 percent a year for choosing stocks over bonds.

In fact, the conventional wisdom is under attack.

In the March/April issue of the Financial Analysts Journal, Robert D. Arnott and Peter L. Bernstein examine two centuries of stock, bond, and economic data. (Yes, you read that right. Two centuries.) They conclude that our expectations have been distorted by the relatively brief 75 years of history in the Ibbotson Associates data. Our expectations have been even more dangerously warped by the experience of the 1975 to 1999 bull market, when stocks returned 17.2 percent, a gigantic 7.8 percent premium over long-term government bonds.

Taking a longer view, they come to a sharply different view of investment return and risk. First, they conclude that the normal risk premium for stocks--- the amount by which stock returns should exceed the return on bonds--- is 2.4 percent, or less. That's half the risk premium usually accorded common stocks.

They also conclude the risk premium isn't constant. Instead, it varies from different price levels. Based on current price levels, they conclude that the risk premium for common stocks is zero, or less.

"The current risk premium is approximately zero, and a sensible expectation for the future real return on both stocks and bonds is 2 to 4 percent, far lower than the actuarial assumptions on which most investors are basing their planning and spending," they conclude.

If Arnott and Bernstein are correct, here are some of the implications:

•           Withdrawal rates from retirement portfolios will need to be reduced                to reflect lower expected returns from common stocks. Failure to                reduce withdrawal rates will result in running out of money. Just                as many retirees of the 60's were threatened by the combination of                high inflation and relatively low stock returns in the 70's,                current retirees may be threatened by a decade of below average                equity returns.

•           While the Ibbotson data indicates good portfolio survival at                withdrawal rates just over 4 percent, the Arnott/Bernstein data                could lower withdrawal rates to the return on long term Treasury                Inflation Protected Securities, or about 3.4 percent.

•           Younger workers will need to save more to adjust for the lower                returns expected from common stocks. There will be some                compensation, however, in that they will be able to invest more in                bonds and suffer less risk.

•           Lower spending by the retired and greater saving by those working                may become part of a self-fulfilling prophecy of slower economic                growth.

•           The investment industry--- brokers, advisors, funds, etc--- is                heading for a deep shakeout.   It has grown fantastically on 25                years of high equity returns. One example: over 100,000 candidates                will take the examinations for Certified Financial Analyst in June.                That's five times as many as took the exams in 1995, only 7 years                ago.   We can expect incredible pressure on financial service                industry revenues as portfolios shift from equities to bonds.                Pressure on fees will be intense.