It's heresy.

Stocks may not have an edge over bonds.

While you and I have lived our entire adult lives with the comfortable idea that stocks always beat bonds and that it takes the real return on stocks to stay ahead of inflation, some new research indicates otherwise.

I learned this recently while visiting with Lacy H. Hunt in Austin. If Mr. Hunt's name seems familiar, it should--- he and Van Hoisington have been the subject of two of my earlier columns for their contrary expectation of a continuing decline in interest rates.

Now Mr. Hunt, the economist at Hoisington Investment Management, has taken a step beyond bonds to examine the hottest issue of the last year--- the so-called "risk premium" that equity investors earn over bondholders. Co-written with David M. Hoisington, the winter issue of the Journal of Portfolio Management contains their article showing that stocks, well, just might not be all we thought they were.

This is not a unique idea.

In March, April, and May of last year I wrote four columns describing the developing challenge to the primacy of equity returns. (The URLs for the columns are listed below.) What's different here is how the historical dissection was performed. While most research starts with real, inflation adjusted returns, Mr. Hunt starts with nominal returns.

As he started to explain I asked how he happened to do the research.

"It really started with a client. A corporate client asked two important questions. Could the post-war returns on stocks and bonds be repeated? And what factors influenced the risk premium?"

Mr. Hunt declined to identify the client but my bet is that it's someone with a big pension plan, someone who is watching their pension liabilities eat their future. If the golden days of equities are over, they need to know about it.

The first thing that Hunt and Hoisington found is that we're all victims of our history. Examining different time periods, the two researchers found that the postwar period was the best period for owning common stocks--- the additional return over bonds was the largest, 6.8 percent. As the time period is lengthened the return on stocks declines but the return on bonds remains virtually unchanged. As a result, the risk premium falls to 4.3 percent.

Long, Long Term, The Equity Risk Premium Shrinks
The longer the historical time period considered, the smaller the advantage stocks have over bonds.
Time Period Equity Return (1) Bond Return (2) Risk Premium (1 minus 2)
Post War (1946-2001) 12.4% 5.6% 6.8%
Ibbotson   (1926-2001) 10.7 5.3 5.4
Authors     (1900-2001) 10.0 4.9 5.1
Schiller     (1871-2001)    9.3 5.0 4.3
Source: "Estimating the Stock/Bond Risk Premium," Hunt and Hoisington

Next they tested the idea that the risk premium was influenced by the rate of inflation during the period plus two starting conditions--- the relative yields of stocks over bonds and the starting price-to-earnings ratio for stocks.

They found that the best 10-year period for stocks was 1949 through 1959. Stocks returned 19.4 percent while bonds lost 0.1 percent, a risk premium of 19.5 percent. Stocks started the period with a yield of 6.8 percent and a P/E ratio of only 6.4 while bonds yielded only 2.3 percent. Inflation averaged 2.4 percent.

Note that none of these conditions are like the present. Today, stock yields are lower than Treasury yields, P/E ratios are much higher, and inflation is lower.

Not so with the 10-year period that was best for bonds, 1928-1938. Then, prices were declining at a 2.4 percent annual rate, stock yields were close to Treasury yields (4.1 percent versus 3.3 percent), and P/E ratios were relatively high at 15.9. Stocks had a negative risk premium. They returned 5.5 percent a year less than bonds.

Today we have similar conditions.   Stock yields are lower than Treasury yields, P/E ratios are higher, and many are talking about impending deflation.

"In high inflation periods, corporations have pricing power and T-bonds deflate. Equities benefit, bonds are hurt," Mr. Hunt commented. "But in low inflation periods, corporations don't have pricing power.

"Now think about another thing. In most periods, stock dividend yields were higher than bond yields. They were in 1871, 1900, 1926, and 1946. But today, stock yields are lower than bond yields. In 2001 stocks yielded 1.5 percent while bonds yielded 5.0 percent, a 3.5 percent difference. That alone is almost the entire risk premium--- that 4.3 percent figure. Worse, stocks are now selling at the highest P/E ratios in history, 34 in 2001."

His prognosis?

If history is an indication, stocks won't be providing the returns most investors have grown to expect as "normal."

Earlier interview with Lacy Hunt on interest rates

March 5, 2002: Trevino and Robertson on P/E ratios and future returns

April 7, 2002: Ibbotson, SBBI 2001 used to compare 1962 to 2002 retrospectively

May 19, 2002: Arnott and Bernstein on the Risk Premium for equities

May 28, 2002: Asness on "Bubble Logic" and future stock returns