FINANCIAL ANALYSTS JOURNAL"
Perspectives on the Equity Risk Premium
Jeremy J. Siegel
The equity risk premium, or the difference between the expected
returns on stocks and on risk-free assets, has commanded the atten-
tion of both professional economists and investment practitioners for
many decades. In the past 20 years, more than 320 articles, enough to
fill some 40 economics and finance journals, have been published
with the words "equity premium" in the title.
The intense interest in the magnitude of the premium is not
surprising. The difference between the return on stocks and the
return on bonds is critical not only for asset allocation but also for
wealth projections for individual investors, foundations, and endow-
ments. One of the most asked questions by investors is: How much
more can I expect to earn from shifting from bonds to stocks?
Academic interest in the equity premium surged after Mehra and
Prescott published a seminal article in 1985 titled "The Equity Pre-
mium: A Puzzle." By examining the behavior of the stock market and
aggregate consumption, they showed that the equity risk premium,
under the usual assumptions about investor behavior toward risk,
should be much lower than had been calculated from the historical
data. Indeed, Mehra and Prescott stated that the equity premium in
the U.S. markets should be, at most, 0.35 percent instead of the approx-
imately 6 percent premium computed from data going back to 1872.
The Mehra-Prescott research raised the following question:
Have investors been demanding—and receiving—"too high" a
return for holding stocks based on the fundamental uncertainty in
the economy, or are the models that economists use to describe
investor behavior fundamentally flawed? If the returns have been too
high, then analysts can justify increased asset allocation to equities
and reduced allocation to bonds; if the models are flawed, economists
need to develop new models to describe investor behavior.
My discussion of the equity risk premium will be divided into
three parts: (1) a summary of the data used to calculate the equity
premium and discussion of potential biases in the historical data, (2)
analysis of the economic models, and (3) discussion of the implica-
tions of the findings for investors and for forecasts of the future
Historical Returns on Stocks and Bonds
In this section, I present historical asset returns since 1802, define the
equity premium, and discuss biases in the historical data that affect
future estimates of the equity premium.
The equity risk
wealth, and the cost
of capital, butwe do
not have a simple
model that explains
]eremy J. Siegel is the Russell E. Palmer Professor of Finance at the Wharton