The Equity Premium: A Puzzle
Source: Mehra, R. & Prescott, E. C. (1985). Journal of Monetary Economics 15(2), 145–161.
TL;DR
Documents the equity-premium puzzle: the historical premium of U.S. stocks over risk-free bonds
(around 6% per year) is far larger than a standard consumption-based asset-pricing model can explain
with any plausible level of risk aversion. Matching the data requires implausibly high risk aversion —
a puzzle that has shaped macro-finance ever since.
What it documents
That the canonical model — a representative agent with time-separable power utility consuming aggregate
consumption — cannot simultaneously fit the high equity premium and the **low, stable risk-free
rate**, because aggregate consumption is too smooth to make stocks seem risky enough.
The argument
consumption growth.
order of magnitude larger than micro evidence supports — and that, in turn, implies a counterfactually
high risk-free rate (the related "risk-free-rate puzzle").
Why it matters
A defining puzzle of asset pricing that motivated the major resolutions covered elsewhere: **habit
formation (Campbell-Cochrane), long-run risk (Bansal-Yaron), and rare disasters** (Rietz,
Barro), as well as behavioral and prospect-theory explanations.
Limitations and risks
each resolution relaxes one of them.