This paper by John Y. Campbell addresses the fundamental questions of how risk should be measured and what economic forces determine the price of risk. It critiques existing models, such as the Capital Asset Pricing Model (CAPM), which measure risk by the covariance of an asset's return with the market return, and argues that these models do not adequately address the issues. Campbell proposes a new model that uses an equilibrium multifactor framework to interpret postwar U.S. stock and bond returns. The model includes factors such as the return on a stock index, revisions in forecasts of future stock returns, and revisions in forecasts of future labor income growth. The main factor determining excess returns is aggregate stock market risk, but the coefficient of relative risk aversion is much higher than the price of stock market risk when human capital or stock market mean reversion is present. The paper develops a simple discrete-time asset pricing model that allows for changing investment opportunities and human capital, which may not be well proxied by a stock index return. The model links time-series and cross-section data, providing a coherent framework for understanding asset returns. It suggests that empirical researchers should identify priced factors by examining the time-series behavior of stock returns and labor income rather than through factor analysis or selection of macroeconomic variables. The model is tested using historical data, and the results show that the cross-sectional variation in returns is primarily explained by the covariances with the market return and news about future returns. The paper also discusses the implications for asset pricing and consumption, emphasizing the importance of human capital in determining the response of consumption to interest rate changes.This paper by John Y. Campbell addresses the fundamental questions of how risk should be measured and what economic forces determine the price of risk. It critiques existing models, such as the Capital Asset Pricing Model (CAPM), which measure risk by the covariance of an asset's return with the market return, and argues that these models do not adequately address the issues. Campbell proposes a new model that uses an equilibrium multifactor framework to interpret postwar U.S. stock and bond returns. The model includes factors such as the return on a stock index, revisions in forecasts of future stock returns, and revisions in forecasts of future labor income growth. The main factor determining excess returns is aggregate stock market risk, but the coefficient of relative risk aversion is much higher than the price of stock market risk when human capital or stock market mean reversion is present. The paper develops a simple discrete-time asset pricing model that allows for changing investment opportunities and human capital, which may not be well proxied by a stock index return. The model links time-series and cross-section data, providing a coherent framework for understanding asset returns. It suggests that empirical researchers should identify priced factors by examining the time-series behavior of stock returns and labor income rather than through factor analysis or selection of macroeconomic variables. The model is tested using historical data, and the results show that the cross-sectional variation in returns is primarily explained by the covariances with the market return and news about future returns. The paper also discusses the implications for asset pricing and consumption, emphasizing the importance of human capital in determining the response of consumption to interest rate changes.