I analyze the sources of U.S. business cycle fluctuations in an estimated Dynamic Stochastic General Equilibrium model with a rich set of nominal and real rigidities and various exogenous disturbances. The model includes a shock to the expected risk-premium, which introduces a time-varying wedge between the policy rate set by the central bank and the cost-of-capital of firms. In the aggregate data, most U.S. corporations finance their investment using internal funds, and stock prices reveal the opportunity cost of this type of financing. I therefore use corporate market value and dividend data in the Bayesian estimation of the model to identify risk shocks. Variance decomposition exercises show that these shocks account for a substantial part of the variation in the stock market, as well as the variation in output and investment, especially at short forecast horizons. The variation of these variables at longer forecast horizons are mainly captured by shocks to investment-specific technological change. Historical decomposition points to the important role played by risk shocks in the run up of stock prices and output in the late 90s, and in the reversal of these variables in the early 2000s and during the recent recession. (JEL E32, E44)