Via the well-known financial leverage effect, decreases in stock prices cause an increase in the levered equity beta for a given unlevered equity beta. However, as growth options are more volatile and have higher risk than assets in place, a price decrease may decrease the unlevered equity beta via an “operating leverage” effect. This is because decreases in prices can be associated with a proportionately higher loss in growth options than in assets in place. Most of the existing literature focuses on the financial leverage effect. This paper examines both effects. Our empirical results show that, contrary to common belief, the operating leverage effect largely dominates the financial leverage effect, even for highly levered firms that presumably have few growth options. We link variations in betas to measurable firm characteristics that proxy for the proportion of the firm invested in growth options. We show that these proxies jointly predict a large fraction of cross-sectional differences in betas. These results have important implications on the predictability of equity betas, hence on empirical asset pricing and on portfolio optimization that controls for systematic risk.