We study the joint dynamics of stock and corporate bond returns using a structural credit risk model, where unlevered asset returns and volatility are driven by systematic risk factors. The model captures key time-series features of stock and bond volatilities, leverage, and credit spreads at the market, industry, and firm levels. It produces return forecasts that exceed Martin’s lower bound for equities and exhibit larger spikes in downturns than the average credit spreads for bonds. These forecasts significantly predict realized returns, outperforming benchmarks. We find that systematic variance risk commands a substantially larger premium in corporate bonds relative to equities.