The empirical connection between financial leverage and equity risk premia is surprisingly weak. We propose a quantitative model linking limited financial flexibility to levered risk premia, where firms make dynamic investment, financing, and default decisions. The model highlights two key variables, leverage gaps and leverage targets, as drivers of risk premia. Firms partially close the gap toward their target, remaining optimally over- or under-levered. Equityholders of overlevered firms face higher debt costs, as their capital structure becomes more exposed to bankruptcy risk. As a result, leverage gaps amplify asset returns. The "lost" leverage risk premium reappears after controlling for targets.

The job search decisions of unemployed workers are forward-looking and shaped by the returns they anticipate from the search process. When expected returns, or discount rates, are high, the discounted benefits from the search process are low. Thus, unemployed workers engage in less intensive job searching. To formalize this discounting channel for explaining the job search behavior of unemployed workers, we build a Diamond-Mortensen-Pissarides search model with variable search intensity and Epstein-Zin preferences. We demonstrate that the job search return for unemployed workers is theoretically equivalent to firms' stock return, as determined by the Nash bargaining process. We calibrate the model and show that variable search intensity quantitatively amplifies both labor market volatilities and stock market risks, relative to fixed search intensity. Consistent with the discounting channel, we show that search intensity negatively predicts stock market returns in the model, aligned with the data.

Firms' payout decisions respond to expected returns: everything else equal, firms invest less and pay out more when their cost of capital increases. Given investors' demand for firm payout, market clearing implies that productivity and payout demand dynamics fully determine equilibrium asset prices and returns. Using this logic, we propose a payout-based asset pricing framework. To operationalize it, we introduce a quantitative model, calibrating the productivity and payout demand processes to match aggregate U.S. corporate output and payout moments. Model-implied payout yields and firm returns match key empirical moments, and model-implied expected returns predict future firm returns in the data.