Macroeconomic factors and long-term risk premia in arbitrage asset pricing based on models with heterogeneous agents and market incompleteness

Abstract

This paper analyzes the impact of macroeconomic factors on long-term risk premia within an arbitrage-based asset-pricing framework built on a dynamic stochastic general equilibrium model with heterogeneous agents and incomplete markets. The theoretical setup is a modified Aiyagari–Bewley–Huggett economy with two types of households differing in patience, facing uninsurable idiosyncratic income shocks and borrowing constraints. The aggregate economy is subject to stochastic shocks to total factor productivity and monetary policy, both featuring time-varying volatility. Model parameters are calibrated via the simulated method of moments to a broad set of macroeconomic and financial series for G7 countries over 1985–2023. The calibrated model successfully reproduces key asset-pricing facts, generating an equity premium of 6.123% versus the historical benchmark of 6.495%, a low real risk-free rate and realistic return volatility. A structural decomposition shows that fundamental aggregate consumption risk accounts for less than one third of the total risk premium, while more than 67% is driven by mechanisms related to agent heterogeneity and market incompleteness, including precautionary savings and endogenous wealth concentration. Sensitivity analysis reveals sizable increases in 5- and 10-year risk premia in response to shocks to GDP-growth and inflation volatility, with a stronger reaction at longer horizons. Regression results further indicate that consumption volatility, inflation volatility and the endogenous wealth Gini coefficient are all significant positive determinants of the equity premium. These findings underscore the importance of distributional channels and macroeconomic uncertainty for long-term asset pricing and macro-prudential policy design.