The impact of fat tails on equilibrium rates of return and term premia

We investigate the impact of ignoring fat tails observed in the empirical distributions of macroeconomic time series on the equilibrium implications of the consumption-based asset-pricing model with habit formation. Fat tails in the empirical distributions of consumption growth rates are modeled as a dampened power law process that nevertheless guarantees finiteness of moments of all orders. This renders model-implied mean equilibrium rates of return and equity and term premia finite. Comparison with a benchmark Gaussian process reveals that accounting for fat tails lowers the model-implied mean risk-free rate by 20 percent, raises the mean equity premium by 80 percent and the term premium by 20 percent, bringing the model implications closer to their empirically observed counterparts. Fat tails also increase the model-implied volatility of the risk-free rate and the equity premium.