Tracing Tails – Large Idiosyncratic Income Shocks in a Heterogeneous Agent Asset Pricing Model

Abstract

I present evidence that large individual income changes can help explaining the size and value premium in a cross-section of portfolio returns. I develop a tail risk measure, the Idiosyncratic Income Risk Factor and estimate it on US income data. The results show that the extreme income shocks vary with the state of the economy. In an augmented Consumption-Based Asset Pricing Model, the Idiosyncratic Income Risk Factor emerges as a priced factor when explaining a cross-section of returns of 25 portfolios sorted by size and book-to-market ratio. The findings support Krebs’ (2004) critique of the Constantinides and Duffie (1996) idiosyncratic risk asset pricing model: central moments of the cross-sectional distribution of income cannot be used to test the implications of the Constantinides and Duffie (1996) model. I present evidence that supports the notion of a fat-tail-generating personal disaster process affecting asset prices.