Dynamic Networks and Asset Pricing

Abstract:     
In the context of an equilibrium model with multiple risky assets, we map the characteristics of the network connecting firms' fundamentals to the cross-section of expected returns. We interpret network connectivity as the ability to transfer a distress state to other firms' fundamentals in a directed and timely manner. We show that 'central' firms, active at transferring but relative immune to distress, have lower P/D ratios and higher expected returns. We use corporate earnings to take the model to the data and estimate the network structure. In accordance with theoretical predictions, we find evidence of a positive centrality price of risk and a sizable centrality risk premium. Furthermore, network centrality helps to motivate the value premium as a distress causality risk premium: part of the expected return of value stocks in excess of growth stocks is a centrality premium in our results, and value stocks severely under-perform during economic downturns.

Number of Pages in PDF File: 33

Keywords: Dynamic Networks, Cross-Section of Expected Returns, Lucas Orchard

JEL Classification: G12, G14

working papers series