Expected Stock Returns Worldwide: A Log-Linear Present-Value Approach

by Akash Chattopadhyay, Matthew R. Lyle, and Charles C.Y. Wang

Overview — Over the last 20 years, shortcomings of classical asset-pricing models have motivated research in developing alternative methods for measuring ex ante expected stock returns. This study evaluates the main paradigms for deriving firm-level expected return proxies (ERPs) and proposes a new framework for estimating them.

Author Abstract

Expected return proxies (ERP) derived from a log-linear present-value (LPV) framework—combining the book value of equity, profitability, and market prices—predict the cross section of out-of-sample returns in 26 of 29 international equity markets, with a highly significant average slope coefficient of close to 1. In contrast, ERPs based on the implied cost of equity or standard factor models—even those with factors based on book-to-market and profitability—fail to exhibit systematic predictive power internationally. LPV models derived using common valuation anchors such as earnings or sales also exhibit predictive ability. LPV ERPs based on the book value of equity and sales subsume the predictive ability of all other ERPs we examine. Collectively, the LPV framework offers a parsimonious accounting-based framework for the estimation of expected returns across international markets.