Understanding the behavior of asset prices and their relation to macroeconomic risks can be considered as one of the most fundamental issues in finance. As is well known, however, the traditional workhorse for studying the link between financial markets and the real economy the consumption-based asset pricing model (CCAPM) has failed to explain a number of stylized facts in finance such as the equity premium (Mehra and Prescott, 1985), asset return volatility (Grossman and Shiller, 1982) or value and size premia in the cross-section of expected returns (Cochrane, 1996; Lettau and Ludvigson, 2001). After a long series of poor empirical results starting with Hansen and Singleton (1982, 1983), more recent studies exploring the basic insights of the consumption-based asset pricing paradigm report encouraging advances (Cochrane, 2007, p.267).
In particular, a recent contribution by Parker and Julliard (2005) suggests to relate asset returns to consumption growth measured over longer horizons within a simple consumption based framework with CRRA preferences. Such reasoning is in line with theoretical literature on long-run consumption risk. Seminal work by Bansal and Yaron (2004) suggests that equilibrium asset returns depend on investors’ expectations about both short and long-run changes in consumption growth. Among other things, this result implies that the covariance of returns with contemporaneous consumption growth may understate the risk perceived by investors. By explicitly accounting for consumption risk over longer horizons, Parker and Julliard’s long-horizon (LH) CCAPM is able to explain a large fraction of cross-sectional variation in expected returns across US size and book-to-market sorted portfolios.
In this paper, we provide new detailed evidence as to whether long-run consumption risk helps explain the cross-section of expected returns in international stock markets. In par-ticular, we modify Parker and Julliard’s empirical approach along two lines. First, we take into account recent criticism about the widespread use of size and book-to-market sorted portfolios in the empirical asset pricing literature (Phalippou, 2007; Lewellen, Nagel, and Shanken, 2007). In order to reduce the adverse effects of strong commonalities in size and book-to-market sorted portfolios, we follow the prescription of Lewellen, Nagel, and Shanken (2007) to include industry portfolios alongside with the conventionally used size and book-to-market portfolios. Second, we provide new international evidence by investigating the model’s explanatory power for the cross-section of equity returns in the United Kingdom and Germany.
Our empirical findings shed new light on the relative merits of the long-horizon CCAPM when it comes to explaining the cross-section of returns in international stock markets. First, we find that under our modified empirical approach accounting for the strong common factor structure in size and book-to-market sorted portfolios, the model’s ability to account for cross-sectional variation in returns is clearly limited. This result suggests that the good empirical performance on US test assets reported by Parker and Julliard (2005) may be somewhat overstated. Tests with size and book-to-market sorted portfolios from the UK and Germany further corroborate the US evidence. Second, we find that measuring consumption risk over longer horizons typically yields lower risk-aversion estimates. Thus, our results suggest that more plausible parameter estimates – as opposed to a higher cross-sectional R2 – can be viewed as the main achievement of the long-horizon consumption-based approach.
Download
Long-Horizon Consumption Risk and the Cross-Section of Returns: New Tests and International Evidence
