Working Paper: NBER ID: w11026
Authors: Ravi Jagannathan; Yong Wang
Abstract: We demonstrate, using data for the period 1954-2003, that differences in exposure to consumption risk explains cross sectional differences in average excess returns (cost of equity capital) across the 25 benchmark equity portfolios constructed by Fama and French (1993). We use yearly returns on stocks to take into account well documented within year deterministic seasonal patterns in returns, measurement errors in the consumption data, and possible slow adjustment of consumption to changes in wealth due to habit and prior commitments. Consumption during the fourth quarter is likely to have a larger discretionary component. Further, given the availability of more leisure time during the holiday season and the ending of the tax year in December, investors are more likely to review their asset holdings and make trading decisions during the fourth quarter. We therefore match the growth rate in the fourth quarter consumption from one year to the next with the corresponding calendar year return when computing the latter's exposure to consumption risk. We find strong support for our consumption risk model specification in the data.
Keywords: No keywords provided
JEL Codes: G12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Differences in exposure to consumption risk (D11) | Variation in average excess returns across firm types (L25) |
| Consumption risk exposure (D11) | Expected returns (G17) |
| Higher consumption risk exposure (E21) | Higher expected returns (G19) |
| Fourth quarter consumption data (Y10) | Stronger empirical support for model (C52) |
| Consumption-based model specification (D12) | Explanation of cross-section of stock returns (G17) |
| Incorporating consumption risk (D15) | Explains cross-section of stock returns (G17) |