Working Paper: CEPR ID: DP10328
Authors: Murray Carlson; David A. Chapman; Ron Kaniel; Hong Yan
Abstract: We use a general equilibrium model as a laboratory for generating predictable excess returns and for assessing the properties of the estimated consumption/portfolio rules, under both the empirical and the true dynamics of excess returns. The advantage of this approach, relative to the existing literature, is that the equilibrium model delineates the precise nature of the risk/return trade-off within an optimizing setting that endogenizes return predictability. In the experiments that we consider, the estimation issues are so severe that simple unconditional consumption and portfolio rules actually outperform (in a utility cost sense) both simple and bias-corrected empirical estimates of conditionally optimal policies.
Keywords: Consumption; Equilibrium; Excess Returns; Hedging; Predictable
JEL Codes: E21; G11; G12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| excess returns are predictable (G17) | rational investors adjust their consumption and portfolio choices (G11) |
| biased point estimates of the data-generating process (C51) | incorrect predictions about investor behavior (G41) |
| habit-endowment ratio increases (D15) | optimal consumption share decreases (D15) |
| simple OLS estimates of the DGP (C51) | biased portfolio rules that overstate sensitivity to dividend yields (G11) |
| bias-correction strategy (C51) | improved estimates (C13) |
| improved estimates (C13) | large utility costs for simulated histories (L97) |
| unconditional policies outperform conditional rules (C52) | lower utility costs (L97) |