Composition of Wealth, Conditioning Information, and the Cross-Section of Stock Returns

Abstract:     
I test conditional implications of linear asset pricing models in which variables reflecting time-varying composition of total wealth drive changes in the investment opportunity set. Conditional models predict that high average return assets, such as value stocks, covary with the risk factors more when factor risk premia are high than do growth stocks, and vice versa. They also imply that conditional expected returns on value stocks are especially high in these states of the world. I estimate conditional moments of returns and factor risk prices nonparametrically and I show while the former prediction is supported by the data, the latter one is not. Thus, exploiting conditioning information to impose joint restrictions on the dynamics of moments of returns and risk factors exposes an additional challenge for the standard asset pricing models.

Keywords: Capital Asset Pricing Model, conditioning information, human capital, size, value

JEL Classifications: G120, G100, C140

Working Paper Series