Changes in the Factor Loadings Across Sectors and Economic Dynamics : Evidence for the Japanese Firms

We estimate the equity premium for Japanese firms using the conditional version of Fama and French three-factor model where only the factor loadings are allowed to change unlike Liew and Vassalou (2000) wherein the factor premiums are allowed to change. Then, with considerations to the cyclical components of outputs with the RBC model we investigate the lead and lag relationships between the equity premium of the industry/sector and the measure of business cycles. We try to find the business-cycles related relationship between the cyclical components of the production index and/or GDP and the expected equity premiums for each industry among different sectors of the economy, notably the consumption goods sector and investment goods sector. The RBC model proposed by Boldrin, Christiano, and Fisher (1995, 2000) suggests that these differences exist. We also add a financial sector into this two sector model and then close the model with suitable equilibrium conditions. Within this model it is inferred that the financial sector will have also different business cycles implications. Furthermore, borrowing from the model proposed by Dias-Gomez et al. (1992), we will also be able to incorporate a banking sector, which would also possess different business cycle implications. The empirical relationship we find between the output fluctuations using a simple H-P filter at this first version of the paper and the equity premium changes for the various sectors of the economy shed a new light into the analysis of the business cycles and the equity premium changes, and hence the evolvement of the economy aggregate risk across different sectors of the economy