A Bayesian Analysis of the Stock Price Decomposition**

This paper employs the state-space model to reexamine the fundamental issue in finance about whether it is the expected returns or the expected dividends growth that is primarily responsible for the stock price variations. We use Bayesian methods to conduct inference and show that there is a substantial uncertainty about the contributions of expected returns and expected dividends to the fluctuation in the price-dividend ratio when using the aggregate returns and dividends data. The substantial uncertainty of the contributions results from the model being weakly identified, in the sense of Nelson and Startz (2007). Our finding challenges the long held notion in the existing literature that it is the expected returns that contribute most to the price-dividend variations and calls for further investigations using more disaggregated data.