Using Bayesian Markov chain Monte Carlo methods, we decompose the log price-dividend ratio into a market fundamentals component and a bubble component. The market fundamentals component depends on expectations of future dividend growth and required returns, while the bubble component is assumed to follow a Markov switching model that allows for the possibility of exploding and collapsing regimes. If prior beliefs allow for the possibility of persistent shocks to dividend growth and/or required returns, the posterior distribution suggests the bubble component contributes virtually nothing to the stock price movements over our sample. On the other hand, if one's priors rule out the possibility of persistent shocks to dividend growth and required returns, the bubble component can have a much larger role to play in stock price movements. However, the regime switching behavior of the bubble bears little resemblance to infrequent switching from an exploding bubble regime to a collapsing or dormant bubble regime.
