The relation between excess returns and the dividend-price ratio is known to be instable. However, there is consensus on the type of instability. We investigate the consequences of different types of break processes on the long-term investor. The break process is estimated with a mixture innovation model using Bayesian methodology. The advantages of this model are that we can estimate the break process, and allow for different break processes for the coefficients and the variances. The estimated parameters show quite some instability, though less than if we were to assume a time-varying parameter model. We show that assuming stable parameters can lead to enormous losses for the long-term investor, even if in reality the break probability is small.