The Markov chain model (MCM) has become a popular tool in the agricultural economics literature to study the impact of various drivers on the structural change of farms, including public support. In order to relax the process-homogeneity assumption underlying the MCM, we consider a mixture of two types of agents, the ‘stayers’ who always remain in their initial size category, and the ‘movers’ who follow a first-order Markovian process. An empirical application to a panel of commercial French farms over 2000-2012 shows that the mover-stayer model (MSM) is a better modeling framework to recover the underlying transition probability matrix.