ABSTRACT
Introduction In Chapter 7, a land-use shares model that is commonly estimated with aggregate data was discussed. To provide more details on this approach, this chapter presents an application of the model to land use in Wisconsin. The material presented here is based on the models in Plantinga et al. (1999) and Plantinga and Wu (2002). Related applications are found in Lichtenberg (1989), Parks and Murray (1994), Parks and Kramer (1995), Wu and Segerson (1995), Hardie and Parks (1997), and Ahn et al. (2000). The purpose of estimating land-use shares models is to quantify the relationship between the shares of land allocated to different uses and the hypothesized determinants of land use, such as the net return to a particular use. The estimation results indicate what land-use determinants are important in explaining land-use choices by individuals. Further, the results can be used to estimate how land use will change if the determinants of land use change. Model Specification The application presented in the next section employs county-level data for Wisconsin. In this case, the land-use share for a given use is defined as the per cent of total county area devoted to that use (e.g., the share of county land in agriculture). The observed land-use share for use k (k=1,…,K) in county i (i=1,…,I) at time t (t=1,…,T) can be expressed as yikt=pikt+εikt, where pikt is the expected share of land allocated to use k, and εikt is a random error term with mean zero. The expected share, pikt, represents the optimal land allocation given the economic and other conditions prevailing in time t. However, since decisions about land use must be made prior to time t, or at least at the beginning of period t, the actual land allocation observed at time t, yikt, may differ from the optimal allocation due to random occurrences such as bad weather or unanticipated price changes. These random events are captured in the term εikt and are assumed to have a zero mean, implying that E[yikt]= pikt.
