Empirical models of land conservation and land use

This dissertation concentrates on land use, both urban and rural, in United States. First, I focus on land conservation and study how state fiscal incentives can affect private land conservation on the eastern region of the country. Then, I move to urban uses of land and analyze how distance to certain amenities in a city can affect residential and non-residential uses, using spatial econometric techniques. Finally, taking advantage of the same methodology, I concentrate on conservation easements and estimate the probability of different levels of public access and how certain variables affect these probabilities. In the first chapter, I study the effect that state level incentives have on land conservation. Private decisions about land conservation are crucial for preservation of endangered species as 80% of their habitats are on private land. I study the efficacy of state tax breaks to promote private land conservation. I use the Protected Area Dataset of United States and construct a county-year level panel of the flow of undeveloped land protected per year. I use fixed effects panel estimations combined with optimal full matching to improve balance on observable covariates between treated and control counties. Results show that, on average, counties in a state with a tax break more than double the yearly flow of conservation after the incentive is in place. These findings suggest that state tax breaks are an effective incentive to promote land conservation. In the second chapter, I concentrate on land use in an urban area and how spatial econometric techniques can help explain land use decisions. I use a large geo-referenced data set and estimate the probability of residential use for individual lots in the urban area. I specifically concentrate on the difficulties that this type of data sets presents and how to overcome them. Spatial data sets pose challenges for discrete choice models because the data are unlikely to be independently and identically distributed. A conditionally parametric spatial probit model is amenable to very large data sets while imposing far less structure on the data than conventional parametric models. I illustrate the approach using data on 474,170 individual lots in the City of Chicago. The results suggest that simple functional forms are not appropriate for explaining the spatial variation in residential land use across the entire city. In the third chapter, I focus only on conservation easements in continental US. Conservation easements generally have a specified level of public access: open, restricted, or closed to the public. This chapter focuses on these particular levels of public access and how location and other variables play a role in that decision. A conditionally parametric multinomial logit model estimates how these variables affect the probability of each level of access. By allowing coefficients to vary throughout space I find effects not capture by standard or spatial logit models. Results show that effects not only differ by level of public access but also spatially. These findings provide useful information for shaping regional policies that are able to address these differences when promoting conservation for specific purposes that allow different levels of public access