Contagious development: Neighbor interactions in deforestation Juan A. Robalino a, ⁎, Alexander Pfaff b a EfD Initiative-CATIE and Universidad de Costa Rica, Costa Rica b Duke University, United States abstract article info Article history: Received 29 November 2005 Revised 21 May 2011 Accepted 6 June 2011 Available online xxxx JEL classification: O13 R12 Q24 Keywords: Instrumental variable Neighbors' interactions Local interactions Social interactions Land use patterns Environmental policies We estimate neighbor interactions in deforestation in Costa Rica. To address simultaneity and the presence of spatially correlated unobservables, we measure for neighbors' deforestation using the slopes of neighbors' and neighbors' neighbors' parcels. We find that neighboring deforestation significantly raises the probability of deforestation. Policies for agricultural development or forest conservation in one area will affect deforestation rates in non-targeted neighboring areas. Correct estimation of the interaction reverses the naive estimate's prediction of multiple equilibria. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Rural areas of developing countries contain most of the world's tropical forest. The demand for the local, regional and global services that forests provide, alongside the poverty in these areas, indicates the need for policies that balance development with forest conservation. The provision of forest services depends upon both the extent and the spatial distribution of standing forest. One important determinant of both the rate and spatial pattern of tropical deforestation is interdependency among deforestation decisions. In general, interdependencies affect the efficiency and optimality of the outcomes of individual choice (Brock and Durlauf, 2001; Cooper and Johns, 1988, and Moffitt, 2001). Exogenous shifts in the determinants of individual choices can have expansionary effects if interdependencies exist (Durlauf, 2001, and Moffitt, 2001). How such spillovers propagate within a population or across space needs to be considered in policy design (Durlauf, 2001). We find significant interdependencies among individuals' deforestation decisions in a developing country. Measuring such interactions is difficult (Bayer and Timmins, 2003; Brock and Durlauf, 2001; Conley and Topa, 2002; Glaeser and Scheinkman, 2001; Manski, 1993, and Moffitt, 2001). One challenge for identifying whether individuals are influenced by their neighbors to take the same action, for instance, is that there are other reasons neighbors behave similarly. Neighbors can have similar unobservable characteristics and be affected by the same unobservable influences (Manski, 1993). Another challenge is that neighbors simultaneously affect each other when interactions exist (Manski, 1993 and Moffitt, 2001). To address these challenges we apply the instrumental variable approach (Moffitt, 2001), using exogenously varying topological instruments: the slopes of neighbors' and neighbors' neighbors' parcels. The literature uses instrumental variables and other approaches to identify interactions in a range of settings such as: education (Crane, 1991; Evans et al., 1992, and Gaviria and Raphael, 2001); employment (Conley and Topa, 2002, and Topa, 2001); crime (Bayer et al., 2004, and Glaeser et al., 1996); and migration (Munshi, 2003). Land-use research has focused on spatial externalities in residential development (Irwin and Bockstael, 2002) and on information networks that may affect technology adoption in agriculture (Case, 1992, and Conley and Udry, 2001). We extend this approach to the study of development and tropical deforestation. Deforestation decisions depend on expected profits that can be affected by neighborhood deforestation for a number of reasons. Local prices may fall if others deforest for agricultural production, reducing Journal of Development Economics xxx (2011) xxx–xxx ⁎ Corresponding author at: CATIE 7170, Turrialba, Costa Rica. Tel.: + 506 2558 2522. E-mail address: robalino@catie.ac.cr (J.A. Robalino). DEVEC-01631; No of Pages 10 0304-3878/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jdeveco.2011.06.003 Contents lists available at ScienceDirect Journal of Development Economics journal homepage: www.elsevier.com/locate/devec Please cite this article as: Robalino, J.A., Pfaff, A., Contagious development: Neighbor interactions in deforestation, J. Dev. Econ. (2011), doi:10.1016/j.jdeveco.2011.06.003