Communal Property Rights and Deforestation

Abstract

Almost a third of world's forest area is under communal management. In principle, this arrangement could lead to a "tragedy of the commons" and therefore more deforestation. But monitoring outsider's deforestation may be easier if the owner is a community rather than an individual. We study the effect of communal titling on deforestation in Colombia using a difference-in-discontinuities strategy that compares areas just outside and inside a title, before and after titling. We find that deforestation decreased in communal areas after titling. Interestingly, we find evidence of positive spillovers of reduced deforestation in nearby areas.

The double fence: Overlapping institutions and deforestation in the Colombian Amazon

“…Once again, Table 6 shows that the treatment was only effective in the large IRs, while there was no effect on the small IRs. A descriptive statistic of deforestation inside small and large IRs, might shed some light on whether this result goes in line with the theoretical predictions of Romero & Saavedra (2019). Table 7 indicates that deforestation was higher for all the periods in small IRs regardless of the definition used to split the sample.…”

“…Moreover, the magnitude of the effect is a little bit larger to the one found using the whole sample, which is consistent with having no effect for the small IRs. This result can be understood with the theoretical model of Romero & Saavedra (2019), where they predict a decrease in deforestation when forests services valuation is high and there are more forest owners. The intuition behind that is very simple:…”

“…Spatial regression discontinuity Designs (RDs) have been also used in the related literature (Bonilla & Higuera, 2019;Pérez-Valbuena et al, 2017;Romero & Saavedra, 2019); this method exploits a discontinuity in the treatment assignment. Nevertheless, there might be important spillovers on the border of the PAs (see for example Robalino et al (2015)).…”

“…In order to determine the impact of collective titling on deforestation in various South American countries, some papers have tried to deal with the selection bias by using fixed effects on the community and year level (Blackman et al, 2017), while others have tried to make the treatment and control groups comparable using PSM (Blackman & Veit, 2018;Buntaine et al, 2015) and some others have used instrumental variables (Liscow, 2013). For the Colombian case, the temporal variation of collective titling has been exploited to use differences in difference (Vélez et al, 2019) and exogenous cutoffs have been used for spatial regression discontinuity designs (Bonilla & Higuera, 2019;Romero & Saavedra, 2019).…”

“…Nevertheless, the framework of Romero & Saavedra (2019) differentiates collective land titles by population and not by the tile's size itself. For the IRs of Colombia there are population data only for 2005 of the Departamento Nacional de Planeación (DNP).…”