Three Essays on the Economics of Reduced Deforestation Policy

Abstract

Reducing deforestation is seen as a necessary part of global efforts to manage greenhouse gas emissions. The opportunities for the current mechanism to reduce deforestation in developing countries, REDD+, appear large; however, there are significant challenges in its implementation. In part, these challenges persist because the tools available to investigate reduced deforestation policy inadequately capture the essence of the economics of deforestation. As a response to this limitation, this thesis makes contributions to theoretical and empirical understanding of the economics of both deforestation and reduced deforestation policy. The thesis investigates the impact of REDD+ on economic growth; carbon 'leakage', or the increase in emissions one region as a response to emissions reductions in another; and the setting of targets for reduced emissions. Firstly, the clearing of forests for alternative land uses, like agriculture, is a well trodden path of economic development. A widespread intervention in reducing deforestation will affect the path of agricultural expansion, and therefore the economic growth of lower income nations. To investigate these processes, a general, dynamic, stochastic model of economic growth with an endogenous process of deforestation and policy to reduce deforestation is presented. The main results in this chapter are firstly that reduced deforestation policy lowers national output. Secondly, compensation for the size of forest stocks does reduce deforestation, but is likely to be prohibitively expensive as a means of reducing deforestation. Finally, additional reduced deforestation can be achieved by offering a fixed compensation rate, rather than a rate tied to a variable carbon price. Secondly, with many reduced deforestation projects having limited spatial coverage, there is the possibility that emissions reductions in one region could be offset by increased deforestation in another. Currently, there is little understanding of the magnitudes of this phenomenon, carbon leakage, for reduced deforestation policy. This chapter presents a new spatial, dynamic modeling approach to investigate carbon leakage under reduced deforestation policy. Numerical approximation of the model, calibrated to match patterns of deforestation in Indonesia, finds leakage rates ranging between 25 and 55 per cent. Leakage is found to be lower where compensation rates are higher; where agricultural productivity is higher and the costs of deforestation are lower. These results occur because more deforestation is reduced under these conditions, and increasing of deforestation elsewhere becomes increasingly costly. Finally, the consequences of assumptions implicit in proposed reference rates for a REDD+ mechanism are highlighted. Reference rates are crucial to the effectiveness of reduced deforestation policy; however, reference rates cannot be systematically set with accuracy. The chapter highlights how the setting of current proposals for reference rates can exacerbate this inaccuracy. It is suggested that cost effectiveness be considered as a criteria in the setting of reference rates. The chapters are linked through a focus on the dynamic process characterising deforestation and reduced deforestation policy. The models developed emphasise how policy interacts with this intertemporal process. The thesis draws out the subtleties of assessing the effectiveness of policy affecting inter-temporal behaviour and the difficulty of setting policy that affects intertemporal behaviour.