Economics of Avoiding Deforestation

Deforestation is estimated to cause about one quarter of anthropogenic carbon emissions. Only late, in 2005, Parties to the United Nations Framework Convention for Climate Change decided to start exploring approaches to reduce emissions from this major source. We carried out a global analysis of the potential effects of financial mechanisms to avoid deforestation. Avoiding deforestation is a low cost option that could have considerable leverage in mitigating climate change. Our model results indicate that a 50% reduction of carbon emissions from deforestation over the next 20 years would require financial resources of some US$33 bn per year. Expectations that financial flows through international climate policy mechanisms would provide a golden opportunity to turn around a 20 year history of gridlock and indecision in international fora addressing deforestation, however, appear inflated. Introduction Deforestation is considered the second largest source of greenhouse gas emissions (1) and is expected to remain a major emission source. The deforestation issue has been at the centre of the international environmental debate for two decades. Yet, despite a large number of studies, commitments, initiatives and strategy papers, this activity has had little impact on deforestation rates: deforestation continues at a rate of about 13 million ha per year (2). Apart from the loss of carbon, deforestation typically is associated with inter alia loss of biodiversity, disturbed water regulation and the destruction of livelihoods for many of the world’s poorest (3). Government and non-governmental attempts to slow down, or even reverse, current trends of disappearing forests have not been successful as the result of many pressures, both local and international. While the more direct causes are rather well established as being agricultural expansion, infrastructure extension and wood extraction (4-6), indirect drivers of deforestation are made up of a complex web of interlinked and place-specific factors. Some see a glimmer of hope for more effective policies with the rise of innovative financial mechanisms under a global climate policy regime (7-10, 20). Indeed, in 2005, Papua New Guinea has proposed to the UNFCCC that carbon credits be provided to protect existing native forests. The proposal triggered a flurry of discussion on the topic. The potential for synergies between forest and carbon policies is quite substantial. For instance, Soares-Filho et al. (11) suggest that protecting around 130 million hectares of land from deforestation in the tropical Amazon could reduce global carbon emissions by 17 GtC over the next 50 years. What is unclear, however, is how much it would cost to achieve this, and which types of policies could be most effective. This paper we use scenario modeling approaches to assess the costs of reducing global deforestation and examine different financial mechanisms to combat deforestation. Costs of cutting deforestation in half 1 FCCC/CD/2005?misc.1 11 November 2005. Baseline deforestation is estimated to be close to 212 million ha or around 5% of today’s forest area between 2006 and 2025 resulting in a release of some 17.5 GtC. The maximum allowable base-line deforestation rates were estimated statistically using forest share, agricultural suitability, population density and economic wealth as independent variables. Sub-Saharan Africa is modeled to be responsible for about 50% of global deforestation emissions over the coming 20 years, while Latin America contributes 35% and Asia 12% respectively (Table 1 and Figure 1). When aiming to reduce the deforestation rate by 50% until 2025, the financial resources required to balance out net present value differences on exactly those forests that would otherwise be converted rise from some US$0.16bn in 2006 to US$2.9bn in 2025 due to increasing geographic coverage of the carbon incentive scheme. The lack of precise information on areas that are about to be cut and principal-agent problems between parties involved, make it impossible to design a perfectly targeted instrument. In the contrarian case of complete absence of information on deforestation pressure, a global forest carbon conservation program aiming at avoiding half of baseline deforestation would require financial resources in the much higher order of US$197 bn in 2006 and US$188 bn in 2025 (i.e. on average US$6/tC/5years (Table 1)). More realistic assumptions of targeted payments to identifiable deforestation agents in areas of high deforestation pressure would cut average annual cost to an estimated US$33.5 bn per year. This large difference in costs indicates the magnitude of costs to be saved by designing targeted incentive schemes. Carbon tax schemes do not suffer as much from an information problem, as global earth observation systems can detect deforestation with some reliability already today. In tax scenarios, e.g. simulated introduction of a forest clearance tax, an average carbon tax of US$9/tC would reduce emissions from deforestation by half if we assume deforestation by slash and burn. If the carbon from the harvested wood is assumed to be temporarily sequestered in a timber products pool, a timber sales tax of US$25/tC would have a similar effect. In practice these two taxes would be additional, i. e. a timber sales tax on top of a land clearance tax. Revenues from such carbon taxes on deforestation would result in annual revenues in the magnitude of US$5.9 bn in 2006 declining to US$4.2 bn in 2025 (Table 1 and Figure 1). Results from the scenario analysis show that almost independent of the financial mechanism (incentive payments or tax), more than half (53%) of the forest carbon would be saved in sub Sahara Africa 30% in Latin America and 16% in Asia. Table 1: Scenarios of forest biomass saved according to financial mechanism. Deforestation Baseline shows the amount of forest biomass (GtC) lost through deforestation over the coming 20 years. Incentive payment and Carbon tax give the amount of avoided deforestation in GtC per 20years at the price/tax levels indicated. Slash-burn and timber sales assume that 100% of the biomass will either be burned on the spot or a harvested wood products pool respectively. Sale/burn is a more realistic and geographically differentiated combination of slash-burn and the wood products pool. The share is region specific, based on empirical evidence of region-specific deforestation drivers (2-3).