Résumé
Safeguarding tropical rainforests is one of the most important challenges for the future, particularly to mitigate climate change. The international community has actively sought international policy solutions to curb deforestation in tropical countries. Debt-for-nature swaps and certification of sustainable forest management have been implemented by NGOs. Some states are currently negotiating the implementation of the REDD (Reduced Emissions from Deforestation and Degradation) mechanism, a North-South financial transfer to compensate countries for avoided deforestation. However, little is known about the efficiency of these instruments. We argue that they may have a double effect: an expected direct impact on deforestation linked to the conditionalities of instruments, and an indirect impact due to their feedback effects on macroeconomic variables, affecting in turn the drivers of deforestation. The second effect is often overlooked by policy makers. The objective of the paper is to disentangle the two effects for different categories of forest countries. We conducted a panel data analysis for the period 1990-2005 and show that cluster analysis of tropical forest countries would be more relevant if it were based on relative forest endowment. On the basis of econometric results, we can recommend differentiating policy instruments according to the relative forest abundance of each country. Debt reduction programs contribute to the reduction of deforestation in all countries. Countries with abundant forests are locked in a development pathway based on overexploitation of their forests making them less responsive to incentive measures. In countries with average forest endowment, we recommend output-based REDD, whereas in countries with low forest cover, either inputbased or output-based REDD mechanisms should be efficient.