Tropical deforestation: a multinomial logistic model and some country-specific policy prescriptions

Three problems-one-way effect hypothesis, data and estimation problems-in the existing econometric models of global deforestation are addressed, robustness of the results is tested and country-specific policy prescriptions, for five countries, are suggested. A theoretical deforestation model is proposed by incorporating two-way effects of all explanatory variables, and hypothesizing that the net effect of a variable may vary across regions. Deforestation is used as qualitative variable to address the data problem. Multinomial logistic model is used to deal with estimation problems, and the results of multinomial logistic are found to be more informative and robust compared to the results of binary logistic and ordinary least square (OLS) methods. Growth in population, forest areas, agriculture and road construction are the main causes of deforestation in high deforesting countries, but debt service growth, in addition to agriculture and road construction, are the main causes in medium deforesting countries.