Files
Abstract
Public policy is a key factor for reducing poverty in developing countries. Designing effective and efficient Pro-Poor-Growth policies requires understanding of growth-poverty and policy-growth linkages. Most existing studies focus only on growth-poverty linkages or rely on reduced form estimations, that require a large set of data. Addressing these shortcomings, we empirically estimate a sector specific, nested two-stage policy impact function. We apply a Bayesian estimation approach that combines existing statistical data with a priori information of political experts, thus reducing data and estimation problems. This is linked with a CGE to model the full link from policy to growth to poverty. We derive a theoretical framework that allows us the definition of indicators for key sectors and key policies of an efficient PPG-strategy. We show that indicators based only on growth-poverty linkages might be misleading. To deal with model uncertainty inherent in the CGE-model application we derive a set of meta models via CGE-simulations conducted under different model parameter-settings. Applying Bayesian model selection allows to draw statistical inferences on competing models to generate relative robust policy-relevant messages even in the presence of model uncertainty. The approach is empirically applied to Senegal analyzing the allocation of public spending on agriculture under CAADP.