Global poverty assessment in GTAP: Improving the data and the model

Maros Ivanic, Purdue University

Abstract

The goal of this dissertation is to enable poverty assessment of trade liberalization at the global level. This objective is achieved by the development of innovative methods, both on the data and modeling fronts, that expand the scope of poverty assessment to respond to global questions. The dissertation concludes with the application of these new methods in a comprehensive exercise that accentuates the contribution to global poverty assessment made possible by this work. The dissertation begins by introducing a method of estimating the relevant distributional household data for those countries where no current household surveys are available. The dissertation shows that the level of national per capita income is a powerful predictor of the main characteristics of national income distribution and is sufficient to draw credible estimates for any given country. This dissertation further includes a method for ex post sampling error correction. The proposed method smoothes the household survey data in a way which mimics the effect of adding more observations. Based on the Hodrick-Prescott filter, it provides a meaningful method for data smoothing and the reduction of sampling error. The dissertation also deals with replacing the representative consumer in the GTAP model with a set of disaggregated households, and bridging of household income and expenditure through household saving. This work is motivated by the problems caused when survey-based household income and expenditure are not properly reconciled. The thesis shows that the inclusion of saving in the household-level final demand structure is particularly important for the poorest households who often dissave. As a final step and to stress the practical nature of its propositions, the dissertation performs a sample poverty assessment exercise, in which the expanded data are used to estimate the level of poverty in forty-six GTAP regions. The success of the exercise in recovering the distribution of poverty around the world validates the introduced innovations as well as the global poverty assessment based on them. This enhanced data set is then combined with the revised model to provide an assessment of the global poverty impacts of multilateral trade reform.

Degree

Ph.D.

Advisors

Hertel, Purdue University.

Subject Area

Agricultural economics

COinS