Poverty Analysis Using an International Cross-Country Demand System

This paper proposes a new method for ex ante analysis of the poverty impacts arising from policy reforms. Three innovations underlie this approach. The first is the estimation of a global demand system using a combination of micro-data from household surveys, and macro-data from the International Comparisons Project. Estimation is undertaken in a manner that reconciles these two sources of information, explicitly recognizing that per capita national demands are an aggregation of the disaggregated, individual household demands. The second innovation relates to a methodology for post-estimation calibration of the global demand system, giving rise to country specific demand systems and an associated expenditure function which, when aggregated across the expenditure distribution, reproduce observed per capita budget shares exactly. The third innovation is use of the calibrated expenditure function to calculate the change in the head-count of poverty, poverty gap and squared poverty gap arising from policy reforms, where the poverty measures are derived using a unique poverty level of utility, rather than an income or expenditure-based measure. We employ these techniques with a demand system for food, other non-durables and services estimated using a combination of 1996 ICP data set and national expenditure distribution data. Calibration is demonstrated for three countries for which household survey expenditure data are utilized during estimation; namely, Indonesia, the Philippines and Thailand. To illustrate the usefulness of these calibrated models for policy analysis, we assess the impacts of an assumed five percent food price rise as might be realized in the wake of a multilateral trade agreement. Results illustrate the important role of subsistence expenditures at low per capita income levels, but of discretionary expenditure at higher per capita income levels. The welfare analysis underscores the relatively large impact of the price hike on poorer households, while a modified Foster-Greer-Thorbecke poverty measure shows that the five percent price rise has a differential effect on poverty across the three focus countries, although it increases the incidence and intensity of poverty in all three cases.