Growth elasticity of poverty: estimates from new data

By using the World Bank's new poverty data that are based on the most recent International Comparison Program report, this research aims to revisit the response of poverty rate to increase in real gross domestic product (GDP) per capita.

The response is summarized in terms of elasticity of poverty with respect to real GDP per capita, which is the ratio of annual percentage fall in poverty rate to annual percentage increase in real GDP per capita. The main calculations are done for the entire group of less‐developed countries (LDCs), poverty‐dense South Asia region, and India, which probably has the highest poverty rate. The periods studied are 1981‐1990, 1990‐1999, and 1999‐2005. The calculations are done for four different poverty measures.

Five major points are noted. First, the elasticities generally show a declining tendency over the period, indicating that poverty‐reducing impact of income growth has been weakening. Second, the elasticities show huge differences across the poverty lines, and generally decline with higher poverty lines. Third, while global elasticities for $1.00 poverty line bear some resemblance to those reported or used by many scholars, elasticities for $2.00 and 2.50 poverty rates are dramatically lower, and reinforce the view that many influential estimates show the effect of income growth on poverty to be much higher than the data indicate. Fourth, elasticities for poverty‐dense South Asia are again seen to be much lower than those for the entire LDC group. Fifth, for India, where $2.00 and 2.50 poverty rates are higher than even in Sub‐Saharan Africa, the elasticities are extremely low and have been declining despite an acceleration in income growth. The overall implication seems to be that income growth has generally been less pro‐poor during the globalization era of the 1990s and the 2000s than during the 1980s. In particular, income growth in India seems to have had an extremely small impact on poverty, and that impact, notably for $1.00 and 1.25 poverty lines, has been declining.

First, although there is a vast literature on growth elasticities of poverty, this seems to be the first study that uses World Bank's new poverty data to judge the impact of income growth on poverty. Second, this is the only study that directly estimates and compares elasticities for the four poverty lines of $1.00, 1.25, 2.00, and 2.50, and shows large differences in the elasticities for different poverty lines. Third, this is probably the only work that compares elasticities for the 1980s, 1990s, and the 2000s. Fourth, although some indication of very low elasticities for South Asia and India does exist in a recent study, $2.50 elasticities reported in the present work for India, and even South Asia, should constitute an eye‐opener for scholars, policy‐makers, and international organizations in regard to the potential role of income growth in poverty reduction. Fifth, the observed decline in most elasticities during the 1990s and 2000s, as compared with the 1980s, despite higher income levels and growth rates, may shed light on the likely role of globalization in reducing poverty.