Do Social Factors Affect Poverty? A Dynamic Panel Data Analysis of 23 Selected Developing Countries

Abstract

This research paper aims at investigating the social factors affecting poverty in selected developing countries which was hardly been investigated, yet. We incorporated income poverty index (poverty headcount ratio and poverty gap) and human poverty index as dependent variables, while social factors have been used as explanatory variables, through 1997-2016, for a group of twenty three developing nations. Whereas poverty indices have been constructed through PCA (Principal Component Analysis). Pesaran and Friedman CD (Cross-Dependence) tests have been applied to confirm the occurrence of cross-dependency in the panel data, while CIPS (second generation panel unit root) test has been applied to verify stationary of the variables. Besides, different first generation panel unit root tests have also been applied to confirm stationary of the variables. We applied Pedroni panel co-integration to check for the existence of long-run linkages among the variables, while System and difference GMM (one step) techniques have been used to find dynamic effects on poverty. The results conclude that age dependency ratio showed positive and significant relationship with poverty, whereas, social globalization depicted negative and significant relationship with poverty. Nevertheless, health (life expectancy at birth) and education (primary school enrolment) have negative relationship with poverty in both income and human poverty models, while, in income poverty model, these variables have insignificant impacts and in human poverty model, they have significant associations. Similarly, population growth has positive associations with poverty in the income and human poverty models.