The impact of credit on income poverty in urban Mexico. An endogeneity-corrected estimation

In recent years, an important number of impact studies have attempted to examine the effect of credit on income poverty; however, many of these studies have not paid sufficient attention to the problems of endogeneity and selection bias. The few exceptional cases have employed econometric techniques that work at the village level. The problem is that the concept of village is inappropriate in the urban context where a large percentage of microfinance organisations in the developing world actually operate. This paper presents an econometric approach which controls for endogeneity and self-selection using data from a quasi-experiment designed at the household level, and conducted in three urban settlements in the surroundings of the Metropolitan area of Mexico City. The paper provides an estimation of the impact of credit, employing different equivalence scales in order to measure the sensitivity of the poverty impact to the intra-household distribution of welfare. We find a link between poverty impacts and lending technology. Group-based lending programmes are more effective in reducing the poverty gap but in doing so, they achieve insignificant impacts on the poverty incidence. By contrast, individual lending programmes reported significant and small impacts at the upper limits of deprivation but insignificant impacts on the poverty gap.