Poverty and Inequality in Chile 1990-1998: Learning from Microeconomic Simulations

This paper contributes to understand the microeconomic determinants of household income dynamics in poverty and inequality in Chile during the 90’s . We use a microsimulation based on the decomposition of distributional changes, developed by Bourguignon et.al. (2000). We estimate a participation model and an earnings equation for each economic agent. We examine how income distribution and poverty would change as a result of a different set of microsimulations. In particular, the distributional structure of 1998 is imposed in 1990. The evidence suggest that while poverty responds strongly to the simulation exercises, the distribution of income appears less sensitive, and is therefore more stable. In particular, we can state that a reduction in poverty would have been observed in 1990 if the returns to education, the regional effects, the structure of non-observables, and the endowments of 1998 had been present. The opposite would have occurred in the case of considering the “returns on experience” and the structure of participation. With respect to the income distribution, in spite of its stability, the most interesting results come from a dynamic perspective. The observed inequality indicators remain at the same level between 1990 and 1998, but if 1998 prices had been observed in 1990, then an increase in inequality would have been registered. The changes in the return to education (and in its convexity) and the structure of participation also increase inequality. The effect of non-observables, in contrast, would have meant a better distributed income distribution in 1990.