Essays in applied econometrics: 1. Nonparametric analysis of wage distributions: An application to Mexico, and 2. recent changes in the United States business cycles.

摘要

Essay 1: Non-parametric analysis of wage distributions: An application to Mexico. This essay uses nonparametric econometrics to examine the extent and cause of the increasing wage inequality in Mexico over time---1984 to 1996. To understand the causes of the rising inequality over time, a simple counterfactual question is asked: "what would the density of wages be if workers' characteristics remained the same, but they were paid according to the wage setting mechanisms of the earlier period?"; The theoretical contributions of this essay are twofold: first, we propose a new optimal window width for the counterfactual distribution; second, we develop nonparametric tests to measure the closeness between the actual and the counterfactual distribution.; Empirically, we find that while wage inequality did increase in Mexico, it is the middle class that is most affected. Doing the counterfactual analysis on the Mexican wage distribution leads us to two interesting findings. First, changes in the endowments (levels) and returns of human capital emerge as the leading cause of the shrinking middle class. Second, once we control for all individual attributes the distribution becomes bimodal, and this bimodality is explained by the gender differences in the wage structure and the wage setting mechanisms.; Essay 2: Recent changes in U.S. business cycles. The U.S. business cycle indicates a substantial decrease in volatility for over a decade. Empirical analysis places this structural break in the dynamics of the U.S. economy at the first quarter of 1984. In this essay, we investigate some possible causes of the observed reduced volatility in the U.S. business cycle. We find evidence that substantial changes in investment and consumption volatility since the second half of the 80s may have been important sources of increase stability in the U.S. GDP. Interestingly, this coincides with the onset of the adult age of the baby boomers. We investigate the possibility of demographics as an explanation for the smoother fluctuations in the U.S. economy.; We also put the whole debate of the decreased volatility of the GDP in an international and a historical context. We look at the GDP data from G-7 countries over the last 150 years, taking into account possible inconsistencies in the series over time. Consequently, structural breaks in the volatility of GDP for these countries are tested, using recursive methods. The findings provide evidence that there have been multiple structural breaks leading to greater stability over time, and that the recent decrease in U.S. GDP volatility is part of a broader long-term trend shared by all the other G-7 countries.