Essays on labor markets in developing countries

In chapter 2, I construct a theoretical equilibrium search model of rural-urban migration in a two-sector economy with closed labor market. I consider both on-the-job search and search from unemployment with search efficiency different across search strategies. I show that a unique steady-state equilibrium exists so long as search costs for workers and vacancy creation costs for firms are sufficiently small. The equilibrium is characterized by a non-degenerate urban wage distribution. In this equilibrium, increases in urban informal earnings prior to getting a formal urban job may induce more migration and higher unemployment in the urban area. Search costs, in contrast, may result in less than socially optimal level of rural-urban migration. Search efficiency relates non-monotonically to migration. As search efficiency increases, migration rises for a while then falls.; In chapter 3, I use a quantile regression decomposition technique to analyze the difference between the urban and rural distributions of household well being in Vietnam. I use two Vietnam Living Standards Surveys, VLSS 1992--93 and VLSS 1997--98, for this study. Real per capita consumption expenditure (RPCE) is used as the measure of household well being. I find that most of the difference between the upper tails of the distributions is due to differences in the marginal effects of covariates such as age or education rather than differences in the covariates themselves, but the opposite is true in the lower tails. Interesting regional and time effects are also identified.; In chapter 4, I use the multinomial logit model to study the distribution of employment across six sectors in Vietnam. The estimates are based on two Vietnam Living Standards Surveys, VLSS 1992--93 and VLSS 1997--98. It appears that, all else equal, older workers, workers with more education and females are more likely to work in the government sector, while younger workers are more likely to work in various types of business enterprises. Education begins to affect the probability of getting a job in a more preferable sector at the level of about 10 years of schooling. No substantial changes in the education effect were observed over time.