Empirical Essays in Earnings and Labor Markets in Developing and Transition Economies

• Main Author: Marku, Marenglen
• Other Authors: Economics
• Published: Virginia Tech 2014
• Subjects: labor force; Albania; gender gap; female; transition; earnings
• Online Access: http://hdl.handle.net/10919/27902; http://scholar.lib.vt.edu/theses/available/etd-05262006-125231/

This dissertation is a collection of three empirical essays on Albania and Iran. In December of 1990, the communist system in Albania came to an abrupt end. The collapse of communism led to a number of macroeconomic reforms that, among other things, brought dramatic changes in the Albanian labor market. This study uses data from the first nationally representative household survey to examine one outcome of a decade-long transition in Albania, the earnings gap between men and women. The average gender earnings gap is calculated at 31 percent, but it is found to be as high as 50 percent in the upper parts of the distribution. The traditional Oaxaca-Blinder method and a recent method that combines quantile regression with the bootstrap are applied to decompose the gender gap into a portion attributable to differences in characteristics and a portion explained by returns to characteristics. Results show that differences in human capital characteristics do not explain any of the existing gap. Furthermore, a large proportion of the gap can be attributed to segregation in occupations and industries. Simulations of female counterfactual wages show that the gender gap is significantly reduced for the entire distribution, and disappears in the higher quantiles of the distribution when occupation and industry are controlled for. The next two essays analyze welfare and female labor force participation in post-Revolution Iran. The Islamic Revolution of 1979 and a number of subsequent macro shocks dealt a huge blow to Iran's economy. In this paper we ask the question of how families and individuals have fared through these tumultuous times. Conventional measures of change in welfare, such as average consumption or GDP per capita, do not accurately reflect the experience of individual cohorts. We utilize annual surveys of expenditures and income conducted between 1984 and 2004 and decompose changes in average earnings and expenditures into cohort, age, and period effects. The estimated period effects accurately reflect the fluctuations in the economy noticeable in the macro data, and the life cycle earnings and expenditures profiles show a typical inverted U-shape. The cohort effects, which compare the position of life cycle profiles of different cohorts, and are of most interest to us, show a rising trend for cohorts born before the 1950s (about 30 years or older at the time of the Revolution). They also indicate that younger cohorts, those born after 1965 and therefore entered adult life after the Revolution, seem to have lost out. We discuss possible reasons for the asymmetrical lifetime experience of the two sets of cohorts. We believe that the disruptions caused by the Revolution itself and the subsequent eight year war with Iraq (1980-88) may have caused lifetime losses for the cohorts who came of age in the early 1980s. The purpose of the third essay is to understand changes in the labor force participation rate of women in Iran after the Islamic Revolution of 1979. Studies consistently show that like other countries in the Middle East and North Africa, Iran has experienced only a modest improvement in female labor force participation rates, despite having gone through the fertility transition and significant improvements in education of women. Utilizing 21 consecutive household surveys from 1984-2004, we decompose changes in the participation rate into age, cohort, and period effects. We find some evidence that the Islamic Revolution of 1979 did indeed have a negative impact on the cohorts that were in their teens or early 20s at that time. However, viewed from a cohort perspective, the evidence shows that women born after 1965 have continuously increased their participation. This is in contrast to the evidence that has been observed by others who have compared cross-section averages over time. === Ph. D.