2005年-世界发展银行全球_Estimating_the_Returns_to_Education_in_Argentina___1992-2002_48页_474kb

Summary of "Estimating the Returns to Education in Argentina: 1992-2002"

Core Content

This paper estimates the returns to education in urban Argentina from 1992 to 2002 using household survey data and both OLS and quantile regression methods. The study aims to understand how the returns to schooling have evolved over time and whether they vary across the wage distribution.

Main Findings

• Overall Increase in Returns: The returns to schooling in Argentina increased significantly over the decade, with the overall rate rising from 9.1% in 1992 to 12.0% for men and 10.8% for women in 2002.
• Gender Differences: Returns to schooling are higher for men than for women. For women, the highest returns are observed at the lowest quantile, while for men, returns increase with higher quantiles.
• Education Levels:
  • Primary Education: Initially had the highest returns but declined over time.
  • Secondary Education: Returns were relatively stable, with some fluctuations.
  • Tertiary Education: Returns increased substantially, with university complete education showing the highest returns by the end of the period.
• Earnings Inequality: The study highlights within-education-level wage inequality, suggesting that the returns to education have increased across the distribution.
• Economic Context: Despite economic fluctuations, including periods of growth, recession, and crisis, the returns to schooling continued to increase, suggesting that demand for skilled labor played a key role.
• Human Capital Theory: The findings support the idea that increased demand for skilled labor drives higher returns to education, even in adverse economic conditions.
• Sample Selection Bias: The paper accounts for selection bias in the estimation of returns for women using Heckman's two-step procedure, which results in slightly higher returns for women compared to the uncorrected estimates, though still lower than those for men.

Key Information

• Data Source: The study uses data from the Permanent Household Survey (EPH) conducted by the National Institute of Statistics and Census (INDEC).
• Methodology:
  • Mincerian Earnings Function: Used to estimate the relationship between earnings and education.
  • Quantile Regression: Applied to detect differences in returns across the distribution of earnings.
• Education Attainment:
  • Average years of schooling increased from 10.1 in 1992 to 10.9 in 2002.
  • The proportion of workers with tertiary education increased significantly, from 9.5% in 1992 to 16% in 2002.
  • Women had more years of education than men, with the gap increasing from 0.8 years in 1992 to 0.9 years in 2002.
• Earnings Disparities:
  • Women earned 7% less per hour than men in 1992, and this gap widened to 13% in 2002.
  • Workers in the informal sector earned significantly less than those in the formal sector.
  • Employers earned more than employees or self-employed individuals.
• Macroeconomic Conditions:
  • The returns to schooling increased during periods of economic growth, recession, and crisis, indicating a strong demand for skilled labor.
  • The returns remained high even when real wages declined, suggesting that education continues to be a key determinant of earnings.

Methodological Notes

• The paper estimates the returns to schooling by comparing different education levels using dummy variables.
• The returns are calculated as the difference in earnings between education levels, normalized by the number of years of schooling.
• The study also considers the impact of experience and experience-squared on earnings.
• The coefficient on experience is positive for both men and women, and the experience-squared coefficient is negative, consistent with human capital theory.

Conclusion

The returns to education in Argentina increased over the decade, especially for tertiary education. The study suggests that this increase is driven by a growing demand for skilled labor, even in times of economic downturn. The analysis also highlights gender differences in returns, with men consistently having higher returns than women, though the gap narrowed slightly over time. The use of quantile regression provides a more nuanced understanding of how returns vary across the wage distribution.