This thesis examines the causal effects of education on earnings across 20 European countries using data from different waves of the European Social Survey (ESS) together with EUROMOD household income data. Although Ordinary Least Squares (OLS) estimates of the return to education are common in the literature, they are prone to bias as a result of endogeneity, especially omitted variable bias and reverse causation. I use an Instrumental Variable (IV) method based on exogenous variation from historical changes in compulsory schooling reforms which raised the minimum school-leaving age in several European nations to address potential endogeneity. The analysis focuses on employed individuals aged 25 to 64 and estimates a Mincer-type earnings function using both OLS and Two-Stage Least Squares (2SLS). Findings show a notable and positive return to higher education. The IV estimates using compulsory education reform show larger causal impact consistent with previous studies indicating OLS underestimates the actual returns. The validity of the instrument is supported by institutional evidence and tested econometrically. I include as age, gender, and occupational skill levels as well as fixed effects for country and birth cohort to isolate within-group variation and account for structural differences across national labor markets. The results have significant policy consequences, especially for education policy in the EU, as they support the function of mandatory schooling changes in fostering income equality and social mobility. In addition, by means of harmonized and representative survey data, the study offers current information across several national settings thus adding to the larger body of work on human capital and income distribution. All things considered, this dissertation emphasizes the need of education not only as a personal investment but also as a structural lever for social cohesion and economic development in Europe.
This thesis examines the causal effects of education on earnings across 20 European countries using data from different waves of the European Social Survey (ESS) together with EUROMOD household income data. Although Ordinary Least Squares (OLS) estimates of the return to education are common in the literature, they are prone to bias as a result of endogeneity, especially omitted variable bias and reverse causation. I use an Instrumental Variable (IV) method based on exogenous variation from historical changes in compulsory schooling reforms which raised the minimum school-leaving age in several European nations to address potential endogeneity. The analysis focuses on employed individuals aged 25 to 64 and estimates a Mincer-type earnings function using both OLS and Two-Stage Least Squares (2SLS). Findings show a notable and positive return to higher education. The IV estimates using compulsory education reform show larger causal impact consistent with previous studies indicating OLS underestimates the actual returns. The validity of the instrument is supported by institutional evidence and tested econometrically. I include as age, gender, and occupational skill levels as well as fixed effects for country and birth cohort to isolate within-group variation and account for structural differences across national labor markets. The results have significant policy consequences, especially for education policy in the EU, as they support the function of mandatory schooling changes in fostering income equality and social mobility. In addition, by means of harmonized and representative survey data, the study offers current information across several national settings thus adding to the larger body of work on human capital and income distribution. All things considered, this dissertation emphasizes the need of education not only as a personal investment but also as a structural lever for social cohesion and economic development in Europe.
Educational Attainment and Earnings in Europe: Estimating Causal Returns to Secondary and Higher Education Using Instrumental Variables
MBOWE, BABOUCARR
2024/2025
Abstract
This thesis examines the causal effects of education on earnings across 20 European countries using data from different waves of the European Social Survey (ESS) together with EUROMOD household income data. Although Ordinary Least Squares (OLS) estimates of the return to education are common in the literature, they are prone to bias as a result of endogeneity, especially omitted variable bias and reverse causation. I use an Instrumental Variable (IV) method based on exogenous variation from historical changes in compulsory schooling reforms which raised the minimum school-leaving age in several European nations to address potential endogeneity. The analysis focuses on employed individuals aged 25 to 64 and estimates a Mincer-type earnings function using both OLS and Two-Stage Least Squares (2SLS). Findings show a notable and positive return to higher education. The IV estimates using compulsory education reform show larger causal impact consistent with previous studies indicating OLS underestimates the actual returns. The validity of the instrument is supported by institutional evidence and tested econometrically. I include as age, gender, and occupational skill levels as well as fixed effects for country and birth cohort to isolate within-group variation and account for structural differences across national labor markets. The results have significant policy consequences, especially for education policy in the EU, as they support the function of mandatory schooling changes in fostering income equality and social mobility. In addition, by means of harmonized and representative survey data, the study offers current information across several national settings thus adding to the larger body of work on human capital and income distribution. All things considered, this dissertation emphasizes the need of education not only as a personal investment but also as a structural lever for social cohesion and economic development in Europe.| File | Dimensione | Formato | |
|---|---|---|---|
|
MBOWE Baboucarr.pdf
accesso aperto
Dimensione
858.48 kB
Formato
Adobe PDF
|
858.48 kB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/89412