Essays on labor economics
- Autores: Anna Houstecka
- Directores de la Tesis: Nezih Guner (dir. tes.)
- Lectura: En la Universitat Autònoma de Barcelona ( España ) en 2020
- Idioma: español
- Tribunal Calificador de la Tesis: Virginia Sánchez Marcos (presid.), Isaac Baley (secret.), Lidia Farré (voc.)
- Programa de doctorado: Programa de Doctorado en Análisis Económico por la Universidad Autónoma de Barcelona
- Materias:
- Enlaces
- Tesis en acceso abierto en: TESEO
- Resumen
In this dissertation, I analyze the determinants and consequences of a mismatch between the skills of workers and the skills required by the occupation they work in. I study how skills mismatch can be measured and how it is affected by unemployment insurance policy through occupational switching. I also investigate the determinants of in vitro fertilization success, which has important implications for women's career and fertility choices.
In Chapter 1, I compute new measures of skills mismatch for literacy and numeracy based on how well workers sort to jobs. Existing measures of skills mismatch based on the Programme for the International Assessment of Adult Competencies (PIAAC) use information only on the worker's side and ignore jobs that workers perform or the sorting between workers and jobs. For 13 OECD countries from the PIAAC, I measure workers' skills by their individual test scores and the skill requirements of their jobs by the occupational requirements from the Occupational Information Network. I then look at 1) the correlation between the skills and skill requirements in each country and 2) the percentage of workers for whom the absolute difference between the percentile rank of their skills and the skill requirements of their jobs is larger than 50. I show that across countries the new measures of mismatch correlate negatively with measures of aggregate labor productivity, and the correlation is stronger than the existing measures.
In Chapter 2, joint with Andrii Parkhomenko, we study the relationship between unemployment benefits and occupational switching. Do unemployment benefits only provide income support for workers during their unemployment spells, or do they also affect post-unemployment outcomes? Using two US data sets, the SIPP and the NLSY79, we document three new facts on the relationship between unemployment benefits and occupational switching. First, unemployed individuals who are eligible for higher unemployment benefits are less likely to switch occupations. Second, conditional on switching, having higher unemployment benefits correlates positively with the cognitive skills requirements of the new occupation. Finally, while the first fact is stronger for workers with longer occupational tenure, the second fact is stronger for workers with shorter occupational tenure. We then build a search model with heterogeneous individuals and jobs to study how unemployment benefits affect skill requirements and wages for workers who experience employment-unemployment-employment transitions. Using the estimated model we find that providing larger benefits to workers with shorter labor market experience results in higher average wages.
In Chapter 3, joint with Fane Groes, Daniela Iorio, Mallory Leung and Raul Santaeulalia-Llopis, we study the determinants of in vitro fertilization (IVF) success using administrative data from Denmark (1995-2009). We find that maternal education significantly determines IVF success (live birth). Compared with high school dropouts, patients with a college (high school) degree have a 21% (13%) higher chance of attaining a live birth through IVF. We argue that the education gradient in IVF reflects educational disparities in IVF productivity (how well women follow the IVF procedure) and the psychological factors (how they are affected by undertaking the treatments). We develop a dynamic model of women using IVF technology in which women differ in IVF productivity and the psychological stress associated with undergoing the treatment. In the model, women face a trade-off between a positive probability of succeeding in getting a child through IVF and the psychological cost associated to undergoing the treatment. The estimated model sheds light on the importance of each of the factors in explaining the IVF educational gradient. In particular, we find that differences in average IVF productivity across education groups account for more than 95% of the observed gradient.
