Análisis de los Factores Determinantes del Desempleo en la Región Central de Costa Rica, 2013-2022
Fecha
2024-07-17
Autores
Bolaños Gutiérrez, Mario Alberto
Mongrillo Barquero, Jean Carlo
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Nacional, Costa Rica
Resumen
El desempleo estructural existe en las economías por la presencia de desajustes entre
las calificaciones de la fuerza de trabajo y los requerimientos de los empleadores; y la
existencia continua de este fenómeno resulta ser un factor generador de problemas
económicos derivados principalmente de las desigualdades sociales, causadas por la
dualización de la sociedad, entre los empleados y los desempleados.
La Región de Planificación Central representa el 21% del territorio costarricense y en
ella residen para el año 2022 el 62% de la población y el 66% de la fuerza laboral de Costa
Rica. Por lo que teniendo en cuenta además que durante el periodo 2013-2022 esta región
muestra tasas de desempleo superiores al 7%, se procede a analizar las principales variables
determinantes de la desempleabilidad mediante el estudio de los microdatos resultantes de
la Encuesta Continua de Empleo para dichos años.
El análisis se basa en la aplicación del Pensamiento Computacional para la resolución
de problemas. Metodología, que presenta cuatro fases: descomposición del problema,
reconocimiento de patrones, abstracción de la información y creación de un algoritmo
(Plan). Por lo que el estudio parte de un análisis exploratorio descriptivo de la oferta de
trabajo en los últimos 10 años, seguidamente se determina la existencia de patrones
repetitivos en las características de las personas desempleadas a través de modelos de
Aprendizaje Automático No Supervisado y en el que se logran encontrar 6 clústers claves.
Posteriormente, se identifican las variables que impactan en mayor medida la
empleabilidad mediante modelos de tipo Logit y se aplica un modelo de Aprendizaje
Automático Supervisado que simula casos hipotéticos de políticas públicas mediante la
predicción de las posibilidades de empleabilidad de un individuo con características dadas.
Resultados cuantitativos que se utilizan finalmente para evaluar las políticas públicas
aplicadas en los últimos años, concluyendo así con una serie de recomendaciones de políticas
dirigidas a obtener un mayor bienestar social.
Structural unemployment exists in economies due to the presence of mismatches between the qualifications of the workforce and the requirements of employers; and the continuous existence of this phenomenon turns out to be a factor generating economic problems derived mainly from social inequalities, caused by the dualization of society, between the employed and the unemployed. The Central Planning Region represents 21% of the Costa Rican territory and 62% of the population and 66% of the labor force of Costa Rica will reside there by 2022. Therefore, taking into account that during the period 2013-2022 this region shows unemployment rates higher than 7%, the main variables determining unemployment are analyzed by studying the microdata resulting from the Continuous Employment Survey for those years. The analysis is based on the application of Computational Thinking for problem solving. The methodology presents four phases: problem decomposition, pattern recognition, information abstraction and creation of an algorithm (Plan). Therefore, the study starts from a descriptive exploratory analysis of the job offer in the last 10 years, then the existence of repetitive patterns in the characteristics of unemployed people is determined through Unsupervised Machine Learning models and in which 6 key clusters are found. Subsequently, the variables that impact employability the most are identified through Logit-type models and a Supervised Machine Learning model is applied that simulates hypothetical cases of public policies by predicting the employability possibilities of an individual with given characteristics. Quantitative results are ultimately used to evaluate public policies applied in recent years, thus concluding with a series of policy recommendations aimed at achieving greater social well-being.
Structural unemployment exists in economies due to the presence of mismatches between the qualifications of the workforce and the requirements of employers; and the continuous existence of this phenomenon turns out to be a factor generating economic problems derived mainly from social inequalities, caused by the dualization of society, between the employed and the unemployed. The Central Planning Region represents 21% of the Costa Rican territory and 62% of the population and 66% of the labor force of Costa Rica will reside there by 2022. Therefore, taking into account that during the period 2013-2022 this region shows unemployment rates higher than 7%, the main variables determining unemployment are analyzed by studying the microdata resulting from the Continuous Employment Survey for those years. The analysis is based on the application of Computational Thinking for problem solving. The methodology presents four phases: problem decomposition, pattern recognition, information abstraction and creation of an algorithm (Plan). Therefore, the study starts from a descriptive exploratory analysis of the job offer in the last 10 years, then the existence of repetitive patterns in the characteristics of unemployed people is determined through Unsupervised Machine Learning models and in which 6 key clusters are found. Subsequently, the variables that impact employability the most are identified through Logit-type models and a Supervised Machine Learning model is applied that simulates hypothetical cases of public policies by predicting the employability possibilities of an individual with given characteristics. Quantitative results are ultimately used to evaluate public policies applied in recent years, thus concluding with a series of policy recommendations aimed at achieving greater social well-being.
Descripción
Palabras clave
MERCADO DE TRABAJO, WORK MARKET, OFERTA Y DEMANDA, OFFER AND DEMAND, DESARROLLO ECONÓMICO, ECONOMIC DEVELOPMENT, DESEMPLEO, UNEMPLOYMENT, EDUCACIÓN EN LÍNEA, ONLINE EDUCATION