Publication: Medir la Brecha Retributiva por Género
Authors
Alvira Baeza, Ricardo
item.page.secondaryauthor
item.page.director
Publisher
Universidad de Murcia, Facultad de Economia y Empresa, Departamento de Sociologia
publication.page.editor
publication.page.department
DOI
item.page.type
info:eu-repo/semantics/lecture
Description
© 2018. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/.
Abstract
Se está produciendo en la actualidad en España un debate, que incluso involucra una posible modificación del marco legislativo, acerca de la dimensión de la Brecha entre la retribución de hombres y mujeres, dimensión que suele cuantificarse como diferencia entre la Retribución Media o Mediana para cada sexo. Sin embargo, sabemos desde hace tiempo que el valor medio y mediano de una distribución puede ser con frecuencia una descripción incorrecta de la misma. La importancia de esta cuestión nos obliga a buscar entonces una caracterización matemática inequívoca de la Brecha que minimice la probabilidad de errores en la caracterización de los grupos y posterior evaluación. Para avanzar en dicha búsqueda/formulación, en esta comunicación se desarrolla un marco axiomático y se muestra la inconsistencia en ciertos casos de las tres fórmulas de agregación utilizadas en la actualidad. Así mismo, estos axiomas implican unas condiciones que nos permite sugerir que cierto tipo de fórmulas basadas en caracterizaciones de la desigualdad de cada distribución pueden ser adecuadas. Una de ellas satisface el sistema de axiomas. Posteriormente contrastamos empíricamente las cuatro formulas [las tres fórmulas utilizadas actualmente y la explicada en este texto], revisando la retribución del conjunto de personas trabajadoras en España, así como las retribuciones en los Consejos de Administración de 22 empresas del IBEX‐35. Los resultados muestran que la formula aquí explicada proporciona resultados consistentes en todas las ocasiones, mientras las formulas actualmente utilizadas presentan resultados anómalos con cierta frecuencia. Esto permite sugerir la idoneidad de la caracterización aquí propuesta.
A debate is currently taking place in Spain, which even involves the possible modification of our legislative framework, about the dimension of the gap between men and women remuneration. This dimension is usually measured as difference between each gender Mean or Median Remuneration. However, we have long known that the mean and median of a distribution often are incorrect characterizations of said distributions. The importance of the question leads us to look for an unambiguous mathematical characterization of the gap which minimizes the probability of errors in the characterization of the groups and later assessment. In order to move forward in this debate, this communication develops an axiomatic framework that allows proving the inconsistency in some cases of three currently used aggregation formulas. Likewise, the axioms imply some conditions that allow us to suggest certain types of formulas based on the inequality of distribution characterization may be appropriate. One of them satisfies the system of axioms. Subsequently, we empirically test the four formulas [the three currently used and herein explained formula], reviewing the remuneration of workers in Spain, as well as the retributions in the Boards of Directors of 22 IBEX‐35 companies. The results show that herein explained formula provides consistent results in all cases, while currently used formulas provide inconsistent results with some frequency. This allows us to suggest the suitability of herein proposed characterization to replace currently used formulas.
A debate is currently taking place in Spain, which even involves the possible modification of our legislative framework, about the dimension of the gap between men and women remuneration. This dimension is usually measured as difference between each gender Mean or Median Remuneration. However, we have long known that the mean and median of a distribution often are incorrect characterizations of said distributions. The importance of the question leads us to look for an unambiguous mathematical characterization of the gap which minimizes the probability of errors in the characterization of the groups and later assessment. In order to move forward in this debate, this communication develops an axiomatic framework that allows proving the inconsistency in some cases of three currently used aggregation formulas. Likewise, the axioms imply some conditions that allow us to suggest certain types of formulas based on the inequality of distribution characterization may be appropriate. One of them satisfies the system of axioms. Subsequently, we empirically test the four formulas [the three currently used and herein explained formula], reviewing the remuneration of workers in Spain, as well as the retributions in the Boards of Directors of 22 IBEX‐35 companies. The results show that herein explained formula provides consistent results in all cases, while currently used formulas provide inconsistent results with some frequency. This allows us to suggest the suitability of herein proposed characterization to replace currently used formulas.
publication.page.subject
Citation
item.page.embargo
Collections
Ir a Estadísticas
Este ítem está sujeto a una licencia Creative Commons. http://creativecommons.org/licenses/by-nc-nd/4.0/