Age quartile effect in determining actual and preferred working hours: Regression discontinuity design

Abstract

It is known that age-stratified quartiles are used in numerous studies, e.g. in epidemiological studies, as an important instrument in the estimation of working life changes and health issues. The aim of this study is to investigate how achieving different age thresholds (32, 42, or 52 years) affects changes in actual and preferred working hours. The data for this study are based on the European Working Conditions Survey (EWCS) for the period from 1991 to 2015. The data for Finland contained 5,646 respondents. We use the regression discontinuity design (RDD) method to predict a categorical outcome (actual or preferred working hours per week) paired with other predictors (age quartile). We found that, on average, the number of actual working hours is 2.46 points higher and the number of preferred working hours is 4.62 points higher for employees reaching 32 years of age. On the contrary, the average number of actual working hours is lower for employees reaching 42 years (1.16 points) and for employees reaching 52 years (3.03 points). To conclude, we found higher dynamics in increasing actual and preferred working hours when achieving the cutoff age of 32 years, while in the ages of 42 and 52 years, dynamics in changes of actual and preferred working hours are lower. We suppose that the age factor is an important instrument in the overall dynamics of working hours with regard to one’s whole working life.