Interrelation between the prevalence rate of suicides and the length of working hours

Abstract

It is known that overworking as well as deficiency of work (plenty of free time) are major factors of a suicide on an individual level which allows when passing to the level of a real social group (employees) to suppose of existence of a certain optimum of working time or a parabolic (U-shaped) connection between the suicide rate and an average duration of working time. From the theoretical point of view the supposed parabolic dependence of the level of prevalence of suicides from an average duration of working time of employees is described from the point of view of suicidology: excessive increase of working time is an external tendency which prevents satisfaction of actual needs of an employee and limits physically the space (off-work time) for their realization. Multidirectional tendencies form a life conflict which has crucial significance when transferring to a suicidal phase. The objective of this article consists in a qualitative assessment of an influence of “an average duration of working time” on the level of prevalence of suicides when other things are fixed (economic, social, religious and others) in a relatively stable social situation. For the econometric analysis, reliable and comparable data of the European database of detailed mortality data of the World Health Organization and Eurostat are used for 22 European countries for the period from 1998 till 2012. Based on analysis of a dynamics of the studied variables different hypothesis have been made: 1) about existence of statistically significant linear or logarithmic dependence of the level of prevalence of suicides from an average factual duration of working time inside a country 2) about existence of a parabolic (U-shaped) dependence of the level of prevalence of suicides from an average factual duration of working time between countries. A set of panel unit root tests and stationarity testify that the examined variables are unsteady variables with integratedness order I(1). The results of Pedroni panel cointegration tests show that the studied variables are likely cointegrated. The models of panel regression of linear, logarithmic and parabolic specifications are actualized: namely the model for mean values, the model with fixed effects and the model with random effects. As a result of econometric analysis it was determined that the difference in the level of prevalence of suicides of employees in timing inside one country is best explained by a logarithmic dependence from an average factual duration of working time by using a model with fixed effects and a difference between countries – by U-shaped dependence by using a model with random effects. As a result it was established that using the minimal solution of the equation of the parabolic model with random effects, the optimal value of the average actual duration of the working week (38,7 hours) is determined, at which the minimum of suicides is reached. Existence of a best value of working time opens up opportunities for scientifically proven regulation of working time towards optimum in order to reduce the number of suicides.