Interaction of working conditions, job satisfaction, and sickness absences: Evidence from a representative sample of employees

Abstract

We study the predictors of sickness absences among 2800 Finnish workers responding to the cross-sectional Quality of Work Life Survey in 1997. The data contain detailed information on the prevalence of adverse working conditions at the workplace from a representative sample of wage and salary earners. We show by using recursive multivariate models that the prevalence of harms at the workplace is associated with job dissatisfaction and dissatisfaction with sickness absences. The policy lesson is that the improvement of working conditions should be an integral part of any scheme aimed at decreasing sickness absence.

Introduction

The European Survey on Working Conditions reveals a large cross-country variation in absenteeism. The share of the EU workforce that has been absent at least 1 day during a year owing to illness or injury varies from a low of 6.7% in Greece to a high of 24% in Finland (Gimeno, Benavides, Benach, & Amick, 2004).¹ Given that Finland has the highest share of sickness absenteeism, it is of interest to analyse the determinants of absenteeism there. Norway and Sweden have experienced substantial changes in the number of sickness absences over the past 15 years (Holmlund, 2004). Absences have been more stable in Finland. Consequently, structural factors (including adverse working conditions) are likely to account for much of the total number of sickness absences and the Finnish evidence is therefore useful for other countries.

This paper examines how the working environment as measured along several different dimensions may affect sickness absences. The contribution of this paper is that we analyse the interaction between adverse working conditions, job satisfaction and sickness absences. This is particularly relevant from the policy perspective, because management always has some control over the working environment and therefore an influence on job satisfaction. A reduction in absences would partly compensate for the shrinking of the labour force owing to the rapid ageing of the population in the industrialised countries.

Our paper fills important gaps in current knowledge. First, there is evidence that job dissatisfaction increases sickness absences (e.g. Brown & Sessions, 1996; Clegg, 1983; Dionne and Dostie, 2007, Farrell and Stamm, 1988). The existing evidence stems from single equation models, however. In this paper, we use a data set, the Finnish Quality of Work Life Survey, that enables us to model the relationship between adverse working conditions, job satisfaction, and sickness absences. The interaction of these variables has not been previously analysed by means of recursive multivariate models. Second, Dionne and Dostie (2007) point out that most of the literature uses data from one company or a very small sample of firms.² This makes it hard to generalise the results obtained. In particular, the focus on small samples means that employer characteristics are not usually included among explanatory variables. We use data that constitute a representative sample of employees. Lastly, we evaluate a wider range of detailed risk factors at workplaces than has been typical in previous research.

Finland has a relatively centralized wage bargaining system, which sets a floor to firm-level pay determination. The system leads to wage compression. This may prevent the creation of wage differentials that would compensate for adverse working conditions. The evidence shows that perceived working conditions have a minor role in the determination of individual wages (Böckerman & Ilmakunnas, 2006). In contrast, adverse working conditions stimulate job dissatisfaction. It is interesting to study whether this dissatisfaction increases sickness absences.