Unpacking the misery multiplier: How employability modifies the impacts of unemployment and job insecurity on life satisfaction and mental health

Abstract

Employability strongly moderates the effects of unemployment and of job insecurity on life satisfaction and mental health. Using nationally representative panel data from Australia, I find that an increase in employability from zero to 100% cancels around three quarters, in some cases more, of the detrimental effect of unemployment. Employability also matters for employees: an increase in men's employability from zero to 100% reduces the detrimental effect of job insecurity by more than half. The effects of extreme job insecurity and of unemployment are large and of comparable magnitudes. The findings are used to compute estimates of the well-being trade-off between increases in job insecurity and increases in employability, relevant to the support of “flexicurity” policies, and of the “misery multiplier”, the extent to which the effect of a rise in aggregate unemployment on those becoming unemployed is supplemented by the effects on others’ insecurity and employability.

Introduction

Luiz Felipe Scolari has shrugged off the pressure mounting on him at Chelsea and declared that another managerial position would always be around the corner for him. “If I lose my job, I have another job …… Maybe tomorrow, maybe after one year or two years. I have worked for 25 years.” (Guardian, 14 January, 2009).

It has been firmly established, in a wide range of empirical studies at individual and country levels, that unemployment is detrimental for health and well-being, both in itself and because it entails a loss of income. At the same time, a large number of psychological studies and a few in economics have found that job insecurity itself also generates substantial losses in well-being. Within both literatures, some studies have uncovered heterogeneous effects associated with scarring and social norms, or across different socio-economic groups. The issue which I address in this paper is that an important reason for heterogeneity in the effects of unemployment and job insecurity is rarely recognised in theory or empirically investigated: namely, that employability matters. The Guardian quotation illustrates one instance of this proposition: Chelsea coach Scolari was reported to be unconcerned by his job being at risk because he felt he was very employable. More generally, the effects on well-being of being unemployed or of the fear of job loss are each potentially mitigated if there are good prospects of finding another job: the question is, how much?

The broad term “employability” refers to the ability of an individual to find and sustain employment. A characteristic of the individual in context, employability is indicated by the probability of obtaining employment, though often proxied by measures of its determinants (skills, adaptability and so on). In this paper I develop a simple conceptualisation of the roles of employment insecurity and employability, with two central features. First, it allows for the uncertainty surrounding unemployment and employment to affect well-being both directly and indirectly through its impact on expected income. The direct effects are justified in psychological and social theory, while the indirect effects are economic. Second, the framework allows for the interaction between unemployment and employability, and between job insecurity and the employability of the employed. To empirically implement this framework, the three key variables – employability of the unemployed, job loss risk, and the employability of the employed – are directly measured by the subjective expectations of the probabilities of future employment transitions.

An understanding of the role of employability in modifying the detrimental impacts of unemployment and job insecurity is greatly relevant to the formation of unemployment and employment policies. European debate, for example, in recent years has focused on “flexicurity”, a strategy to devise employment and welfare legislation that will optimise the ability of employers to redeploy labour (thereby, other things equal, raising job insecurity) while at the same time providing generous support and training for the unemployed (European Commission, 2007). “Flexicurity” policies are argued, not only to be efficient, but also to provide a political compromise by protecting the welfare of the unemployed. There is, however, no empirical evidence through which the impacts of job insecurity and of employability could be compared, and any trade-off evaluated from the perspective of the well-being of workers.

My findings provide new estimates of the impact of unemployment and of job insecurity, in the context of a model that takes account of the effects of the interacting transition risks. These findings are gleaned using fixed effects estimation on panel data, and are therefore more confidently interpreted as causal than in the many cross-section studies in the literature. I examine how the magnitude of the effects of insecurity among employees compares with the effects of being unemployed.

It turns out that, as predicted, unemployed people with little hope of finding a job enjoy the least well-being by a considerable margin, while employed people who are both highly employable and in a secure job enjoy the most. In between there is substantial differentiation according to employability, job insecurity and their interaction. The estimates imply that there are considerable gains from raising the employability of an unemployed person. Meanwhile, high job insecurity substantially lowers subjective well-being, but less so if the employee is more employable. Relative to a secure job the deleterious effects of a high level of job insecurity are comparable in magnitude with the effects of unemployment. I compute crude estimates of the “misery multiplier” ranging between 3.2 and 3.5 – this being the ratio of the total impact of a rise in unemployment on well-being to the impact on just those made unemployed. It is this broader impact of unemployment, deriving from its extended impact on job insecurity and employability, that accounts for the society-wide impact of recessions. The estimates also allow the trade-off between greater job insecurity and improved employability to be computed, thus providing a first step for a potential evaluation of “flexicurity” policy.

The paper proceeds as follows. Section 2 overviews the two literatures on unemployment and job insecurity, and sets up the simple framework and specification that takes account of the interactions among the uncertainties. Section 3 describes the data and Section 4 my findings, and I conclude in Section 5 with the policy implications.