Job stability in Europe over the cycle

• Date: 01 September 2018
• Published date: 01 September 2018
• Author: Rahel FELDER,Ronald BACHMANN
• DOI: http://doi.org/10.1111/ilr.12117

International Labour Review, Vol. 157 (2018), No. 3

Copyright © The authors 2018

Journal compilation © International Labour Organization 2018

Job stability in Europe over the cycle

Ronald BACHMANN* and Rahel FELDER**

Abstract. This article investigates the evolution of job tenure for the time period

2002–2012 using microdata from the European Union Labour Force Survey

(EU-LFS). Overall, the data show a slight increase in average job tenure at the EU

level, which can be explained by disproportional lay-offs of short-tenured workers

during the crisis. When controlling for changes in the demographic composition

of the workforce, an underlying negative trend in mean tenure becomes visible.

Job tenure evolved very differently across the EU before and during the crisis,

highlighting the importance of institutional frameworks, especially employment

protection legislation (EPL).

Changes in the economic environment over recent decades have led

to growing concerns about decreasing job stability. In particular, the

potential decline in the prevalence of jobs that last for a long period of time

(a “job for life”) has been intensively discussed in both academic research

and the media (Hall, 1982, is a seminal article in this regard). The fear is that

globalization and technological progress, such as advances in communica-

tion technologies, have induced changes in the labour market that require

employees to be more flexible. Workers have to adapt to more frequent tran-

sitions between jobs and intermittent spells of unemployment. These changes

in the labour market are likely to adversely affect job satisfaction and worker

well-being (European Commission, 2001). In this context, job tenure, or

the length of time a worker has been continuously employed by the same

* RWI – Leibniz Institute for Economic Research, Essen, Düsseldorf Institute for

Competition Economics (DICE), Heinrich-Heine-Universität Düsseldorf, and Institute for the

Study of Labor (IZA), email: ronald.bachmann@rwi-essen.de. ** RWI – Leibniz Institute for

Economic Research, Essen, and Ruhr-Universität Bochum (RUB), email: rahel.felder@rwi-essen.de.

This article is based on the project “Labour market transitions in turbulent times”, which

was carried out by RWI for Eurofound. We thank Donald Storrie and Carlos Vacas-Soriano from

Eurofound for their support at various stages of the project. We also thank Hanna Frings, Matthias

Giesecke and Sylvi Rzepka for their helpful comments and suggestions. The views and opinions

expressed in this article are those of the authors and do not necessarily reflect the official policy or

position of Eurofound.

Responsibility for opinions expressed in signed articles rests solely with their authors, and

publication does not constitute an endorsement by the ILO.

482 International Labour Review

employer, is of paramount interest to workers, as it can be interpreted as a

measure of job stability (Neumark, 2000).

Aside from long-term trends, evidence suggests that the labour turno-

ver rate was strongly affected by the recent financial and economic crisis, with

potentially detrimental consequences for job tenure (Bachmann et al., 2015).

This crisis has led not only to a large and persistent increase in unemployment

in many European countries, but also to a divergent development in labour

markets across the European Union (EU). Since turnover is closely related

to the length of time workers spend in a job, it is expected that the Great

Recession also had an effect on job tenure.

Against this background, our article analyses the evolution of job tenure

as measured by the length of uncompleted employment spells with the same

employer1 for the time period 2002–2012 for a large number of European coun-

tries using worker-level data from the European Union Labour Force Survey

(EU-LFS). In so doing, we provide evidence both for longer-term trends and

for recent developments which took place during the Great Recession. In par-

ticular, we provide aggregate evidence on the evolution of job tenure at the

European level, but also for specific countries. Furthermore, the richness of

the EU-LFS data allows us to analyse heterogeneities with respect to both

worker and job characteristics. Finally, we analyse cross-country differences,

pointing out the important role played by institutional frameworks, in particu-

lar when in the form of employment protection legislation (EPL). This study

thus fully exploits the richness of the microdata from the EU-LFS from 2002 to

2012 in order to examine how job tenure evolved across EU countries during

the pre-crisis and crisis periods.

Our analysis is related to several strands of the literature. Job tenure was

analysed in an international context by Auer and Cazes (2000) and Cazes and

Tonin (2010). Both studies, the first one for the 1990s and the second for the

period from 1996 to 2006, find that mean tenure remained relatively stable in

most European countries and increased only slightly in a few countries during

the observation period. However, the authors report pronounced differences

in levels between countries, which they attribute to heterogeneous labour mar-

ket institutions and workers’ labour market behaviours. Examining data for

the mid-1990s for eight EU countries, Japan, the Russian Federation and the

United States of America, Burgess (1999) found that the United Kingdom and

the United States had relatively low levels of tenure in the working relation-

ship. These results suggest that tenure is generally low in countries that are

characterized by flexible labour markets. Additionally, results from Burgess

(1999) suggest that EPL has a positive effect on mean tenure. Furthermore,

Boockmann and Steffes (2010) find that labour market institutions play an

important role in reducing mobility and thus prolonging tenure.

1 This definition comprises accumulated durations of subsequent jobs at the current firm

and is sometimes referred to as “employment tenure” in the literature (for instance, by Auer and

Cazes, 2000).

Job stability in Europe over the cycle 483

The relationship between tenure and job characteristics such as the prev-

alence of temporary contracts has also been examined. This is of paramount

interest in the context of job stability because temporary contracts are encoun-

tered with increasing frequency in the EU. However, this does not necessarily

imply any immediate effects on mean tenure, since temporary contracts are

designed differently with respect to termination time and contract renewals

from one country to another. Auer and Cazes (2000) detect no clear pattern of

association between temporary work and job tenure.

The literature identifies different patterns and trends before the crisis in

the relationship between socio-demographic characteristics and mean tenure,

and these hold true for the majority of EU countries. Cazes and Tonin (2010)

show that young workers do not experience a systematic decline in tenure

over time, except for those in Central and Eastern European (CEE) countries.

After controlling for age, however, they report some reduction in average ten-

ure in a majority of EU countries. Against this background, our study explicitly

analyses the role played by demographic change in relation to tenure.

As mentioned above, there is detailed research on job tenure in the pre-

crisis period across European countries. However, the impact of the Great

Recession on job tenure has not been examined for European countries.

Studies for the period before the Great Recession show that job tenure behaves

countercyclically and moves with the unemployment rate (Auer and Cazes,

2000). It decreases during an economic boom, when unemployment falls and

job creation increases, leading to new hires and voluntary job-to-job transition

(for the United States, see Shimer, 2005; for Germany, see Bachmann, 2005).

The opposite is found during a recession, when exit flows from employment

increase as firms dismiss workers. As a result, unemployment rises, and job

tenure tends to increase (Eurofound, 2014). Importantly, workers are affected

differently along the tenure distribution, with those who have little seniority

being more likely to lose their job during a recession than high-tenured workers

(Abraham and Medoff, 1984; Jovanovic, 1979).

Long-run trends and more recent developments such as the Great

Recession may have adversely affected the job tenure of different subgroups,

meaning that analysing tenure at the aggregate level may mask changes for

subgroups of the population. In order to reveal variation in job stability within

each group, we additionally investigate the evolution of job tenure for different

worker groups and job types. We analyse changes both at the EU aggregate

level and for specific countries, initially focusing on a comparison of mean job

tenure across countries and subpopulations. Our study therefore contributes

to the literature by creating a complete picture of the changes in job stabil-

ity across countries and subgroups, taking both a longer-term perspective and

looking at its evolution during the recent financial and economic crisis.

The remainder of this article is structured as follows. The next section

provides a brief description of the EU-LFS data set. The third section displays

an overview of the aggregate evolution of job tenure with respect to both

trends and cyclical features for the EU at the aggregate level and for specific