6 Conclusion
In summary, at the EU level, average job tenure increased slightly from 116.5 months in 2002 to 123 months in 2012. As this observation period includes the Great Recession, cyclical factors are likely to be an important explanation: First, short-tenured jobs were disproportionately destroyed during the crisis; second, there was less job creation during the crisis, leading to a lower number of newly created jobs (Bachmann et al, 2015). This is in line with evidence from the US provided by the Bureau of Labor Statistics (BLS) which reported that median tenure increased between 2006 and 2014 (BLS, 2014). Consequently, mean tenure is characterised by a strong cyclical component that has to be separated from long-term trends.
However, at the individual country level, strong heterogeneities prevail before the Great Recession and in the reaction to the recession. Possible reasons for diverging pre-crisis levels of mean tenure include composition effects in terms of the workforce or industry structure, different labour market institutions or country-specific preferences of workers in terms of the ‘lifetime employment relationship’. In terms of labour market institutions, employment protection legislation, and the prevalence of temporary contracts and, more generally, labour market flexibility appear to be particularly important.
