The participation of older workers in the labour market is a key concern as it points to important economic issues in terms of demography and the financing of unemployment compensation and pensions.
When analysing the labour market participation of older workers in France, two major periods may be considered. The first focuses on the period between 1970 and 1990 and is characterised by the Malthusian outlook regarding the division of labour in society. During this period, older workers were encouraged to leave the labour market through early retirement schemes to facilitate youth employment. Given the ineffectiveness of these Malthusian policies and the underlying economic issues, a second period emerged in the 2000s. This period was characterised by successive employment policies put into place to change how society was organised and encourage the return or retention of older workers in employment. For instance, older workers were allowed to combine professional activities and part of their retirement pensions. Other European countries put similar changes into place as early as the 1990s. This period showed an improvement in labour market indicators among older workers. The rate of employment of workers aged 55-64 years increased by 10 points between 2003 and 2011 to stand at 42%. In 2013, this rate reached 45%. The Treaty of Lisbon advocated a senior employment rate of 50% among the population aged 55-64 by 2010. Compared to neighbouring European countries, the employment rate of older workers in France was low.
We assume that the low increase in the employment rate of older workers has been the result of two obstacles: labour supply and labour demand. Among the obstacles linked to labour supply are higher costs of labour, lower productivity and approaching retirement synonymous with short-term employment. Among the obstacles linked to labour demand are the constraints of some institutional measures, the stereotypes associated with older workers by virtue of their characteristics in professional situations, and the discriminatory behaviour arising from stereotypes. Moreover, based on the number of complaints (according to human rights advocates), age discrimination has increased and has become the third criterion of discrimination since 2005 despite the policies put into place to raise companies' awareness of the value of older workers in the labour market.
This chapter is divided into four sections. The first section provides an in-depth analysis of the changes that have occurred in the labour market participation of older workers in France over the long term. These changes, notably the low rate of senior employment in France compared to other European countries, can be explained by different factors that result from labour supply on the one hand, and the defined labour demand on the other. The second section measures the employment gap between the senior population and the median age of the population using a model that evaluates the probability of being employed based on data from French labour force surveys. The determinants of employment are evaluated based on age. In the third section, we measure age discrimination in employment using data from surveys. We use an extension of the Blinder-Oaxaca decomposition technique for binary dependent variables: the Fairlie decomposition method. This method enables us to respond to the following question: Can the differences between labour market participation and age be explained by individual characteristics or do they result from residuals including age discrimination? Based on the measurement of the unexplained portion of the decomposition--to which discrimination belongs,--the final section focuses on the link between the evolution of this unexplained portion and a business cycle indicator.
Labour market participation of older workers over time
Depending on the different conceptions of employment policies, the labour market participation of older workers has undergone many changes over time. The two periods that questioned the role of older workers in the labour market were both accompanied by public policies oriented towards their respective objectives. Two types of policies can be used: These can be perceived as "structural labour market interventions that seek either to increase the level of employment in the economy and/or to improve the functioning of the labour market (so-called active measures), or to alter the negative effects of unemployment on wellbeing (so-called passive measures)" (Erhel, 2005).
The years between 1970 and 1990 were characterised by an "early retirement culture" which undertook a specific organisation of professional activities seeking to adjust the employment of older workers in the event of economic changes. The retirement age initially set at 60 was therefore extended to the 55-59 population: between 1981 and 1984, the employment rate among men decreased from 80% to approximately 60-65%. This organisation was made possible by a social consensus between economic stakeholders and a relatively negative collective representation of seniors perceived, for instance, as less adaptable and less productive in the face of difficult working conditions. In the early 1980s, there was a significant increase in unemployment (a 4-percentage-point rise and a rate of 10% in 1984). Early retirement measures were seen as a means of protecting those in the median age of the population. An early and compensated retirement was thus possible for those aged between 55 and 59 years of age (Brun-Schamme & Janod, 2007). It seems, however, that the mechanism through which the early departure of older workers was to enable the access of youth to the labour market failed to achieve its objective. Indeed, the rate of youth unemployment developed inversely to the rate of older workers' employment across the period 1970-2002 (D'Autume et al., 2005). Over the last four decades, there have been more than ten early retirement mechanisms (Appendix 1).
During the 1990s, a number of changes in the business cycle led to the reversal of employment policies. These included weak economic growth and an increase in the dependency ratio, i.e., the number of those in the labour force in relation to the number of retired persons decreased when baby boomers went into retirement. An increase in life expectancy also meant that people could remain in employment for longer periods. Moreover, with a reduced labour force, the number of contributors to the pension scheme fell yet pension payments increased. Given this context, it was thus necessary to increase the labour force by promoting the return of previously excluded groups. European policy also developed in this direction. In 2000, the Treaty of Lisbon advocated, in the member states of the European Union, a senior employment rate of 50% among the population aged 55-64 by 2010. As a result, France implemented several pension reforms (the Balladur law of 1993, the Fillon law of 2003 ...). The age of entry into early retirement was raised; to obtain a full pension, workers needed to contribute longer.
These policies relied on a legal framework based on two fundamental texts: The National Inter-professional Agreement of 13 October 2005 on "the employment of older workers to promote their retention and return to work" and the concerted National Action Plan for the employment of senior workers between 2006 and 2010, which applied the agreement. This plan reflected the three key issues facing French society: one of the lowest older workers' employment rates among industrialised countries; a sharp increase in the average age of the labour force; and high rates of long-term unemployment among older workers because of a shrinking working-age population from 2006. Given this context, the agreement planned to establish measures to link economic performance and age management. The objective was to show companies that it was in their best interests to associate the two measures. Title II of the agreement addressed career security through the proactive management of jobs and skills (Article 4), professional interviews during the second part of individuals' careers (Article 5), the organisation and improvement of working and employment conditions (Article 6), training approaches (an individual's right to training (DIF), training plans, personal skills assessment, accreditation of past professional experience (VAE), orientation period) and the transmission of knowledge and know-how (Article 10). The agreement also planned to address seniors' return to work by entirely phasing out the age criterion from job opportunities, encouraging professional development contracts for older workers, establishing fixed-term contracts targeting them, etc. Lastly, end-of-career management was also addressed; employees could request a retirement appraisal and the re-organisation of working hours and of the end-of-career in the event of recognised hardship.
1.2. Grounds for the reluctance to hire older workers
There are many non-discriminatory factors behind the reluctance to hire older workers. Some of these are inherent in labour supply and are based on individual behaviour in labour market participation. Labour market regulations conducted by government can modify these behaviours. Other factors associated with labour demand --and within which discriminatory behaviours fall--illustrate the social consensus that has existed among economic stakeholders on the early retirement of older workers from the labour market since the 1980s and the misconceptions faced by these workers. This social consensus has persisted in the behaviour of companies where discrimination is residual. Discrimination occurs when there is unequal treatment (for instance, in the access to employment) between two people with identical productive characteristics because of a criterion identified as a recognised source of discrimination.
1.2.1 Labour supply