The Belgian Context

Traditionally, social protection in Belgium has been developed according to the corporatist model, as in Germany, France or Austria. The Bismarckian model has been in place since the Belgian government established its modern social welfare system in December 1944. The model is based on a community of reference that focuses on the worker, who has to be protected from the vicissitudes of life – and above all from temporary or permanent exclusion from the workforce. Therefore, the Belgian model is based on a generous social welfare system that is closer to the Scandinavian universalist model in some aspects. However, it is also characterised by a relatively low level of decommodification and defamiliarisation (Esping-Andersen, 1999). The welfare system is undoubtedly generous, but it also favours the model of the male breadwinner by providing assistance differentially according to gender: cohabitants, predominantly women, systematically receive less generous benefits than heads of household, predominantly men.

In the early 1970s, this model was affected by the economic crisis. It was then shaped by the politically motivated desire to encourage early exit from the labour market in order to support the hiring of young workers (Sanderson, 2015). A new norm emerged, with retirement occurring earlier and earlier (Sanderson & Burnay, 2017). However, since the early 2000s, the public sector management of the end of working life has aimed at increasing employment rates amongst older workers. The argument put forth is primarily economic, based on the sustainability of the social welfare system. Indeed, employment rates for seniors in Belgium are still amongst the lowest in the industrialised countries, although they have increased significantly in recent years (primarily due to the increase of professional activity amongst women).

In 2018, the employment rate amongst workers aged 55 to 64 years was 50.3% in Belgium, compared to 58.7% for the European Union as a whole (Eurostat, 2019). Moreover, the Belgian employment market is also characterised by a high proportion of part-time work at all stages of professional life, including at the end of the career. This is particularly true for women: in 2015, one out of two women aged over 50 years in Belgium works part time compared to 13.3% for men in the same age group. The measures taken by the federal and regional authorities in recent years regarding the end of working life are directly influenced by this objective of increasing employment rates: those aim indeed to encourage workers to remain on the labour market (e.g. through incentives) or to reduce the legal possibilities to leave the labour market prematurely, i.e. before the legal retirement age (65+) (Table 10.1).

Table 10.1 Evolution of employment rates of 55–64-year olds in Belgium from 2014 to 2018
Full size table

The aim of this chapter is to show how the recent measures taken by the government, especially to increase the employment rate for older workers, do not fundamentally change life courses in Belgium. Both in terms of early withdrawal from the labour market and extending the working life, Belgium is still characterised by a standardisation of life courses and, recently, by a form of de-institutionalisation of life courses. The main hypothesis is that standardization has been maintained as a result of factors which lead to early retirement and to the extension of working life in Belgium. This article is constructed into two empirical parts. The first part highlights the factors that lead to early retirement. The second part attempts to determine the profiles of workers who continue their professional activity after the legal retirement age. These two parts will allow us to return to the notions of standardisation and institutionalisation of life courses.

Evolution of the Belgian Institutional Model: Towards a Form of De-institutionalisation of Life Courses?

Belgium is historically characterised by a high degree of institutionalisation and standardization of life courses. According to Cavalli (2007: 61), standardisation refers to the degree of regularity shown by idiosyncratic life courses, especially with respect to the timing of major life experiences; institutionalisation refers to the structuring of lives by social institutions, the state and its policies. It is through a strong institutional context that life courses are structured, clearly defining what is possible. In fact, this institutionalisation by federal and regional authorities leads to a form of chronology of individual lives by establishing ages of transition from one state to another (particularly compulsory schooling and retirement ages) with relatively precise ages at which these events occur (Guillemard, 1986). Indeed, for Mayer and Schöpflin, “in the Welfare State, the continuous flow of life is transformed into a series of situations all of which have a clear formal definition… Periodisation of life and proliferation of sharp transitions which derive from the social insurance system combined with a lifelong biographical pattern” (1989: 198).

In Belgium, all the legislative measures are based on the maintenance of defined age thresholds, in particular for older workers. For example, in his government statement of 14 October 2014, the new Belgian Prime Minister, Charles Michel, announced an increase of the retirement age to 67 in Belgium by 2035. This measure is therefore clearly part of a form of the institutionalisation of life courses: the legal retirement age is being increased but remains effective. Indeed, traditionally in Belgium, the legal retirement age means a complete termination of professional activity: the age at which it takes place is a real cut-off point. However, still in 2015, extending employment beyond the statutory retirement age was possible. Until then, there were some possibilities to retire early, but under such conditions that few workers were entitled to them: for example, above a certain income ceiling, they would lose part or all of their pension. According to Eurostat figures, people aged 65–69 who were still working represented only 3.7% in 2009 and 4.3% of all people in this age group in 2013, which was still far from the European average: in the 28 EU Member States, 11.2% of people aged 65–69 were still active on the labour market. However, this recent measure taken by the federal government that made working past the statutory retirement age possible did not have a large impact on the Belgian labour market since in 2017 only 5% of persons aged 65–69 years were still active (Conseil Supérieur du Travail, 2018). This rate is extremely low compared to other European countries. For example, in Sweden it is 23.4%, in Denmark 18.0%, in Germany 16.1% and in France 6.6%. Belgian older workers clearly do not take advantage of this measure to work beyond the legal retirement age.

Even more important than a statistical increase, this legal change breaks with the system of institutionalisation of life courses, since it allows, without restriction today, to continue to engage in a professional activity without really taking into account the legal retirement threshold. It is therefore not just a minor adaptation but also a real breakdown in the Belgian institutional paradigm.

From De-institutionalisation of Life Courses to De-standardisation?

The Belgian institutional system is accompanied by a form of standardisation of life courses leading individuals to experience the same events at the same ages (Cavalli, 2007). Even if professional trajectories are today increasingly interrupted, the normative representations associated with “a good job” remain attached to the unlimited-term contract.

In 1978, Smelser and Halpern developed the idea of the triangularisation of life between school, family and work. For them, the industrial period was built on a division of the stages of life into three periods: youth (a period of training/education), adulthood (a period of professional activity), and retirement (a period of rest after a lifetime of work). These three periods, with their underlying logic, have constructed lives and structured our daily rhythms in an important gender dynamic, most often focused on the masculine career with women being relegated to domestic work (Sapin et al., 2007). Moreover, they have defined our social policies by establishing the rights and responsibilities of citizens according to age thresholds, which are largely inspired by this division and are also based on a gendered vision of society (Vrancken & Thomsin, 2009).

While entry into the labour market occurs later and later, in particular due to the lengthening of education, career paths are increasingly marked by discontinuity, and the end of a career is above all characterised by early retirement. Therefore, and according to Kohli (1989), in terms of the transition to retirement, it is not flexibility that is required, but a longer transition period towards retirement: workers are released earlier but without questioning the triangularisation of the life course. This would then be more a matter of lengthening the transition period than a real blurring of temporal benchmarks, which would not therefore lead to forms of de-standardisation of life courses (Sanderson & Burnay, 2017). Indeed, de-standardisation can be analysed either by a form of complexification and diversification of individual life courses, making them less and less predictable (Widmer et al., 2009), or by the ‘disruption of the social clock’ with a change in the limits and a diversification of the ages at which transitions are experienced (Kohli, 1986). If, in Belgium, the end-of-career translates into diversified forms – preretirement but also permanent departure for health reasons or older unemployment (Burnay & Falez, 2009) – it is not a form of de-standardisation of life courses but rather the fact that people leave the labour market more quickly (Burnay, 2002).

From Early Exit to Extended Working Life

As discussed above, the standardisation of life courses is directly linked to the reasons for an early departure, both in its individual and family dimensions and in the public policies implemented, favourable (or not) to early retirement. Individual and family reasons refer either to a desire to leave the labour market prematurely or to constraints. For the majority of workers, this premature exit is perceived above all as a positive transition that responds to personal, family or social aspirations: the spouse’s employment situation, involvement with children or grandchildren, free time for leisure activities (Hardy & Hazelrigg, 1999; Schultz et al., 1998; Szinovacz & De Viney, 2000).

Financial constraints and health condition also appear to be good predictors of early retirement (Adam et al., 2002; Barnes-Farrell, 2003; Feldman, 1994; Walker, 1985; Gratton & Haug, 1983, Friedman & Orbach, 1974; Parker, 1980; McGoldrick & Cooper, 1980, Saurel-Cubizolles et al., 1999, Barnay & Jeger, 2006). The greater the financial constraints, the longer older workers stay on the labour market: in this case, the loss of income due to retirement cannot be borne by the household, forcing workers to remain in employment. A deterioration in health status contributes to the exclusion from employment, especially when a career path has been characterised by heavy and arduous working conditions (Lund & Borg, 1999). Health problems are thus one of the most important reasons for withdrawal from the labour market, either through specific sickness-related schemes or disability schemes (Molinié, 2006).

While older workers can leave the labour market prematurely, they can also, with the new legal measures (2015), remain in employment after the statutory retirement age, through the bridge employment, defined as any kind of paid employment (e.g., part-time, full-time, or self-employment) that employees engage into after they retire from a career job (Beehr & Bennett, 2015: 113). In the Belgian context, this bridge employment can therefore be defined as the transition period between the statutory retirement age and the definitive cessation of activity. Those recent changes to the Belgian system make it possible to broaden the possibilities of bridge employment.

According to Beehr and Bennett (2015), there are two major explanations for the development of this transition period. The first refers to the increase of life expectancy; the second to the disastrous consequences in some countries (including the United States) of the financial crisis of 2007–2008. Given these two conditions, bridge employment has been more perceived as necessary to ensure a decent standard of living (Maestas, 2010).

However, this bridge period can also be an opportunity to develop new skills and new career opportunities (Deal, 2007). Aspirations and motivations, but also health status (Kerr & Armstrong-Stassen, 2011), can be defined as push or pull factors. Family situations can also play a role in the desire to continue one’s professional life. Social roles, especially traditional ones, are primarily a means for men to pursue their professional career (Wang et al., 2008); the expression of their social identity is more likely to be achieved through work than for women.

The same indicators can explain both early exit from the labour market and the extension of working life. A major question is therefore whether decisions to retire early or to extend the working life are based on choices or constraints. The answer to this question must of course take into account, directly or indirectly, individual, family, social or cultural reasons.

Data

The data used to explore retirement patterns are from the ‘Gender and Generation’ survey carried out in Belgium in 2008. This survey is part of the large programme of surveys developed by the UNECE (United Nations Economic Commission for Europe) and includes a number of European countries, such as France, Germany, Italy, etc. For Belgium, 7163 interviews concerning individuals living in private households and aged between 18 and 79 were conducted. When only the persons belonging to generations where all the members are no longer (or not at all) active on the labour market are considered, this represents a total of 1498 persons declaring themselves to be (pre-)retired. This sample includes only those who have left the labour market. They were either employed or registered as jobseekers, whether or not they had opted for early retirement. These individuals left the labour market between the age of 39 and 73, with a median age of 60 and a mean age of 58.3.

Several questions concern the departure from the labour market, including the year of departure and the reasons for it. Other questions deal with the existence of a complementary activity, which were asked regardless of the status of the respondents. Finally, these questions can be linked with the socio-demographic characteristics of the persons concerned. The survey makes it possible to apprehend the situation before 2015 when the measures relating to the maintenance in employment of senior citizens were taken. It is also possible to analyse the social desirability of these transformations and to determine whether they fulfil an expectation in the population.

Different statistical methods were used. First, some basic descriptive analyses are presented in order to characterise the sample. A binary logistic regression was then carried out to highlight the characteristics of individuals according to their motivations for retirement.

The choice of binary logistic regression was justified by the fact that the dependent variable (the motivation for retirement) is a nominal variableFootnote 1 that could be decomposed into several binary variables (see Table 10.2). This work on the variable was intended to determine the extent to which specific profiles corresponded to each motivation. Three models were constructed: the first is based on respondents whose retirement had been constrained; the second, those whose departure was a deliberate choice; and the third, those who had left because they had reached the statutory retirement age. The explanatory variables were selected on the basis of the findings in the literature: in addition to age, gender and level of education (Wang et al., 2008), the presence of children aged 14 or over in the household, the presence of a spouse in the household and the characteristics of the last profession were added (full-time or part-time, type of employer, type of profession) (Hardy & Hazelrigg, 1999; Schultz et al., 1998; Szinovacz & De Viney, 2000).

Table 10.2 Reason for leaving the labour market
Full size table

Results

Leaving the Labour Market: Between Constraint and Opportunity

Amongst the 1498 retirees, almost half report that they took voluntary early retirement (in other words, they left the labour market before reaching the legal retirement age), 18% report that they had to leave their job because they had reached the legal retirement age and 16% report that they were forced to take early retirementFootnote 2 (either their business went bankrupt and they were encouraged to leave the labour market, or they were pushed out by their managers who wanted to downsize or rejuvenate their workforce…).

These three categories were built from the responses given by respondents to the question: “What was the main reason why you stopped doing this activity?”. These first results highlight the emergence of early retirement as a norm, as it is perceived more as an opportunity than a constraint. The construction of early departure as a social norm was therefore opposed to policies in Europe aimed at keeping workers in the labour market as long as possible.

In the following analysis, the reasons for departure were clustered into three categories: forced departures, departures at the legal age and voluntary departures. In order to construct profiles by category, three logistic regressions were carried out, using as the main characteristics those retained in the literature: family, social and demographic characteristics.

Overall, only the first two models give significant results. The third (retirement at the legal age) is more difficult to interpret and the explanatory variables used for each of the models give poorer results in this case. The models were applied to generations that had reached age 65, which made it possible to retain only cohorts that had left the labour market.

The following results were obtained through a model-by-model analysis.

  • For the constraining factors model:

It concerns more men and younger people; conversely, it is experienced less as a constraint if the person has a low level of education, works full time, has blue-collar status and works in the public sector.

  • For the choice model:

It concerns more women, households with young children, part-time workers in their last job and older people.

These results highlight various factors that contribute to the transition to retirement:

  • working conditions or, more precisely, the status of the person (a less qualified person will more often find himself or herself in less valued subordinate jobs);

  • a part-time worker would experience her or his departure from the labour market less as a constraint than as a deliberate choice. In this sense, either the move to part-time work could be a way of preparing for retirement, or part-time work would reflect a lower attractiveness of work (in particular, through a lack of consideration or interest in the profession practised);

  • Contrary to the analyses conducted by Szinovacz and De Viney (2000), the family environment seems to have little influence on the decision, except for individuals choosing to retire, whose choice is influenced by the presence of a child;

  • Finally, gender is an important factor in understanding how early departure is experienced: women see it more as a deliberate choice, men more as a constraint. This result refers to the social construction of gender and the symbolic role of each individual (Table 10.3).

Age is an interesting factor to analyse. Even if age is not significant in the first model, model 2 and model 3 show the impact of legal measures and the impact of changes in employment policies effected for more than 20 years in Belgium. Data come from a 2008 survey, just at the beginning of policies that extended the working life.

Table 10.3 Profile of respondents by types of reason for leaving the labour market
Full size table

In model 2, voluntary early departures can be found amongst the oldest generations, those who could have benefited from early departure policies. Model 3 confirms this interpretation. Retirement at the legal retirement age would have been more common amongst the younger generations in the sample. However, this result should be nuanced: the average age at which people leave the labour market has steadily declined from 60 years for the oldest generations to 54 years for the most recent generations, as noted in recent work (Bourguignon et al., 2017). Moreover, amongst those who reported that they left work at the legal age, only 44% left the labour market at around 65 years of age. The others left the labour market earlier. Indeed, it is questionable whether, in model 3, some respondents report leaving at the legal age in the sense of the age at which they could apply for early retirement. This behaviour would confirm that early retirement has become the norm (Sanderson & Burnay, 2017). As a result, these individuals would have come to confuse the statutory retirement age with the age at which the person could apply to leave the labour market.

What About Bridge Employment in Belgium?

Bridge employment was at the centre of the reforms carried out in 2015 when the Belgian government took measures to facilitate and encourage the maintenance of a complementary activity beyond the legal retirement age. Data analysed in this chapter were collected before these measures were implemented. Based on the results presented in the previous section, the hypothesis is that the measures taken in 2015 did not correspond to an aspiration of the workers.

The hypothesis can be verified from the number of workers retaining a complementary activity (within the limits provided for by law in 2008) after retirement. The higher this number, the more significant the bridge employment will be. The very low number of workers involved in the labour market after retirement threshold has confirmed this hypothesis: only 12 out of the 1498 people in the sample kept a complementary activity after the statutory retirement age.

This result was confirmed by the intention to develop a professional activity within 3 years, from a specific question included in the GGP survey. Out of the entire sample, 4.7% of the respondents indicated that they intend to develop an activity to supplement their retirement. This percentage rises to 8.9% amongst early retirees under the age of 65. There is a clear overrepresentation of men (66%) in this group, confirming the presence of gendered social roles in which men invest in their professional activity more than women.

These figures concerning bridge employment are low and probably very far from the objectives of the Belgian government to extend the working life. However, they could increase both in terms of intention and in terms of people in supplementary employment due to new constraints such as cost-cutting measures taken on pensions and the weakening of career paths. These results confirm the analyses of Maestas (2010) on bridge employment, since it is utterly a response to a constraint. However, if we examine the socio-demographic profile of retirees wishing to return to work, 69% of them have a very high level of education, i.e. those who probably held the most remunerative and rewarding jobs. Consequently, in 2008, the people concerned by bridge employment were probably not people in need of a supplementary income but rather people seeking to maintain a certain status acquired through their professional activity, contrary this time to the observations made by Maestas (2010). For the time being, therefore, the Belgian results do not fit in with a logic of necessity, but rather with a dynamic of choice.

Discussion and Conclusions

Traditionally, Belgium is characterised by a high level of institutionalisation of life courses. Thresholds are fixed to separate beneficiaries from other categories. However, recent measures run counter to this tradition. By relaxing the rules on working after retirement, the Belgian government is giving less weight to the legal retirement age and thus moving away from a form of institutionalisation of life courses.

This kind of de-institutionalisation does not lead to a form of de-standardisation of life courses. Thus, life is still based on a tripartite model where adulthood corresponds to a time of work and retirement to a time of rest. Although the boundaries between these times have shifted in recent decades, they remain present and meaningful for understanding life courses. Both in the analysis of the reasons for leaving the labour market and in the analysis of bridge employment, the results show the extent to which end-of-career life courses continue to be standardised.

These results show the strength of the social norm built around early retirement and not around extending the working life. Nearly one out of two retirees reports that their withdrawal from the labour market is voluntary, even though they have a specific profile: voluntary retirement would be more feminine, would involve more people working full-time, with children in the household, while voluntary retirement would be more masculine.

The norm for early retirement remains strong, even though public policy has changed dramatically over the past 20 years. There is therefore a mismatch between policies to retain older workers on the labour market and older workers’ willingness to leave the labour market prematurely. The extension of working life is a political reality, but one which is still not endorsed by older workers.

These first analyses on bridge employment show how career extension does not seem to be a concern for Belgian workers. The measures decided in Belgium in 2015 are still too recent to be able to assess their impact. They will have to be analysed and evaluated in the coming years in order to determine the extent to which they have changed retirement practices, reinforcing a form of de-institutionalisation already under way. However, it is already apparent how much the political will runs counter to entrenched cultural practices, which will make change difficult.