1 Introduction

Many countries have carried out over the past decade a series of reforms and measures to encourage longer working lives and to respond to the looming challenges of rapid population ageing. But have these steps gone far enough and have the necessary measures been taken? Much of the focus of this policy action has been on old-age pension reform but, as stressed in the report Live Longer, Work Longer (OECD 2006), a more comprehensive set of reform may be necessary to encourage work at an older age. This includes policy action in three broad areas to: (1) reward work, (2) change employer practices, and (3) improve the employability of workers. This agenda of age-friendly employment policies and practices has emerged from a major multi-country review conducted by the OECD in 2003–2005, which, for each participating country, sought to identify the key obstacles to working at an older age and the key policy measures that should be taken to tackle these obstacles. A follow-up review launched in 2011 allows assessing the extent to which countries have implemented the OECD’s specific policy recommendations.

Our paper is focused on four countries (France, The Netherlands, Norway and Switzerland) which participated in both reviews (OECD 2003, 2004, 2005a, b, 2013a, 2014b, c, d). Its aim is to provide an overview of recent policy initiatives to give older people better work incentives and choices implemented in these four countries as well as to identify areas where more could be done, covering both supply-side and demand-side aspects. The paper is divided into two sections. Section 1 presents recent labour market developments for older workers in the OECD area as a benchmark to assess the comparative performance of the four countries under review. Section 2 summarises the main lessons that can be drawn from the policy initiatives taken in these four countries since 2006.

2 Recent Labour Market Developments for Older Workers

2.1 The General OECD Context

In terms of employment, older workers weathered the global financial crisis far better than youth and prime-aged workers in most OECD countries (Fig. 1). This reflects a remarkable change. While older workers tended to early withdraw in large numbers from the labour market following major recessions in the 1970s, 1980s and early 1990s, this time round they have stayed in the labour force and even increased their participation since the start of the crisis (OECD 2013b). This is similar to the pattern observed in the aftermath of the (shallower) recessions in the early 2000s, suggesting it may be part of a longer-term trend.

Fig. 1
figure 1

Older workers have fared better than young and prime-aged workers since the crisis in the OECD area (unweigthed averages for 34 OECD countries and 21 European OECD countries). Source: OECD estimates based on national labour force surveys

Full size image

The considerable job losses experienced by youth during the crisis has raised concerns that the larger numbers of older workers staying on in the labour market may be crowding out youth from employment. However, an econometric analysis using data for 25 OECD countries over the period 1997–2011 shows that, on average, increases in the employment rate of older workers are either associated with increases in the youth employment rate or have no impact at all (OECD 2013b). Moreover, there is no evidence that the relationship between older worker employment and youth employment is significantly different during periods of recession, or that it has changed since the onset of the financial crisis.

Focusing on employment rates for older workers by age group, gender and education helps to disentangle composition and cohort effects behind aggregate trends. Across OECD countries, well-educated people increasingly work longer than the less-skilled (Fig. 2). The gap is part of a deepening divide between those who are well-educated and well-off and those who are unskilled and poor, which cuts across all age groups (The Economist 2014). Irrespective of age, women also face dimmer job prospects than men.

Fig. 2
figure 2

Large socio-demographic disparities in employment, workers aged 50–69, OECD countries, 2011. Source: OECD estimates based on national labour force surveys

Full size image

Employment and retirement decisions of older people are driven by a mix of supply- and demand-side factors. To date, the majority of studies of these decisions have focused on supply-side factors (see an overview in Blöndal and Scarpetta 1998 and Duval 2003). Supply-side factors comprise the generosity or the ease of access to replacement income in the event of leaving employment, including old-age pension schemes, early-retirement schemes, disability benefits, unemployment benefits and social assistance benefits. In the past, early retirement schemes were often used, mainly in continental Europe, in the hope that this would open up jobs for youth. However, most studies suggest that such policies have been ineffective in creating jobs for youth (OECD 2013b). In general, early retirement pathways have been substantially reduced over the past decade, as a result of pension reforms and some tightening of eligibility criteria to—or outright phasing-out of—other social transfer programmes that had been used as alternative early labour market exit pathways in the past. But specific displacement effects between unemployment, disability and other welfare benefits that are used as alternative early retirement pathways, remain important and need to be evaluated.

2.2 Evaluation Studies on the Impact of Benefits Criteria on the Older Unemployed

A number of studies have recently been undertaken to evaluate the impact that changes in benefits parameters for the older unemployed have had on their labour market situation (see a broad presentation of these parameters in Venn 2012). Researchers generally employ the experimental evaluation methodology, which compares the situation of those affected by the change with those who are not affected. The main lessons to be drawn are the following.

First, reduced unemployment benefits for older persons have produced little in the way of immediate return to work in France, but yielded a medium term improvement in employment and earnings of “young seniors” in Switzerland. Fremigacci (2010) analyses the impact on transitions out of unemployment that resulted in France from the sharp cuts (from 60 to 42 months or from 45 to 36 months depending on the period of past contributions) in maximum benefits duration for unemployed persons aged 50 and over, introduced by the 2003 reform. He uses official data from the National Employment Agency and identifies two groups, pre and post-reform. He finds a significant structural effect through the displacement of jobseekers from unemployment insurance (UI) to social assistance when their UI benefits run out. The effect on transitions to employment is, on the contrary, quite modest. Degen and Lalive (2013) examine how the 2003 reduction in the duration of unemployment benefits from 24 to 18 months affected jobseekers younger than 55 years in Switzerland. They find that this reduction increased employment and earnings for jobseekers aged 50–54 years not only in the first 24 months after entering unemployment but even up to 50 months later. This was not the case for the control group, unemployed persons aged 55 and over. The positive effects were particularly pronounced among jobseekers who were previously employed in R&D-intensive industries and whose previous occupation involved the acquisition of cognitive skills. These findings suggest that the positive effects of reducing the depreciation of human capital and the stigma of long-term unemployment among older jobseekers outweigh the negative effects of a lower reservation wage.

Second, extending UI for older persons with facilitated access to other early-retirement provisions has had little effect on re-employment in Austria. Inderbitzin et al. (2013) analyse the impact on take-up of different welfare benefits (early retirement, disability and social assistance) flowing from more generous unemployment benefits for older workers. In Austria, between 1988 and 1993, the duration of UI benefits was increased in some regions from 1 year to 4 for persons aged 50 and over. There is programme complementarity (sequential take-up of unemployment and disability benefits) among the different benefits for workers from 50 years up to retirement age, as well as programme substitution, but only for workers aged 55 and over, who are less likely to claim disability benefits. The authors conclude that the best approach is a set of measures that allow for co-ordinated reforms in unemployment and disability insurance funds, together with incentives for firms to hire and retain older workers.

Third, action on several early-retirement provisions has increased employment in Finland. Kyyrä (2010) follows the career paths of workers aged 51 born between 1941 and 1948 and analyses the joint effects of several reforms to the eligibility age for three early-retirement provisions: the “unemployment tunnel”, the “disability pension”, and the “part-time pension” (a subsidy for persons who switch from full-time to part-time work). The “unemployment tunnel” or “disguised early retirement” in Finland means that after receiving UI for the maximum period of 24 months, an older unemployed person is entitled to an unemployment pension until retirement. The eligibility rules from the 1990s (53 years for the unemployment tunnel; 58 years for the disability pension; and 56 years for the part-time pension) have been changed: (1) the age of eligibility for the unemployment tunnel has been increased by 2 years, to 55 years (1997 reform); (2) the disability pension has been eliminated (2000 reform); and (3) the age of eligibility for the part-time pension was postponed by 2 years, to 58 years (2003 reform). In fact, these reforms have served to raise the average age of exit from the labour force by 3.9 months. The most important impact has to do with the disappearance of the disability pension (+3.4 months), compared with the impact of the unemployment tunnel (+1.3 months) and the delayed access to the part-time pension (\(-0.6\) months). While in the last case, the average exit age has declined somewhat, this has served, overall, to increase the number of hours worked by older persons who have continued to work full-time. However, the impact of each reform has not been uniform across different groups of older workers. In particular, the impact of changes in access and eligibility to the unemployment tunnel has been felt primarily by men in their early 50s working in the manufacturing sector.

Fourth, abolition of the job search exemption for the older unemployed has increased employment but also disability claims in the Netherlands. Hullegie and van Ours (2014) examine the impact of the job-search exemption that existed until 2004 in the Netherlands, for unemployed persons aged 57.5 years and over. They examine the job-finding rate of workers who became unemployed in 2001 between the ages of 55.5 and 57.5 years. They find a markedly negative effect at the exemption age and also ex ante before reaching that age. According to Lammers et al. (2013), the end of the exemption for persons aged 55–59 led to an increase of 6 % points and 11 % points, respectively, in the number of men and women who find a job within 2 years after becoming unemployed. On the other hand, this effect was accompanied by an increase of 4 and 9 % points, respectively, in the number of men and women receiving a disability allowance.

Fifth, more activation during unemployment produces a return to work after a delay in Sweden. Bennmarker et al. (2013) analyse the reform which reduced the duration of UI from 90 to 60 weeks for the older, long-term unemployed who had exhausted their benefits but which allowed them to continue receiving benefits if they agreed to participate in a work experience programme. The authors find that return to work increased with a certain delay, as the unemployed take some time to be fully persuaded to participate in active programmes. There appears to be no effect in terms of lower earnings, which suggests that the higher rate of return to employment results from more intensive job searching rather than any reduction in the reservation wage.

In short, a measure that is limited solely to unemployment compensation is not effective for returning the older unemployed back to work unless it is part of an overall strategy that covers all provisions for early exit. In fact, displacement effects frequently appear among the different provisions, and these can create a “tunnel” for older jobseekers until retirement, without helping them find a job. This effect is particularly noticeable in countries that are facing a structural deterioration of their labour market. On the other hand, steps to ensure a swift return to work for the older unemployed and to make the allowance truly conditional on participation in an active labour market programme (ALMP) would seem more effective.

Nevertheless, while a significant pension and welfare reform process has undoubtedly strengthened incentives to continue working at an older age, further action is still required on the demand side to ensure that those older workers who wish to work longer can do so (D’Addio et al. 2010). Indeed, recent research has tested for policy complementarities between supply-side reforms and demand-side efforts. In particular, Neumark and Song (2013) find that in the United States, stronger state age discrimination protections enhanced the effects of the social security reforms, complementing the supply-side incentives created by these reforms by reducing demand-side barriers. They find that stronger age discrimination laws increase hiring of those older workers who were trying to work longer in response to social security reforms and that these laws may make extended work lives more viable by boosting transitions to bridge or partial retirement jobs, in contrast to older workers simply staying a bit longer in their career jobs.

2.3 The Puzzle of the Age-Productivity Profile

To what extent are there employment barriers in firms such as age discrimination, seniority rules in wage setting, low training participation and employment protection rules? The evidence suggests that older workers encounter major difficulties in changing jobs or in finding a new one being unemployed. One of the main obstacles perceived by employers has to do with labour costs rising more with age than with productivity. However, the conclusions of recent empirical studies of older workers’ productivity are rather mixed on this point, although they do highlight the positive impact of older workers’ experience for maintaining productivity at work.

There are two important weaknesses in the studies conducted on the issue of older workers’ productivity: they deal only with older persons in employment, whose productivity is on the whole bound to be higher than that of older persons who have withdrawn from the labour market; and they are confronted with the difficulty of measuring productivity gaps between individuals within a firm. This explains why the conclusions from the most recent research differ and are highly dependent on the choice of measures. This is demonstrated by Bloom and Sousa-Poza (2013) in their summary review of this issue. Nevertheless, recent research has yielded some significant findings.

First, a bias against age could persist in innovative sectors with evolving skills. According to Aubert et al. (2006), older workers represent a smaller portion of the workforce in firms in France that are characterised by considerable technological change. Workers over 50 years also tend to be more concentrated in declining sectors, automatically finding themselves in firms that structurally destroy more jobs, rather than in firms operating in sectors that are currently dynamic. This “technological and organisational change bias” also appears within different skills categories. On the other hand, Romeu Gordo and Skirbekk (2013) show that, in Germany between 1986 and 2006, older workers who remained in employment were quite successful in adapting to technological change. They were more likely to be employed in tasks that demand intensive use of cognitive faculties, which may explain the significant increase in their wages over this period.

Second, for older workers, the more firm-specific their experience the lower will be their productivity if they change jobs or employers. The fact that older workers’ productivity relies primarily on skills specific to the firm may go a long way to explaining their low rate of recruitment into firms (Daniel and Heywood 2007).

Third, creating mixed teams of workers of different ages can boost the productivity of all workers. Zwick and Göbel (2013) analysed good practices in employee management in a panel of German firms over 1997–2005. One of their conclusions is that working arrangements need to be adapted to the capacities of older workers, for example by changing equipment to deal with auditory and visual problems or assigning them tasks that are physically less demanding. They find that as a result of such measures, older workers’ productivity is on average no lower than that of prime-age workers.

Fourth, adjusted working hours and specific training for older workers have no significant effect on the age/productivity profile. Zwick and Göbel (2013) find that these arrangements have not been adequately implemented in Germany. Thus, flexible working time has been used for older employees primarily as a pre-retirement measure, with the possibility of working full-time in order to take retirement earlier. Similarly, training opportunities have not taken into account the specific needs of older participants. It also seems that, in France, offering training to older workers is not sufficient in itself to overcome the bias against age (Behaghel et al. 2014).

Fifth, thanks to the experience acquired by older workers, the average age/productivity profile rose to age 60 in an automobile plant. Börsch-Supan and Weiss (2013) measured errors occurring on an automobile assembly line in Germany, according to age. They show that the experience of older workers and their capacity to optimise teamwork help them to avoid severe errors, in contrast to younger workers.

2.4 Effective Labour Force Exit Age

The effective labour force exit ageFootnote 1 can reflect the combination of supply-side and demand-side policy reforms and measures as well as changes in the characteristics of older workers in terms of improved education levels and health. In almost all OECD countries, the effective labour force exit age has declined substantially since the 1970s and despite an increasing trend since mid-1990, it has remains below the levels prevailing 40 years ago (Fig. 3). For men, it fell from 68.6 in the early 1970s to 64.2 in the 5 years to 2012, while for women, it dropped from 66.7 to 63.1.

Fig. 3
figure 3

Effective labour force exit age [the effective labour force exit age is derived from labour-force survey data. It is calculated as a weighted average of the exit age of each 5-year age cohort, starting with ages 40–44, and using absolute 5-year changes in the labour-force participation rate of each cohort as weights. The 5-year change in participation rates is simply the difference between the rate for each age group (e.g. 55–59) at the beginning of the period minus the rate for the corresponding age group that is 5 years older (e.g. 60–64) at the end of the period] by gender, OECD countries, 1970–2012. Source: OECD estimates based on national labour force surveys

Full size image

2.5 Comparative Labour Market Prospects

Labour market prospects for older workers have remained particularly dynamic over the past decade in Norway and Switzerland. By contrast, even if encouraging results have been achieved in France and in the Netherlands, these two countries, and particularly France, remain well behind the best achievers. Nevertheless, raising the employment rate of people aged 65 and over is a challenge in all four countries. In addition as also shown in the figures below, older workers are significantly affected by a risk of long-term unemployment, hiring opportunities after the age of 55 remaining particularly low, even in Norway and in Switzerland.

The overall employment rate for the 55–64 age group stood at 54 % on average in the OECD area in 2012, well above the European average (48 %) (Fig. 4a). France (44.5 %) is below international averages but while the Netherlands (59 %) is above, it remains well below the rate recorded by Switzerland (70.5 %) and Norway (71 %), both countries located among the very best OECD achievers with Iceland, New Zealand, and Sweden. Even during the crisis from 2007 to 2012, employment rose on in all OECD countries, except in eight countries (Iceland, Greece, Ireland, Portugal, Japan, United States, Spain and Slovenia). While the employment rate of the OECD area increased on average by just 2.9 % points, the Netherlands experienced among the largest increases among OECD countries after Germany (respectively, 9.8 and 10.2 % points).

Fig. 4
figure 4

Employment rates of workers aged 55–64 and 65–69, OECD countries (unweigthed averages for 34 OECD countries and 21 European OECD countries), 2007 and 2012. Source: OECD estimates based on national labour force surveys

Full size image

Employment rates after the age of 65 are generally higher in OECD countries outside Europe (Fig. 4b). The European average (11 %) in 2012 is well behind the OECD average (19 %). There is a large disparity in the employment rates for the 65–69 age group in the four countries ranking from 6 % in France, 13 % in the Netherlands, 20 % in Switzerland to 26 % in Norway, even if the increase from 2007 to 2012 was quite similar in the four countries (about 3–4 % points).

The unemployment rates for persons aged 55–64 are higher in 2012 than in 2007 (or at least equal) in all OECD countries, except in Germany (Fig. 5a). In 2012, among the four countries under review, France is the only one with an unemployment rate for that age group above the OECD average (respectively 7 and 6 %) and with a significant increase (2 % points) since the crisis. But importantly, the incidence of long-term unemployment for the same age group was higher in 2012 than the OECD average (47 %) not only in France (61 %), but also in the Netherlands (57 %) and in Switzerland (59 %) (Fig. 5b).

Fig. 5
figure 5

The older unemployed, OECD countries (unweigthed averages for 34 OECD countries and 21 European OECD countries), 2007 and 2012. Long-term unemployment or unemployed for 1 year or more. Source: OECD estimates based on national labour force surveys

Full size image

The unemployment rate only gives part of the picture about the labour market situation of older workers. A broader indicator of non-employment of older persons can be estimated by adding up the unemployed and the inactive willing to work and refer them as a proportion of the whole age group (Fig. 6).

Fig. 6
figure 6

People aged 55–64 neither in employment nor completely in retirement (this group is estimated by combining the answers to different questions of the survey on the main status and the willingness to work), specific OECD countries (unweighted average of countries shown), 2002 and 2012. Source: OECD estimates based on the European Union Labour Force Survey

Full size image

The size and the composition of the indicator “People aged 55–64 neither in employment nor completely in retirement’ are helpful to understand the alternative early retirement pathways in each country (Fig. 7). It also reveals a “reservoir” of older workers potentially available for the labour market. This finding is particularly useful in countries where the labour force is declining or where there is a shortage of workers like in Norway and Switzerland.

Fig. 7
figure 7

People aged 55–64 neither in employment nor completely in retirement [this group is estimated by combining the answers to different questions of the survey on the main status and the willingness to work. Inactive persons willing to work are decomposed in each inactivity category based on their self-defined status (i.e. whether disabled, (pre)retired or inactive for other reasons)], France (data for France do not allow the identification of disabled people who are so included in the “other inactive” category), Netherlands, Norway (data for Norway do not allow the identification of disabled and (pre)retired people who are so included in the “other inactive” category) and Switzerland (2001 instead of 2002 for Switzerland), 2002 and 2012. Source: OECD estimates based on European Union Labour Force Survey

Full size image

The size and the composition of the number of persons aged 55–64 neither in employment nor completely in retirement differs by country. The number of persons neither employed nor completely in retirement is particularly low in France. The reason for this is that the French tend to retire early and there is only a small group of retirees who want to work. In addition, the increasing number of older people registered as unemployed are often narrowly perceived by employers and society as in a kind of disguised early retirement and could be better activated to return to work. In the Netherlands, the “reservoir” of older workers potentially available for the labour market increased over the past decade. The main challenges are to enhance the labour mobility of older workers characterised by a comparatively low hiring rate and a high incidence of long-term unemployment. In Norway, the “reservoir” of older workers potentially available for the labour market decreased over the last decade, particularly among the older inactive willing to work. This could be partly related to the flexibility of the transition to retirement included in the 2010–2011 pension reform. In Switzerland, the “reservoir” of older workers potentially available for the labour market decreased over the last decade, particularly for the number (pre)retirees from the generous second pillar willing to work. The “reservoir” remains nonetheless important.

Indicators of retention and hiring can provide some information on the labour dynamics of older workers and the extent to which they remain in the same firm or are likely to be hired if seeking a new job (Fig. 8). Norway, Switzerland and the Netherlands rank among the countries with the highest retention rate after the age of 60, but their hiring rate of older workers is below the average.

Fig. 8
figure 8

Hiring [the hiring rate for 55–64 is the number of employees in the age group with job tenure of less than 1 year as percentage of the total number of employees aged 54–63 the year before] rates of older workers (55–64) and retention [employees currently aged 60–64 with tenure of five years or more as a percentage of all employees aged 55–59 5 years previously] rates after 60, selected OECD countries [unweighted average of countries shown], 2012. Source: OECD estimates based on national labour force surveys for retention rates and on the European Union Labour Force Survey for hiring rates

Full size image

The shift from work to retirement is not always abrupt and can go through periods involving a shift from full-time to part-time employment. It is particularly the case for women aged 55–64 in the Netherlands and Switzerland (Fig. 9). Even 16 % of men aged 55–64 work part-time in Switzerland, early retirees in this country having the opportunity to combine a wage and a pension without any penalty.

Fig. 9
figure 9

Full-time/part-time employment rates by gender for workers aged 55–64, OECD countries, 2012. Source: OECD estimates based on national labour force surveys

Full size image

Participation in lifelong learning is likely to have an important positive impact on the employability of older workers. Training participation declines with age, and in most countries, the age gap is relatively high. Many papers consider the question whether and why training measures for older employees are less effective than for younger employees. Zwick (2011) shows that, in Germany, training of older employees is less effective with respect to key dimensions such as career development, earnings, adoption of new skills, flexibility or job security. Older employees also pursue less ambitious goals with their training participation. An important reason for these differences during the life cycle might be that firms do not offer the “right” training forms and contents. By contrast, Picchio and van Ours (2013) in the Netherlands find that older workers who receive on-the-job training are more likely to keep employment.

Skills development is an essential policy instrument for keeping workers employed. Results from the new OECD International Survey of Adult Skills (PIAAC) show, in particular, that the less-skilled workers who have not received training to stabilise their cognitive capacities are more vulnerable to technological change (OECD 2013c). In addition, the link between skills and wages changes with age, partly reflecting skills acquired through work experience (OECD 2014e). For youth, the level of education is the biggest single contributor to the variance in young people’s hourly earnings, followed closely by generic skills and those related to specific fields of study and training. Information-processing skills appear to play only a minor role for this age group. For prime-aged and older workers, specific field of study skills account hardly at all for variations in hourly wages, while the contribution of information-processing skills grows in importance, eventually exceeding the influence of education. The contributions of education and generic skills increase slightly with age. Yet skills differences between generations are often large starting from a low base for the older generations, particularly in France compared with the Netherlands and Norway (Fig. 10)Footnote 2. Korea has made the greatest progress in terms of the skills of younger generations, in contrast to the United Kingdom which has seen none. The question that should be examined in future work based on the survey is whether older workers in fact left the education system with a mediocre level of skills, or whether their skills have deteriorated over the course of their working life.

Fig. 10
figure 10

Progress in literacy skills for two generations, selected OECD countries, 2012 (UK covers here only England and Northern Ireland). Source: OECD (2013c), OECD Skills Outlook 2013: First Results from the Survey of Adult Skills, OECD Publishing, Paris

Full size image

3 Main Policy Lessons from France, The Netherlands, Norway and Switzerland

This section takes a closer look at the progress made in France, the Netherlands, Norway and Switzerland to improve labour market prospects for older workers, and the extent to which these four countries have implemented measures put forward in the report Live Longer, Work Longer (OECD 2006). Three broad areas to encourage work at an older age are identified:

  • Strengthen financial incentives to carry on working: Enhance incentives to remain in work longer. Eliminate incentives to retire early. Ensure that unemployment, disability and other welfare benefits are not used as alternative pathways to early exit from the labour market. Give better options for phased retirement.

  • Tackle employment barriers on the side of employers: Monitor closely the effects of anti-discrimination legislation. Move ahead with abolishing mandatory retirement. Review pay practices based on age or length of service. Help protect employment opportunities not jobs. Encourage the social partners to promote best practice.

  • Improve the employability of older workers: Provide suitable training opportunities at all ages. Help private and public employment agencies give better assistance for older jobseekers. Improve the work environment.

This policy agenda was based on the main lessons that emerged from the 21 country reviewsFootnote 3 undertaken in 2003–2005 and published by the OECD under the series on Ageing and Employment Policies.

A summary assessment of the extent to which the four countries have followed the OECD’s recommendations is given in Table 1. The four countries have taken some action in the three broad areas, but in general, more could be done particularly to tackle barriers on the employer’s side. The key messages for these four countries developed in each country report Working Better with Age are summarised below.

Table 1 Summary of ageing and employment policies between 2006 and in 2014
Full size table

3.1 Strengthening Financial Incentives to Carry on Working

Priority has been given to this area. The four countries have implemented or adopted measures to strengthen financial incentives to carry on working at an older age. Recommendations such as to adjust the retirement age according to demographic trends, to reduce incentives to retire early and to reward work, even after the statutory minimum age of retirement, have been high on the policy agenda of France, the Netherlands and Norway. In Switzerland, the first reform of the public pension (first pillar) since 1997 is currently under discussion. These countries, except France for the older unemployed, have also tightened alternative pathways to early exit from the labour market, often combined with measures to improve worker employability through ALMPs and, in line with the OECD’s recommendations, to reinforce the “mutual obligations” of older jobseekers. Despite the job crisis that hit particularly France and the Netherlands since 2007, both countries have resisted pressure for reintroducing any formal early retirement measures, although France did allow since end-2012 more persons with long careers to retire early.Footnote 4

Except for Switzerland,Footnote 5 France, the Netherlands and Norway have carried out some forms of pension reform in the first pillar since the 2006 review was undertaken (OECD 2013d). In France, the 2003 pension reform extended the duration of contributions necessary to qualify for a full pension with life expectancy adjustment (from 40 to 41 years and three-quarters), effective as of 2009 until 2020, and the 2010 reform pushed back, between 2011 and 2017, the qualifying age for pension entitlement (from 60 to 62 years) and automatic access to the full rate (from 65 to 67 years). The 2013 pension reform extents to 43 years the time required to achieve full pension entitlement as of 2035. In the Netherlands, the statutory pension age is stepwise increased from 65 in 2012 to 67 in 2021. After that, the retirement age would, according to the 2012 proposal, be adjusted according to changes in life expectancy every fifth year. In Norway, a pension reform was implemented in 2010–2011 establishing flexible retirement between the ages of 62 and 75. The pension benefit is actuarially calculated, with life expectancy adjustment an explicit element. In Switzerland, a pension reform is currently under discussion (Réforme Prévoyance vieillesse 2020). It includes a proposal to increase the legal age of retirement of women from 64 to 65 (then becoming the reference age for both genders) and measures to promote a more flexible transition to retirement between 62 and 70 in the two pillars.

There are less formal early retirement schemes but alternative pathways to early exit from the labour market still exist. The possibilities for older workers to leave the labour market prematurely have become increasingly restricted as a consequence of the retrenchment and gradual phasing-out of public early-retirement schemes or options in France, the Netherlands and Norway. In France, unemployment rules after the age of 50 have not been tightened very effectively. One effect of the suspension of public funding for early retirement since 2006 as well as the complete end of the exemption for the unemployed older than 57 from the obligation to look for a job in January 2012 has been to boost the number of older persons registered as unemployed. The maximum time during which unemployment insurance benefits may be claimed is 3 years for persons over 50 (as opposed to 2 years for those under 50), and it rises to 7 years after age 58, i.e. it is maintained until the person is eligible for a full pension. For this reason, many older workers quit their jobs by agreement with their employer at that age, through the negotiated contract termination (Rupture conventionnelle) introduced in 2008.

The Netherlands and Norway have eliminated special unemployment retirement pathways as special eligibility rules for unemployment benefits for older people have been abolished. In the Netherlands, a policy reform implemented in 2004 put an end to previous special rules exempting unemployed people over the age of 57.5 from reporting job-search efforts to the Public Employment Service (PES). Those over 64 are still exempt from job search. In Norway, the special rules for older people in the unemployment benefit regulations were suspended in 2011. Prior to 2011, unemployment benefit recipients aged 64 years were allowed to receive benefits until the age of 67 years (the prior pension age). Following the reform, the unemployment benefit regulations for people aged 64–67 years are the same as for people younger than 64. However, people older than 64 who are no longer eligible for unemployment benefits have a guarantee of a place on an active labour market programme if they have been fully or partly unemployed over the previous 6 months.

In Switzerland, private early retirement options tend to decrease but special UI rules exist for the older unemployed. The duration is 24 months after age 55, with uninterrupted contributions for 22 months over the last 24 months (12 months for those under 55 with no dependents, after 12 months of registration). They are eligible for 6 additional months if they became unemployed during the 4 years preceding retirement age and for whom job placement is impossible or very difficult.

In the Netherlands, Norway and Switzerland (but not in France), disability pensions have been used since the 1980s as alternative pathways to early retirement, reflecting a weak work focus of benefits with limited emphasis on fulfilment of mutual obligations, rehabilitation and use of remaining work capacity (OECD 2010). However, over recent years these three countries have sought to strengthen the work focus of disability benefits and the assessment of remaining work capacity. Nonetheless, more could be done to further encourage people older than 50 on disability benefits to take up suitable work opportunities.

Phased retirement allows for continued work but on a part-time basis, with a first portion of the pension paid at end-of-career and a second portion at some optional point in time. Many workers find that working part-time while receiving a partial pension is more attractive than taking full retirement. According to the Eurobarometer survey of 2012, this was the case for two-thirds of French citizens over 15 currently working (Eurobarometer 2012). In France, even though it has been possible since 30 years to take phased retirement (retraite progressive) which can improve the worker’s future pension rights, this option has never taken off, mainly because there is little value attached to part-time work in France. By contrast, phased retirement is much more frequent in the Netherlands, Norway and Switzerland, countries with a strong part-time culture. However, in Switzerland, highly educated male retirees tend to work more often part-time after 55 as a result of favourable early retirement options from their occupational pension schemes (second pillar), in particular in the finance and banking sector.

3.2 Tackle Employment Barriers on the Side of Employers

The four countries have launched a number of ad hoc initiatives to change attitudes of employers to older workers and for the prevention of age discrimination. They include legislation (except in Switzerland) and information campaigns to promote age diversity. Concerning the need to tackle objective reasons for employer reluctance to hire or retain older workers (e.g. high seniority wages and employment protection legislation), little has been done, partly because the four countries tended to adopt age-neutral measures on the demand-side. However, in the four countries, the social partners have not reviewed extensively pay practices based on age or length of service.

The concrete actions taken to change employer practices have mostly included awareness campaigns, development of “tool kits”, promotion of best-practices and consultation and co-operation with the social partners. Consultative forums or councils involving actors across sectors and disciplines are now quite widespread. However, there have been few solid evaluation studies of the effectiveness of these measures.

Age discrimination is banned by law in France, the Netherlands and Norway in line with the 2000 EU directive, but not in Switzerland where a proposal of an anti-discrimination law was rejected by the Federal Council in 2009. The Netherlands has launched very concrete measures towards age discrimination, such as the screening of vacancy announcements. In France, the PES (Pôle emploi) has developed the “simulation” recruitment method enabling new hires to be selected through aptitude tests without regard to age or previously held jobs. Yet neither this tool nor the use of anonymous résumés is guaranteed to eliminate discrimination in hiring, for they can influence only the first stage of the selection process and they also eliminate the possibility for positive discrimination, if this is present in the contracts to be promoted.

Mandatory retirement often is still informally in place. Age limits in some collective labour agreements remain a barrier to working at an older age in the Netherlands, Switzerland and Norway, as well as giving the signal that work ability is limited beyond a certain age. In France, the mandatory retirement age has been pushed back to 70 years for private-sector workers. On the other hand, for public-sector workers, there is still a full-pension age limit (67 years in 2017) even if there are some exceptions.

The wage that older full-time workers receive in comparison with that of younger workers over a given year is slightly higher in France, the Netherlands and Switzerland than the international average. The ratio for persons aged 55–59 to those aged 25–29 was 1.5/1.6 in 2010 in these three countries, compared with an OECD average of 1.3 and a ratio of 1.2 in Norway. Does this mean that the wage profile by age is too steep in France, the Netherlands and Switzerland and may be reducing job opportunities for older workers? Answering this question involves tracking the profile of a cohort of workers in order to eliminate the composition effect related to the fact that less-skilled workers often leave the labour market earlier. This is the approach taken in France by Andrieux and Chantel (2012) to describe the career paths and the wages received at the end of working life for persons born in 1942, in the public and private sectors. The principal message is that wages stagnate for more than half of salaried workers between age 50 and the end of their career, except for those who remain at work full-time until retirement.

Wage policy is determined primarily by the social partners. The levers for action available to the government in terms of fixing wage are therefore limited. They relate basically to the official minimum wage in France and in the Netherlands and more generally, to incentives for the social partners to negotiate. In particular, the authorities could encourage the social partners to adopt wage practices that reward experience, which is a factor in maintaining the productivity of older workers. On the other hand, there is a need to rethink the automatic nature of seniority premiums and the consideration of seniority and age in salary schedules. Public authorities could give an example in public-sector wage-setting arrangements by introducing performance pay and limiting automatic rises in salary with tenure.

There is no employment protection legislation specific to older workers in the four countries under review. In France, the Delalande contribution, which the employer had to pay after 1987 when terminating the contract of a worker aged 50 or over, was eliminated as of January 2008 because it was found to be a disincentive to the hiring of older workers (see the evaluation of Behaghel et al. 2008). What is important is that older workers’ jobs are protected de facto, because more than 90 % of them have a permanent contract in the four countries under review. In addition, severance pay very often increases with age/tenure and so incumbent older workers are implicitly more protected than younger workers. Consequently, the Netherlands is moving in the right direction with the proposed reform of employment protection legislation: severance pay would be changed to a training allowance, called a transmission budget, with a maximum of EUR 75,000 or 1 year’s salary, whichever is less. The focus would be more on transferring to other jobs and less on possible pathways out of the labour market.

Employers and employees are often in the best position to identify and promote best practice. Co-operation with and involvement and commitments from the social partners is therefore crucial to change attitudes and to increase the esteem of older workers. The social partners play a central role in the Netherlands and Norway. For example, the social partners participate actively in the activities of the Centre for Senior Policy in Norway to raise awareness of the value of older workers’ resources, identify what is needed to motivate workers to be longer at work and stimulate age diversity at work (Box 1).

The social partners often advocate mutually reinforcing strategies to promote better employment prospects for both younger and older people. In France, the “generation contract” was created by the law of March 2013, following the national multi-sector agreement signed by all the social partners on October 2012 (Box 1). However, the implementation of this new contract has been much slower than expected and a rigorous evaluation of its effectiveness is planned.

3.3 Improve the Employability of Older Workers

Action to improve the assistance given to older jobless people by the PES and access to ALMPs have been one of the most frequently reported measures (apart from pension reform), even before the Great Recession. An important decision was taken in many countries to have first a mainstreaming strategy and, second, to design well-targeted schemes only for the most vulnerable groups. Training measures to reduce the age-gap between older and adult workers were also implemented, but without significant impact, as well as measures to improve working conditions.

Training must be made more attractive in the eyes of older workers and at the same time employers must be persuaded to provide training even after the age of 45. Even if in Switzerland participation in training up to 60 is high, it is not the case for less qualified older workers. In France, age merely accentuates unequal participation in vocational training that is essentially linked to the level of qualifications and diplomas.

Obtaining recognition of skills acquired through experience, i.e. the Validation of Acquired Experience (VAE), is especially valuable for older workers whose initial qualifications may be outdated. They often acquire skills on the job but do not have certificates to prove it. Currently, these skills are not sufficiently assessed or measured in all four countries. Validation is necessary to be able to recognise those skills adequately, and to render them transparent to potential employers. In the Netherlands, the instrument to validate skills acquired on the job is the Ervaringscertificaat (Experience Certificate). In recent years, its use has increased through campaigns (television, radio and billboards), a quality code, and regional infrastructures for learning and working. The Certificate is also included as part of collective labour agreements in several sectors, and is paid for by a number of training and development funds.

Table 2 Country-specific examples of policies for older workers
Full size table

One question which remains highly topical is whether the needs of older unemployed people are best met by specially targeted measures or by an all-age mainstreaming approach. The four countries have introduced or strengthened enforcement of mutual obligations for older unemployed people. It is a step towards mainstreaming of employment services for older people rather than having separate programmes or special exemptions. In all four countries, there are no special programmes for those aged 50 and over, but instead the PES focuses on an early response to unemployment and on meeting individual needs. Individual action plans are part of the services provided by the PES to respond to the special needs of older long-term unemployed people. An experimental evaluation assessed the positive impact on the older unemployed in Switzerland of a special programme based on early intervention, intensified coaching, and individual targeting (Box 1).

Working conditions can have an important bearing on decisions by older workers to keep working or retire as their impact on health is considerable. The four countries aim to prevent the effects of bad working conditions in a preventative way. In Norway, anticipating and reducing sickness-related absenteeism more effectively is one of the three goals of the “Inclusive Workplace Agreement” which was introduced in 2001 and by 2012 covered nearly 60 % of workers. The networking initiative developed in the Netherlands among volunteer “pioneer” firms to promote sustainable employability and better health for their workers may provide some useful lessons to go beyond a narrow focus on older workers (Box 1). In France, as a headline measure of the 2013 pension reform, the personal account for preventing strenuous work exposure is to come into force in January 2015 for private-sector workers. Depending on the number of points accumulated in this account, a worker can “convert” to a new occupation, take on a part-time job without loss in pay and/or take earlier retirement. Implementation of this account is proving to be a complex affair, and the government has set up a commission to implement it progressively. In Switzerland, according to the results of the European Surveys on Working Conditions, the situation has been deteriorated between 2005 and 2010. The recent OECD report Mental Health and Work in Switzerland highlights also the fact that employers are not well-equipped to deal with mentally ill employees and sickness monitoring and management practices are highly variable (OECD 2014a).

4 Conclusions

The labour market situation of older workers has improved over the past decade. Despite the severity of the most recent recession, the labour force participation of older workers has held up well in comparison with previous recessions and relative to other demographic groups. These outcomes may to some extent reflect pre-crisis reforms in pension systems and the closing of routes to early retirement. However, this has also meant that an increasing number of older workers who lost their jobs during the crisis became unemployed with a high risk of long-term unemployment. Therefore, it is essential that older unemployed individuals be encouraged to seek jobs actively and that governments ensure the provision of suitable employment services, career advice and re-employment incentives to assist them in this task following the “mutual obligations” approach towards more inclusive growth.

In France, the Netherlands, Norway and Switzerland, the top priority in terms of policy action to encourage work at an older age has been in the first area of the OECD’s policy agenda, i.e. to “Strengthen financial incentives to carry on working”. Policy action in the second (employers’ barriers) and third (employability) areas has been much less frequent and sometimes only at the margin. In many instances, awareness campaigns have been the only measure taken to tackle employment barriers on the side of employers, and the issue of improving the employability of older workers has often been addressed at too late a stage, i.e. when they are already long-term unemployed, to be very effective.

In fact, it is timely in the wake of the recent economic crisis, and the rise in the unemployment among older workers, to pay particular attention to demand-side issues and appropriate measure to strengthen the employability of older workers. In particular, if individuals lose their job in their late 50s, they are most at risk of long-term joblessness. They will likely face a chaotic transition to retirement, being too “young” to retire but too “old” to find a new job. Therefore, more focus should be been given to improving the hiring of older workers and on facilitating greater labour mobility as a way of promoting employment at an older age. It is crucial to know more about the factors driving or hindering mobility of older workers and the evidence base on this should be improved. Employability issues should also go beyond a narrow focus on older workers to Work Better with Age. In particular, better training opportunities and recognition of prior learning should be enhanced, working conditions should be improved at all ages and finally family–friendly employment policies should encourage more stable and durable working careers for women, particularly in Switzerland.