Keywords

Introduction

The Greek financial crisis , started with per capita incomes falling in 2008, necessitated the largest international state bailout in history in 2010, and did not formally end until the expiry of the third consecutive bailout in 2018. With a duration of 10 years it is the longest financial crisis in the Western World—longer even than the interwar Great Depression (Meghir et al. 2017). The crisis is usually viewed through a macroeconomic lens, focusing on austerity and its impacts. This emphasis obscures structural factors, both as causes of the crisis and as determinants of differential impact among demographic groups. Long term secular developments, such as population ageing , combined with an unreformed pension system to necessitate a rising stream of fiscal support to pension providers; grants to finance pension outlays were seen as a key cause of the country’s fiscal collapse (Nektarios and Tinios 2020). In this setting, older women were in the cross-fire—as a group directly targeted by structural reforms, as victims of austerity but also as affected by changes in health and long term care .

The turbulent political economy of the period, as well as the urgency accorded to structural changes in the pension system by the ‘troika’ of international creditors (the International Monetary Fund, the European Central Bank and the European Commission) imply features that differentiate an account of the status of older people in Greece from that in other countries. First, the pace and content of reform was directed from outside with fiscal considerations uppermost; it was not the result of balancing social policy options. Second, the prioritisation and design, as well as the speed and implementation of reforms were matters of contention between the authorities and the troika; they involved bargaining, some backtracking and the virtual absence public debate . Third, the pension system was radically transformed; it shed its key characteristic of fragmentation , retaining however the almost exclusive reliance on State-provided first pillar pensions financed by PAYG. Fourth, the absence of gradualism distinguishes Greece from equivalent social reforms elsewhere. Changes that in other settings have taken decades were ‘telescoped’ in Greece over a few years (Tinios 2020).

In consequence, the story of the bailout period, 2010 to 2018, was characterised by an interplay between austerity policies and a forced rebalancing of pension policy . When austerity was first imposed, early retirement was still possible, resulting in a wave of early retirement by women. Cuts repeatedly imposed on existing pensions proved insufficient to prevent rises in the share of pensions in GDP , which in 2018 still absorbed the highest proportion of GDP in the EU (Nektarios and Tinios 2020). This was despite a succession of structural laws that restricted access to pensions by increasing vesting periods and rising retirement ages , both affected women more than men . Consequently, many women who could have retired under previous rules are after 2016 blocked from exiting the labour market. These radical changes were not matched by policies to make the labour market friendlier for older workers , the need for which has been ignored. In technical terms, the supply of labour was forcibly increased by blocking early retirement , while compensating changes on the demand side in the labour market are absent. This imbalance, combined with gaps in care policies, should give rise to concern about the future wellbeing of older Greeks , and of women in particular.

Older women face an uncertain future . Those who managed to retire before 2016 must live with low pensions, possibly punctuated with further cuts. Those who retire after 2016 must remain in the labour force for much longer. Older workers who lost their jobs in recent years confront ageism and the near nonexistence of a labour market for older workers , potentially contributing to an inability to qualify for a pension once they reach the new, older, retirement ages . Both groups, whether with or without an early pension, are called upon to meet the burden of caring for older people on their own—the State has all but resigned responsibility for long-term care (LTC). It is indicative that in the 2018 Ageing Working Group report (AWG 2018), the Greek State’s fiscal outlays on LTC are given as 0.1% of GDP , while the Netherlands with a lower proportion of people aged 80 + in the population spends 3.5%.

Gender, Ageing and Employment in Greece

The generation retiring in 2009 was that of the Greek baby boom, born in the 1950s and known as the ‘Polytechnic Generation’ —referring to their role in the student rebellion against the dictatorship in 1973. Women of that generation were very different from their mothers: more educated and with a greater stake in paid employment. Although the need for social policy to recalibrate itself away from the familial system had been recognised, policy was lagging behind official intentions (Lyberaki and Tinios 2014). Consequently , caring responsibilities overwhelmingly fell on female members of the family. They included caring for older relatives and spouses, as the new programme for home help was provided at levels far below demand , (Tinios 2016). Childcare provision was similarly lacking which had implications for the ‘sandwich generation’ of women providing intensive care to children and grandchildren and their parents (Lyberaki 2017).

The picture of labour market outcomes for older workers (Table 21.1) is one of a traditional labour market, which was nevertheless changing fast in the pre-crisis period (2001–2009): employment gender gaps were wide but shrinking. The crisis is seen in an explosion of unemployment for both genders (for women it trebled from 7 to 20%). This led to the already low employment rates for women falling further; employment rates in 2015 were no higher than a decade earlier. To interpret the figures, one must take on board that there was no single retirement age ; retirement ages differed by occupation and other characteristics.

Table 21.1 Labour market outcomes of older workers by gender in Greece 2001–2019
Full size table

The sharp rise in the activity rate after 2015 is not a sign of economic take-off; women’s unemployment had peaked in 2017 and was twice its 2009 level in 2019. It is instead an indication of labour market changes and restraints in access to retirement (Lyberaki et al. 2017). The third bailout in 2015 overturned the previous policy of tacitly encouraging early retirement (Nektarios and Tinios 2020). Retirement ages of all groups increased sharply and abruptly between 2015 and 2016, blocking labour market exits for older workers . In consequence the employment situation for older women in Greece after the end of the bailout was the worse among the EU-28; gender gaps were more than twice as wide as the EU28 average. To interpret the implications of the effects of the bailouts on older workers, we need to factor in three pension factors affecting gender (Panageas and Tinios 2017; Lyberaki 2018):

  1. 1.

    Periodisation. The period of build up to the crisis (2001–2009) can be distinguished from the crisis itself (2010–18). Legislation up to 2015 (the first two bailouts) introduced new pension arrangements for younger workers, aiming in various ways to exempt (that is to ‘grandfather’) older workers . The third bailout reversed this approach, retroactively extending new regulations to everyone, even those who had already retired and to survivors’ pensioners.

  2. 2.

    Fragmentation . The key characteristic of pre-crisis pension arrangements was fragmentation (Tinios 2020). Old age protection, despite being supplied by state bodies and financed by PAYG, was fragmented by provider (over 300 organisations supplying primary, auxiliary pensions and separation payments), financing , retirement ages and modes of pension calculation. Access to old age pensions (vesting rules and retirement ages ) was exceptionally uneven by occupation, allowing retirement from 42 years of age to over 70 for some occupations. Retirement ages were systematically lower for women, especially in the public sector and for larger enterprises. As a result, the use of averages can be seriously misleading: in the largest pension provider the ‘rule’ for retirement ages (65 for men and 60 for women) was followed by 15% of men ; 85% made use of more favourable exceptions (30 and 70% respectively for women -Tinios 2010). Pension reforms tried to stop this fragmentation through organisational consolidation and enforcement of common rules. These moves would inevitably have a greater impact among women, as the distance between pre-crisis and new system parameters are greatest.

  3. 3.

    A strong cohort effect. The cohort to retire during the crisis and future cohorts are considerably different from previous cohorts, an effect particularly important for women. From the ‘Polytechnic generation’ on, women were more educated , more involved in the labour market, less likely to be in farming or self-employed or unpaid workers in a family business , and more likely to be in public sector employment.

The reform of the pension system left important legacy effects: Firstly, even after retirement ages increased, there remain many women pensioners who were under 60: based on administrative data (Helios System), 11% of women pensioners in 2017 were under 60, a figure underestimated as this total does not include widows and other recipients of survivors’ pensions. Secondly, many women have no access to pensions in their own right, relying instead on derived rights from their husband. Thirdly, there is a wide pension gender gap (Betti et al. 2015). The averages obscure the fact that early retirees have systematically lower pensions—which will create a problem of low pensions in the future as they will not be able to benefit from any post-crisis recovery.

The reliance on early retirement is well documented before the crisis (Tinios 2010): already at age 50, 17% of Greek women and 6.5% of men were pensioners. Siegrist et al. (2007) have shown that an effort reward imbalance and low control at work are factors behind early retirement . This is corroborated by Siegrist and Wahrendorf (2011), adding that poor quality of work is associated with poor health after retirement. People who had experienced poor quality of work were more likely to report reduced health post retirement, compared to those who had not experienced poor quality of work . Life satisfaction among older women with a full career in Greece is much lower than in other countries (Lyberaki et al. 2013).

Other studies find evidence of age discrimination affecting older people . Bratt et al. (2018) investigated perceived age discrimination across different age groups with data collected in 29 countries in the European Social Survey (European Social Survey 2019). Their analysis indicates that in Greece older people are the victims of age discrimination ; moreover, this affects older respondents to a greater extent.

Pension Systems Pension Policies and Extended Working Life

The Greek pension system in 2019 is very different from 2009, when the bailout reforms commenced. Pension ages are at the high end of the EU (67 for both genders and 62 for a 40 year career). New state pension calculations result from applying the same two-tier pension scheme (composed of a fixed rate ‘national pension’ with a portion proportional to career length) to all pension applications . Whereas previously different occupations faced very different arrangements, after 2019 very little occupational differentiation remained (Panageas and Tinios 2017). However, the way these changes were implemented created side effects and legacy issues that will mark the next two decades.

A flavour of how the priorities of pension reform resulted from an interplay between the Greek authorities and the troika of international lenders can be gleaned from the case of Heavy and Hazardous Occupation (HHO) , and the role played by its most quoted example, hairdressers . Rules for HHOs were originally designed for groups with demonstrably lower life expectancy , such as workers in underground mines. This was vastly expanded (sometimes stretching credulity) to include people whose claim to occupational health deterioration was tenuous or fictitious—swelling to account for 40% of all pension applications in the private sector (Tinios 2010). Criteria for inclusion were never specified; occupations were added case-by-case. For instance, Church cantors were included using the argument that participating in funerals was depressing. Hairdressers were added to the list of heavy occupations in 1966 and could thus receive an old age pension at age 50. Their specific case was unwittingly drawn in the high politics of the bailout . As Germany was the main source of bailout funds, the German government in 2010 was sensitive to comments in the German press; the paradox of German taxpayers (whose retirement had been recently raised to 67) paying for Greek hairdressers (among others) to continue retiring at 50, was used to argue against the bailout . Their example increased pressure to insist on strict and specific conditionality of bailout funds—including as a priority the overhaul of the HHO system.

Commentary on egregiously early retirement led to across the board retirement age increases in the first law passed after the bailout in 2010. The system of HHOs was next in line in early 2012. Hairdressers , among many other, though not all, occupations, were withdrawn from the list of HHOs ; however, those with more than 10 years’ working experience retained existing rights. Consequently, the vast majority of active hairdressers were exempted, or using the technical term ‘grandfathered’ (a technical term covering the exemption of individuals with ‘legacy rights’ from changes). The pension reform passed in the third bailout in 2016 partly reversed this privilege in May 2016: the reference retirement ages were increased to 67, (62 with 40 years’ employment), which automatically pulled up all retirement ages . Nevertheless (until further notice) hairdressers could still retire 5 years earlier than the new reference ages.

‘Grandfathering’ was only part of the story. Many women in the public sector were encouraged to retire early as a substitute for being made redundant between 2010 and 2015, through the following mechanism : mothers of underage children had previously been entitled to retire at 50. At that age, most offspring were over 18, implying little impact. In 2010, the age test was moved earlier (to the age when the mother had completed 20 years’ contributions). So, a woman who started work at 20 would need to prove that her children had been underage when she was aged 40; if so, she could retire on completion of her 50th year. In this way, thousands of women suddenly became ‘mothers of underage children’ with a right to retire at 50, even though their offspring could have been close to 30 when their mothers retired.

Another unique feature of the Greek pension reform was applying cuts to pensions already paid out. As employment earnings in Greece were in freefall, and given that there was no inflation, adjusting the pension bill could be implemented only by nominal cuts to pensions already in payment. Such cuts occurred repeatedly until 2015; as the criterion for cuts was the value of pensions, larger pensions were cut by more than 50%, while lower pensions were largely spared. Given that private earnings collapsed, low income pensioners paradoxically found themselves in a privileged position. Despite cuts, pensioners’ risk of poverty halved, as the median incomes on which the poverty depends fell by more than the value of pensions. Gender pension gaps also fell for the same reason, since women’s pensions were typically on the lower end of pension values. However, notwithstanding major changes in pension entitlements, the underlying structure of the system remained unchanged: nearly 100% of pensions are still provided by State bodies, while little opening was created for private or collective pillars to grow (Tinios 2020).

The complex political economy of forced reform during a deep fiscal crisis means that the impact on particular groups was not the product of reasoned discussion and weighing of alternatives but could emerge as an unintended residual following more urgent considerations–a kind of ‘collateral damage’. The case of widows and survivors pensions is an example (Tinios et al. 2020). Widows , both existing and future, were a group who disproportionately lost out: they bore the full brunt of pension cuts , while survivors’ pensions are recalibrated as if the deceased had spent his/her life in the system inaugurated in 2016. The inherent tensions and instabilities of the Greek reform can be seen in age restrictions (55 years of age) on the permanent receipt of survivors’ pensions: These were imposed in 2010, and further tightened in 2016. Some months after the departure of the troika, in May 2019, all age restrictions were abolished, including some that had been in force since 1997. The explanatory report of the 2019 legislation justified this, as ‘in Greece surviving spouses (mostly women) stay out of the labour market during marriage, with the result of encountering great difficulties in seeking employment, especially at relatively old ages’ (Hellenic Parliament 2019, p. 14).

The upshot of pension reform was twofold: Firstly, those who had managed to retire were faced with insecurity. The 2016 Legislation enforced a kind of bailout legacy, in the form of further cuts mandated to occur after 2019. That date is when all pre-2016 pensions must be recalibrated as if all careers had taken place under the new system. Any excess value using the new calculation formula will be gradually eliminated by being excluded from pension indexation . The group most affected by recalibration are widows , who are treated as recent pension applicants, implying much larger falls (Tinios et al. 2020). Secondly, those who had not retired before 2016 were faced with the prospect of working for as much as 17 years more than they had originally planned. As women had to make up the largest distance in retirement ages , they represent the largest group forced to remain in the labour market. Their number will start rising cumulatively after 2018, after the impact of the rush to retirement before 2016 wears off.

Extended Working Life Employment Policies

The reform of pensions will force a major rise in the supply of labour , chiefly on the part of older women . This, however, has yet to be matched by equivalent compensating initiatives on the demand for older workers by employers .

To be sure , lip service to active ageing had been a feature of action plans for employment since at least 2000. However, except for a few demonstration projects, mostly in the area of training, very little has been achieved. A labour market for older workers is essentially non-existent ; as a result, if an older woman loses her job, the prospect of finding another is very small. During the crisis older worker unemployment was exacerbated by the closure of large numbers of small businesses. Lyberaki and Tinios (2017) argue that the distinguishing feature of this crisis is that the closure of firms was not counterbalanced by the opening of new enterprises (as it had previously in more positive socioeconomic contexts). The large increase in social insurance contributions of the self-employed after 2016 accelerated closures and made seeking employment harder.

The second major factor discouraging employment of older workers are collective agreements linking minimum and average pay rates to age and seniority. This means that older job seekers are at a considerable disadvantage . Thus, if older workers are insiders , employment protection enables them to hold on to their jobs. Should they lose them, they join the queue of outsiders , albeit at a considerable disadvantage . Behind these policies is an ingrained and generalised ageism . Most commentators do not hide their belief in the ‘Lump of Labour fallacy ’ i.e. that the labour market is akin to a game of musical chairs where if an older person holds a job, it is taken from a younger unemployed person.

Health and Long Term Care Policies

The ‘system’ of HHO has already been mentioned. This at the outset was designed as actuarial compensation to accommodate occupations with demonstrably lower life expectancies . It was quickly abused to include hundreds of occupations some with only spurious claims to health impacts. Its use as a clientelistic device can be illustrated by contrasting by its application in the Public Power Corporation: waiters were deemed an HHO , while pylon repair crews were not. In the case of real health-threatening conditions , this system operated as an incentive to retain unhealthy working conditions, discouraged improvement, and maximised risk exposure (Tinios 2010). Although the list of such occupations was reduced in 2012, no general operating principles of inclusion were debated or included in the law.

Disability pensions have been a problem for private sector pensions. They were used, sometimes fraudulently by men of retirement age with a contribution record less than the fifteen years necessary to vest an old age pension i.e. they operated as a device complementary to contribution evasion . As women had access to other early retirement routes , such as regulations on underage children and some HHOs , they did not use invalidity pensions to the same extent as men (Tinios 2010).

The problem was scaled down after 1992, when minimum pensions were linked to the extent of disability, which led to the proportion of disability pensions falling by more than half. As part of the bailout , disability-checking procedures were made tougher.

Much crisis commentary, such as Kentikelenis et al. (2014), a study widely reproduced in the international press, laments a ‘Greek public health tragedy’, of deteriorating health outcomes as a direct consequence of retrenchment. A recent study (GBD-G 2018) provides a careful analysis of all available morbidity and mortality indicators : both (though morbidity to a lesser extent) sharply deteriorated for older individuals during the crisis . This was not due to austerity but was caused by the acceleration of ageing and by unhealthy behaviour (smoking, BMI, diet), especially among recent cohorts of women. Even so, the fall of per capita health expenditure , combined with ‘rooted inefficiencies’ of healthcare could still be indirectly linked to health outcomes. Their overall conclusion is uncertain as to causation but is unequivocal as to the existence of ‘a disproportionate decrement in the health of Greeks , which parallels the course of the economic crisis’ (GBD-G 2018, p. 404).

How did austerity and the crisis affect Long Term Care ? Survey evidence (Lyberaki and Tinios 2018) shows that, despite worsening self-reported health , the extent of unmet care needs fell. Interestingly, this was not due to rising family provision but due to a rise in professional care . As municipal LTC services (the ‘Help at Home programme’) shrank (Tinios 2016), and there was no increase in other state-provided services , the increased use of LTC was mostly due to a rise in purchased professional care services paid for out of pocket by families. Lyberaki et al. (2019) speculate that a reason for this is that pensions fell by less than the wages of carers. In consequence, the overwhelming share of responsibility for care remains with the family; a larger part of this than pre-crisis now falls on family budgets.

Debate on Extended Working Life

Pension reform was the first reform bill of any kind to be passed in the first bailout in 2010; it was also the last in December 2018, three months after the third and last bailout ended. Pension policy through this long period unfolded against the background of a relentless recession , through an unsteady interplay between the international creditors, on the one hand, and successive Greek governments, on the other. The creditors pushed for fiscal retrenchment while Greek governments of the day were concerned with containing political fallout . In this tug of war, blame avoidance displaced open debate . The unsubstantiated assertion that extended working lives contributed to unemployment, which had been used before the bailout to prevent rises in retirement ages , was given a new lease of life in bailout period when early retirement was encouraged. Further, the entrenchment of ageist beliefs extended to political parties and to social partners , while the dearth of discussion about ageism or the plight or possibilities for older workers meant that there were no opportunities for these views to be rebutted (Tinios 2018). The fact that major changes to retirement took place independently of social conditions was responsible for four paradoxes that our analysis has uncovered:

Paradox 1: The baby boom generation received lavish pension promises which were later dramatically reversed in the crisis , by rises in retirement ages and retroactive cuts of entitlements. The kind of transformations that were phased in over decades elsewhere, were forcibly and quickly implemented in Greece . Cohorts of women now in their sixties and late fifties, who played by the rules of a Europeanising fast growing country and who prided themselves in modern careers, are in for a surprise as reality does not conform to the retirement path they were led to expect.

Paradox 2: Policy first encouraged and then penalised early retirement . In the first phase (2010–2015), considerable effort focused on exempting those near the age threshold, leading to early retirement . This was financed by cuts in pensions affecting older pensioners with relatively high pensions. The second phase (late 2015) reversed the early retirement course with large increases in the retirement age . At the same time, obstacles and disincentives to working were enforced for older workers and for the employment of pensioners.

Paradox 3: The forced increase of the labour supply as a result of delaying pension entitlement was not matched by measures to raise employment demand . This paradox is glaring in official projections for the Greek pension system included in the Ageing Working Group report (2018). The viability of the new Greek pensions system rests on the assumption of a substantial increase in women’s work (by approximately 10 percentage points or 20% increase) by 2030. Mobilising women’s labour reserves is the main source for financing the pensions system and restraint of the deficit . However, few measures to support and justify such a drastic increase are mentioned or apparently planned.

Paradox 4: The dramatic future increase expected in the need for long term care (LTC), particularly if women’s labour force participation increases to levels needed to sustain Greek pensions in the future, is not matched by a credible increase in supply. The projections already pledge today’s informal carers to also be in paid employment, while no action is taken to support LTC. Until recently, the care gap was filled by a private, largely informal, care market supplied mainly by immigrant Eastern European women (Lyberaki 2011). Declining pensions and wages in Greece , in combination with soaring unemployment, are limiting the prospects of such jobs in the future; this is mirrored in a trend favouring the return of many carers to their countries of origin.

The four paradoxes will place older women in a maelstrom of developments over the next decade. Women are expected to shoulder the lioness’s share of adjustment in pension reform , whilst being denied the means to do so. The underlying cause of their predicament is an insufficient understanding of the interplay of the role of older workers in labour markets and the normative expectations for women’s unpaid care work. The costs of early retirement on the part of a large privileged minority were cushioned by external debt finance before the crisis . The costs of early exit from the labour market was also disguised by the use of averages and the lack of statistics documenting differences by gender. These combined with an implicit belief by policy makers in the ‘lump of labour fallacy ’ which complemented existing sexism with heavy doses of ageism .

In consequence, Greece repeatedly went against the European mainstream, as far as active ageing is concerned. While there was token acknowledgment of European priorities in measures for older workers financed by structural funds, ageism in the labour market remains rampant. At the end of the bailouts , the costs of a policy ignoring the potential of older citizens are beginning to be apparent.

Policy Recommendations

Greece is ageing fast. Nevertheless, ageing is not seen as an integrated challenge in policymaking circles. Policy is absorbed with pension reforms and ignores the flipside , older workers . Incremental increases in retirement age have implications going far beyond legacy costs .

First, the early retired will have to live with permanently lower pensions. Under current policies, they will not share in post crisis output gains; rather, their entitlements will be permanently locked to crisis levels. For women, this means lower independence , a greater reliance on men , and a greater future poverty risk . These could be ameliorated if policies such as the following can be contemplated:

  1. (a)

    Policies encouraging economic activity by early retirees , including the possibility of recall to the labour market.

  2. (b)

    Involvement of women pensioners in providing LTC services .

Second, the labour market and many enterprises were emptied of female staff during the economic crisis and bailout period. Current hiring limits mean that the public sector will no longer be able to exert an equalising impact that improves prospects for women’s employment. In the private sector , the hollowing out of the labour market by early retirement will mean fewer senior female cadres available for higher positions. The net result could be a lower representation of women, and a strengthening of gender stereotypes .

Third, those trapped in the labour market by the rapid rise of retirement ages will face growing difficulties to navigate a hostile labour market. To overcome this, a number of initiatives must be considered:

  1. (a)

    Change in seniority loaded pay scales

  2. (b)

    Measures promoting a market for older workers , including job search.

  3. (c)

    Training focussed on older workers

  4. (d)

    Loans for starting a business should not exclude the older self-employed . Small enterprises are the most likely source of extra employment.

Fourth, arrangements for LTC are of critical importance. There is a glaring inconsistency between planning for a major increase in involvement of today’s carers in paid employment, without planning to assist them in any meaningful way. To correct for this the following may be needed:

  1. (a)

    Policies assisting informal carers in their LTC roles (e.g. respite leave, training) .

  2. (b)

    Quality assurance to aid the operation and quality improvement of a private or 3rd-secror LTC market.

Greece will have to adapt from being a country where older women’s place was in retirement (even if not in the home) to one where they can be active members of society.