The "Great Recession" and sovereign debt crises in several EU countries in the wake of the 2008 financial market crisis have triggered drastic reforms in old-age security systems. Almost exclusively, the reforms meant retrenchments with often severe and immediate consequences for the living conditions of present and future pensioners. This paper deals with the contents and circumstances facilitating or enforcing the reforms in nine EU countries: Greece, Hungary, Ireland, Italy, Latvia, Portugal, Romania, Spain, and the UK. Cross-national comparison reveals similarities and differences and also sheds light on the social consequences that are already visible today.
Keywords: financial market crisis, pension reforms, Europe.
JEL codes: E32, J26, N34.
The 1990s and early 2000s have shown that in democratic polities reforms of pension systems - parametric (path-dependent) as well as systemic (path-departing) changes - were not impracticable as was suggested by research on the "new politics of the welfare state" [Pierson 1994,2001]. Reforms came about when incumbent governments were able to shift or share the blame for enacted retrenchments, to hide the true impact of changes, or could even reap credit for reforms that put pension systems on a more sustainable footing in view of advancing population aging [Hinrichs 2011]. After 2008, however, in the wake of the "Great Recession" in a number of European countries plagued with high budget deficits and mounting sovereign debt, pension reforms came to the fore that were different in two respects. First, their magnitude was large, particularly when changes are taken together, and they (will) cause a substantial and immediate negative impact on the living condi - tions of present and future retirees. In a situation where austerity is no longer simply "permanent" but rather "pervasive", it is hardly surprising that pensions became a prime target for saving on expenditure because, almost everywhere, they are by far the largest item of welfare state spending which itself amounts to around half of total public outlays [Obinger 2012].
Second, the political process that brought about these changes deviated from previous attempts to retrench, re-finance or recalibrate old-age security systems. The post-2008 reforms in crisis-shaken EU countries were indeed large and sometimes also changed the hitherto pursued policy direction, swiftly passed the legislative process and were (or: will be) implemented with a short time lag. Hence, they can be considered as "rapid policy changes" [Rüb 2012]. Mainly, this reform pattern sprang from the pressure exerted by financial markets and supranational actors (IMF, European Commission) which urged governments to neglect voteseeking objectives within the well-known credit claiming/blame avoidance framework for the sake of attaining rapid savings on public expenditure [Bonoli 2012]. Consequently, in a number of countries reforming politicians were punished and lost power during subsequent elections because voters rarely appreciated retrenchments designated to overcome a "major crisis situation".
In this contribution eight countries will be analyzed - four Southern European countries (Greece, Italy, Portugal, and Spain), three CEE states (Hungary, Latvia, and Romania) and Ireland. All of them have conducted pension reforms after 2008 in order to ensure their schemes' financial viability in the short and long term or to realize notions of intergenerational equity. Most urgent, however, was regaining room for fiscal maneuver and obtaining financial aid from supranational organizations (IMF, EU). Seven of the eight countries had to seek such aid in the wake of the financial market crisis (2007/08) triggering an economic slump and, as one immediate outcome thereof, a sovereign debt crisis. The causal relevance of these events on the reform process can be read off from concrete recommendations issued by the European Commission or detailed reform demands attached to bailout agreements ("memoranda of understanding"), whereas the intensified reform effort of deeply indebted Italy was driven by the rising spread over German government bonds.
In the following, the content of reforms will be delineated1, but attention will also be given to the circumstances which led to the respective changes. It will be shown that the challenges these countries were confronted with enforced or facilitated drastic reforms which otherwise had not been practicable or, in view of the political consequences, had not been initiated by the respective governments.2 Moreover, the social consequences of the policy changes will be looked into, although an evaluation is limited because not everywhere has the reform process come to an end yet and the actions taken have not brought to bear their full effects.
1. Eight countries - three types of pension systems
The eight European countries belong to different welfare-state regimes. Ireland is representing the (Anglo-Saxon) liberal cluster, whereas Greece, Italy, Portugal and Spain appertain to the conservative-corporatist type, but are also said to form a specific Southern Model. According to Ferrera  these welfare states show a "clientelistic" structure (privileging the labor force in certain occupations and economic sectors) and remain "rudimentary" because, among others, family and labor market policies are still underdeveloped. Hungary, Latvia and Romania and further CEE countries belong to the post-socialist cluster which, due to these countries' diverse welfare state designs, is hardly and coherently an (emerging) "Eastern Model" [Hacker 2009].
There is a certain arrangement of old-age security attached to the different welfare state types. Ireland (like the UK) has followed the Beveridge model, i.e. state responsibility is limited to universal basic security in old age, while status maintenance is left to (state-regulated) private provision by firms and individuals. In Southern Europe, social insurance schemes of the Bismarck type play the pivotal role. Access to and the level of public pensions is dependent on prior earnings-related contributions. The accruing benefits are meant to ensure status maintenance. However, the schemes are institutionally differentiated along occupational lines - most strongly in Greece and least in Portugal - as of which benefit generosity varies. Despite being "rudimentary" welfare states, the social expenditure ratio is comparatively high but, because old-age security constitutes the central pillar of their social policy arrangements, the structure of expenditure is strongly "age-biased" (much more than in Ireland - Table 1, rows 7 and 8), and this imbalance has grown in recent years. The causes are the hitherto very generous pension payments (at least for insiders), as well as broad access to early retirement (discernible in the low employment rates among those 55-64 years of age) and the already high and further increasing longevity (Table 1, rows 2, 3, 5,9 and 10).
Originally, CEE countries had followed the Bismarck model, and certain elements remained intact during communist rule. The social insurance approach was revitalized after 1990, before Hungary (1998), Latvia (2001), Romania (2007) turned to a multi-pillar pension system as propagated by the World Bank.3 They established a second pillar that was mandatory for the younger part of the work force (the middleaged could join voluntarily). While the total contribution rate remained unchanged, the share allocated to the pay-as-you-go first-pillar scheme was reduced (and also the accruing pension level). The remaining part was transferred into the second pillar, i.e. private pension funds for investment. This systemic change was expected to deliver a higher total benefit level and to further capital market development.
The pre-2008 reforms in the CEE states and other European countries have been mainly motivated by advancing population aging showing up in an increasing oldage dependency ratio or system dependency ratio (Table 1, row 3 and 4), the latter being more relevant for the financing of old-age security systems. Both developments are the result of rising longevity at higher ages (row 5) and declining birth rates. At the latest since the 1990s, mainly the OECD, the European Commission and the International Monetary Fund have purveyed the conviction that high social security contributions and taxes have a negative effect on employment levels, thereby supporting national actors in their policy efforts to at least keep the tax and contribution burden constant. Thus, when the contributions of employers and employees to the first pension pillar are supposed not to rise (further) in future or when (higher) tax-financed payments have to be made to pension schemes, only a limited number of levers remain by which further spending increases due to demographic change can be contained. First, the ratio between pensioners and contributors can be changed by lifting the pensionable age. Moreover, it would be possible to lower the level of newly awarded pensions by modifying benefit calculation or changing the way in which pensions in payment are indexed.
2. Post-2008 pension reforms
Irelands basic state pension is contribution-financed and flat-rated (but means-tested for retirees with an incomplete contribution record). The level amounted to about 37 percent of average wages. Benefit increases in real terms during the 2000s have significantly reduced the risk of old-age poverty - from over 40 to 10.6 percent in 2010. During the last two decades, occupational pension plans constantly covered about one half of the employees. In 2001, a National Pension Reserve Fund (NPRF) was established in order to ease the financing of the basic state pension and the occupational scheme for public service workers after 2025. Every year the government paid one percent of GDP into the NPRF.
The financial market crisis has hit hard the assets of the NPRF and occupational pension funds. In 2011 80 percent of the defined-benefit (DB) plans were still in deficit [OECD 2013, p. 53]. The under-funding has tobe removed by 2015 and, additionally, the pension funds have to build up a risk reserve. At the same time, they have to compound a special levy of 0.6 percent (p.a.) on the accumulated assets. The larger part of the (shrunken) NPRF assets have been used to bail out and recapitalize the banking sector. Before, the government had already stalled the payments into the NPRF, not least because it had to cover the growing deficits of the social insurance fund with taxes. Considerable savings stem from changes of the DB-type public service pension schemes (Table 2): Pensions in payment have been cut in progressively; the accrual of entitlements is capped at 40 years of service; newly awarded pensions turn out lower due to pay cuts; public service workers have to pay higher contributions themselves; new public servants have to reckon with more unfavorable rules for calculating and adjusting their (future) pensions [OECD 2013, pp. 41-3]. These changes were part of the program of measures agreed with the Troika, another was the increase of the normal retirement age from 66 to 68 (legislated in 2011), and already since 2009 there has been no adjustment of the basic state pension.
2.2. South European countries
Greece is clearly the straggler among the Southern European countries with regard to social policy reforms in general and the adaptation of old-age pension systems to changed circumstances in particular. As early as the 1990s there were complaints about the ineffectiveness and inefficiency of the Greek welfare state and disparate benefit levels - with public employees and some groups of self-employed benefiting disproportionately - and an inability to reform leading to crisis was identified [Katrougalos 1996; Venieris 1996].
Drastic pension reforms came about only in 2010 and thereafter due to obligations related to the bailouts [Nektarios 2012]. A significant structural reform of the extremely complex Greek old-age pension system - consisting of an incomerelated general compulsory scheme and a (largely) compulsory earnings-related supplementary insurance - started as early as 2008, however. The merger of occupationally differentiated schemes and standardized rules are supposed to bring about greater transparency and fairness and to save on administrative costs. From 2012 all supplementary pension schemes were brought together into a single pension fund.
In 2010 the Greek Parliament decided to lower the accrual rate, most decisive for the replacement ratio, to 0.8 to 1.5 per cent for one year of contribution payments (depending on the total length of insurance). Furthermore, the pensionable ages of men and women were equalized at 65 until 2013.4 In future, entitlement to a full pension will require 40 instead of 35 insurance years and pensions will be calculated on the basis of the whole working life. Without reductions (6 per cent a year) only those who can prove 40 insurance years can take early retirement (from 60 ).5 From 2021 the normal and early retirement age will be adjusted every three years according to the development of life expectancy. Two of the previous 14 monthly payments were abolished and replaced by a (largely) uniform bonus of 800 euros only for pensioners above 60. In future, the indexation of pensions in payment must not be higher than the increase in consumer prices; for the period 2011-2015 the adjustment was suspended entirely. Moreover, further measures are to be taken if projections show that there will be an increase of pension expenditure of more than 2.5 percentage points of GDP by 2060 in comparison to 2009.6
Subsequently, in 2012 a new (NDC look-alike) benefit formula with a built-in sustainability factor for the supplementary pension scheme was decided upon; pensions of the general scheme higher than 1,300 euros were cut by 12 per cent (in 2010 those above 1,400 euros already by 8 per cent); access to invalidity pensions was made more difficult; and disproportionately high one-off payments in the area of supplementary pensions were cut. By 2015 supplementary pensions will no longer be guaranteed by the state, in other words, there will be no subsidies provided for covering deficits [European Commission 2012a, p. 98].
The measures predating another support package for Greece in November 2012 deprived the social insurance funds of a large part of its reserves held in Greek government bonds, and the liquidity problems of the pension system were aggravated due to lower government subsidies and fewer workers paying contributions. Obligations related to the support package included a rapid increase of pensionable age to 67 , allowing workers with 40 insurance years to retire not before 62 (with deductions) and to cut pensions of retirees who benefited from the previously more generous calculation formula and/or prematurely claimed their pension.
In Italy, the first attempt to contain the steep rise of public spending on pensions was undertaken in 1992, and in 1995 the basis of the pension system was changed to the notional defined contributions (NDC) model which, within the framework of a lingering pay-as-you-go scheme, emulates the working of a fully funded plan. All subsequent reforms aimed at an accelerated implementation of the NDC rules, tighter eligibility criteria for "seniority pensions" (which, originally, could be claimed after a 35-year insurance period regardless of age), and advancing the harmonization of the differentiated public pension system. Until 2007, these attempts were only partly successful, but massive changes took place between 2009 and 2011 when the financial standing of the Italian state became endangered.
The harmonisation of still occupationally differentiated pension arrangements was taken further, including an alignment of contribution rates (Table 2; [see also Ministero 2012, pp. 41-3]). Furthermore, the alignment of womens pensionable age with that of men was speeded up and will be completed 2018 (in the public sector already realized in 2011 and then - as for all men - raised to 66 in 2012). Also, from 2013 the standard retirement age and the age of eligibility for seniority pensions will be linked to the development of further life expectancy. Thus, for 2019 but in 2021 at latest a pensionable age of 67 is expected for men and women in both the private and public sectors, rising to just below 70 by 2050. "Seniority pensions" - hitherto available either after 40 years of contributions or at 62 years of age after 35 contribution years - are de facto abolished since the conditions follow the rising age limits and early retirement is possible only with deductions and if the pension level exceeds the social minimum pension by one and a half times. On the other hand, corresponding increments are expected to result in pensions that ensure a standard of living for those who continue to work up to the age of 70. Considerable short- and medium-term savings arise from accelerated implementation of the NDC system. From 2012 new pensions will be calculated pro rata according to the contribution periods before 1995 in the old (DB) system and the contribution years under NDC rules after 1995. Finally, the adjustment of pensions to price development for pensions over 1,400 euros is suspended for 2012 and 2013.
In Spain, there was already a switch from expansion to consolidation in 1985, and since 1995 all pension reforms have been based on the (repeatedly renewed) Toledo Pact between the respective government and the social partners. As a result, between 1998 and 2010 the Spanish social security system ran surpluses, so that in 2011 there were 66 billion euros (around 6.3 per cent of GDP) accumulated in the reserve fund, increasingly invested in Spanish government bonds.
The most substantial changes of the 2011 reform, which will come into force mainly between 2013 and 2027, include a rise in the statutory retirement age from 65 to 67 for workers showing less than 38.5 years of contribution payments. Early retirement rules have again been tightened (beyond the 2011 legislation) in 2013. The age range will move from 63 to 65 and the required insurance years from 33 to 35. For unemployed persons (involuntary early retirement) the conditions change from age 61 to 63 and 31 to 33 insurance years. Premature claiming of benefits goes along with deductions (7.5 per cent per year) while a deferring retirement age is rewarded with supplements (between 2 and 4 per cent a year). In future, a "full" pension will require 37 instead of 35 contribution years (it will still be the case that 15 years qualifies someone for a half pension), and from 2022 pensions will no longer be calculated on the basis of the last 15 but rather the last 25 insurance years. Finally, from 2027 a sustainability factor will be introduced through which the relevant system parameters - for example, the requisite insurance years for a full pension or the statutory retirement age - will be adjusted to the development of life expectancy every five years. After Spain had to apply for loans from the ESM in 2012, it is quite possible that the effective date of the legislative changes will be accelerated.
In the wake of an "Excessive Deficit Procedure", Portugal negotiated a pension reform package with the social partners in 2007. It included increased pension decrements for retirement before 65; the already legislated pension calculation on the basis of the whole working career was brought forward to 2017; a sustainability factor was introduced, linking the level of newly awarded pensions to longevity increases; indexing of pensions in payment was debased; incentives for the (continued) employment of older workers were strengthened; and normal retirement age for public service workers will rise from 60 to 65. The combined effects are quite significant: a comparison of projected pension spending in 2050 on the basis of calculations from 2005 and 2008 shows that Portugal had taken the biggest leap of all EU countries. Instead of 20.8 per cent of GDP only 13.6 per cent pension expenditure was estimated for 2050 [European Commission 2009, p. 104].
Therefore, in order to reduce its public deficit in the short term, Portugal was obliged to only take a few pension reform measures when it sought financial assistance in 2011. Pension indexation was partly suspended, higher pensions are burdened with a special levy, and early retirement of employed workers is ruled out until 2014 whereas older unemployed cannot take out a public pension before 62 (Table 2). Furthermore, employees of state-owned enterprises - banks, telecommunications - were integrated in the pay-as-you-go pension insurance system and a total of 9.3 billion euros of the capital reserves of special schemes was transferred to the state budget7 and, hence, reduced the present deficit, but increased the implicit debt of the general scheme. Finally, in May 2013 the Coelho government announced the normal retirement age to be raised by one year, to 66.
2.3. CEE countries
The parametric pension reforms during the 1990s in Hungary, Romania and Latvia were less a reaction to demographic change than to the economic transformation. The successive paradigmatic restructuring of their pension systems followed a widespread comprehension of "privatization" as an expression of "modernization" [Orenstein 2008; Cerami 2011; Fultz 2012]. These (and further) countries became amenable to the transnational policy campaign launched by the World Bank. A multi-pillar pension system including a mandatory and fully funded private pillar promised to ensure adequate as well as financially sustainable old-age security. However, the problems with this new approach came to the fore during the economic slump and subsequently triggered significant changes.
Hungary was a forerunner in introducing the multi-pillar pension model. Already highly indebted before 2008, it became dependent on international loans as a result of the economic crisis. These were granted on condition of structural reforms, including pensions. The package contained a gradual though rapid increase of the standard retirement age (from 62 to 65), a less favourable indexation of pensions in payment, and the abolition of the 13th monthly pension payment [Simonovits 2011]. The 2010 elected Orbán government rejected (further) external rescue measures and turned to the structure of the pension system for regaining fiscal latitude. Due to the diversion of 8 percentage points of the total contribution rate into the second pillar, the pay-as-you-go first pillar lacked the necessary resources to honour the entitlements of present retirees. The shortfall had to be balanced by subsidies out of the state budget which, ultimately, increased public debt. Soon after coming into office the government started a turnaround which meant a factual abolition of the multi-pillar system: contribution payments into the second pillar were stopped, and the contribution rate of the first pillar was increased accordingly. The already accumulated assets of the second-pillar pension funds were "confiscated" and transferred into the state budget, immediately reducing the deficit and the amount of public debt. The entitlements the participants had earned in the second pillar were shifted to the first pillar which increased the long-term obligations (the "implicit debt") there. In order to extenuate the future spending increase, opportunities to retire early via disability pensions or due to long service have been eliminated almost completely, except for women (Table 2).
Romania was also hit hard by the recession and had to solicit for a 20 billion euro loan from the IMF, but it left its recently established and still small second-pillar scheme (contribution rate in 2009: 2 percent) intact and even introduced a minimum pension in 2009. Reforms which started to take effect in 2011 continued along the legislation of 2000 [Ghinararu 2011]: normal retirement age will increase further (but still be lower for women); the years of contribution required for pension eligibility and for a full pension will rise once more. Furthermore, pensions of the first-pillar scheme which applies a point system for calculating benefits have not been adjusted in 2011 and 2012, and until 2030 the indexation of a pension point will gradually shift to price level changes. The traditionally far more generous special pension regimes will be integrated into the general scheme and, finally, the eligibility criteria for early and disability pensions have been tightened, and recipients face high decrements until they reach normal retirement age.
Of all EU member states, Latvia suffered most from the economic slump (Table 1, row 6) and also had to seek external aid. Thus, it was first priority to ensure the financing of pensions in the short run. Instead of finalizing the planned split of contributions between the first and second pillar in 2010 (10 percent for each), in 2009 the rate diverted to the fully funded pillar was lowered from 8 to 2 percent in order to close the financing gap of the first pillar (implying correspondingly higher entitlements in future).8 Additionally, the limit on earnings subject to social insurance contributions was lifted from 2009 to 2013, and for the same period the adjustment of pensions in payment according to consumer price changes was suspended. Further savings accrued from the functioning of the NDC-type first pillar: fewer contributors and significantly reduced average earnings devaluate automatically the notional "assets" available for conversion into a pension. Otherwise identical employment careers assumed, claiming a first pension in 2009 or 2011 made a difference of more than 30 percent [European Commission 2012b, p. 296]. Changes aimed at the long-term sustainability of the first pillar have been legislated in 2012: normal retirement age will increase to 65, and the minimum period required for an old-age pension rises to 20 years.
3. Commonalities and differences
After 2008, all countries looked at the legislated similar changes of their pension systems for attaining savings on public pension expenditure in the short and long run. Suspended or less favorable indexation rules, applied in all eight countries, ease the financial troubles of public schemes most rapidly but, due to the base effect, will ripple through subsequent years (Table 2). Nominal cuts of pensions in payment (as happened in Hungary and Greece) are even more effective but can be problematic. Corresponding legislation has been ruled unconstitutional in Portugal, Romania and Latvia. Short term savings also result from closing early retirement pathways, tightening entry conditions or computing decrements when the pension is claimed prematurely. Such changes happened in all eight countries as was also true for increasing normal retirement age which, depending on the length of the phasing-in period, delivers medium or long term savings. However, the target ages set in the CEE countries remains (much) below those in the Italy, Spain or Ireland which appears comprehensible in view of further life expectancy at 65 being about three years shorter (Table 1, row 5). Likewise, long-term savings will mainly accrue from changes of the benefit formulae. In particular, tightening the contribution/benefit link as in NDC schemes (Italy and Latvia) or taking into account more insurance years (up to the whole employment career) lead to lower pensions for the newly retired. Whereas NDC schemes operate with a built-in life expectancy factor (when converting notional "assets"), all Southern European countries have introduced a sustainability factor that automatically changes system parameters (normal retirement or number of insurance years required for a "full pension") upon longevity developments. All eight countries have made steps forward to harmonization, be it by unifying different schemes in order to save on administrative costs and/or by removing existent privileges for certain occupational groups (like a lower retirement age or higher accrual rates). Predominantly, those equalizing reforms focused on public service workers and on women when they were still entitled to a lower pensionable age (however, Hungary and Romania remain exceptions in that respect).
As a result of the intensified reform activity, the predicted growth in public pension spending by 2040 will be considerably lower than was calculated for the Ageing Report 2009 (Table 1, row 8; [European Commission 2012a, pp. 142-4]). Since the most recent reforms were not included in the projections the increase should be even smaller.
The financial market crisis of 2008 posed serious problems for countries whose pension system relies heavily on private, fully funded components. In our sample this was Ireland. Because these events - occupational and sovereign pension funds suddenly losing much of their assets - are unforeseeable and cannot be ruled out in future, this vulnerability also affects the CEE countries where fully funded private pensions should play a larger role in the retirement income mix. The development of the mandatory second pillar there reveals a specific challenge that is related to the well-known "double payment problem": While the younger cohorts (including the middle-age who joined the 2nd pillar voluntarily) build up financial assets for a part of their retirement income out of the diverted share of insurance contributions, for several decades the pension entitlements of present-day retirees and older workers have to be honored. The lowered contribution revenues of the first pillar are insufficient to meet these obligations [Holzmann & Guven 2009, pp. 170, 230-1], and arising deficits could be covered out of the state budget only as long as it was not under pressure itself. However, this was exactly the case after the economic slump in 2009 and forced the governments in Latvia and Hungary (and elsewhere) to take action. The financial market crisis shook not only the public s confidence in fully funded pensions. It also triggered a rethinking among political actors in CEE countries, leading to a revision of the implemented multi-pillar model [Orenstein 2011; Drahokoupil & Domonkos 2012; Socha 2013].
4. Social impact of recent pension reforms
A look at the income situation of older people (65+) in 2011 reveals a diverse and surprising picture. The at-risk-of-poverty (AROP) rate, that means less than 60 percent of the weighted median income, was lowest in Hungary, followed by Latvia and Ireland. In these three countries and Romania, the AROP rate for the elderly is below that of the adult population below age 65. The opposite is true for Greece and Portugal (Table 1, row 12). It is surprising that nowhere has the AROP rate for the 65+ not changed much since 2008 and even fell considerably in Latvia (51.2% in 2008, 8.9% in 2011). This is mainly due to the fact that for example in Ireland, Greece and Latvia the median income threshold decreased as a result of unemployment and wage cuts, whereas at first pensions remained largely stable and indeed functioned as "automatic stabilizers". However, it does not mean that the elderly have become more "wealthy". Rather, as yet they have merely incurred smaller income losses than the population of employable age.
The altogether not too bleak present situation (at least in relative terms) might change when the reforms take full effect. The downside of containing the long term growth of public pension spending is lower replacement rates (Table 1, rows 9 and 10). These figures, however, only provide a rough clue because the calculations are based on standardized assumptions (e.g. 40 years of average earnings and drawing a pension from 65) regardless of the actual prevalence of "full" employment careers in a given country, participation in supplementary pension schemes etc. OECD data [2011, p. 129] show that in Greece and Portugal about 80 percent of the retirees received no more than the (contributory or means-tested) minimum pension and in Italy and Spain still more than 30 percent. These figures qualify the informative value of "replacement ratios".
In order to evaluate the living conditions of present and future retirees, a few further aspects have to be considered. In all eight countries the homeownership rate among the elderly is very high (Table 1, row 13), and few are still burdened with mortgages. If imputed rent is taken into account, regularly, income inequality and the AROP rate are reduced whereas disposable income increases [Sauli & Törmälähto 2010]. Thus, homeownership may partly compensate for low minimum benefits (Table 1, row 11). Effects in the opposite direction emanate from co-payments when utilizing the health care system. Those individual out-of-pocket expenses amount to more than one third (Greece and Latvia) or one quarter (Hungary and Portugal) of total health care funding [OECD 2012b, p. 129]. If pensioners are not exempted their disposable income is reduced. Furthermore, in none of the eight countries is the adjustment of pensions in payments linked to the development of wages anymore. The switch to consumer price indexation (or even temporary suspension) decouples retirees from gains in prosperity and exposes them to an increasing risk of relative poverty the older they get [European Commission 2012b, p. 83]. Finally, high unemployment not only means lower contribution revenues today, but also incomplete insurance records and, hence, less pension entitlements in future. Particularly since the contribution/benefit link has been tightened all over by taking account of more insurance years or the entire working life when benefits are calculated, "atypical" employment biographies imply a high poverty risk.
The austerity consensus now prevailing in Europe exerts hard pressure through ex- ternal actors on governments of European countries confronted with high public deficits and sovereign debt to also reform their pension systems. These pressures have been incomparably stronger than the "soft governance" that emanated from the Open Method of Coordination and have brought about a rapid policy change in the pension domain. To a large extent, the reforms legislated after 2008 have fun- damentally changed the course of national pension policy. We have observed (sec- tion 3.) a reversal from an established early retirement policy towards the prolon- gation of working lives, or the shift from orderly increases of pensions in payment towards more unfavorable indexing formulae, adjustment moratoria and even ben- efit cuts. In CEE countries, the multi-pillar path taken before was reconsidered or even completely left.
The post-2008 reforms have either been imposed on governments by actors in charge of lending money to those ailing countries (IMF, the Troika) or in the nation- al context defined as an irrefutable necessity (Italy and the 2011 reform in Spain). In either case, "normal" pension politics was suspended, and the reforms passed in an accelerated decision-making process without being prepared by expert commis- sions or negotiated with social partners (exception: Spain).
Political actors striving to implement pension reforms that aim at savings on expenditure face a dilemma: "grandfather clauses" and long phasing-in periods until complete implementation reduce their effectiveness, in other words, the short-term savings potential. In contrast, rapid implementation of drastic cuts meet resistance from labor unions, senior citizens and others, especially when the measures are unilaterally imposed by the government and are not based on compromises. Almost without exception, the post-2008 reform legislation was followed by a swift implementation process by which it was drastically intervened into the future plans of older workers (e.g. the rapid rise of normal retirement age in Italy) or the consumer customs of pensioners (e.g. the nominal pension cuts in Greece). Thus it cannot be taken for granted that the pension policy changes - including the automatic ad- justment mechanisms - will in fact be implemented as legislated, especially not if unemployment, both overall and in particular among young people, remains high and a lower retirement age appears as an outlet.
1 If not indicated otherwise, information on the contents of pension reforms was obtained from the International Updates of the US Social Security Administration (http://www.ssa.gov/policy/index.html), the country reports of the project Analytical Support on the Socio-Economic Impact of Social Protection Reforms (ASISP) (http://www.socialprotection.eu) and publications of the European Commission [2010 (Annex 6), 2012b] and the OECD [2012a].
2 The politics of pension reforms that occurred during the two decades prior to 2009 have been studied in detail for individual countries or in comparison [see e.g. Bonoli & Shinkawa 2005; Arza & Kohli 2008; Inglot 2008; Natali 2008]. Such fine-grained analysis of recent reforms in eight countries that takes into account the involved actors, their interests, and the political-institutional conditions is, due to limited space, not possible here.
3 It would have been quite possible to include further CEE countries in the comparison, for example Poland or Estonia whose retirement income system come close to the one in Latvia, and they concluded similar changes after 2008.
4 Beginning in 2015 (but possibly postponed), a basic pension of euro360 will be taken into account when the individual benefit is calculates. For new retirees with fewer than 15 contribution years the basic pension is means-tested.
5 Several hundreds of professions had been listed as "heavy" or "hazardous" implying the entitlement to a full pension after 35 years of contributions as early as 55. The 2010 reform stipulated that a revised Hst must not cover more than 10 percent of the labor force and a full pension should not be available before age 60 and less than 40 years of service.
6 During the period 2005 to 2008, Portugal (see below) did best in containing projected spending increases until 2050. When comparing the estimates of2008 and 20011 Greece is ranked first: The increase would be lowered by 8.7 percentage points (from 24.1 down to 15.4 per cent of GDP) if the legislated reforms were actually implemented [European Commission 2009, p. 291; 2012a, pp. 143, 328].
7 Already in 2005 3 billion euros had been transferred from the capital reserves of the state-owned banks' special pension scheme to cover the deficits of the general pension scheme.
8 In 2013, the division of the total contribution rate between the first and second pillar was changed to 14/6 percent and is expected to remain stable.
Arza, C, Kohli, M. (eds.), 2008, Pension Reform in Europe: Politics, Policies and Outcomes, Routledge, London/New York.
Bonoli, G., 2012, Blame Avoidance and Credit Claiming Revisited, in: Bonoli, G., Natali, D. (eds.), The Politics of the New Welfare State, Oxford University Press, Oxford.
Bonoli, G., Shinkawa, T. (eds.), 2005, Ageing and Pension Reform around the World: Evidence from Eleven Countries, Edward Elgar, Cheltenham.
Cerami, A., 2011, Ageing and the Politics of Pension Reforms in Central Europe, South-Eastern Europe and the Baltic States, International Journal of Social Welfare 20(4), pp. 331-343.
Drahokoupil, J., Domonkos, S., 2012, Averting the Funding-Gap Crisis: East European Pension Reforms since 2008, Global Social Policy, vol. 12, no. 3, pp. 283-299.
European Commission, 2009,2009 Ageing Report: Economic and Budgetary Projections for the EU-27 Member States (2008-2060), European Economy, no. 2, Office for Official Publications of the European Communities, Luxembourg.
European Commission, 2010, joint Report on Pensions: Progress and Key Challenges in the Delivery of Adequate and Sustainable Pensions in Europe, European Economy, Occasional Papers, no. 71, November, Brussels.
European Commission, 2012a, The 2012 Ageing Report: Economic and Budgetary Projections for the EU27 Member States (2010-2060), European Economy, no. 2, Brussels.
European Commission, 2012b, Pension Adequacy in the European Union 2010-2050, Publications Office of the European Union, Luxembourg.
Ferrera, M., 1996, The "Southern Model" of Welfare in Social Europe, Journal of European Social Policy, vol. 6, no. 1, pp. 17-37.
Fultz, E., 2012, The Retrenchment of Second-tier Pensions in Hungary and Poland: A Precautionary Tale, International Social Security Review, vol. 65, no. 3, pp. 1-25.
Ghinararu, C, 2011, Romania, in: Hirose, K. (ed.), Pension Reform in Central and Eastern Europe in Times of Crisis, Austerity and Beyond, International Labour Organization, Budapest.
Hacker, B., 2009, Hybridization instead of Clustering: Transformation Processes of Welfare Policies in Central and Eastern Europe, Social Policy & Administration, vol. 43, no. 2, pp. 152-169.
Hinrichs, K., 2011, Pension Reforms in Europe: Directions and Consequences, in: Eren Vural, I. (ed.), Converging Europe: Transformation of Social Policy in the Enlarged European Union and in Turkey, Ashgate, Aldershot.
Holzmann, R., Guven, U, 2009, Adequacy of Retirement Income after Pension Reforms in Central, Eastern, and Southern Europe: Eight Country Studies, The World Bank, Washington (D.C.).
Inglot, T., 2008, Welfare States in East Central Europe, 1919-2004, Cambridge University Press, Cambridge.
Katrougalos, G.S.-P., 1996, The South European Welfare Model: The Greek Welfare State in Search of an Identity, Journal of European Social Policy, vol. 6, no. 1, pp. 39-60.
Ministero delTEconomia e delle Finanze, 2012, Economic and Financial Document 2012. Section I: Italy's Stability Programme, Roma.
Natali, D., 2008, Pensions in Europe, European Pensions: The Evolution of Pension Policy at National and Supranational Level, Peter Lang, Bruxelles et al.
Nektarios, M., 2012, Greece: The NDC Paradigm as a Framework for a Sustainable Pension System, in: Holzmann, R., Palmer, E., Robalino, D.A. (eds.), Nonfinancial Defined Contribution Pension Schemes in a Changing Pension World, vol. 1, The World Bank, Washington (D.C.).
Obinger, H., 2012, Die Finanzkrise und die Zukunft des Wohlfahrtsstaates, Leviathan, Vol. 40, Nr. 3, pp. 441-461.
OECD, 2011, Pensions at a Glance: Retirement-Income Systems in OECD and G20 Countries, Paris, OECD.
OECD, 2012a, OECD Pensions Outlook 2012, Paris.
OECD, 2012b, Health at a Glance Europe 2012, Paris.
OECD, 2013, Review of the Irish Pension System, Paris.
Orenstein, M.A., 2008, Out-liberalizing the EU: Pension Privatization in Central and Eastern Europe, Journal of European Public Policy, vol. 15, no. 6, pp. 899-917.
Orenstein, M.A., 2011, Pension Privatization in Crisis: Death or Rebirth of a Global Policy Trend, International Social Security Review, vol. 64, no. 3, pp. 65-80.
Pierson, R, 1994, Dismantling the Welfare State? Reagan, Thatcher and the Politics of Retrenchment, Cambridge University Press, Cambridge.
Pierson, R, 2001, Coping with Permanent Austerity: Welfare State Restructuring in Affluent Democracies, in: Pierson, P. (ed.), The New Politics of the Welfare State, Oxford University Press, Oxford.
Riib, F.W., 2012, Rapide Politikwechsel in der Demokratie: Gründe, Akteure, Dynamiken und Probleme, in: Kersten, J., Schuppert, G.F. (eds.), Politikwechsel als Governanceproblem, Nomos, Baden-Baden.
Sauli, H., Törmälehto, V.-M., 2010, The Distributional Impact of Imputed Rent, in: Atkinson, A.B., Marlier, E. (eds.), Income and Living Conditions in Europe, Eurostat Statistical Books, Publications Office of the European Union, Luxembourg.
Simonovits, A., 2011, Impact of the Economic and Financial Crisis on Pension Systems in Central and Eastern Europe: Hungary, Zeitschrift für Sozialreform, Vol. 57, Nr. 3, pp. 287-298.
Socha, B., 2013, Government Presents Three Options for Future ofOFEs, Warsaw Business Journal, June 26.
Venieris, D.N., 1996, Dimensions of Social Policy in Greece, South European Society and Politics, vol. 1, no. 3, pp. 260-269.
University of Bremen