Pension Reform in Europe

Politics, Policies and Outcomes

Camila Arza, Martin Kohli

• Publisher: Taylor & Francis
• Year: 2007
• eBook ISBN: 9781134134366
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

This new book provides a cross-country comparative analysis of the key issues shaping the latest pension reforms in Europe: political games, welfare models and pathways, population reactions, and observed and expected outcomes.

Pension reform has been a top policy priority for European governments in the last decade. Ageing populations, changing labour market patterns and the process of European integration are the 'irresistible forces' pushing for reform throughout the region.

The Political Economy of Pension Reform evaluates the political forces that make pension reform viable in different national and institutional contexts and the nature of political bargains, actors and cleavages surrounding policy change. The volume also examines the nature and outcomes of pension reform experiences in Europe, searching for a solution to the financial challenge posed by growing pension budgets. By addressing the nature of change, the pathways of reform, and the outcomes of the new pension mix in the region, the authors conclude with an analysis of people's perceptions and attitudes towards pension policy and their acceptance or otherwise of different reform options.

This book will be of interest to students and scholars of international political economy, European politics, and social policy.

1 Introduction

The political economy of pension reform

Camila Arza and Martin Kohli

The issues

In current accounts of pension policy, and of the welfare state more generally, there is something of an implicit consensus that emphasises path dependence and obstacles to reform. This book takes issue with such accounts. By focusing on the most recent reform experiences, it observes a trend towards some convergence between different paths and towards substantive change. It offers several approaches to explain this process.

The idea of a limited number of welfare regimes in terms of specific patterns of institutions, and of path dependence in terms of change processes that would deepen rather than flatten this specificity, has been a powerful corrective to the earlier assumption that the dynamics of capitalist modernisation would eventually make all countries converge on a single institutional model. But the new consensus has in its turn decreased our capacity to observe and make sense of what is going on today, by unduly limiting our attention for changes that do not fit the assumed paths, and our tools and concepts for giving them form.

The decades since the end of the Second World War have throughout been a period of major transformation in pension policy. Pension reform has become central to the European social policy agenda – first in terms of construction and expansion, then increasingly in terms of consolidation and retrenchment. The high levels of pension expenditures experienced in the past few years, and projected for the coming decades, have become a key concern for fiscal and labour market policy and economic growth.

Pension systems thus need to be viewed in a broader political economy framework. Their major purpose is to provide income security to retirees. In addition to such redistribution across the life course, they may also aim at redistribution across population groups, such as lifting the low-income elderly out of poverty. But beyond these goals, they are linked up with a range of other issues.

• They are typically the largest public transfer programmes, and thus the source of major fiscal pressures (and sometimes opportunities).
• They influence financial markets by creating or impeding the accumulation of funds and the rate of personal savings.
• They regulate labour markets by facilitating an ordered transition out of employment.
• They enable employers to manage their workforce by offering instruments for the shedding or replacement of workers.
• They contribute to the institutionalisation of the life course by creating a predictable sequence and timing between work and retirement.
• They provide workers with a legitimate claim to compensation for their ‘lifelong’ work, and thus with a stake in the moral economy of work societies.
• They produce new social and political cleavages by creating large groups of actual and potential beneficiaries.
• They structure the agenda of corporatist conflict and negotiation.
• They offer opportunities for administrative offices and jobs.
• They weigh in on election outcomes.

Through all these issues, they form a major part of the political economy of current societies.

In this introduction, we take up the issues first through an examination of the historical evolution of pension systems (and by that, of retirement as a universal new life stage), and second, through a discussion of the research literature. In the third section, we present an overview of the book’s contributions, and finally come back to its overall results.

The evolution of pension systems

The origins of modern public pension policy can be traced back to the last few decades of the nineteenth century, when the first two pension schemes (which would become the two key models for pension policy-making) were set up: the (work-based earnings-related) German scheme in 1889, and the (universal flat-rate) Danish scheme in 1891. Following these examples, either universal or work-based pension systems were created over the first half of the twentieth century in all European countries. The early schemes followed different models or regime types, but they all tended to provide low benefits at rather high retirement ages. The idea of ‘retirement’ was just starting to be constructed and pensions were often conceived more as disability allowances than as retirement benefits as such (cf. Kohli 1987). Retirement age was around 70 years in most countries, and life expectancies were low, thus making the period of benefit receipt rather short. As a result, expenditures in pensions remained modest relative to current levels.

The big expansion of pension systems came after the Second World War. In countries where pensions were restricted to some specific occupational sectors, coverage was broadened to the entire working population. Countries with only basic income protection (flat-rate or means-tested benefits) also expanded coverage, sometimes eliminating the means-testing, sometimes including new earnings-related layers in the public scheme, or mandating occupational schemes. Eligibility became more generous, normal retirement ages were reduced, and early retirement options were introduced in many countries (cf. Arza and Johnson 2005). In some cases, easier-to-meet eligibility rules were only applied to some occupational categories, reflecting the hazardous nature of some occupations, but also political power and influence. By and large, with the expansion of coverage and benefit generosity, pensions became a comprehensive system of income protection in old age. Their role for public policy broadened as they increasingly became a key instrument for industrial restructuring and for managing unemployment (Kohli et al. 1991).

Retirement as a universal new life stage thus became fully institutionalised in the second half of the twentieth century only. It was fuelled by the economic boom of the 1950s and 1960s when many countries started to provide pensions at a level of wage substitution – either through public pay-as-you-go systems or through broad occupational pensions – which allowed for a full exit from the labour force at a specific (and increasingly early) age. The long-term evolution of retirement has been striking, as can be demonstrated for Germany, which in 1889 introduced the first public pension system available for large parts of the population. Between 1881–1890 and 2002–2004, the proportion of men surviving to age 60 has increased from 33.5 to 87.8 per cent, and the average life expectancy at age 60 from 12.4 to 20.1 years. These added years are now increasingly spent in retirement: the labour force participation rate of men aged 60 or more has dropped from 67.9 per cent in 1895 to 14.4 per cent in 2004. In other words, retirement has become a life stage of its own, to be expected by the majority of the population, of considerable length and structurally set apart from gainful work (see Kohli 2000).

The long periods involved in the maturation of the new pension rules introduced after the Second World War meant that in many countries their full financial impact would not be observed immediately, but only some decades later, when the generations under these schemes started to retire. More importantly, the age structure was still that of a young and growing population, with a broad base of young and a narrow top of older ages. As a result, pension expenditures in the 1960s and 1970s were still rather low. In 1960, they amounted to 3.6 per cent of GDP in Sweden, 3.3 in Italy, 3.7 in the Netherlands, 2.6 in France, 5.9 in the Federal Republic of Germany, 1.2 in Spain and 3.1 in the UK. In 1980, they had more than doubled in most countries, reaching 9.6 per cent of GDP in Sweden, 6.9 in Italy, 11.4 in the Netherlands, 7.6 in France, 9.6 in the Federal Republic of Germany, 5.7 in Spain and 5.5 in the UK. Pension expenditures generally continued to rise over the 1980s to reach, by 1989, over 9 per cent of GDP in France, Germany and Greece, and over 11 per cent in Sweden, Italy and the Netherlands.¹ Further expenditure growth projected for the new century started to be seen as a risk for the sustainability of public finances and the competitiveness of national economies.

Over the 1950s, 1960s and early 1970s, in an environment of sustained economic growth, the impact of pensions on public finances was less of an issue than it is today. Governments allocated a significant part of their budgets to welfare expansion, in a context where public expenditure and aggregate demand were seen as key ingredients in the economic growth strategy. But things started to change after the mid-1970s. Growth rates fell, population ageing became more pronounced, and pension systems matured. Economic ideas started to shift away from Keynesianism, towards new supply-side policies which stressed productivity, international competitiveness and stringent public finances. The high wage contributions required to finance the growing level of pension expenditures (especially with pay-as-you-go financing) did not fit these ideas. The rising level of future pension commitments was also seen as a risk. Political attention thus started to shift to pension reform: in economic as well as in political terms the key puzzle was how to make existing pension schemes financially sustainable for the future while maintaining their effectiveness.

The rhetorics of ‘reform’ have also been important. Most of the policy steps that currently go under this label consist of ‘retrenchment’, in other words, of cuts in existing welfare state programmes. More neutral terms for what is going on would be ‘change’ or ‘transformation’. The choice of terms is clearly not innocuous. As Vivian A. Schmidt has observed, ‘no major and initially unpopular welfare-state reform could succeed in the medium term if it did not also succeed in changing the underlying definition of moral appropriateness’ (Schmidt 2000: 231), and changing this definition requires convincing rhetoric and discourse. We speak of ‘reform’ here because it has become the general terminological currency. It also has some basic arguments going for it, in the sense that the existing pension schemes face increasingly stringent financial challenges, partly through the combined demography of low fertility and increasing life expectancy, and partly through the stronger exposure of national economies to global competition. But the translation of fiscal pressures into specific institutional changes owes much to the new mainstream of economic thinking and lobbying about the need for welfare state retrenchment and the merits of privatisation.² This should be kept in mind when speaking of ‘reform’.

Solving the financial puzzle was certainly not easy in the new demographic context. The proportion of the total population over 65 years of age, which in 1950 was at only 9.1 per cent for the EU15 average, reached 14.3 per cent in 1990, and is projected to grow up to 20 per cent in 2020 and 27.6 per cent in 2050.³ As to expenditure projections, estimations in the 1980s were that, in a nochange scenario, pension expenditures would soar to 20.8 per cent of GDP for the EU15 average in 2050.⁴ Although some doubts have lately emerged on how precise these estimations were, it is agreed that a major growth of pension expenditures was to be expected. In any case, these projections granted pension reform a privileged place on the policy agenda. Governments throughout Europe started to discuss reforms which could go from small parametric adjustments all the way to major structural change.

Since the late 1980s, many reform plans have been approved and implemented. About 25 pension reforms affecting either specific groups of the population or the entire pension system were passed in EU countries over the period 1986–1990.⁵ In the following five-year period (1991–1995), this number increased to 36, and in the past few years (from 1996 to 2002), to 55.⁶ Although the number of reforms is not a comprehensive indicator of the magnitude of the changes taking place, it gives some illustrative idea of the importance of the pension issue on the policy agenda. The direction of reform has been, more often than not, towards a reduction of future expenditures, and increasingly so as time went on. While just over half the reforms introduced in the 1986–1990 period tended to reduce the generosity of the system, over three-quarters of the reforms passed between 1991 and 1995 moved in that direction. This was the period of ‘cost-containment’ during which indexation rules were modified, access to early retirement schemes was restricted, retirement ages were raised and a number of other specific parametric adjustments were introduced with the aim to restrict eligibility and reduce future benefit levels. In the most recent period (1996–2002) reforms have continued in this direction although in many cases they have also introduced a new feature: the creation of compulsory or voluntary private individual pension accounts (encouraged through tax incentives), with the aim to at least partly compensate for the projected fall in public benefits.

Beyond bare retrenchment, one of the key strategies of reform throughout Europe has been to operate via incentives: incentives to work, incentives to save, incentives to retire later. Labour market activation has become a key feature of European social policy. Early exit from the labour force, long encouraged by consensual strategies of all labour market actors, has come to be considered one of the central problems facing pension finances. While raising the retirement age limit beyond the traditional threshold of 65 remains highly contentious, raising the labour force participation below this threshold is now a broadly consensual goal, as stated, for example, in the Lisbon and Stockholm agenda of the EU which asks member states to reach a labour force participation rate of at least 50 per cent among the population aged 55–64 by the year 2010. Institutional incentives embedded in new eligibility rules and pension formulas are aimed at pushing the retirement decision up to later years. The critical issue that has generated conflict here is to what extent this is indeed a free decision by the worker, and to what extent it is enforced by, for instance, health reasons or labour market conditions. Defined-contribution arrangements, in which the level of benefits depends on contributions made, are often advertised partly because of their expected impact on labour market participation, even though the same goals could be reached through defined-benefit arrangements with actuarially fair discounts for early retirement. The decision to introduce tax incentives for private pension schemes (e.g. in Germany, UK, Italy) has also been a political one. Incentives can be appealing for all because, in principle, there is no forced change in behaviour: those who just wish to ignore them can do so. In practice, the introduction of incentives has often been combined with some forced change, for instance, when rejecting the ‘incentive’ to work longer under defined-contribution formulas means receiving a lower benefit. Moreover, as tax incentives have a public budget cost which is paid by the whole population, individuals in their role as tax payers cannot really opt out of them: they continue to pay for the incentives taken by others.

These reforms have entailed a new role for the market in the provision of pensions. Privately administered funded pensions, investing workers’ contributions in the financial market, were not a European invention. Already in the 1980s, Chile had established the first large-scale private system of individual pension accounts. In 1994, the influential World Bank report Averting the old age crisis advocated the development of these schemes as one of the pillars in the ‘three-pillar model’ that has been largely applied in Latin America first, and Central and Eastern Europe later on (World Bank 1994). Although the World Bank prescriptions did not have a direct impact on Western European policy-making, the idea of pension funding and savings has gained greater attention by policy-makers as a new instrument to deal with the sustainability of public pension finances. In a step by step process, European countries have started to introduce voluntary or compulsory funded schemes, which are in most cases privately administered (the exemption being Sweden) but with various degrees of public regulation – an issue of sustained contention. This has occurred as much in countries with a long history of private provision (like the UK), as in those where a public single-pillar pay-as-you-go system was dominant (like in Italy and Germany). The idea of funding has not been restricted to private sector pensions. Some countries have introduced (or developed existing) ‘buffer funds’, i.e. a system of asset accumulation for public pensions aimed at guaranteeing financial sustainability over the demographic transition. As a result, total pension assets have been projected to increase markedly in many countries over the period from 2004–2050: from 135 to 243 per cent of GDP in the Netherlands, from 39 to 61 per cent in Sweden, and from 52 to 73 per cent in Finland.⁷ Similar rises of pension assets have been projected for the Central and Eastern European countries where the shift to funded pensions was compulsory.

The research literature

Just as pension systems and pension reform have been central to policy agendas, they have been central to the welfare literature. Since Esping-Andersen’s seminal Three worlds of welfare capitalism (Esping-Andersen 1990), the welfare regime concept has influenced most welfare research. ‘Regime theory’ has been oriented to evaluate how the diversity of institutional designs across countries was affected by different political orientations (and hence, ideas and power resources of different groups), and has produced different welfare outcomes – thus creating clusters of countries with similar institutional design, similar political orientations and similar outcomes. A substantial part of the research that followed was aimed either at empirically evaluating these welfare clusters (e.g. Goodin et al. 1999), at generating new clusters and typologies (Castles and Mitchell 1993; Ferrera 1996, 1998), or at revising the overall methodological approach (Kasza 2002). For pension systems themselves, recent reforms have put both ‘regime theory’ and the earlier Bismarckian-Beveridgean classifications under greater scrutiny, and new analyses have emerged to address the most recent reform pathways and the new ‘pension regimes’ (Myles and Quadagno 1996; Bonoli 2003; Natali 2004b, 2005; Ebbing-haus 2006; and Arza 2006). The reforms have changed what originally was defined as models and ‘regimes’. As countries with different systems have adopted similar design features, the original classifications are no longer appropriate for comparison. Chapters 6 and 7 in this book are concerned with pension models and their outcomes. But rather than aiming at an overall classification of countries in clusters, they examine the specificities of system design, and use the results to comparatively evaluate reform pathways and impacts. This takes account of the increasing complexity of pension systems (and of the combination of ‘layers’ and ‘pillars’ within e...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Illustrations
• Contributors
• Preface
• Abbreviations
• 1 Introduction: The political economy of pension reform
• Part I The politics of pension reform
• Part II Reform options and outcomes