%20-%20https://sketch.com%20--%3e%3ctitle%3e3ECBBF50-FD08-412A-943D-2CFC5E025A16%3c/title%3e%3cdesc%3eCreated%20with%20sketchtool.%3c/desc%3e%3cg%20id='Stylesheet'%20stroke='none'%20stroke-width='1'%20fill='none'%20fill-rule='evenodd'%3e%3cg%20transform='translate(-926.000000,%20-668.000000)'%20id='BUTTONS'%3e%3cg%20transform='translate(100.000000,%20597.000000)'%3e%3cg%20id='BTN-/-RIGHTS'%20transform='translate(816.000000,%2065.000000)'%3e%3cg%20id='ICON-/-Rights'%20transform='translate(10.000000,%206.000000)'%3e%3cpath%20d='M13.5752841,0%20L1.34943182,0%20C0.59375,0%200,0.586666667%200,1.33333333%20L0,22.6666667%20C0,23.4133333%200.620738636,24%201.34943182,24%20L17.6505682,24%20C18.40625,24%2019,23.3866667%2019,22.6666667%20L19,5.36%20L13.5752841,0%20Z%20M13.5752841,1.89333333%20L17.0838068,5.36%20L13.5752841,5.36%20L13.5752841,1.89333333%20Z%20M1.34943182,22.6666667%20L1.34943182,1.33333333%20L12.1988636,1.33333333%20L12.1988636,5.36%20C12.1988636,6.10666667%2012.7926136,6.69333333%2013.5482955,6.69333333%20L17.6235795,6.69333333%20L17.6505682,22.6666667%20L1.34943182,22.6666667%20Z'%20id='Shape'%20fill='%23595959'%20fill-rule='nonzero'%3e%3c/path%3e%3ccircle%20id='Oval'%20stroke='%23595959'%20stroke-width='1.5'%20fill='%23595959'%20cx='9.5'%20cy='13.5355339'%20r='5.5'%3e%3c/circle%3e%3cpolyline%20id='Path-2'%20stroke='%23FFFFFF'%20stroke-width='1.5'%20transform='translate(9.500000,%2013.035534)%20rotate(45.000000)%20translate(-9.500000,%20-13.035534)%20'%20points='8%2016.0355339%2011%2016.0355339%2011%2010.0355339'%3e%3c/polyline%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/svg%3e)
Introduction
In EU countries, national pension systems have differed greatly in the past. While all of them were based on a public scheme as the most important means to old-age income, some of these public schemes were based on a residency-related ‘people's pension’, with huge differences in the level of the flat-rate pension, while others were based on labour-market-related insurance systems. And although pension systems were clustered – following a Beveridgian versus Bismarckian setup, or a setup related to welfare regimes – differences within these clusters themselves were enormous. Market principles however, in general, played a marginal or even negligible role in past pension-system setups.
As a consequence of a variety of developments, among which were demographic challenges, social changes, highly competitive global markets, the European Economic and Monetary Union (EMU) and ideological shifts, the question was raised as to how pension systems should be designed to offer ‘adequate and sustainable pensions for all’ also in the future, as the European Commission terms it (European Commission, 2010). This question was answered by different international organisations, such as the World Bank and the Organisation for Economic Cooperation and Development (OECD), with a seemingly technical pension design that mixes different ‘pillars’ or ‘tiers’, and a particular sympathy for ‘funded’ and market-based pensions to reduce demographic risks and positively influence the financial markets (this argument is to be found in all national and international reports, e.g. OECD, 2009: 89). Although this perspective was widely criticised by academic experts (Barr, Reference Barr2002; Gilbert, Reference Gilbert2002; Sinn, Reference Sinn2000), pension reforms that have taken place since the early 1990s in European countries (EU and Central and Eastern European Countries (CEEC)) and elsewhere in the world (e.g., China, Chile) were deeply influenced by these viewpoints. In the end, framed by the ‘Open Method of Coordination’ and restrained by legal and economic limitations, European pension systems were redesigned, with non-public, market-based pensions introduced or reinforced (see Ebbinghaus and Gronwald, Reference Ebbinghaus, Gronwald and Ebbinghaus2011; Hyde and Dixon, Reference Hyde and Dixon2009; OECD, 2011). Notwithstanding this systemic transition from public to hybrid systems of old-age protection, pension institutions’ ingredients, characteristics and nexus are far from being homogeneous; as a consequence, the strengthening of market principles in pension systems has different impacts on the attainment of pension rights and future pension incomes.
This contribution analyses the manner in which market principles are institutionalised in the hybrid pension system and how it determines the degree of social protection afforded to the various social citizens and the number of future pensioners with adequate pensions.
State of the art
The literature on pension reforms is enormous; however, nuanced analyses of the institutionalisation of hybrid pension systems and its impact on social-risk spreading are lacking. Of main importance are two streams of analysis of the market components in pension systems: the first concerns the market structures in hybrid pension systems, and the second the application of market principles to them, and both in regard to social risks. In addition, a reflection on the analysed pension level is called for.
One of the main changes in pension design is the establishment of hybrid pension systems by (1) outsourcing parts of the public system to market-based sub-systems while (2) applying the former objectives of social security. As to (1), much of the formerly social contributions to the public scheme are now invested in market-based personal schemes (the insurance market, stock market, etc.), which are offered by profit-oriented providers. This development is interpreted by the OECD (2009: 97) as ‘implicit or explicit “privatisation” of part of the retirement-income system’. However, marketisation is not the same as privatisation (Barr and Diamond, Reference Barr and Diamond2009; Ebbinghaus and Wiß, Reference Ebbinghaus, Wiß and Ebbinghaus2011; European Commission, 2010; Hill, Reference Hill2007; Hyde and Dixon, Reference Hyde and Dixon2009), since the latter might concern the privatisation of risks, the collection and payment of financial resources or the responsibility or administration of pensions. The conceptual differentiation of public and private schemes is, therefore, blurred (Harvey and Maier, Reference Harvey, Maier, Clasquin, Moncel, Harvey and Friot2004; Hyde and Dixon, Reference Hyde and Dixon2009).
Privatisation does not adequately cover the actual dynamics, since strengthening private pensions, for instance, does not necessarily result in strengthened market principles; this depends on the degree of state regulation (Berner, Reference Berner and Leisering2011; Ebbinghaus and Wiß, Reference Ebbinghaus, Wiß and Ebbinghaus2011; Leisering, Reference Leisering2011). Therefore, the regulation of market-based pension sub-systems is at the heart of this contribution. Three decades ago, Myles (Reference Myles1984) already showed that welfare-, quasi- (Le Grand and Will, Reference LeGrand and Will1993) or social-markets (Gilbert, Reference Gilbert2002) construct new kinds of social rights. And indeed, those market-based pension sub-systems that this contribution focuses on have been institutionalised by the welfare state with the explicit aim of enabling citizens to reach the target replacement rate identified by the welfare state. Therefore, they are highly regulated and co-financed by public means (Gilbert, Reference Gilbert2002). In short, within the institutional frame for reaching adequate pensions there is no purely market-based sub-system. As opposed to welfare markets for social care, however, analyses of welfare markets for social security are rather scarce (Hacker, Reference Hacker2002, for the US; Hyde and Dixon, Reference Hyde and Dixon2009, for mandated private pensions). They are very different from welfare markets for social care since they engender other forms of markets and long-term monetary transfers (see Berner, Reference Berner and Leisering2011; Harvey and Maier, Reference Harvey, Maier, Clasquin, Moncel, Harvey and Friot2004) and other linkages between public and market-based sub-systems.
Regarding (2), taken together, the very schemes of hybrid pension systems are explicitly assumed to enable citizens to build up adequate social protection (European Commission, 2010). Consequently, the linkage of the sub-systems matters; it is, however, constructed with great variation in the nationally specified schemes and measures (Ebbinghaus and Wiß, Reference Ebbinghaus, Wiß and Ebbinghaus2011; Hyde and Dixon, Reference Hyde and Dixon2009). Some authors (Ebbinghaus and Gronwald, Reference Ebbinghaus, Gronwald and Ebbinghaus2011; Palier and Martin, Reference Palier and Martin2008) interpret these developments as path departures from Beveridgean and Bismarckian systems or from welfare-regime types. Indeed, differences in pension traditions have decreased since market-based sub-systems have been integrated into all pension systems. To identify a global development towards convergence in multi-pillar systems (Bridgen and Meyer, Reference Bridgen, Meyer, Meyer, Bridgen and Riedmüller2007; Bonoli and Shinkawa, Reference Bonoli and Shinkawa2005) is however meaningless, since multi-pillar systems are extremely diverse. Explorative studies support this criticism (e.g., Aysan, Reference Aysan, Vanhuysse and Goerres2012). The ‘additive’ concept misses important details in the schemes (and in particular in their combination) that are of particular importance for building up pension rights. Therefore, I oppose the argument that introducing market-based schemes into pension systems or reinforcing them necessarily reduces social risk-sharing, as argued by Blackburn (Reference Blackburn2006) and Rowlingson (Reference Rowlingson2002). What matters is – as my first hypothesis – the concrete regulation of the welfare market and its linkage with other pension sub-systems.
Additionally, multi-pillar concepts of pensions do not shed light on the principles that are applied by the hybrid systems. International and comparative studies have shown that institutional regulations based on economic principles, on the one hand, and institutional regulations aiming to provide adequate social security, on the other hand, might not be compatible (Barr and Diamond, Reference Barr and Diamond2009; Hill, Reference Hill2007; Hyde and Dixon, Reference Hyde and Dixon2009; Pension Commission, 2005). This is interpreted by Mabbett (Reference Mabbett2008) and Taylor-Gooby (Reference Taylor-Gooby2004) as the result of incompatible values underlying the different institutions. According to these authors, a diversity in values or underlying principles between the market and public institutions implies per se that they are not complementary and thereby incommensurate. International comparative empirical studies which prove that public and market-based institutions indeed have opposite effects are lacking. In fact, this interpretation misses two particularities of factual welfare-state reforms. First, within the hybrid pension systems there are no market-based pensions which build purely on market principles, since these welfare markets have been highly adapted to welfare institutional aims. Secondly, in public pension schemes, market principles exist and have been strengthened (already shown by Myles, Reference Myles1984; Frericks, Reference Frericks2010).
Market principles as applied in hybrid pension systems differ in kind and degree and thereby in effect. They differ in kind because providers establish particular schemes, and in degree because they apply market principles each to a particular extent. Market principles include profit-orientation, competition, consumer choice, allocation through the ability to pay and individual responsibility. The latter refers to the individual's choice of an appropriate work–life balance and appropriate savings. The individualisation of benefits thereby often goes hand in hand with an individualisation of risks (Marshall, Reference Marshall1964; Saunders and Wong, Reference Saunders and Wong2011).Footnote 1 Those market-based schemes that were established or reinforced when public schemes were reduced, i.e. the schemes that are of interest in this analysis, however, also apply principles which are untypical of markets, such as collective responsibility for resources or different forms of redistribution, including, for instance, unisex life-tables (Barr and Diamond, Reference Barr and Diamond2009). Concurrently, public pension systems emphasise some principles normally referred to as ‘market principles’, such as various forms of individual responsibility including the individual obligation to continuously work (Frericks, Reference Frericks2007). Market principles – as my second hypothesis – differ in effect since they are interlinked with other principles and institutionalised in particular ways.
The distribution of social risks among the population is affected by processes of marketisation in social protection (Bonoli, Reference Bonoli2005; Myles, Reference Myles, Esping-Andersen, Hemerijck, Myles and Gallie2002). However, the concept of social risks – and more so its opposite, that of adequate social protection – lacks precision in the literature (see Frericks, Reference Frericks2011). For old age, adequate pensions are mostly interpreted as social rights to universal, non-means-tested basic pensions (Leitner and Lessenich, Reference Leitner and Lessenich2003; Twine, Reference Twine1992). This, however, conflicts with the concept of social rights put forward by Marshall in two ways: the level of social security and the institutionalisation of rights (Marshall, Reference Marshall1964). Marshall understood social rights as historically and geographically differing institutional implementation of a socially acknowledged ‘idea’ of income security. This idea, resulting from the perception of shared needs, manifests itself in the right to income security on a level defined by society (Marshall, Reference Marshall1964). Apart from some exceptions (e.g., Denmark), all European countries apply such a socially defined and legally regulated level of income security in old age (Berner, Reference Berner and Leisering2011).
‘Adequate and sustainable pensions for all’ is a key goal of European pension reforms, but the concept of adequate pensions is by far not used in a uniform way and, often, adequacy is not concretised (e.g., European Commission, 2010; Hill, Reference Hill2007; OECD, 2011; WB Report, 2005). In empirical studies, mainly microsimulations, adequacy is in general referred to as the predictable level of pension revenue (European Commission, 2010; Dekkers et al., Reference Dekkers and d'Italia2009; Palme, Reference Palme, Goodin and Ware1990). Simulations thereby apply some standard biographies, such as the ‘base case’ (EU) or the ‘average production worker’ (OECD). The calculated replacement rates are then referred to as an ‘inclusion threshold’ (Bridgen and Meyer, Reference Bridgen, Meyer, Meyer, Bridgen and Riedmüller2007), such as the poverty line of 60 per cent of average income, which is interpreted as adequacy of pension income (OECD, 2009; European Commission, 2010). For my analysis, this method is inadequate since it reflects the distribution of future pension income instead of the institutional framing of pensions. In short, there is too little reflection on the institutionalised target level of social protection in pension literature.
A comparative analysis of the current hybrid state–market Dutch and German pension systems will show that what matters in terms of adequacy is the institutional setup of the hybrid systems and the way market principles are implemented in pension systems. In other words, market principles might be incommensurate with social rights principles; however, market-based and public pensions are not necessarily incommensurate. The setup determines how risks are attributed to the individual or a collective, whether (and in what way) intra- and intergenerational redistribution takes place and influences different risks, such as those related to demographic change, different forms of labour-market participation, financial investments, biographical configurations, and so forth. The approach of this institutional analysis seems obvious; it is, however, innovative. Therefore, the clarification of aims and methods is crucial.
Aims and methods
The aim of this contribution is to analyse the impact of strengthened market principles in pension systems on social-risk spreading. To do this, the conceptual design of this study answers the following questions: Where in the pension system do we observe a strengthening of market principles (point of reference)? What do we mean by strengthening market principles (independent variable)? And how might we assess the impact of this on social-risk spreading (dependent variable)?
As stated above, empirical and conceptual literature on the target replacement rate of pension systems is rather limited and insufficient to capture what is meant by ‘adequate’ pensions. Often misinterpreted as a vague or normative idea of a researcher, the concept of adequate pensions forms an essential part of various pension systems. Here, the concept is used as the institutionalised and nationally defined target level for old-age protection, i.e. the pension calculation norm. It serves as a conceptual tool to analyse institutional functioning (with regard to the institutional aims), and not to exactly measure or even evaluate national calculation norms. Adequacy was, in most European countries, related to the family or the family wage, to the ‘team’ of the married couple as termed by Marshall (Reference Marshall1964). In Germany, a term was created to cover the national calculation norm of the adequately insured: the Eckrentner; other countries created their specific calculation norms, often still referred to as ‘full pensions’ (see also the National Strategy Reports to the EU – European Commission, 2013). In the context of various reforms and adaptations of pension systems, the conditions for building up adequate pensions are being changed by the increasing emphasis on the individual and involvement of different pension sub-systems. Together these modifications are explicitly assumed to enable citizens to build up adequate social protection. The different schemes are institutionalised by the welfare state (through taxes, regulations, etc.), and building up pension rights in the different schemes is necessary to achieving an adequate pension. Therefore, together they form an institutional constellation with regard to adequate pensions. To guarantee adequate pensions ‘for all’, the institutional constellation needs necessarily to enable the various citizens to comply with the conditions applied in it. The focus of this contribution is, therefore, the institutional setup of hybrid systems with regard to market principles, and the chances of the various citizens to reach the ‘adequate’ level of pensions and maintain a nationally defined standard of adequacy. Conceptually, I differentiate between three levels of pensions: (1) a basic level to prevent poverty, (2) the institutionalised and nationally defined target level for old-age protection (adequate pensions), and (3) pensions at a level intended to maintain previous living standards. I focus on the second level only – corresponding to Marshall's understanding of social rights. Consequently, the concept of ‘adequate’ pensions takes into consideration only a part of factual pensions; additional pension investments, whether occupational or private, do not form part of the institutional constellation of adequate pensions.
The strengthening of market principles is used in this contribution as the independent variable – understood as the increased application of market principles in public pensions and the (explicitly or implicitly) outsourced part of public pensions in the form of market-based pensions. As the dependent variable, I analyse the resulting (greater or lesser) social risks. The institutional regulations reflect and co-determine the construction of the ‘normal citizen’ and, as a consequence, citizens’ opportunities to comply with the conditions for adequate old-age security. In particular, social risks related to these opportunities might increase whenever market principles are reinforced in the welfare institutions that frame old-age social security, and thus social risks would be spread differently among the population.
My analysis is theoretically based on the combination of two concepts. First, my approach builds on the concept of social citizenship as proposed by Marshall (Reference Marshall1964), with its focus on public responsibility for adequate social protection as it is now called (formerly, employment-related opportunities to build up entitlements, and family-related derived rights). Secondly, the concept of flows of resources related to social rights (Harvey and Maier, Reference Harvey, Maier, Clasquin, Moncel, Harvey and Friot2004) underlines the importance of intergenerational and intragenerational redistribution and differences in financial opportunities to build up pensions as set up in the national pension design. This concept focuses on resource flows as a dimension of social citizenship, as put forward by Marshall, related to the redistribution of public means, the inclusion of low-income earners and the inclusion or the degree of marginalisation of non-labour-market-related biographical periods (e.g., those spent in caregiving).
The analysis of social citizens’ opportunities to build up adequate social protection in old age is operationalised in terms of the following indicators: the marginalisation of some groups and changed redistribution, the guarantees on old-age investments and risk-sharing, and the share of people covered by the various necessary schemes. The latter tells us about the overall inclusion of social citizens in the social protection schemes necessary to build up adequate pensions; guarantees on the necessary investments indicate public responsibility taken to secure these investments and to collectively share investment risks of old-age investments that are particularly difficult to foresee (Barr, Reference Barr2002; Harvey and Maier, Reference Harvey, Maier, Clasquin, Moncel, Harvey and Friot2004); the marginalisation of some groups shows us who is indeed marginalised, since non-participation in those schemes does not necessarily result in social risk or inadequate social protection because some groups build up other resources, such as, in the German case, civil servants or the self-employed. The redistribution, thereby, provides an insight into the financially supported groups in society and the groups that lack financial support to build up pension protection.
The empirical part of this paper focuses on two countries: the Netherlands and Germany. The choice is based on their very different social-protection traditions as manifested in their differing institutional setups for old-age security, which are contribution-based social insurance linked to labour-market participation in Germany, and a payroll-tax-financed ‘people's pension’ in the Netherlands, in which entitlements derive from residency. The analysis will centre on the impact that the kind and degree of strengthening of market principles, and the concrete linkage of market-based and public institutions in each regulatory system, have on the spreading of social risks.
The comparison of the linkage of public and market-based pension protection in the Dutch and German cases is based on the analysis of policy documents (national and European), publicly available national and international data (e.g., OECD, relevant national databases) and documents (e.g., the old-age security report, ASIB, for Germany). To better understand the different functionings of the partly marketised regulative systems, interdisciplinary studies were consulted and expert interviews with national experts (ministries, advisory boards, pension providers and academia) were undertaken. With this comparison of two very different institutional constellations, which frame adequate social protection in old age, I want to deliver a first elaboration of the characteristics of institutional constellations in which market principles have been strengthened – which imply different forms of risk-spreading.
Analysis: strengthening market principles and its impact on social-risk spreading
Differences in the design of hybrid pension systems significantly determine the degree of social protection of the various social citizens and the number of future pensioners with adequate pensions. To analyse the different impacts of the two institutional setups of current pension systems as a combination of public and (differently) market-based components on social risk spreading, I analyse the following parameters: (1) the pension ingredients that determine the marginalisation of some groups and the redistributive effects, (2) the pension characteristics that determine risk-sharing principles and mechanisms including the guarantees on investments and (3) the pension nexus determining the coverage of the population in the necessary sub-schemes. Conclusions will be drawn about the opportunities social citizens actually have to build up adequate pensions in their national hybrid pension system.
Ingredients in current hybrid systems: marginalisation and redistributive effects
In both the Dutch and the German systems, additional market-based old-age protection schemes were introduced, or further marketised, with the aim of compensating for cuts in public pension provision. In the Netherlands, the absolute and relative pension level provided by the public people's pension, which is financed through any form of income tax (and in case of need supplemented by general means), was indirectly reduced since it was detached from wage growth in 1980 (the decrease between 1980 and 1998 was already 25 per cent, Delsen, Reference Delsen2000). Therefore, it became necessary for people to build up so-called second- and possibly third-pillar entitlements to ensure adequate pensions, interpreted in the Netherlands as 70 per cent of the average individual income. The so-called first pillar is independent of participation in the labour market; it depends on the number of years of residency. The proportions of the different pension schemes changed from up to 90 per cent covered by the people's pension in 1980, to 50 per cent covered by this pension scheme in 2004 (Börsch-Supan, Reference Börsch-Supan2004). Thus, labour-market participation rose in importance for building up adequate pensions, and currently about 50 per cent of the means to old-age protection is circulating in pension funds and individual schemes.
In Germany, the contribution-based public pensions were explicitly partly outsourced, i.e. reduced and replaced by ‘voluntary private investments’, the so-called Riester pension in particular (Altersvermögensgesetz, 2001), while the wage indexation of public pensions was further eased by the ‘pension insurance sustainability act’ (RV-Nachhaltigkeitsgesetz, 2004). These outsourced pension investments are pension policies contracted by the individual employees, provided by insurances and banks and co-financed by the public purse by means of special subsidies (basic, spouseFootnote 2 and child allowances) and tax incentives (they are tax deductible). Qualifying for the full subsidies presupposes an ‘optimal’ investment of now 4 per cent of gross earnings. At the beginning of this decennium, about 82 per cent of the employed were insured within the public scheme that covered about 78 per cent of old-age income (DRV45, 2003), and the proportions between the various schemes were as follows: public – 85 per cent, additional occupational – 5 per cent and private – 10 per cent. The pension calculation norm for adequate pensions referred to the public scheme only. The Riester reform, phased in from 2002, was meant to change this ratio to 60–70 per cent of public entitlements and up to 40 per cent of additional occupational and market-based entitlements (Rürup, Reference Rürup2003), while the outsourcing of parts of the public pension to market-based schemes should possibly even overcompensate for the losses in public pensions (ASIB, 2008). Currently, the pension norm of an ‘adequate pension’ is 67 per cent of the average lifetime wages.Footnote 3
The market-based schemes, thereby, have their own effects on marginalisation and redistribution. First, some groups of citizens are differently marginalised in the two market-based sub-systems. In the Netherlands, additional occupational pension provisions or state-subsidised ‘third-pillar investments’ are disadvantageous for low-income earners and the part-time employed. In particular, this disadvantage is due to the threshold for additional occupational pensions, called the Franchise, which is still applied by many pension funds. This threshold marks the minimum income a person must earn to participate. Only income above this threshold, which might differ much from people's factual pension levels (Frericks et al., Reference Frericks, Maier and De Graaf2006), can build up additional pension entitlements and also benefit from the employer's contributions. In Germany too, lower-paid employees are marginalised, as are those who are simply less well-informed. This is the case due to, secondly, different redistributive effects of the market-based sub-systems. In the Netherlands, minimum income is guaranteed via the redistributive ‘people's pension’ scheme. For adequate pensions however, redistribution measures hardly exist and pensions are assumed to be increasingly individualised (e.g., partner allowances were eliminated). Redistribution to the advantage of the better-off, who profit particularly from tax incentives (Cox, Reference Cox2000), has been reduced since 2006. The German market-based sub-systems are first and foremost redistributive for those disposed to invest (let's say, the well-paid and well-informed), as well as for non-working spouses, families and, in principle, the lowest paid; they, however, are quite unlikely to profit since the minimal amount of own investment is rather high, as a percentage of low wages, and therefore quite unrealistic. Indeed, only a small portion of low-income earners participate in Riester schemes (Börsch-Supan et al., Reference Börsch-Supan, Coppola and Reil-Held2012; Corneo et al., Reference Corneo, Keese and Schröder2009).
In Germany and the Netherlands, the strengthening of market principles in pension systems has reinforced the dependency on non-interrupted income and work biographies which generally reduce redistribution. This reduced redistribution is reinforced for those with low income in both countries by means of the Franchise in the Netherlands, and the non-additive pension construction in Germany. Both pension systems build on individual investment, which in general is less favourable to low-income earners and the part-time employed. However, the German system introduced additional redistributive mechanisms for families and low-income earners. Obviously, the effect of partly marketising old-age protection in both countries has different effects on those that participate and those that do not, as well as on those who are additionally supported by public means and those who are not.
Characteristics of current hybrid systems: risk-sharing
In both countries, resources were shifted from pay-as-you-go to funded sub-systems. The characteristics of the hybrid systems which frame the risk-sharing vary considerably between the countries. In this context, it is important to notice that both public and market-based social protection have changed in considerable ways. Public schemes are increasingly adopting criteria of efficiency and economic sustainability traditionally attributed to the market, whilst the new market in social-protection products is changing greatly in characteristics through increasing state subsidies and new regulations (see above). As a consequence, criteria for risk-sharing and principles of resource flows are changing, possibly in contradictory ways even within one system.
In the Dutch calculation norm of adequate pensions, the link between contributions and pension entitlements has become stricter as the level of people's pensions decreased and market-based pensions gained in importance. Occupational pensions, which are generally financed by employer and employee contributions (comparable to the German public scheme), have also changed, so that pension entitlements are calculated on a DC (defined contribution) basis (and lifetime average income) rather than a DB (defined benefit) basis (and related to final wages), as was the case before.Footnote 4 The current Dutch DC schemes are, however, a somewhat hybrid form of linking benefits and contributions (Barr and Diamond, Reference Barr and Diamond2009), since they still incorporate collective elements in benefit calculation and are highly regulated in various ways. Since these funds are arranged and partly managed by the employing company or sector, membership in these occupational schemes is, in a way, a collective insurance or investment. The investment strategy of Dutch pension funds is oriented towards the global financial markets, a development that gained impetus during the 1990s. After periods of financial market difficulties, mainly in 2001, they are highly regulated, so that – despite the current crisis in the financial market and after some necessary adjustments to the pension schemes – they are still quite wealthy (Rijksoverheid, 2013). Regulations oblige them to submit yearly actuarial and liability reports to a state supervisory body. Pension funds may invest no more than 5 per cent of assets in the company should a pension fund be restricted to a single company, and must cover 105 per cent of all pension obligations to current and future pensioners; otherwise, a readjustment plan must be developed and accepted by the state supervisor. In addition, pension funds with many older participants (already retired or close to retirement) are obliged to avoid risky investments. In the current financial crisis, the percentage of coverage of many pension funds has been less than 105 per cent. The obligatory re-adjustment plans implemented in 2010 used different strategies to redress the coverage. First, pension levels were frozen so that they neither follow possible wage increases nor cover possible inflation. Secondly, a majority of the pension funds (for example, for public employees and all employees in the health sector) increased their levels of contributions. And, finally, some small pension funds reduced the level of actual pensions they paid to pensioners.
So-called private or third-pillar pensions are numerous in the Netherlands; however, they play a marginal role in the institutional calculation norm of adequate pensions (see below). Often offered by the employer or by the economic sector, or by pension funds in arrangement with particular insurance companies, this informal collectivisation of third-pillar individual pension schemes offers some guarantees against risky products. Though no-one is obliged to participate in these products, a high percentage of Dutch people (estimates vary at 60–90 per cent) use third-pillar products. The high number of third-pillar contracts is also related to substantial subsidies through tax relief, some of which were implemented in such a way that people could invest in savings schemes from income before taxes (Spaarloon), and these contributions were made tax-deductible if invested in pension plans, with the result being a more-or-less free additional pension investment (Cox, Reference Cox2000).
In Germany, the public scheme, which is contribution-based, i.e. occupational, is now after various changes more strictly linked to a continuous individual labour-market biography. This is mainly the case after the elimination or reduction of non-contributory entitlements to pension schemes. Although some pension credits (i.e. state-financed replacements for pension contributions lost during periods of employment interruption, such as for childcare and care of frail older relativesFootnote 5) were introduced and further developed, others were reduced or eliminated, such as pension entitlements for periods of education or long-term unemployment and widow(er)s’ pensions. Additional occupational pensions (Betriebsrenten), which play a far less important role in Germany than in the Netherlands, are not considered here, since they do not form part of the institutionalised pension calculation norm for adequate pensions as indicated above.
The Riester pension, however, does form part of this calculation norm and is, as other market-based pensions, subject to regulations prescribing allowed forms of their usage and takeup. These regulations also apply to investment in home ownership, which is now, after a short interruption, again recognised as a proper form of old-age investment and protected by law. Regulations on explicit pension savings schemes have established a variety of state-certified products, the criteria for which are listed in the ‘Old-age Investment-Contract Certification Act’ (AltZertG, 2001). Since 2006 (Alterseinkünftegesetz), these criteria have been reduced and simplified, providing for a guarantee of the nominal value of the contributions as well as the obligation to pay out benefits as life-long unisexFootnote 6 annuities. The various regulations result in a rather complex differentiation of the products, of which some are evaluated as being ‘supportable’, i.e. they may be tax-subsidised (also so-called Eichel and Rürup pensions) and, in addition, subject to allowances (‘Riester pensions’, all named after the respective ministers or expert in charge), and some are guaranteed to be ‘non-deductible’ investments, i.e. in such cases as long-term unemployment they do not have to be used up before the client is entitled to means-tested unemployment benefits (‘hartzsicher’). These classifications do not necessarily include the same products, i.e. ‘supportable’ schemes are not necessarily ‘non-deductible’.
The characteristics of the market-based sub-schemes of both countries mainly differ with regard to guarantees. Although there is no financial guarantee given by the Dutch state on the level of additional pensions in so-called third-pillar pensions (that in some cases might be used to build up adequate pensions, see below), there is highly regulated state supervision of pension funds, supplemented with penalties. In addition, customers are not as highly individualised as they are in Germany, since it is the employer or the sector that offers the pension fund, and in many cases the third-pillar schemes as well (collectivisation). In Germany, guarantees are given with respect to some life-course risks, i.e. some of the certified pensions chosen by individuals cannot be reduced if they receive means-tested welfare benefits, and they are also highly subsidised for families. Social risks to the employed are higher and risk-sharing is less in Germany since the additional contributions in necessary pension schemes are invested in various individually contracted insurance policies (fragmentation, see also Frericks, Reference Frericks2011). The market risks vary per policy and depend on the individual's choice between a more or less risky product. Thus, while in both pension systems the emphasis on ‘funded’ pensions acquired by labour-market income has increased, the characteristics of the hybrid systems and their risk-sharing capacity differ.
Nexus of current hybrid systems: coverage
It is the concrete nexus of the sub-systems which determines the role market principles play in hybrid systems. The share of people covered in the different schemes differs with this distinction.
In the Netherlands, participation in (supplementary) occupational pension schemes is compulsory. And apart from the so-called white spot of less than 10 per cent of all employees (those with disastrously insufficient provisions in labour contracts, or possibly some well-off self-employed), all employees build up entitlements in pension funds. People naturally have better or worse pension-fund investments, larger or smaller respectively, depending on the employment sector, quality of contracts and employer contributions. Most part-time employed face huge difficulties in building up occupational pensions due to the institutional linkage of the ‘people's pension’ scheme and the occupational pension schemes (the Franchise). The voluntary component of the Dutch pension combination can be found in so-called third-pillar pensions, which depend on the willingness to invest, the individual's literacy in taxation and financial products, the personal decision in favour of a more-or-less risky product and, of course, on the resources disposable to be invested. Special tax incentives to invest in these products are provided for those who do not reach the pension norm of adequate pensions of 70 per cent (SZ, 2012). In principle, this measure could have been advantageous for people with shorter work biographies, i.e. fewer than the obligatory forty years of contributions into the pension funds (e.g., due to family phases) or migrants with less than the fifty years of residency required to build up a full ‘people's pension’. But in fact the incentives mainly advantaged high-income earners rather than those with limited resources and were therefore changed and reduced in 2006.
In Germany, the part of the public scheme outsourced to the Riester pension, though forming part of the pension calculation norm of adequate pensions, is voluntary. In the first years of the arrangement, very few such individual pension contracts were signed; however, in recent years the percentage has increased for several reasons. First, there was a massive promotion undertaken by the Ministry of Labour and Social Affairs (BMAS), different banks and insurance companies. Secondly, after the changes of 2006 (Alterseinkünftegesetz), insurance agents were more motivated to promote the contracts since they could spread their profits over five instead of ten years, and, finally, the simplification of certification criteria for customers and providers increased the motivation to sell and buy Riester contracts (Börsch-Supan et al., Reference Börsch-Supan, Coppola and Reil-Held2012). However, there are still obstacles to investing in this necessary sub-scheme of the pension system, and 90 per cent of all Germans above eighteen years of age said they had not entered into a Riester contract by the end of 2008 (Corneo et al., Reference Corneo, Keese and Schröder2009). This seems to be the case because regulations, far from being transparent, are highly complex, such that customers do not really understand which pension product is regulated in what way. The state certificates do not say much about the pros and cons of the different policies that profit-oriented providers offer, and there are no free public advisors (as exist in Italy or Sweden) to help citizens who are forced to then individually find the most suitable product in order to invest in old-age provision. While Riester pensions were explicitly created as an outsourced part of the public pension insurance, other additional and subsidised pensions, such as the Rürup pension and the Eichel pension,Footnote 7 were also introduced but intended rather for the self-employed or as additional occupational pensions. Not surprisingly, it is particularly the better-paid who can and do afford additional state-subsidised pensions. To make Riester attractive to lower-paid employees as well, specific subsidies and regulations were introduced, for instance the offer of incentives to employees holding a so-called Mini-job (at €400 per month). However, as Schmähl and colleagues (Reference Schmähl, Himmelreicher and Viebrok2003: 7) calculate, the minimum additional contribution of low-income earners working full-time amounts to €30 per month for thirty years continuously, just in order to balance the losses resulting from the 2001 reform. Low-income earners, in addition, have to fear their benefits being set off against means-tested pensions. The institutional setup, therefore, leads to a so-called savings trap for people not investing at all and consequently ending up below adequate pension levels (Börsch-Supan et al., Reference Börsch-Supan, Coppola and Reil-Held2012).
Thus, coverage in the sub-systems that together frame the building up of adequate pensions differs in both countries. The number of Dutch covered in both the necessary schemes is very high, while in Germany the coverage rate is much lower in both schemes, which might be unproblematic for civil servants and those in liberal professions since they build up pensions in separate schemes. Germans invest much less in the necessary voluntary pension schemes than the Dutch do in their additional voluntary pension schemes, although the tax incentives for these schemes, relative to benchmark savings, is more than double in Germany than it is in the Netherlands (OECD, 2009: 110). This difference in old-age investments is important since the Dutch, on average, are already well covered by the people's pension and the obligatory additional occupational pensions, while the Germans, whose adequate pensions depend on additional investments, are not.
To sum up, one could say that the Dutch hybrid pension design guarantees a high level of coverage (benefits might differ tremendously for individuals), and participation in pension funds is almost universal for employees, while the German hybrid pension design, by contrast, creates a low level of coverage in the necessary parts of the system.
Conclusions
The pension institutions analysed above in terms of ingredients, characteristics and nexus of the hybrid systems show that market principles have been implemented in very different ways. To understand the effects of strengthening market principles in pension systems, two aspects are of importance: the role of market-based sub-systems within the hybrid system and the implementation of market principles in the overall system. This is the case because pension schemes do not follow pure principles: market principles in market-based schemes have been eased by regulations and collective structures while public schemes apply market principles. The concrete institutionalisation of hybrid pension systems determines social-risk spreading, because the institutional framing of adequate pension income depends on three effects of the institutional constellation: the marginalisation and the financial support of some groups, the risk-sharing with regard to guarantees and the coverage of citizens within the necessary sub-systems. I showed that, first, the marginalisation of some groups and the redistributive effects of the systems can be identified by an analysis of the ingredients of current hybrid pension systems. Secondly, the institutionalised forms of risk-sharing with regard to guarantees of pension contributions are determined by the characteristics of the institutional constellation and its sub-systems. And, finally, the coverage of citizens in the different sub-systems depends highly on the nexus of the hybrid system: insufficient coverage within one of the necessary schemes might be compensated for by an institutionalised possibility of more intense coverage within another scheme. This possibility is institutionalised in the Dutch voluntary sub-scheme, although not consistently. So in order to grasp the impact on social risks understood as underinsurance, one needs to analyse both the mix of the schemes and the principles applied in them by detecting the details of the ingredients, characteristics and nexus of the pension systems. In other words, it is the degree and the kind of market principles that are applied in the pension systems which determine whether marketisation and social risk-sharing are incommensurate. This conclusion is derived from three empirical results of this study.
First, partly marketised pension systems do not necessarily have to lead to increased social risks, as both cases show. Germany's Riester pensions, though ‘marketised’, involve strong redistributive mechanisms. Market principles in this institution were modified and therefore it fits the definition of Marshall's (Reference Marshall1964) welfare institutions. In the Netherlands, obligatory occupational market-based schemes are highly regulated, while the voluntary market-based pensions, known as the ‘cocoa on the Dutch pension-cappuccino’, function per definitionem as a supplement. Therefore, the Dutch, in terms of their pension norm and except for some of the self-employed, do not depend on this scheme, but can consider it a surplus.
Secondly and consequently, the overall composition of the institutional constellation of adequate old-age pensions shows that offering some basic pensions, which are not subject to means-tests, does more than simply prevent poverty for those who are being economically marginalised. In the Netherlands, old-age poverty rates are low, and since occupational pension schemes are mandatory, the percentage of people relying solely on basic pensions is, and shall be particularly in the future, very low. Since these are necessary to build up adequate pension entitlements, their being compulsory within the institutionalised pension-calculation norm of adequate pensions is consistent. The level of coverage of people building up entitlements to adequate pension provision in a combination of basic, occupational and/or voluntary pensions is very high in comparison to other countries. In Germany, however, guaranteed pensions are means-tested, which is meant to prevent poverty for those who have failed to build up sufficient pension claims in different schemes. People however hardly invest in the necessary sub-systems. The manner in which market principles were strengthened in pensions, therefore, leads to savings traps, which are currently being addressed by policy-makers (Zuschussrente).
Thirdly, the concrete institutionalisation of hybrid systems seems to have an impact on the willingness to invest in voluntary schemes. Organising voluntary pension supplements as (informally) collective, as the Dutch do, seems to increase the willingness to invest. In particular, the institutionalisation of additional market-based pensions on a voluntary basis might profit from a design of surplus rather than one with countless ifs and buts, which is particularly troublesome when these supplements are necessary in the sense of the above definition of adequate pensions.
Recent reports and studies (e.g., Barr and Diamond, Reference Barr and Diamond2009; Hill, Reference Hill2007; Pension Commission, 2005) make suggestions in accordance with the results of this analysis. In a nutshell, they call for a kind of pay-as-you-go-financed basic pension to prevent poverty and to secure rights to resources over time, supplemented by schemes designed to guarantee adequate pensions. Most importantly, for the components of the ‘necessary’ pension provision (following the above definition), they call for less choice of pension providers and more centralised, less individual pension plans to reduce administration costs and the ‘problems of imperfect consumer information’ (Barr, Reference Barr2002: 4). There is, as pension experts agree, still a lot to be elaborated on hybrid, pension systems and the implementation of market principles in pension setups, in order to overcome both the fault-lines of past systems and those in the currently dominant discourse.
