Squeezing social partners to embrace reforms: The case of private earnings-related pension schemes

Abstract

Spanish

French

How do changes to public schemes and ongoing economic difficulties impact private earnings-related pension schemes (PERPS) governed by social partners? The decreasing generosity of public schemes does put strong pressure onto social partners to improve their PERPS; however, PERPS face challenges of their own related to their integration within the pension system and their financing mechanisms. Based on a comparative analysis of Finland, France, the Netherlands and Sweden, this contribution demonstrates that PERPS have all enacted measures to reduce the generosity of their scheme. Yet, the emerging policy measures responding to these challenges are quite different and depend on the previous public/private mix and the financial structure behind their PERPS.

Keywords

Introduction

Social partners have always featured predominantly in welfare state debates. Unions and employers cooperated to entrust social insurance to the state or keep the state away from pooling risks (Baldwin, 1990; Mares, 2003). Following in this tradition, this study analyses recent changes to private earnings-related pension schemes (PERPS)¹ managed by social partners in four European countries (France, Finland, the Netherlands and Sweden). Most of the scholarship on social partners and the politics of retrenchment focuses on public pension schemes. This neglects social partners’ PERPS, which play an important role in many industrialized countries. PERPS secure a higher replacement rate for current and future retirees, often compensating for the weak generosity of public schemes (Rein and Wadensjö, 1997). The ongoing reforms of public pension schemes have put tremendous pressures on social partners to consider changes to their PERPS. Social partners can ignore adjustments made to public schemes, adapt their schemes independently, or pursue different alternatives, such as adding new collective schemes. The state can also accentuate or alleviate this pressure depending on whether or not it seeks to offer alternatives to maintain high replacement rates.

Inspired by the literature on path dependency and policy feedbacks (Pierson, 1993, 2000; Weaver, 2010), this contribution seeks to analyse recent policy developments and explain the presence (or absence) of drift, layering (Hacker, 2004) and extension. I argue that the interplay between prior institutional arrangements and the financing mechanism of PERPS explains whether the state and/or social partners will engage in drift, layering, or extension. First, in pre-existing pension arrangements where social partners have operated PERPS independently from the state, social partners are unlikely to coordinate their efforts with state officials to provide compensation for less generous public pension schemes. However, the likely response of social partners is highly dependent on their funding mechanisms. On the one hand, if social partners have operated pay-as-you-go (PAYG) schemes, they are unlikely to improve the generosity of their pension because they suffer from the same conditions that prompted state authorities to decrease the generosity of public schemes. In this case, the state is likely to develop an alternative pension mechanism – a state layering response. On the other hand, if social partners manage funded defined contribution (DC) schemes, it is relatively easy to offer the possibility – to both employers and employees – to increase contributions, since there are no explicit pension outcomes guaranteed by these schemes. This is akin to an extension response. Contrary to a formal revision, layering, or conversion, the additional provision does not jeopardize or replace prior institutional structures.

Second, policy responses will be very different where social partners and the state have been actively cooperating to develop PERPS to produce a single earnings-related pension scheme managed and operated by social partners at arm’s length from governmental authorities, with the latter providing a generous basic pension. In these cases, pension reforms involve the state, but the role of social partners extends far beyond that of a defensive veto player. This institutional arrangement is highly conducive to drift, since it features low levels of policy discretion and multiple veto players (Hacker, 2004). Nonetheless, social partners are far more likely to pursue a layering strategy by seeking new funded solutions when PERPS operates on a PAYG system; the creation of a funded top-up is unlikely to be appealing where the core earnings-related pension scheme is already fully funded.

These theoretical expectations are empirically assessed in a four-country comparative case study of France, Finland, the Netherlands and Sweden.² These countries were selected for a variety of reasons. The PERPS offered in these countries have reached maturity and offer near-universal coverage. As such, they can present some insights into the future politics of pensions in countries, such as Belgium and Norway, which have recently implemented PERPS to increase replacement rates for future retirees. Very few other cases can present similar characteristics, because countries in the liberal regime only offer voluntary schemes with limited coverage in the private sector and many Bismarckian states have not (or have only done so very recently) developed PERPS atop their public occupational pension schemes (Esping-Andersen, 1990).

This article is divided into two broad sections. The first discusses the importance of social partners’ PERPS and their current challenges. This is followed by a theoretical discussion on expected changes in policy modes considering both the state and social partners. The second section demonstrates the validity of these claims via a comparative analysis of France, Finland, the Netherlands and Sweden.

Social partners and PERPS

The state has multiple policy instruments to enhance social provisions. As succinctly described by Rein and Rainwater (1982: 41–44), the state can: enact legislation mandating a third party to provide social benefits under conditions specified by the state; induce individuals, groups, or firms to assume social responsibilities by offering a stimulus, such as a tax credit; subsidize private sector activities that have broader social objectives deemed desirable by the state; and pay and provide for social provisions itself.

Most of the comparative welfare state literature centres on ‘pay and provide’ provisions, which often downplays the importance of other types of state action, such as subsidizing, stimulating and mandating. The use of these policy instruments has given rise to various public/private mixes where both public and private provisions are encouraged to achieve social goals (Béland and Gran, 2008; Rein and Rainwater, 1982; Rein and Wadensjö, 1997).

Although established within specific national settings, the state has encouraged the development of PERPS in two different ways. First, some countries, often referred to as latecomers, expanded the coverage and generosity of their pension systems in the 1960s and 1970s. Contrary to the various PAYG schemes established in most countries soon after the Second World War, funded solutions became more acceptable and were deemed more suitable pension instruments. This prompted countries such as the Netherlands, Denmark and Switzerland to facilitate the expansion of large pension funds instead of creating mandatory public earnings-related pension schemes (Myles and Pierson, 2001).

Second, PERPS grew because the public earnings-related pension schemes have a low-income ceiling. As a result, employees with income stretching above the ceiling sought an additional mechanism to obtain higher replacement rates. They took the form of PERPS that were made compulsory or semi-compulsory in countries such as France, Sweden and most recently Belgium; they became voluntary in liberal welfare states such as Canada and the United States. These PERPS represent an important component of collective labour agreements (see Rein and Wadensjö, 1997).

Whether or not PERPS act as the main earnings-related pension scheme or as a top-up, the state is relieved of the primary managerial and financial responsibilities associated with PERPS. It is nonetheless highly involved in the regulation of these schemes by providing financial – often in the form of tax credits – and administrative assistance. Also, states can, in a neo-corporatist fashion, facilitate the resolution of conflicts between social partners and achieve broader socioeconomic goals by cooperating actively with social partners. Lastly, citizens still expect the state to act when PERPS run into difficulties, even if they have delegated pension policies to a third party. For all these reasons, it remains in the interest of the state that PERPS function well and alleviate the need to increase state intervention to secure a generous pension for its citizens.

Reforming PERPS and pressures for alternatives

In the past decades, cost containment has become a key policy objective for public pension systems (Haverland, 2001). Within this new economic environment, the importance of social partners, and most particularly unions, has been a hotly debated issue in the comparative welfare state literature. On the one hand, unions are no longer considered an important player, since they have been replaced by beneficiary groups, such as pensioners’ associations, which are highly organized and successful in protecting current benefit provisions (Pierson, 1994). On the other hand, unions fulfil the function of beneficiary groups in many European countries and can even act as a veto point (Green-Pedersen and Haverland, 2002). These contributions, however, are primarily focused on public pension schemes and rarely discuss the complex public/private mix that is present in many jurisdictions (for exceptions see Béland and Gran, 2008; Ebbinghaus, 2011; Rein and Wadensjö, 1997).

With the generosity of most public pension plans in decline (OECD, 2010), there is increasing pressure on PERPS to fill the gap to preserve the replacement rate of pension systems. Based on recent contributions (Clark and Whiteside, 2005; Ebbinghaus, 2011), we expect the state to embrace an extended regulatory role and promote solutions that do not increase its liabilities. As such, this implies a move away from ‘pay and provide’ solutions in favour of policy actions through which the state stimulates, mandates and/or subsidizes pension provisions.

The expansion of private alternatives represents an opportunity for both unions and employers to accentuate their influence on pension policies. Rein and Wadensjö (1997) emphasize that PERPS can actually offset changes to public schemes, but they also stress that public schemes and PERPS can follow a parallel development and trigger a ‘double decline’ in pension provisions. Social partners are not necessarily interested in expanding their role since many PERPS are also facing difficulties of their own.

With demographic, economic, financial and political pressures on PERPS, what kind of responses do we expect from social partners? Pension policies are highly path-dependent, with previous choices resulting in ‘trajectories that are inherently difficult to reverse’ (Hacker, 2002: 54; see also Pierson, 2000). Across public policies, pensions feature many characteristics associated with path-dependent outcomes. Prior contributions create a strong sense of entitlement that help foster powerful interest groups (Pierson, 1994). In addition, pension systems relying mainly on PAYG financing mechanisms face the ‘double payment’ problem with any significant shift towards funding instruments requiring contributions to both PAYG and funded schemes by current workers (Myles and Pierson, 2001). Funded pension schemes have been struggling to tackle the challenges arising from the recent financial crises.

Recently, more attention has been granted to negative policy feedback, which focuses on ‘the consequences of policy that tend to undermine rather than reinforce the political, fiscal or social sustainability of a particular set of policies’ (Weaver, 2010: 137). The major Swedish pension reform undertaken in the mid-1990s is a prominent example of negative policy feedback. It was triggered by an actuarial report stressing that the underlying mechanism of its public scheme made reform an inevitable outcome. On the one hand, the scheme would have required substantial increases in contribution rates, if the economy continued to generate limited (or negative) economic growth. On the other hand, strong economic growth would have resulted in most Swedes reaching the contribution ceiling, transforming the earnings-related scheme into a flat contributory pension (Marier, 2008). Interestingly, however, Weaver’s (2010) findings exhibit strong stability for countries that have developed a public/private partnership. He claims that they have ‘limited opportunities for exit to another type of pension regime’ (2010: 149).

Opposed to the claims that pension regimes are locked in paths and limited in their capacity to change, recent contributions have stressed the importance of incremental changes on (future) policy outcomes (Ebbinghaus and Gronwald, 2011; Hacker, 2004; Hinrichs and Kangas, 2003). For example, Ebbinghaus and Gronwald (2011) argue that the recent wave of pension reforms might represent a path departure outcome.

Theoretical expectations

Two key parameters explain the kind of policy adjustments that will be favoured in pension systems where PERPS play an important role (see Table 1). First, do PERPS act as a top-up or as the core earnings-related pension scheme? A top-up scheme implies that the state operates a public earnings-related pension scheme with features such as a contribution ceiling that encourage the growth and expansion of PERPS. Both schemes follow different reform dynamics. The public scheme faces the issues associated with the politics of retrenchment described above, while modifications to PERPS are embedded in the cyclical negotiations of broad collective agreements. As a result, policy responses are likely to be uncoordinated between the state and social partners. Social partners jealously guard their autonomy whenever the state tries to intervene and negotiations surrounding these PERPS are removed from public pension discussions (Marier, 2008; Reynaud, 1997; Trampusch, 2010). Social partners can thus offset or generate a double decline in response to retrenchment measures introduced by the state in its earnings-related pension scheme. Although social partners may not wish to harmonize their PERPS with changes to state benefits, these may ultimately have consequences on PERPS that may require revisions by social partners. France and Sweden exemplify this pension arrangement.

Table 1.

Social partners’ PERPS and expected mode of policy change.

Financing/Type of PERPS	PAYG	Funded
Top-up	(1) State layering	(2) Social partner extension
	France (AGIRC & ARRCO)	Sweden (ITP-1, SAF-LO, PA03, KAP-KL)
Core	(3) Social partner layering	(4) Status quo
	Finland (multiple schemes)	The Netherlands (sectoral and company pension funds)

PERPS can also represent the core earnings-related scheme within a pension system. This pension arrangement implies that the state provides a basic (universal or means-tested) pension leaving social partners with a broad mandate to maintain and develop substantial PERPS. Contrary to the previous case, social partners are involved in all discussions to alter the pension system, including public programmes, since they have a direct impact on the functioning of their PERPS. Negotiations to resolve pension issues are likely to be more complex, since they involve a higher number of veto players. Finland and the Netherlands fall into this category.

The second parameter involves the financing mechanism of PERPS. The choice of pension instrument generates policy feedbacks that can either entrench policy solutions (Pierson, 1993), or facilitate their demise in the case of negative policy feedbacks (Weaver, 2010). The key element is whether or not PERPS were originally built on a PAYG basis or fully funded. Both financing mechanisms generate different risks and encourage their own set of solutions in times of duress.

In the case of PAYG financing, the income of pensioners is tied into the contributions generated by current workers. These schemes require a ‘reasonable balance between earned income and entitlements’ (Clark, 2003: 39). With the ratio of workers to retirees decreasing, an economic crisis accentuates the consequences of population ageing by reducing further contribution revenues as a result of a rise in unemployment and the premature retirement of older workers. There are diverse solutions to tackle financial difficulties in public and private PAYG schemes. These include raising contribution rates, reducing the generosity of benefits, altering indexation mechanisms, raising the retirement age and strengthening the relationship between contributions and benefits. Some solutions are clearly more viable than others: for example, raising contribution rates can have negative spill-over consequences on the costs of labour, which are already high in Continental Europe (Scharpf, 2000). PERPS in both France and Finland operate primarily on a PAYG basis.³

In the case of fully funded schemes, recent financial crises have generated highly volatile returns, which raises concerns about their ability to deliver expected pension benefits. In DC schemes, the risks are assumed primarily by individual contributors. Thus, increases in contributions or a sharp reversal of fortune in the performance of the portfolio are required to make up for losses, in order to sustain a specific replacement rate. The impact of financial crises on DB (defined benefit) schemes is more complex, because it depends on the nature of the formula utilized to calculate the value of the pension benefit and the regulation surrounding these schemes. A key policy issue is maintaining sufficient funding to cover (current and future) liabilities. Multiple schemes, including the major ones in the Netherlands, have been dealing with funding ratios below 100% in the wake of recent financial crises. Potential solutions include a raise in contribution rates, decreased benefits, changing the underlying formula for benefits, altering the indexation mechanisms and requiring additional funding via a one-time payment or an increase in the retirement age. In addition, regulatory measures likely to be adopted include increasing the period to restore the funding ratio to 100% in the event of a financial crisis and introducing new investment rules.

Combining both PERPS type and financing mechanism generates four pension constellations resulting in different expectations with regard to modes of policy change. Before outlining theoretical expectations, it is first imperative to discuss the dependent variable: policy change. Hacker (2004) claims that there are four key modes of policy change based on whether there are high or low barriers to internal policy conversion and barriers to authoritative policy change. In the case of pensions, there are high barriers to internal policy conversion. Therefore, only two potential outcomes are likely to occur: drift and layering. Drift is likely to occur when there are no formal changes in policies, but their operation or effect are transformed. Hacker (2004) cites the growing importance of 401(k) in the USA as a good example of drift. Its importance increased as a result of a declining coverage in DB pension schemes. No actions were taken by governmental authorities to steer this change, which occurred as a result of benign neglect. Layering occurs when governmental authorities opt to introduce a new policy or programme alongside pre-existing ones.

It is important to stress that all PERPS have undergone noticeable reforms, which have had the result of diminishing their generosity. Still, there is limited support to suggest that these changes amount to a policy revision of current PERPS. As defined by Hacker (2004), a revision implies a formal replacement or elimination of a policy and it is ‘quite obviously not the normal state of affairs in welfare state politics’ (2004: 247). The PERPS under study have altered some policy objectives, such as targeting a specific replacement rate. However, the broad policy objective to contribute towards retirement income remains strongly embedded in each of the PERPS analysed in this study.

With a mandate to provide PERPS, social partners also have the ability to influence policy directions. Thus, they can also opt to engage in policy drift and layering. For example, social partners can foster layering by offering new PERPS atop existing ones. Social partners can also expand the range of current PERPS. This can take multiple forms, such as offering the possibilities – to both employees and employers – to improve the value of their pensions within an existing scheme. I label these changes extension. Contrary to layering, this policy change does not require the creation of a new public programme.

Following a decline in the generosity of public pension programmes, there are four potential outcomes related to PERPS based on the pension arrangement in place and the financing mechanism. First, a PAYG top-up PERPS is conducive to generating a layering policy response with the state adding stimulus or subsidies to create an expansion in fully funded pension schemes. This is the policy response embraced among multiple countries that have relied strongly on PAYG financing for their earnings-related schemes (Ebbinghaus, 2011). Social partners are unlikely to be active in generating an alternative. Improving a PAYG PERPS remains difficult with slow economic growth and population ageing. Employers are unlikely to support new mandatory social contributions and, as in public PAYG schemes, difficult decisions are expected with regard to preserving current PERPS.

Second, the presence of funded top-up PERPS offers the best possibility of offsetting the potential loss of retirement income that results from the retrenchment of a public scheme. In a DC scheme, it is relatively easy for both employer and employee to offer something beyond what is stated in a collective agreement, since this simply requires a higher contribution rate. Employers and employees still benefit from lower administrative fees as a result of their attachment to a large scheme, hence the potential for an extension of PERPS making it less appealing for the state to pursue a layering strategy. Although a similar outcome can arise for a DB scheme, additional contributions require additional guarantees, which impact the entire DB scheme. Sweden illustrates this dynamic.

Third, the presence of PAYG financing with a core PERPS is likely to trigger multiple discussions between social partners and the state in a search for new alternatives to increase potential replacement rates. With social partners clearly representing veto players in this case, policy change is expected to be very incremental, with policy drift being a likely outcome for the state. With social partners assuming most pension responsibilities, they are the ones likely to initiate a layering response for current losses in replacement rates. This type is represented by the Finnish case and its highly centralized PERPS.

Fourth, a core-funded PERPS should result in the state and social partners working together to produce a comprehensive response to the ongoing difficulties surrounding public and private schemes. With a very high number of veto players and a strong commitment to funded schemes, any policy change towards a new institutional arrangement is least likely to occur. The status quo is far more likely than the alternatives, including layering, since employers and employees are already strongly invested in a funded solution. This is the case of the Netherlands and its funded PERPS.

Comparative case study: France, Finland, the Netherlands and Sweden (see also Table 2)

France

The French pension system features mandatory top-up PERPS, with AGIRC and ARRCO the most noteworthy. With the cadres refusing to join the régime général without guarantees that their previous occupational schemes would be continued, the compromise solution consisted of establishing a less generous PERPS (limited at a 40% replacement rate for the average worker), while legislating supplementary pensions financed by contributions above the ceiling of the régime général. This led to the creation of AGIRC in 1947, as part of a national collective agreement, and ARRCO for non-cadres employees in 1961 (Baldwin, 1990; Reynaud, 1997). Participation was made compulsory and universal for workers covered by the régime général in 1972 (Reynaud, 1997: 71–72).

Table 2.

Public programmes, social partners’ PERPS and modes of policy change.

	Basic pension	Public earnings-related pension scheme	Social partners’ PERPS	Modes of policy change
France	Minimum provisions in régime général	Régime général, multiple régimes in public sector and other specialized professions	AGIRC, ARRCO	State layering: creation of voluntary individual and collective savings schemes with tax subsidies
Sweden	Guaranteed pension	NDC scheme (16% contributions) and funded private accounts (2.5% contributions)	Private white-collar (ITP-1); private blue-collar (SAF-LO); municipal and regional government (KAP-KL); civil service (PA03)	Marginal social partner expansion & state drift
Finland	Universal national pension		Mostly sectoral schemes	Social partner layering
The Nether-lands	Universal basic pension (AOW)		Company and sectoral funded schemes	Status quo

Core programmes only.⁴

These schemes are directed and managed exclusively by the social partners. They operate on a PAYG basis using a point system resembling what is now called a notional defined contribution (or ‘non-financial defined contribution’, or NDC). A decline in labour force participation has accentuated the upcoming financial challenges associated with the retirement of the baby boomer cohorts. In 2009 alone, revenues from contributions shrank by 1%, while expenditures rose by 4%. This prompted authorities to utilize funds from their reserve, which had not been expected before 2015–2016. Multiple measures were introduced to decrease the size of the actuarial deficits experienced by both regimes, the most noteworthy being a gradual rise in the retirement age from 65 to 67 (see Cahiers, 2011).

Social partners have demonstrated on multiple occasions that they are under no obligation to align their schemes (AGIRC and ARRCO) with the reforms being implemented by the régime général. This explains why governmental efforts to increase private pensions rarely refer to AGIRC and ARRCO, but rather focus almost exclusively on private savings vehicles. The two programmes introduced to increase private pension savings avoid any linkage to mandatory supplementary occupational pensions; the plan d’épargne retraite populaire involves banks and insurance companies and the plan d’épargne retraite collectif is offered primarily by individual firms (Bonoli and Palier, 2007: 565). Thus, in line with theoretical expectations, the French government is pursuing a layering policy response without the involvement of social partners and their PERPS.

Sweden

The Swedish case features top-up PERPS that are truly private without any interference from the state. Social partners can alter the design of their scheme as they wish (Overbye, 1998) and the role of the state has been limited to ensuring that PERPS comply with financial regulations. There are four main occupational schemes (covering blue-collar workers and white-collar workers in the private sector, municipal and regional employees and civil servants). The schemes are highly centralized and part of collective bargaining. In 2006, they covered 90% of the workforce and represented 27.7% of retirement income for men and 19% for women (Sjögren Lindquist and Wadensjö, 2011). They owe their existence to the presence of a ceiling in the previous public earnings-related pension scheme (ATP). The previous occupational pension scheme was altered for white-collar workers immediately after the adoption of the ATP in 1959; blue-collar workers obtained something similar in 1972, and civil servants soon followed suit (Overbye, 1998: 177; Palmer, 2002: 38–39). The indexation mechanism of the ATP gradually increased the importance of occupational schemes (Kangas et al., 2010: 275). They all underwent important transformation moving primarily from DB to DC (Palmer, 2002).

In spite of their large degree of independence, social partners have had to deal with the consequences of the overarching pension reform introduced in 1994–1998. First, some members of the pension working group advocated the inclusion of an occupational pension as part of a new means test (as in Finland). This sparked strong opposition from the social partners, who argued that their schemes are private pensions. This idea was abandoned and the means test applies exclusively to the NDC element of the new scheme (Marier, 2008: 133). Second, the introduction of the public scheme reform facilitated the transition to DC, since it would have been very expensive to maintain a DB scheme on top of the newly created NDC scheme (Palmer, 2002).

Contrary to theoretical expectations, there is no evidence to suggest that social partners have embraced an extension of existing schemes. The replacement rate of the combined earnings-related pension schemes (public + occupational) has traditionally been quite generous with voluntary private solutions often being employed by those with less attachment to the market (see Sjögren Lindquist and Wadensjö, 2011). With most reforms targeting younger workers with a long transition period, it will take time before one can observe a trend towards either increasing contributions within an occupational pension scheme or within an individual pension account. Currently, evidence suggests that drift rather than extension is a more likely outcome, since there are no indications that employers and/or employees have embraced the possibilities of an extension.

Finland

Although the Finns consider their occupational pension regime as semi-public and part of the first pillar, the European Union and many international observers consider it akin to a PERPS (see Kangas, 2007). Contrary to France, however, these large sectoral pension schemes represent the only mandatory earnings-related schemes; they are also decentralized and managed by private financial institutions (Overbye, 1998: 174). With the state providing only an income-tested pension following the 1996 reform, the role and importance of these schemes have resulted in genuine multi-level governance in which the social partners have cooperated closely with governmental authorities in pension policy. Of all the reforms introduced since the early 1990s, only one had no explicit reference to labour market partners (Kangas, 2007: 281). For the 2002 and 2004 reforms, the government even commissioned the labour market partners to present propositions, which were later adopted by the government (Kangas, 2007: 284–285).

Reforms have had a significant impact on future benefits. Among the most important measures, the latest reforms eliminated the 60% replacement rate target, introduced a life-expectancy based indexation mechanism, and altered the calculation of the pension formula to consider contributions over the entire working career. The cumulative impact of these reforms may result in replacement rates in the 35–50% range, which is far below expectations of 60–70% (Kangas and Luna, 2011).

These reforms have opened the door to new voluntary occupational schemes and individual pensions, which seek to replace this (future) retirement income. The employment coverage of supplementary occupational schemes has been growing at a rate of 20% per year since 2008 (Kangas and Luna, 2011: 228). As expected, new funded alternatives have been introduced in Finland. Contrary to France, however, the Finnish government has not been developing new voluntary savings schemes; rather, social partners have embraced a layering strategy by adapting previous regulations to generate new voluntary pension schemes.

The Netherlands

The Dutch pension funds are very different from Finland: they are pre-funded resulting in a different regulatory framework and are far more decentralized, with close to 500 sectoral and company funds. A long divisive political battle between Beveridge and Bismarck resulted in a somewhat unique pension structure: a public pension with a universal flat rate scheme (AOW) coexists with a quasi-mandatory private occupational pension scheme (Haverland, 2001: 314). The Dutch governance structure includes large pension funds that are not steered by private financial institutions or employers, but rather by social partners under the principle of ‘social solidarity’, which is enshrined in collective agreements (Clark and Bennett, 2001). The pension funds operate at arm’s length from both government and employers with a legal status giving fund managers a large degree of operational autonomy (Ponds and Riel, 2009: 93).

The governmental influence is quite pronounced because public authorities provide generous tax subsidies representing 2.1% of GDP (Anderson, 2007: 727). In the past, this influence compelled social partners to close early retirement options and adopt cost containment measures for their pension schemes (Clark and Bennett, 2001; Mabbett and Bolderson, 1998). This financial input has been employed in negotiations with social partners to achieve other pension objectives, such as ensuring the portability of pensions and expanding coverage. The state also has a high stake in ensuring the sustainability of occupational pensions via its role as employer, because rising pension premiums result in a loss of revenue via tax credits (Van Riel et al., 2005: 83).

The three financial crises combined with rising life expectancy, low interest rates and poor investment decisions have had a noticeable impact on the Dutch multi-level governance structure. The Pension Authority, which operates within the independent central bank, has emerged as a core player. Its role has increased substantially in the wake of poor stock market performance with funding ratios plummeting from a nominal average of 200% to 100% in 20 years (DNB, 2011: 96). The first financial crisis in the early 2000s resulted in the introduction of many reforms, such as the creation of a 105% nominal minimal threshold for pension funds and the transformation of multiple schemes into hybrid DB-DC plans with new pension formulas based on average wage, rather than final salary (Ponds and Riel, 2009: 103).

The subprime and euro crises have accentuated the financial difficulties of many schemes, which will likely result in further cuts in the pension benefits of both current and future retirees. The nominal funding ratio averaged 98% by the end of 2011, representing a real funding ratio of 80%. Despite measures such as increasing the time period to facilitate the recovery of funds (now ranging from five to 15 years), there are currently 103 (out of 454) pension funds that will face either a decrease in pension rights or cuts in benefits, if their financial situation fails to improve by the end of 2012 (DNB, 2012). None of the four largest schemes is currently above the minimum threshold of 105% (Cobley, 2012).

Negotiations are ongoing to reform multiple features of the Dutch pension system, which will likely feature a rise in retirement age, stabilization of contribution rates and a new assessment process for pension funds. The reform is expected towards the end of 2012. As expected, policy developments concerning the Dutch pension system are all geared towards preserving the status quo, rather than presenting or promoting new pension programmes.

Discussion and conclusion

With public schemes becoming less and less generous, there is increasing pressure on PERPS to improve, or at least maintain current benefits. However, as illustrated in this study, PERPS are also facing important challenges. As a result, all PERPS have made multiple adjustments to ensure that they will be sustainable in the long term. In the wake of recent changes to public pension programmes, PERPS have contributed to the generation of a ‘double decline’ (see Rein and Wadensjö, 1997) in pension generosity. This article stresses that, like the state, social partners have the capacity to influence modes of policy change within a pension system.

The institutional arrangement and the financing mechanisms of PERPS in the four countries discussed above shed light on ongoing policy developments. In both cases where PERPS were financed on a PAYG basis, policy layering is highly noticeable. With social partners already operating the core earnings-related pension scheme in Finland, they continued actively to develop new schemes (albeit on a voluntary basis) to compensate for lost income associated with the reforms to both basic pension and their PERPS. In France, where PERPS represents a top-up, it was the state that followed a layering strategy.

The cases of funded PERPS are different. The recent economic and financial crises have provoked multiple difficulties for Dutch pension funds. With such a strong investment in funded solutions, there is little desire to add new funded alternatives, as in the case of France and Finland. On the contrary, reform discussions are geared to secure the future of the funded schemes and there is increasing pressure on the state to improve the generosity of its basic pension scheme.

In the case of Sweden, top-up funded PERPS have been transformed into DC schemes (a DB portion remains for high earnings in the public sector). To compensate for the loss of potential income resulting from both public and private pension reforms, employers can supplement their mandatory contribution to a PERPS or individuals can engage in private pension savings. Both of these options are subject to a tax credit. Nonetheless, these options are currently fairly marginal. A key reason is that current replacement rates remain high, with many retirees having a large portion of their pension calculated under the previous public and private schemes. However, this could change rapidly, as an increasing number of individuals retire under the new rules. Weak returns on DC investment and lower public pension benefits, which could occur as a result of increases in life expectancy or slow economic growth, could prompt individuals to be more active in seeking additional sources of revenue for their retirement.

Beyond modes of policy change, this article suggests that social partners play a different political role depending on the importance of their PERPS. In addition, the governance of pension systems is increasingly becoming complex and multi-layered. In the case of both Finland and the Netherlands, the social partners play a leadership role in pension policies, which occurs as a result of their mandate to provide core earnings-related pension schemes. They also have a great deal of influence on pension issues beyond their own PERPS. While social partners in Finland have been quite active in expanding voluntary funded schemes, the Dutch are engaged in complex negotiations involving the state, social partners and various financial actors. The role of the state is multifaceted, since it acts in many capacities, such as regulator, employer, subsidizer and provider of the basic pension. Conversely, social partners in France and Sweden are highly independent when it comes to their own PERPS. However, these findings suggest that they do not have the same broad input on pension matters as their Dutch and Finnish counterparts. Further research on governance would be necessary to evaluate the degree to which the input of the partners varies across these cases.

In closing, this contribution focused on four cases in which social partners are actively involved in PERPS. Further studies would be needed to assess whether the arguments presented in this contribution apply to other countries. There are, unfortunately, few other cases where social partners are involved in these kinds of PERPS. However, with the creation (or expansion) of top-up funded PERPS in countries such as Belgium and Germany, future studies could compare their development (after maturation) with these findings.

Footnotes

Funding

This paper was partly funded by the EU-FP7 project GUSTO, based at the University of Warwick (PI Colin Crouch). Further funding was provided to the author by Ministère du développement économique, de l’innovation et de l’exportation (Québec), L’évolution des régimes sociaux et d’emploi au Québec et au Canada dans une perspective comparative and Social Sciences and Humanities Research Council of Canada, Grant number: 410-2008-1960.

Notes

Biographical note

Patrik Marier is Canada Research Chair in Comparative Public Policy at Concordia University, Montréal, Canada. His research interests focus on the policy challenges related to population ageing and the role of civil servants in reforming the welfare state. He has published in journals such as Governance, West European Politics and the American Journal of Political Science. He is the author of Pension Politics: Consensus and Social Conflict in Ageing Societies (Routledge).

References

Anderson

(2007) The Netherlands: Political competition in a proportional system. In: EM

Immergut

Anderson

Schulze

(eds) The Handbook of West European Pension Politics. Oxford: Oxford University Press, pp. 713–757.

Baldwin

(1990) The Politics of Social Solidarity: Class Bases of the European Welfare State 1875–1975. Cambridge: Cambridge University Press.

Béland

Gran

(2008) Public and Private Social Policy: Health and Pension Policy in a New Era. Basingstoke: Palgrave Macmillan.

Bonoli

Palier

(2007) When past reforms open new opportunities: Comparing old age insurance reforms in Bismarckian welfare systems. Social Policy and Administration 41(6): 555–573.

Cahiers (2011) Un nouvel équilibre pour les régimes: L’accord du 18 mars 2011 vise à pérenniser les régimes Agirc et Arrco. Cahiers de la retraite complémentaire 4.

Clark

(2003) European Pension and Global Finance. Oxford: Oxford University Press.

Clark

Bennett

(2001) Dutch sector-wide supplementary pensions: Fund governance, European competition policy, and the geography of finance. Environment and Planning A 33(1): 27–48.

Clark

Whiteside

(2005) Pension Security in the 21st Century. Oxford: Oxford University Press.

Cobley

(2012) Dutch facing more pensions cuts despite regulator’s leniency. Financial News, 19 January.

10.

DNB (2011) Annual Report 2010. Amsterdam: De Nederlandsche Bank.

11.

DNB (2012) 103 pension funds may curtail pension benefits. Press release, 20 February.

12.

Ebbinghaus

(ed.) (2011) The Varieties of Pension Governance: Pension Privatization in Europe. Oxford: Oxford University Press.

13.

Ebbinghaus

Gronwald

(2011) The changing public-private mix in Europe: From path dependence to path departure. In: Ebbinghaus

(ed.) The Varieties of Pension Governance. Oxford: Oxford University Press, pp. 23–51.

14.

Esping-Andersen

(1990) The Three Worlds of Welfare Capitalism. Princeton, NJ: Princeton University Press.

15.

Green-Pedersen

Haverland

(2002) The new politics and scholarship of the welfare state. Journal of European Social Policy 12(1): 43–51.

16.

Hacker

(2002) The Divided Welfare State. Cambridge: Cambridge University Press.

17.

Hacker

(2004) Privatizing risk without privatizing the welfare state: The hidden politics of social policy retrenchment in the United States. American Political Science Review 98(2): 243–260.

18.

Haverland

(2001) Another Dutch miracle? Explaining Dutch and German pension trajectories. Journal of European Social Policy 11(4): 308–323.

19.

Hinrichs

Kangas

(2003) When is a change big enough to be a system shift? Small system-shifting changes in German and Finnish pension policies. Social Policy and Administration 37(6): 573–591.

20.

Kangas

(2007) Finland: Labor markets against politics. In: Immergut

Anderson

Schulze

(eds) The Handbook of West European Pension Politics. Oxford: Oxford University Press, pp. 248–296.

21.

Kangas

Luna

(2011) Finland: From statutory pension dominance towards voluntary private schemes. In: Ebbinghaus

(ed.) The Varieties of Pension Governance. Oxford: Oxford University Press, pp. 211–240.

22.

Kangas

Lundberg

Ploug

(2010) Three routes to pension reform: Politics and institutions in reforming pensions in Denmark, Finland and Sweden. Social Policy and Administration 44(3): 265–284.

23.

Mabbett

Bolderson

(1998) Devolved social security systems: Principal-agent versus multi-level governance. Journal of Public Policy 18(2): 177–200.

24.

Mares

(2003) The Politics of Social Risk: Business and Welfare State Development. New York: Cambridge University Press.

25.

Marier

(2008) Pension Politics: Consensus and Social Conflict in Ageing Societies. London: Routledge.

26.

Myles

Pierson

(2001) The Political Economy of Pension Reform. In: Pierson

(ed.) The New Politics of the Welfare State. Oxford: Oxford University Press, pp. 305–333.

27.

OECD (2010) Pensions at a Glance. Paris: OECD.

28.

Overbye

(1998) The politics of voluntary and mandatory pensions in the Nordic countries. Labour 12(1): 169–189.

29.

Palmer

(2002) The evolution of public and private pension insurance in Sweden in the 1990s. In: OECD (ed.) Regulating Private Pension Schemes. Paris: OECD, pp. 37–50.

30.

Pierson

(1993) When effect becomes cause: Policy feedback and political change. World Politics 45(3): 595–628.

31.

Pierson

(1994) Dismantling the Welfare State? Reagan, Thatcher, and the Politics of Retrenchment. Cambridge: Cambridge University Press.

32.

Pierson

(2000) Increasing returns, path dependence, and the study of politics. The American Political Science Review 94(2): 251–267.

33.

Ponds

EHM

Riel

(2009) Sharing risk: The Netherlands’ new approach to pensions. Journal of Pension Economics and Finance 8(1): 91–105.

34.

Rein

Rainwater

(1982) The institutions of social protection. In: Rein

Rainwater

(eds) Public/Private Interplay in Social Protection. London: ME Sharpe, pp. 25–56.

35.

Rein

Wadensjö

(1997) Enterprise and the Welfare State. Lyme: Edward Elgar.

36.

Reynaud

(1997) France: A national and contractual second tier. In: Rein

Wadensjö

(eds) Enterprise and the Welfare State. London: Edward Elgar, pp. 65–98.

37.

Scharpf

(2000) The viability of advanced welfare states in the international economy: Vulnerabilities and options. Journal of European Public Policy 7(2): 190–228.

38.

Sjögren Lindquist

Wadensjö

(2011) Sweden: A viable public–private pension system. In: Ebbinghaus

(ed.) The Varieties of Pension Governance. Oxford: Oxford University Press, pp. 241–263.

39.

Trampusch

(2010) How and why industrial relations influence party effect upon welfare state retrenchment: A comparison of the Netherlands, Denmark, Germany and France. Zeitschrift für Vergleichende Politikwissenschaft 4(1 supplement): 103–128.

40.

Van Riel

Hemerijck

Visser

(2005) Is there a Dutch way to pension reform? In: Clark

Whiteside

(eds) Pension Security in the 21st Century. Oxford: Oxford University Press, pp. 64–92.

41.

Weaver

(2010) Paths and forks or chutes and ladders?: Negative feedbacks and policy regime change. Journal of Public Policy 30(2): 137–162.