Gender and power in occupational pensions: Trade union strategies and institutional constraints

Abstract

The shift from state to occupational pension plans has raised concerns among scholars because occupational plans tend to lack the redistributive elements found in state plans. This disadvantages workers with interrupted careers and low pay, who are usually women. However, occupational pensions also enable trade unions, including unionised women, to exert influence. Using Norway as a case study, we investigate how male- and female-dominated unions shape pension policy. We argue that the structure of wage bargaining and the potential for coalitions between male-dominated unions and employers have sidelined female-dominated unions and privileged fully individualised occupational pension models.

Introduction

A gender pension gap of 26.1% in the European Union (EU) has recently been reported (European Commission Directorate-General for Employment, 2024). Women receive lower pensions than men in all EU countries, and this phenomenon has persisted over time, leading to women being significantly more vulnerable to poverty in old age than men (European Commission Directorate-General for Employment, 2024: 118). The reasons behind the pension gap are complex and include women’s lower labour market participation, lower wages in female-dominated jobs and the gender pay gap (Foster, 2010; Foster and Heneghan, 2017). The slow pace at which this gap is narrowing has raised concerns about how ongoing pension reforms may affect women’s standard of living in old age in the future (Arcanjo, 2019; Frericks et al., 2009; Gardiner et al., 2016; Ginn, 2004; Ginn and Foster, 2023; Halvorsen and Pedersen, 2018; Meyer and Bridgen, 2008). In particular, scholars point to the de facto privatisation of pension systems, as occupational pension schemes assume a greater role while state pension schemes are scaled back (Natali and Pavolini, 2017). Feminist scholars argue that occupational pension schemes often lack the redistributive elements of state pension schemes, disproportionately disadvantaging workers with interrupted careers and low pay – typically women (Frericks et al., 2007; Ginn, 2004; Ní Léime and Street, 2016).

The aforementioned developments underscore the need to examine pensions not only as social policy instruments but also as integral components of the wage relationship. Pensions can be understood as deferred wages – that is, they represent earnings that workers are entitled to based on their labour, but which they can access later in life. In this sense, pensions are linked directly to the structures and inequalities of the labour market itself. As occupational pensions assume a more prominent role, social partners obtain greater influence as key agents in negotiating this pension scheme (Anderson, 2019; Gonzales and Fernandez, 2024). This shift creates new opportunities for cooperation between employers and employees but may also give rise to conflict between them. This development also carries the risk of exposing underlying disagreements within broad federations on either side, as this article demonstrates.

A relatively small number of studies have examined employer and employee preferences in the design and regulation of occupational pension schemes (Bridgen and Meyer, 2005; Grødem and Hippe, 2018; Keune, 2018; Trampusch, 2013; Wiß, 2015). These studies indicate that the development of occupational pension schemes is highly context-dependent and that trade unions have fought a largely defensive game in recent years to preserve existing rights in some form (Naczyk and Seeleib-Kaiser, 2015). There have also been studies on how the increasing reliance on occupational pension schemes potentially exacerbates female workers’ vulnerabilities (e.g. Foster, 2010; Ginn and Arber, 1993; Grødem and Hippe, 2021). However, to our knowledge, no study has explored whether male- and female-dominated unions advocate different policies regarding occupational pensions. This article addresses this gap by asking the following question: What interests have trade unions historically pursued in relation to occupational pensions, and do these interests align across unions representing sectors with different gender compositions? Furthermore, what institutional barriers in negotiation structures marginalise female-dominated unions?

We use a qualitative research design to examine the evolution of Norway’s occupational pension design and two recent strikes led by the Norwegian Union of Municipal and General Employees (Fagforbundet), a strongly female-dominated union. These strikes were the first in modern Norwegian history to focus on occupational pensions and, hence, unique historical events. Strikes are a form of political action, with the capacity to induce economic disruption, challenge the legitimacy of established institutions, reconfigure policy agendas, and exert agency of those marginalised from policymaking processes (Tilly and Tarrow, 2015). In circumstances where alternative channels of representation are unavailable, strikes can function as a means to exert influence from the bottom-up. Our aim is to demonstrate how the two strikes examined in this article emerged from long-standing discontent regarding the degree of risk sharing, considering women’s longer life expectancy.

The gender-progressive context of our Norwegian case study allows us to analyse the dynamics of pensions in a setting in which unions, managers and workers are likeliest to be aware of them. Moreover, unionisation rates are high in Norway among both male and female workers, but because the Norwegian labour market is highly segregated, women and men dominate different sectors of the economy and, correspondingly, are organised in different unions.

Norway implemented a pension reform in 2011 (Stølen et al., 2020), which will lead to reduced state pensions and allow occupational pension schemes to play a greater role over time. The debate over workplace-based pension schemes during and after the reform process will be relevant to other countries undergoing similar processes – that is, most European countries (Hinrichs, 2021). Our findings may be particularly relevant for other counties with gender-segregated labour markets, where women disproportionately work often in public health and care services. These countries include the Nordic countries (Nordic Council of Ministers, 2025).

Our findings suggest, first, that trade unions will speak with a unified voice on some issues, such as the right to negotiate occupational pensions rather than just accepting the terms imposed by employers. On other issues, however, they may have different priorities or direct disagreements. Priorities and interests follow both sectoral and gender lines, but in a segmented labour market such as the Norwegian one, these two dimensions can be conflated. Second, how the negotiations are structured has a strong impact on which results are possible. We find that debates over occupational pensions bring to the fore issues that are typically seen as tangential, at best, in the general debate on pension scheme design. In our case, these issues are the inheritability of pension rights and the duration of annuities.

This article proceeds as follows. First, we review the literature on the social protection of women, with a focus on their various roles as mothers, carers, wives, workers and citizens. Second, we examine the evolving debate on pension policy, particularly the shift from state to occupational pensions and the institutions involved as well as the gendered challenges these pose. Third, we examine the context of Norway before conducting an empirical analysis, which is structured into two parts. The first part traces the development of the current occupational pension system and provides a historical overview from the late 1990s onwards. Drawing on policy documents, previous research and 16 interviews with key stakeholders, this section builds on comprehensive analyses published earlier (e.g. Grødem and Hippe, 2018; Pedersen et al., 2018). The second part of the analysis focuses on two strikes that took place in 2022 and 2023, respectively. This section is based on communications from the parties involved, including press releases and media statements. Through this historical and empirical analysis, we aim to broaden the debate on occupational pensions beyond the narrow focus on coverage rates (Grødem et al., 2018) and contribute to a more nuanced discussion of durability, risk-sharing mechanisms and regulatory frameworks.

Social protection of women: Mothers, carers, wives, workers and citizens

As late as the 1990s, women in the then-EU countries generally had much lower employment rates than men. The Nordic countries were partial exceptions and were seen as ‘leaders’ towards a future with higher female employment rates, longer parental leave and comprehensively subsidised public childcare (Ellingsæter and Leira, 2006). However, women in Nordic countries worked disproportionately in the public sector, typically in jobs related to health, care and education, and this gender segregation in the labour market raised questions about how successful the labour market mobilisation of women had been in the struggle for gender equality (Mandel and Semyonov, 2006). Moreover, the structures that covered the labour market remained largely unchanged, including the practices for collective bargaining.

As workers in the public sector, women benefited from strong employment protection, regulated working hours and, in many cases, generous and well-regulated occupational pensions. Incidentally, however, women’s entry into the labour markets coincided with a shift in public policies from public to private provision when many previously public enterprises were privatised (Keune et al., 2008). This was mainly true for infrastructure services, such as transport, mail and telecom services, but was also observed in health, care and education services. Women’s public-sector jobs – and the protections that come with them – are thus increasingly vanishing and reappearing in the private sector. A third main trend since the late 1990s has been the shift in social policies from ‘passive protection’ towards ‘active social investment’ (Morel et al., 2012), which has led to cuts in programmes designed for protection during periods and life phases of nonwork. In old-age pensions, this shift has led to a series of pension reforms designed to prolong working life and mobilise older workers’ resources (Hinrichs, 2021).

Taken together, the aforementioned developments bring back a key concern from the feminist social policy debate in the late 1990s: On what basis should women be protected in social policy? Should they be protected as mothers through benefits linked to childbearing and childrearing, as carers with benefits for all caring work, as wives through derived benefits, as workers with accrued benefits or as citizens through universal pension schemes (Ciccia and Sainsbury, 2018; Lewis, 1992; Sainsbury, 1999)? In societies based on the male breadwinner model, men are protected as workers, with their own rights accrued through employment (Lewis, 1992), but for women, the picture has traditionally been much more complex. Over time, with increasing female employment rates and demands for gender equality, the trend has moved in the direction of protection for individual workers of both genders, combined with minimum pension schemes protecting those without sufficient accrual. Survivors’ pensions have been decreased or abolished in many countries, including Norway (Organisation for Economic Co-operation and Development, 2018).

Employed women are covered by occupational pension schemes on the same terms as their male colleagues. Given the tendencies towards gender segregation in labour markets in many countries and women’s higher propensity to have shorter, interrupted careers with spells of part-time work, concerns have been raised over women working in industries with poor pension schemes or their exclusion from existing pension schemes due to a lack of accrual (Ginn, 2004; Ginn and Arber, 1993). In purely actuarial systems, pensions are calculated using mathematical logic, which creates dilemmas in terms of gender. As explained by Frericks et al. (2009: 713):

If the principles of such a system were to be fully applied there would be a risk of negative consequences for women: gendered life-tables work out very negatively for women due to calculations of their higher life expectancy, that is, a longer period of retirement. To get the same monthly benefits women would have to pay higher contributions.

In practically every country in the world, women have longer life expectancy than men. In traditional pension schemes, this discrepancy is managed through risk sharing, whereby those who die early ‘subsidise’ those who live longer. In actuarial pension schemes, however, the systematic difference in expected longevity poses a conundrum with no perfect solution. One potential solution would be to lower the annual payout or delay the retirement age for women, but this is politically impossible in most contexts. Another solution would be requiring higher contributions for women, as suggested by Frericks et al. (2009), or imposing a cap on payouts, such as that no one receives benefits for more than 10 years. The former option would make women more costly to employ and therefore less attractive to employers, while the latter would leave women with more uninsured years at the end of their lives.

The conundrum described above revives the debate on gendered pension policy, this time focusing on women in old age. Occupational pension schemes must involve some form of risk sharing to efficiently protect women as workers throughout their lives without making them less attractive employees. As an alternative, women can be protected as survivors in a purely actuarial pension scheme if their partner dies before their occupational savings account is emptied.

If a female worker can inherit her partner’s remaining savings, she can build a nest egg for the period in which she is not insured. However, this approach would not benefit single women, and for women with partners, it would make their financial situations in their old age dependent on their partners’ choices and circumstances. These are situations that the women’s movement has tried to avoid (Ciccia and Sainsbury, 2018; Sainsbury, 1999), and it is bound to be controversial if measures are introduced to protect women as wives in an attempt to soften the impact of pension schemes that fail to fully protect women as workers.

Occupational pension models, union strategies and gendered challenges

Occupational pensions are negotiated between social partners, but the state can play an important role as a regulator. Trade unions – organisations that are built on solidarity and aim to promote their members’ interests – want decent pension incomes for unionised workers. As noted by Keune (2018), occupational pensions can be both opportunities and threats in this regard. On the one hand, the fight for better occupational welfare can be a source of solidarity (Trampusch, 2007). It can allow trade unions to have a genuine influence on a benefit that is of great value to their constituents and to have the opportunity to directly influence future pension levels. On the other hand, the shift from public to occupational pensions can be seen as a form of privatisation and individualisation of social risks (Keune, 2018: 465). In this sense, occupational pension schemes represent a threat, and trade unions often act as defenders of the status quo and opponents of pension privatisation. Naczyk and Seeleib-Kaiser (2015) suggested that trade union strategies have changed over time. In the 1980s, with the liberal turn and increasing pressure from international organisations to develop a multipillar pension system, trade unions unanimously opted to defend the existing public system. However, as their efforts failed, they instead demanded extending the coverage rates of supplementary pensions to larger sections of the workforce ‘through either collective bargaining or legislative changes’ (Naczyk and Seeleib-Kaiser, 2015: 365).

As mentioned earlier, while occupational pensions may lead to new disputes between employers and employees, they also open up opportunities for cooperation between them. For instance, trade unions may push for collectively organised pension schemes because they offer generous benefits, while employers may be drawn to such pension schemes to promote wage moderation (Anderson, 2019). Moreover, broad, industry-wide pension schemes lower the risk of unfair competition, where enterprises can use occupational pension schemes to compete for labour or cut costs. It follows that both employers and unions may embrace collective pension schemes anchored in wage bargaining and underpinned by strong state regulation (Morgan and Orloff, 2017). However, this can play out differently across sectors, and different unionised workers can have different interests.

In situations in which well-regulated collective pension schemes are established, trade unions will want as high a coverage rate as possible for their members. However, there may be trade-offs between coverage rates and the quality of pension schemes; this is where different unions can have different priorities. Basically, there are two core models of pension calculation: defined benefit (DB) and defined contribution (DC) (Bodie et al., 1988). In ideal-typical DB pension schemes, pensions are calculated as a proportion of the employees’ wages, typically aiming for a replacement rate of about 67%. Payout is lifelong, and if a member of the DB pension pool dies, their savings revert to the pool. In contrast, ideal-typical DC pension schemes are saving schemes that do not aim at a set replacement rate. Levels of payout depend on how much is saved and on any dividends from the investment of the savings. DB pension schemes are typically administered by employers and can be costly because of the guarantees they imply and because administrative costs can be high. Conversely, DC pension schemes can be administered by the workers themselves and generally imply lower administrative costs.

In recent decades, there has been a massive shift across countries from DB to DC pension schemes (Bridgen and Meyer, 2005; Perrin, 2025) in a move that is actively encouraged by supernational organisations, such as European institutions (Perrin, 2025) and the World Bank (World Bank, 1994). However, the impact of this shift in terms of gender has rarely been discussed. As transitioning to purely actuarial (DC) pension schemes has important implications for women, particularly in light of women’s longer life expectancy, there is a need to analyse what this implies for women as well as how the unions representing female employees – the key policy actors in this area – are responding. In the next section, how this shift has played out in Norway is discussed.

Occupational pensions and the gendered labour market in Norway

Labour market and pension system

Norway has high female labour market participation and unionisation rates (Nergaard, 2024) but has a highly segregated labour market (Barth et al., 2023). Both men and women are unionised, with unions often clearly divided into male- and female-dominated sectors (Nergaard, 2023). This segregation leads to distinct bargaining areas with different needs (Wagner and Teigen, 2021). Occupational pensions with minimum savings rates are mandatory for all, which implies that Norwegian unions can focus on pension models rather than just coverage rates. Hence, occupational pensions have been contentiously discussed in male- and female-dominated sectors. The design of such pensions, the actors involved and the right to strike are key topics discussed in this section.

In Norway, annual national negotiations take place through sequential bargaining at the sectoral level, where the sectors are largely defined by employers’ organisations. The three main employer confederations in the private sector are NHO (Næringslivets Hovedorganisasjon [Confederation of Norwegian Enterprises]) (private industry), Spekter (deregulated state businesses) and Virke (services), while smaller organisations include the National Association of Private Kindergartens (PBL).

Out of a labour force of just over 3 million, nearly 1 million are organised in the Norwegian Confederation of Trade Unions (Landsorganisasjonen [LO]). This makes LO the largest umbrella organisation for organised workers in Norway. Other major union confederations include Unio (with approximately 400,000 members) and Akademikerne (with 255,000 members) (Nergaard, 2024). These confederations negotiate separately with NHO, Virke and Spekter each year. Negotiations start with the LO-affiliated Fellesforbundet and the NHO-affiliated Norsk Industri. These actors organise employees and employers in export-oriented industries and, in accordance with the Norwegian model of pattern bargaining, set the pace for all subsequent negotiations. Thus, when LO or NHO wishes to bring up an issue in the negotiations, it must first persuade Fellesforbundet or Norsk Industri to give it priority. This gives Fellesforbundet, a male-dominated union, a significant influence within the trade union movement. (An overview of the main players in the field of Norwegian industrial relations is given in the Appendix.)

Strikes are strictly regulated and can occur only during central-level negotiations. Local-level negotiations cannot involve strikes; hence, unions have a strategic interest in pushing key issues to the central level. However, unlike their counterparts in Sweden and Denmark, NHO and Virke have consistently vetoed negotiations of occupational pensions centrally (Trampusch, 2013). The third main confederation, Spekter, includes occupational pensions in its agreements. ¹ These differences among employer confederations allow strategic manoeuvring among trade unions while also limiting their scope for action.

Different occupational pensions and life expectancy

Norway currently has three types of occupational pension schemes: DB, DC and hybrid. Before 2001, DB pension schemes covered all public-sector workers and about half of all private-sector workers (Hippe et al., 2007). By 2023, however, DB pensions had been phased out in the public sector and covered fewer than 1% of employees in the private sector (Finans Norge, 2024). DB pension schemes, with risk-sharing and lifelong guarantees, are quickly disappearing, marking a transformation in Norway’s pension system over two decades.

DC pension schemes now dominate the private sector, covering nearly 2 million workers (Finans Norge, 2024), and they have been mandatory since 2006. Employers must contribute at least 2% of an employee’s wages to their pension. In the dominant DC variation, contributions are paid into a personal account that can be transferred between jobs and topped up with private savings. Payouts must last for at least 10 years or until age 77. Accumulated assets can also be used to buy a lifelong annuity, but this option is rarely used, and the cost of the annuity will differ between men and women. If a recipient dies with funds still in the account, the balance goes to surviving children first and the surviving partner second; if no such beneficiaries exist, it is included in the deceased’s estate. Due to their longer life expectancy, women face more uninsured years, but those with a partner are likely to inherit their partner’s remaining pension funds. If the deceased partner worked in the public sector, the spouse will receive a survivor’s benefit.

In 2023, hybrid pension schemes covered 71,000 private-sector employees (Finans Norge, 2024). While similar to DC pension schemes in savings rates, hybrids differ in that they promise lifelong payouts. Due to gender differences in life expectancy, this requires higher savings rates for women, and because they include risk sharing, they incur some administrative costs for employers. If a member dies before their savings are exhausted, the remaining funds revert to the pool, meaning those who die early ‘subsidise’ those who live longer. Table 1 summarises the key features of the DB, DC and hybrid pension schemes.

Table 1.

Key features of the defined benefit, defined contribution and hybrid pension schemes, as currently existing in the private sector in Norway.

	Defined contribution	Defined benefit	Hybrid
Risk sharing	None	High	Medium/low
Level of benefits	Varies by years worked and lifetime income	Depends on final wage and is reduced if the fixed minimum contribution years have not been met	Varies by years worked and lifetime income
Costs to employers	Depends on savings rate	High and increasing with the worker’s age	Depends on savings rate
Cost control for employers	High	Low	Medium
Duration	At least 10 years	Lifelong	Lifelong
Survivor’s benefits included	No	Sometimes	No
Inheritability of accumulated assets	Yes	No	No

Methods

This article examines the reasons why the female-dominated union Fagforbundet went on strike twice in two years for better occupational pensions – the first and only strikes to date on this issue in Norway.

The analysis is divided into two sections: first, the evolution of occupational pensions between the late 1990s and 2018, and second, two recent strike actions. The first part, which focuses on the transition to DC pension schemes in Norway, is based on policy documents, previous studies and information obtained in qualitative interviews with 16 key actors in the pension reform process. These interviews have been thoroughly presented and analysed in previous contributions (e.g. Grødem and Hippe, 2018; Hagelund and Grødem, 2019) and form the background for the discussion in this section. In this first part, we highlight the actions of two unions, Fellesforbundet (representing male-dominated professions) and Fagforbundet (representing female-dominated professions), both of which are affiliated with LO. Fellesforbundet, the largest private-sector union in Norway, with 170,000 members (83% male), first conducts annual wage negotiations with Norsk Industri. Fagforbundet, the largest union overall, with 400,000 members (77% female), represents workers across sectors such as health and care services, postal services and hairdressing and personal services. In this article, we consider Fellesforbundet as representing the male voice in Norwegian unions and Fagforbundet as representing the female voice, but we acknowledge that the real world is more complex than this.

The second part discusses the development of the two pension-related strikes. For this, we analyse public documents and union statements. We are interested in how the actors communicated their positions to the general public – that is, their communicative discursive strategies (Schmidt, 2008). Communicative discourse aims at legitimising political ideas to the general public, as opposed to coordinating discourse, which takes place among the policy actors themselves. Analysing communicative discourse provides insights into how technical issues can be translated into recognisable political ideas and which arguments and concerns key actors consider legitimate. We assume that the coordinating discourse within unions – and between unions and employers – is more technical, but we do not have direct information about it (but see Grødem and Hippe, 2018). However, public documents offer insights into key actors’ positions (see Bowen, 2009).

We analyse the communicative discourse of key actors in two strikes: one in the cultural sector (2021) and another by private kindergartens (2023). We focus on documents issued in relation to the two strikes: 53 documents collected from unions and five from the employers during the cultural-sector strike, and five documents from trade unions and two from employers during the kindergarten strike. A lower number of documents from the kindergarten strike was used partly because this strike was of smaller scale and was less controversial than the cultural-sector strike.

We also searched Atekst, a database of articles from 2970 Norwegian media sources, using the search term ‘Organisation name AND strike’ for the durations of the strikes. This resulted in many duplicates, mainly news reports on the strikes’ durations and consequences. We manually selected articles featuring statements from key spokespersons from both parties.

Throughout the analysis, we employed an inductive analytical approach to determine how the key actors articulated the interests of female workers in relation to occupational pensions, with particular attention given to two pivotal moments: the structural shift from DB to DC pension schemes and the subsequent industrial action. Key themes included rising tensions within the union movement and the growing focus on the trade-off between lifelong pensions and the inheritability of pension rights as DC pensions became increasingly prominent.

Findings: Historical developments and two strikes

From state to occupational pension schemes: Simmering discontent

Occupational pensions in Norway underwent a complete transformation between the late 1990s and the 2020s from solely DB pensions to an overwhelming domination of DC pensions. In this section, we discuss how this process unfolded, by whose initiative and the positions taken by the trade unions and employers involved.

Introduction and normalisation of DC pension schemes

In the late 1990s, only DB occupational pension schemes received tax rebates, and such schemes covered about 50% of the employees in the private sector. Fellesforbundet considered the low coverage rates a problem, while also realising that getting more employers to adopt expensive DB pension schemes with unpredictable costs would be impossible. Thus, in the 1998 negotiations, they partnered with the NHO-affiliated Norsk Industri to lobby for tax rebates for DC pension schemes (Pedersen et al., 2018). Female-dominated unions were sceptical even at this early stage and feared that making DC pension schemes more attractive options would undermine the established DB pension schemes. These unions, the predecessors of Fagforbundet and Unio, mainly organised public-sector workers already covered by DB pension schemes; hence, coverage rates were not a key concern for them (Pedersen et al., 2018). Given the structure of Norwegian pattern bargaining, Fellesforbundet and Norsk Industri negotiate first and set the pace, and important strategic issues tend to be settled in their negotiations. The parties sent a joint letter to the government demanding tax rebates for DC pension schemes, and the government agreed and introduced new tax regulations that took effect in 2001.

NHO, unlike its counterparts in other Nordic countries, refused to negotiate occupational pension designs and levels during central negotiations (Trampusch, 2013). This stance was motivated by the fear of a competitive imbalance between firms with and without collective agreements and by lobbying from small firms that feared that they would be burdened with occupational pensions they could not afford. Decisions about occupational pensions would thus be made at the local level, where there is no right to strike. This gave the upper hand to employers; thus, the main impact of the 2001 tax regulations was the mass conversion of existing DB pension schemes into DC pension schemes. Coverage rates remained largely unchanged (Hippe et al., 2007). Frustrated, Fellesforbundet demanded central pension negotiations in 2002 and 2003 but received a hard ‘no’ from NHO.

In 2004, Fellesforbundet changed its tactic and invited Norsk Industri to collaborate on another letter to the government, this time demanding a mandatory occupational pension for all. Pensions for all had been its chief priority all along, and if such pensions were mandatory, employers would not have to worry about competitive imbalances. The government responded favourably to the demand and made occupational pensions mandatory in 2006. Since then, all employees in the private sector have had access to occupational pensions, but savings rates vary widely. All the new pension schemes were of the DC type, which explains why DC pensions are now the vastly dominant form of pension in the private sector (Pedersen et al., 2018).

The fight for DB pension schemes and the introduction of hybrid pension schemes

The transformation of occupational pensions in the private sector caused concern in the trade unions that prioritised generosity and risk sharing in pension schemes as highly as, or higher, than full coverage. When the state tried to fundamentally transform the public-sector DB pension scheme in the 2009 negotiations, the unions refused (Hagelund and Grødem, 2019). Some large, deregulated state enterprises also retained their DB pension schemes, and in private kindergartens, unions negotiated a deal that guaranteed employees the same pensions as their colleagues in public kindergartens. However, this situation became increasingly unsustainable because ‘first-pillar’ pensions in the National Insurance Scheme were reformed in 2011 to be calculated using a DC-like formula (Stølen et al., 2020). Retaining occupational pensions on a DB formula in this situation would lead to immense financial strain on DB providers over time. The Ministry of Finance attempted to find a compromise by presenting legislation for hybrid pensions in 2014, which was initially met with a lukewarm response. Unio, Fagforbundet and other female-dominated unions were more interested in an ongoing attempt to develop a DB model that fitted the new public pension (Official Norwegian Report, 2015: 5). However, this attempt fell flat due to strong and unified opposition from employers. Meanwhile, Fellesforbundet, committed to DC pension schemes, explored a model for the full individualisation of occupational pension savings.

The mounting tensions between trade unions, many of which were affiliated with LO, came to a head at the LO representative board meeting in 2016. The female-dominated union Handel og Kontor, backed by Fagforbundet and others, raised a motion to instruct Fellesforbundet to initiate negotiations on ‘broad, encompassing pension schemes’ in the upcoming collective bargaining round (Pedersen et al., 2018). However, what ‘broad, encompassing pension schemes’ meant was unclear, and it was understood as code for a hybrid insurance scheme paying lifelong, gender-neutral benefits. Against the explicit wishes of Fellesforbundet and the recommendation of the LO leader, the motion passed. Loyal to LO, Fellesforbundet raised the demand but was rebuked by Norsk Industri, which again insisted that occupational pensions were not to be negotiated at the central level. However, Norsk Industri was willing to work again with Fellesforbundet for legislative changes, this time to make DC pension savings fully individualised through individual savings accounts.

In the 2018 negotiations, public-sector unions caved and agreed to replace their existing DB pension schemes with hybrid pension schemes (Hagelund and Grødem, 2019). This was the final nail in the coffin for DB pension schemes and made hybrid pension schemes much more visible. In the 2019 round of negotiations, Fagforbundet and Unio negotiated with PBL and established a hybrid pension scheme nicknamed kindergarten pension (barnehagepensjon). Achieving this result required a long process and the assistance of independent pension experts, but in the end, the trade unions Utdanningsforbundet, Fagforbundet and Delta celebrated the outcome as a victory and a new gold standard for private-sector pensions (Sandgrind and Karlsen, 2020).

The shift from DB to DC pension schemes in the private sector and the refusal of NHO to negotiate alternatives in central negotiations gave unions representing female-dominated sectors few options beyond raising the struggle for occupational pensions with risk sharing in their own negotiations outside NHO. The breakthrough of the hybrid pension scheme in the public sector also gave them a blueprint: they could easily argue that they deserved what their colleagues in public-sector units had. In the next section, we discuss the room for manoeuvre expressed in two strike actions. First, as discussed in the cultural-sector strike, if Fagforbundet wanted to establish hybrid models at the sectoral level, this would have to happen in areas in which employers accepted central negotiations over occupational pensions. The only employer organisations that allowed this were PBL – where a hybrid pension had already been negotiated – and Spekter. In the next section, we analyse the subsequent strike against Spekter. Second, the different occupational pensions in different sectors brought the risk of employers ‘shopping around’ and joining employers’ organisations in which they would not be bound by sector-wide agreements to provide hybrid pensions. This, as we shall see, prompted a strike in kindergartens.

Strikes for occupational pensions in Norway

In this section, we briefly describe the cultural-sector and kindergarten strikes and how the involved parties communicated their concerns to their members and the public. An overview of the two strikes is provided in Table 2.

Table 2.

Overview of the two strikes.

	Cultural-sector strike	Kindergarten strike
Period	3 September–25 October 2021 (7 weeks)	13 February–17 March 2023 (4 weeks)
Size	900 employees	63 employees
Trade unions	NTL, Fagforbundet, Creo	Fagforbundet
Employer association	Spekter	NHO
Main issue	Transition from the DC pension scheme to the hybrid occupational pension scheme	Transition from the DC pension scheme to the hybrid occupational pension scheme
Outcome	Agreed to establish a ‘lifelong and gender-neutral’ pension scheme from 1 April 2022. The word ‘hybrid’ is not mentioned in the agreement. Agreed to work together with independent experts to resolve contentious details by 1 February 2022.	Agreed that kindergartens that joined NHO should retain their existing pension arrangements until 30 April 2026. Agreed to work together with independent experts to resolve contentious details by 1 February 2024.

The cultural-sector strike

Causes of conflict

In 2021, Fagforbundet, together with fellow LO organisations NTL (Norsk tjenestemannslag, the Norwegian Civil Service Union) and Creo, called a strike in the cultural sector during the sectoral-level negotiations with their employer counterpart Spekter. This was the first-ever strike in Norway solely on occupational pensions; thus it was a unique event. Most of the cultural institutions whose employees were represented by Fagforbundet were deregulated public-sector units in which employees historically had DB pensions. During the 2016 negotiations, unions were under pressure and accepted a shift to a DC pension scheme after employers cited financial difficulties and threatened the closure of theatres and orchestras. In a protocol, the parties agreed to revisit ‘lifelong and gender-neutral’ pensions later. Five years later, unions claimed that the protocol was a binding promise for action, while the employer side argued that it was only aspirational, with no firm obligation.

Demands and responses

The above historical background heavily dominated trade unions’ communications in the early days of the cultural-sector strike. Union spokespersons and a series of press releases hammered the point home: workers in the sector were promised ‘lifelong and gender-neutral pensions’, and that was all that the strike demanded (e.g. LO stat, 2021). Thus, the trade union side relied heavily on a gender and equity discourse in its demands, something that had not been explicitly promoted by the union before in the area of pensions.

Conversely, the employers framed the negotiation in the context of economic and social strain, as opposed to gendered impacts. First, they argued that the cultural sector was already under immense pressure due to the prolonged impact of closures during the COVID-19 pandemic. After 18 months of closures, cancellations and limitations on audience capacity, the sector was in a fragile state, struggling to recover financially and operationally. Moreover, they emphasised three points: (1) employees in the cultural sector had the highest savings rate that the DC pension scheme allowed, (2) the costs of the scheme were low and predictable for employers and (3) saved means could be paid out to the families in the event of death (Spekter, 2021). Spekter refrained from engaging with a gendered narrative, treating the design of occupational pensions as a technical matter that was, in fact, regarded as beneficial for the workers.

The strike went on for seven weeks, which is long in the Norwegian context, with several meetings between the parties and the state’s arbitrating body. The lack of progress led to increased frustration on both sides, and the rhetoric was ramped up. On 21 October, 32 leaders of cultural institutions published an open letter claiming that their industry was being used for a broader agenda. They stated that occupational pensions for cultural workers were fine, but LO had chosen their institutions as the battleground to spearhead hybrid pensions in the private sector (Scenekunst, 2021). This framing fitted the overall rhetoric of Spekter, which had already claimed that LO tried to force a pension scheme with high and unpredictable costs on cultural institutions for political reasons. Hence, the debate was moving away from ‘what was promised in 2016’ and towards more technical discussions over ‘the pros and cons of hybrid versus pure savings schemes’. This shift prompted Unio, which had long been a staunch defender of hybrid pensions, to enter the field in late October to emphasise the importance of the strike for female-dominated sectors of the labour market. The Unio leader declared, ‘We are an organisation for large female-dominated groups and see this strike as a struggle for gender equality in occupational pensions’ (Unio, 2022) and announced that Unio considered sympathy action. However, the strike ended on 25 October 2021, before the alliance between Unio and the LO-affiliated unions on strike solidified.

Results

The parties involved in the cultural-sector strike reached an agreement after seven weeks. The collective agreement established ‘lifelong and gender-neutral’ pensions from 1 April 2022 (i.e. a year later). However, it did not explicitly include the unions’ demand for hybrid pensions, with the term ‘hybrid’ conspicuously absent in the agreement. Details of the 2022 solution were to be settled through collaboration with pension experts. This process turned out to be difficult, and there were threats of new strikes in 2022, but a compromise was eventually reached, in which hybrid pensions were to be the default, with employees allowed to choose DC pensions. This achieved the unions’ key goal but still left the door open for DC pension schemes.

The cultural-sector case shows the potential for strikes over occupational pensions. In the kindergarten sector, a female-dominated field with an existing hybrid pension, a strike arose when employers threatened to join an association that opposed hybrid pensions and banned central-level occupational pension negotiations.

The 2023 kindergarten strike

Causes of conflict

Similar to the cultural sector, privately owned kindergartens were ‘public sector-adjacent’ enterprises. Unlike in the cultural sector, however, the fight for hybrid pensions was already largely won in the sector. By 2022, about 50% of the kindergartens in Norway were owned by municipalities and thus covered by the public-sector hybrid pension scheme by default. Among privately owned kindergartens, about 70% were organised under the employer’s organisation PBL, which established the hybrid occupational pension scheme in 2019. Kindergartens organised under Virke or Spekter followed suit, leaving only a few NHO-affiliated kindergartens with a DC pension scheme.

In the 2021 budget, the government proposed cuts in the reimbursement for occupational pensions in PBL. A few months later, a large private kindergarten chain announced that it would leave PBL and join NHO instead (Fagbladet, 2022). Trade unions feared that this large actor’s exodus was just the beginning and that private providers would flock to the only organisation where they would not be required to offer hybrid occupational pensions – NHO. Closing this loophole thus became crucial.

Demands and responses

In NHO, occupational pensions are not part of collective agreements. The unions on strike thus knew that the best they could hope for with regard to occupational pension models would be a declaration of intent. Fagforbundet’s rhetoric from the commencement of the strike to the very end was straightforward: kindergarten pension was now standard in kindergartens, and it was absurd and unreasonable that some kindergarten workers should not have access to it. It asked, ‘Why can the NHO-affiliated kindergartens not manage such pensions for the employees when all other kindergartens can?’ (Fagforbundet, 2023). In addition, Fagforbundet also employed the rhetoric from the cultural-sector strike just over a year earlier and repeatedly used the phrase ‘lifelong and gender-neutral’ in interviews with the press (e.g. Dagsavisen, 2023). One of the official posters for the strike read, ‘Kindergarten employees deserve a pension that is gender neutral, lifelong [and] safe’ (poster, published on Fagforbundet’s webpage, 2023).

NHO released a short statement on 14 February 2023, in which it expressed regret for the consequences of the strike for families and emphasised that the kindergartens affiliated with it already offered DC pensions with high savings rates. The statement maintained that hybrid pensions were unpredictable, unsustainable and incurred high administrative costs, and that accumulated funds would not be inherited by the family when the insured person died. These points were repeated in subsequent interviews (e.g. Dagsavisen, 2023). Similar to Spekter in the strike described earlier, employers refused to engage with the ‘hybrid or DC’ discussion and opted instead to highlight the generosity of the existing DC pension schemes – thus hinting that the unions on strike were privileged – and the inheritability of DC pensions. As in the cultural-sector strike, Fagforbundet highlighted that hybrid occupational pension schemes offer the best protection for female workers because they offer lifelong payouts. Here, unions once again emphasised life expectancy as a political issue, while employers emphasised seemingly gender-blind technical details of comparatively good DC pension schemes and the lack of inheritability in hybrid pension schemes.

Results

The kindergarten strike ended after four weeks, on 17 March 2023, when what could be considered a compromise was reached. LO considered it a victory that the parties agreed that new enterprises joining NHO would retain their existing pension agreements until 30 April 2026. This dampened the acute fears of opportunistic ‘shopping around’ between different industry organisations to obtain cheaper pension commitments. However, LO did not succeed in negotiating hybrid pensions for employees in NHO-affiliated private kindergartens. Instead, the parties agreed to establish a working group to outline the options available and the pros and cons of the respective models in preparation for the 2024 negotiations. Similar to what happened in the cultural-sector strike, the negotiated solution allowed those involved to kick the can down the road in the hope of reaching a consensus through deliberation and the involvement of experts. However, this process amounted to little, and as of early 2026, employees in NHO kindergartens still have DC pensions.

Discussion

We have analysed the interests that trade unions have pursued in the occupational pension area and discussed how and to what extent such interests differ between unions representing different sectors with different gender compositions. Moreover, we have asked whether there may be institutional obstacles that prevent the interests of female-dominated unions from being properly articulated and promoted (Wacjman, 2000). Our findings show the importance of differentiating the kinds of unions involved in the occupational pension policy debate, who they represent and the role they play in the institutionalisation of occupational pensions. We find that unions representing female- and male-dominated sectors have consistently pursued different priorities in the debate on occupational pensions and that unions representing male-dominated sectors have been more successful in achieving their goals. We argue that this is at least partly due to the institutional design of the collective bargaining process. The discussion addresses issues central to the debate on trade unions’ policies on occupational pension (e.g. Keune, 2018; Naczyk and Seelieb-Kaiser, 2015) but adds a gender perspective that has largely been absent in this debate.

Occupational pensions in Norway have undergone a massive shift in the last 25 years: from limited coverage of solely DB benefits before 2001 to full (mandatory) coverage and overwhelming dominance of DC pension schemes in the private sector by 2026. In the public sector, a hybrid pension scheme has replaced the old DB pension scheme. This overall shift to DC pensions served the mutual interests of powerful unions and employers. Unions secured broader coverage, while employers moved away from costly and unpredictable DB pension schemes. Although mandatory occupational pension coverage and individualised accounts were presented as efficient and fair, these measures simultaneously marginalised concerns about risk sharing and redistribution, which were primarily raised by female-dominated unions. These unions never obtained sufficient institutional strength to seriously challenge the emerging consensus.

The findings based on the cultural-sector strike in 2021 and the kindergarten strike in 2023 illustrate the strategic constraints and limited discursive traction available to female-dominated unions. The 2021 strike, launched under unfavourable conditions, was easily delegitimised by employers who cited recent pandemic hardships and the sector’s already strong pension provisions. The unions sought to reframe the issue as one of gender fairness but struggled to counter the dominant economic narratives. The 2023 kindergarten strike was more defensive, aiming to prevent the further deterioration of pension quality. Ultimately, neither strike achieved its aims, and a broader coalition in support of hybrid pension models has failed to materialise. The limited resonance of gendered fairness arguments, even among unionised women, highlights the entrenched power imbalances in pension governance. Employers’ insistence on firm-level decision-making and cost control – tacitly supported by male-dominated unions – appears to have had a greater long-term influence than the strikes themselves. While the strikes demonstrated resistance, they also exposed the limited political space within which female-dominated unions operate, both discursively and institutionally.

Our findings highlight the relationship between the gender composition of union membership and preferences for specific occupational pension models. This relationship has not been thoroughly examined in studies despite the increasing significance of occupational pensions as a substantial component of income for retirees. The analysis reveals that the institutional design of the bargaining process can marginalise women’s interests (Dawson, 2014; Wacjman, 2000; Wagner and Teigen, 2021), especially in systems where unions representing male-dominated sectors still dominate the political or wage-setting landscape, such as those in Sweden (Erikson, 2021), Belgium (Lemeire and Zanoni, 2021) and Germany (Artus, 2025). Consequently, female-dominated unions face considerable challenges in advocating for their interests.

In both strikes examined herein, female-dominated unions articulated their vision clearly by demanding gender-neutral, lifelong pensions based on individual entitlement. Their employer counterparts countered this by highlighting the inheritability of DC pensions – an argument consistently rejected by the unions, refusing to engage with the idea that women should rely on others’ entitlements for retirement security. In doing so, these unions aligned with a long feminist tradition that centres individuals as rights-bearing workers and citizens (Ciccia and Sainsbury, 2018; Ginn, 2004; Lewis, 1992).

The struggle over occupational pensions in Norway unfolded in the context of a broader structural shift. Recent years have seen a widespread move from DB to DC pension schemes across Europe, including in countries historically dominated by DB models, such as Germany and the Netherlands (European Insurance and Occupational Pensions Authority [EIOPA] 2025; Perrin, 2025). EU-level institutions, including the European Commission and EIOPA, have acknowledged and facilitated this transition, offering roadmaps and regulatory adaptations to accommodate the growing dominance of DC pension schemes. In Norway, the mandatory introduction of occupational pensions in 2006 improved coverage for women, especially in the lower-paid sectors. However, women are far more vulnerable than men to experiencing poverty in old age, and time-limited annuities will exacerbate this risk, given women’s longer average life expectancy. This is the concern that female-dominated unions in Norway have attempted to raise, and one that policymakers across Europe will have to grapple with as DC pensions increasingly become the norm (Frericks et al., 2007, 2009).

The strikes discussed in this article thus offer a rare lens through which to examine how gendered union interests are expressed, contested and ultimately constrained within corporatist policymaking arenas. The 2021 cultural-sector strike and the 2023 kindergarten strike were both defensive moves aimed at resisting the further erosion of collective, redistributive pension principles. However, despite drawing on strong egalitarian and gender-equality discourses, the strikes had a limited political impact. Employers either delegitimised the strikes publicly or refused to engage at all – a communication strategy that proved effective in maintaining the status quo. Attempts to build broader coalitions around hybrid pension alternatives in the wake of these strikes have so far failed to gain traction, even among unionised women.

This study has some limitations. First, it is a case study of only one country. As occupational pensions are deeply embedded in social dialogue and within national social policies, the extent to which the arguments presented here can be made relevant in other contexts is unclear. Nevertheless, we argue that while every national context may be unique, general insights about gendered arrangements are useful for theory building. Second, we have relied on published documents and press releases to chart the communicative strategies of the parties involved in the examined strikes. Interviews with the main players might have supplemented these with more insights into the coordinative strategies – that is, the key debates among the main players – and, hence, provided more information about strategies and priorities. Third, an explicit lens on the employers’ side is missing and useful to explore in further research because employers largely refused to engage publicly during the strikes discussed in this article.