Setting the scene: Privileging marriage, institutionalising inequality

The welfare state set up in Britain following the Second World War, with social security entitlement linked to employee-contributions, assumed full (male) employment, a male pattern of continuous full-time employment over the life course, and stable heterosexual marital unions. Patriarchal (and imperialist) logic excluded women from citizenship (Andersen, Reference Andersen2020; Lister, Reference Lister1990, Reference Lister and Hobson1999), with the Beveridge report emphasising women’s roles as wives and mothers with derived rights through the marriage contract. The institutionalisation of the male breadwinner model was explicit in many policies in the early decades. In fact, married women were initially ineligible for a benefit for caregivers introduced in the mid-1970s because such care was viewed as part of a wife’s normal duties (Lewis, Reference Lewis1992).

From 1975, legislation brought formal equality to the labour market by regulating the behaviour of employers, albeit retaining the androcentric view of the worker. Formal marriage bars, permitting automatic dismissal of female employees on marriage, initially abandoned across much of the public sector in the 1940s were outlawed across the private sector through the Sex Discrimination Act (1975). The Equal Pay Act, coming into force from 1975, and initially requiring equal pay for men and women doing the same job was in 1983 extended to cover equal pay for work of equal value (following a European Court of Justice judgement). However, despite such formal equality, the assumption remained in state policy that paid work would not be a (married) woman’s primary activity or responsibility and that her wages would be of secondary value in the household finances (Lewis, Reference Lewis1992). This assumption is exemplified by joint taxation of spouses, whereby a married man received a higher tax allowance than a single person. In effect boosting his take-home wages, this state-subsidised ‘family wage’ system (Seely, Reference Seely1995) privileged heterosexual marriage over other relationships and living arrangements and explicitly supported a gendered division of labour. Although individual taxation was introduced in 1990, the change included a married couple’s tax allowance option explicitly set at a level to fully compensate for the drop to the husband’s tax threshold (and thus his net income) from the removal of the married man’s allowance (Seely, Reference Seely2019). This allowance was abolished by New Labour a decade later but in 2015 the Conservative government introduced a new marriage allowance, whereby the lower earning spouse (or civil partner) could transfer a portion of their tax allowance if the higher earner was on the basic tax rate. Given that the lower earner in different-sex relationships tends to be the woman due to the gender pay gap and/or part-time hours, government projection was that among eligible couples the vast majority of transfer recipients would be men (Seely, Reference Seely2021), thus reaffirming continued state support for unequal division of paid and unpaid work within couple households and the continued gendering of ‘the worker’ as the male breadwinner.

Benefits: from unemployed breadwinners to ‘activation’ of caregivers

The male breadwinner model bias is even stronger in the benefits system. Writing in the early 1980s, Amina Mama (Mama, Reference Mama1984) highlighted the ‘legislated sexual discrimination’ of joint assessment, whereby jobseeker status was assigned to one partner paid the benefit (usually the man) in couples where neither was employed. Not only did this reiterate the man as breadwinner whether employed or not, joint assessment with assumed female financial dependency also invisibilised the lived reality of many Black and Asian women who to a greater extent have been full-time workers and equal earners or heads of households throughout the period (Brah, Reference Brah1996; Carby, Reference Carby and Mirza1982). During the 1980s, eligibility for unemployment benefits also became more strongly linked to recent employment (Lewis, Reference Lewis1992), creating a barrier for re-entry after time spent caring and thus reasserting the worker as male and without care responsibilities. Despite policy rhetoric changing from the 1990s to increasingly encouraging individual self-provisioning through labour market activity, the benefit system continued to assume and reinforce women’s dependence on their (assumed male) partners for income (Lewis, Reference Lewis2002a; Rake, Reference Rake2001). This family model requires dependable and sufficiently high-paying jobs for the man, not in reality guaranteed in many areas especially for many racialised working-class men (Brah, Reference Brah1996; Carby, Reference Carby and Mirza1982; Dean and Shah, Reference Dean and Shah2002; Reynolds, Reference Reynolds2009). The concept of financial dependency is thus premised on a white and middle class, as well as hetero-patriarchal, ideal.

With increasing neoliberal focus on reducing state welfare and increasing individual responsibility and economic self-provisioning, more recent employment activation policies aimed at women may be interpreted as a shift to an adult worker model with expectation of employment for all. However, from the New Deals for lone parents and ‘partners of the unemployed’ under New Labour to Universal Credit’s extension to also include partners in low income households, these moves neither destabilise the male breadwinner family ideal nor its related ‘maternalist logic’ (Orloff, Reference Orloff, Orloff and Morgan2017) of caregiving. First, the strong commitment to joint means-testing for benefits and the preoccupation with ‘workless’ households reveal the continued assumption of the male breadwinner/female caregiver dichotomy and eclipse any focus on employment support needs of caregiving (potential) workers. As an example, New Labour’s activation strategy for partners of Jobseeker’s Allowance claimants provided individualised support but tied eligibility to being a financial dependant on a Jobseeker’s Allowance recipient rather than own support needs (Rake, Reference Rake2001), highlighting continued state endorsement of women’s financial dependence on (male) partners. This focus on ‘workless’ households, with its inherent privileging of the male breadwinner family, continues under Universal Credit. When criticised for its disincentives for dual earning and access to independent sources of income for many women in claimant couples, the Universal Credit design was justified by the government on the basis that the priority is on reducing households without at least one earner (Bennett, Reference Bennett2021). The single Universal Credit payment for couples, ostensibly to mimic a salary, also reveals the underlying assumption of a single-earner/full-time carer family model, ‘lead’ carer work requirements notwithstanding. Second, the encouragement of maternal part-time work under Tax Credits and compulsion under Universal Credit for both lone parents and nominated lead carers in couple-families to look for at least part-time work do not challenge the breadwinner/caregiver dichotomy. Since part-time employment rarely provides financial security or pay progression in the short- or long-term, it does not provide a route to full citizenship (Andersen, Reference Andersen2020; Millar and Ridge, Reference Millar and Ridge2018; Nightingale, Reference Nightingale2020). Disregarding existing care responsibilities and gendered household divisions of labour create role contradictions that are difficult to navigate for many (Andersen, Reference Andersen2020; Bennett, Reference Bennett2021; Millar and Ridge, Reference Millar and Ridge2018), but in the context of gendering effects encouraging part-time work for women with children may best be viewed as compatible with upholding the primacy of their caregiver role. It absolves the state of either supporting the combination of full-time work and caregiving for all or valuing childrearing through adequate levels of social security and pension protection (see e.g Ginn and Arber, Reference Ginn and Arber1999; Ginn and MacIntyre, Reference Ginn and MacIntyre2013, on how classed and androcentric life course assumptions in UK pension policy perpetuate working age inequalities into old age).

Work-family ‘reconciliation’?

Over the decades, as tax and arguably also aspects of benefit policies have taken some steps in the direction of assuming a self-provisioning adult worker, ostensibly gender-neutral yet modelled on an androcentric ideal life course, policies to support the combination of paid work with caring and childrearing have not to date abandoned the gendered worker/caregiver distinction. Policy efforts to encourage fathers to share childrearing have been largely absent, with UK policy makers’ stated reluctance to interfere with private matters of how families organise their paid work and care thus reinforcing the gendered status quo of division of labour. As a result, as over the decades mothers increasingly re-entered or remained in the labour market following childbirth, childrearing and care remained firmly primarily their responsibility, and by extension in practice their primary responsibility. In fact, a ‘maternalist logic’ (Orloff, Reference Orloff, Orloff and Morgan2017) underpins ‘reconciliation’ policies that, despite their gender-neutral terminology, provide low or no payment for parental leaves and no incentives for men to make adjustments to their paid work. While such policies can assist some mothers to combine paid work and care, they simultaneously reinforce childrearing as a female responsibility.

UK maternity leave policy is a prime example of maternalist policy logic. The Social Security Act (1973) introduced 18 weeks of maternity allowance and the Employment Protection Act (1975) introduced 29 weeks of statutory job-protected maternity leave and protection against unfair dismissal. However, stringent eligibility criteria excluded many by requiring two years of service with the same employer (five for those working short part-time hours; Sigle-Rushton, Reference Sigle-Rushton2009). By stipulating such male-typical employment patterns, the effects of widespread ineligibility for job-protected leave effectively continued to position mothers as primarily full-time caregivers (and workers as male). These statutory provisions have over time been extended in coverage, length and the proportion of leave that is paid, but have remained firmly attached to the mother (or in the case of adoption, a nominated main carer), setting the UK distinctly apart from other European countries (Lewis et al., Reference Lewis, Knijn, Martin and Ostner2008). Further, the low level of maternity pay relative to earnings, which has persisted to the present day, demonstrates the assumption that the mother will have a (male) breadwinning partner to financially support her during maternity leave. It was not until 2003, when the Employment Act (2002) came into force, that second parents gained two weeks of statutory ‘paternity’ leave at birth or adoption, and not until 2011 that it became possible for parents to share ‘maternity’ leave (renamed ‘shared parental leave’ from 2015). The silence regarding assumed family form in leave policy is notable as, consistent with the gendering commitment to the nuclear family seen in other aspects of UK policy, most of the terminology has largely remained heteronormative even after adoption rights were extended to same-sex couples and single adults in 2002.

Until the late 1990s, public provision for childcare was limited and primarily targeted at children ‘in need’ (Lewis et al., Reference Lewis, Knijn, Martin and Ostner2008; Pascall, Reference Pascall1997). Importantly parental employment did not constitute a ‘need’, particularly impeding lone parents’ prospects of entering or sustaining paid work (Lister, Reference Lister and Hobson1999). In fact, women’s increased paid work specifically in part-time employment has been attributed to (especially Conservative) governments’ resistance to policies supporting maternal paid work (Warren et al., Reference Warren, Pascall and Fox2010). ‘In the early 1960s, the State was still busy trying to encourage (white) women to stay home and embrace domestication and consumerism. It wasn’t prepared to offer any childcare support to Black women who had to work’ (Bryan et al., Reference Bryan, Dadzie and Scafe1985, p. 29), despite often having come to Britain as workers in their own right (many recruited directly into the newly established NHS). Thus because of the intersection of class, race and gender, many Caribbean women, on arrival in Britain found that the lack of council childcare – especially if they were married – forced them either into home-work, characterised by particularly poor pay and lack of employment rights and protection, or anti-social hours such as night work (Bryan et al., Reference Bryan, Dadzie and Scafe1985; Carby, Reference Carby and Mirza1982; Klug, Reference Klug, Yuval-Davis and Anthias1989; Mama, Reference Mama1984). Childcare arrangements and costs being firmly positioned as an individual responsibility is thus not only gendering with regard to ‘the worker’ but also classing and racializing with regard to the ‘right kind’ of parenting (mothering).

It was not until the late 1990s and early 2000s that a broad suite of work-family ‘reconciliation’ policies were initiated, including the launch of the first National Childcare Strategy in 1999 with its early years education entitlement (initially for 4-year-olds, extended to include 3-year-olds from 2002). The stated policy goals were two-fold: to promote maternal employment (the entitlement was often referred to as ‘free part-time childcare’) as well as children’s early learning (Daly, Reference Daly2011; Lewis et al., Reference Lewis, Knijn, Martin and Ostner2008), yet covered only term-time and amounted to less than the minimum weekly hours of part-time work recognised for Tax Credit purposes. The 2016 Children Act extended the entitlement to 30 hours in England and Wales for (already) working lone parents and dual-earner couples. However, with almost a quarter of the year not covered and eligibility neither beginning at the end of post-birth leave nor covering required job search activity, UK childcare policy continues to fall short of genuinely assisting parents (mothers) to enter and sustain employment.

Mary Daly (Reference Daly2011) has argued that the policy model emerging in Europe during the first decade of the 2000s was not that of the autonomous adult worker but rather a gender-specialised dual earner family model. Similarly, this tracing of ‘the worker’ in UK policy over time shows that despite both activation and reconciliation moves towards an assumption and endorsement of dual-earning, or lone parent earning, the underlying family model remains gender-specialised and ultimately the changes have not altered the assumed (androcentric) shape of the worker’s life course.

The ‘main carer’

By the time of the introduction of the post-war welfare state, feminists had long campaigned in favour of a substantial amount to be paid to all mothers in recognition of their caring role (Lewis, Reference Lewis1978). The 1945 Family Allowances Act introduction of Children’s Allowances paid directly to mothers (replaced in the late 1970s with Child Benefit) was thus welcomed as a partial success despite being neither universal (due to pronatalist motivations) nor sufficient to challenge wives’ economic dependency. While putting all women in a vulnerable position, and inaccurate in relation to the paid work of many women of colour, the rigid construction of women as main carers financially dependent on a male breadwinner also resulted in some devastating ‘lived effects’ for migrant women and their children. Examples include Asian women threatened with deportation following divorce or widowhood (Brah, Reference Brah1996), and a disproportionate number of Black children taken into foster care due to the combination of low income, lack of childcare and poor housing, along with racializing and classing assumptions about ‘normal’ mothering and family arrangements (Mama, Reference Mama1984). Whereas white women’s paid work was socially accepted on the condition that it not adversely impact family (child) wellbeing, ‘black women were seen to fail as mothers precisely because of their position as workers’ (Carby, Reference Carby and Mirza1982, p. 49). Decades later, (white) middle-class models of the ‘right kind’ of mothering can similarly be seen to underlie family policies, albeit using gender-neutral language.

Family support policies have increasingly adopted an individualising frame, both under New Labour and subsequently with the austerity agenda of the Coalition and Conservative governments, disregarding and obscuring underlying structural issues of poverty and labour market precarity and directing attention instead to correcting ‘poor’ parenting (Jensen, Reference Jensen2012; Lehtonen, Reference Lehtonen2018; Rake, Reference Rake2001). Parenting orders introduced by New Labour, which place the responsibility for children and young people’s truanting, antisocial or criminal behaviour on individual parents, provide an example. Despite gender-neutral terminology, mothers are much more likely to receive parenting orders and be summoned to court. Upholding and naturalising gendered family dynamics, officials permit paternal lack of engagement on the basis of work responsibilities and assume and reinforce the primacy of maternal caregiving among both lone parent and couple families (Holt, Reference Holt2009; Peters, Reference Peters2012). Effectively, the parent held responsible is the mother.

Continued gendering of the parent despite a shift to gender-neutral policy language is also evident in the last couple of decades of benefit policy aimed at families. Both Tax Credits and Universal Credit explicitly assign ‘main’ or ‘lead’ carer status to only one parent (Bennett, Reference Bennett2021; Howard and Bennett, Reference Howard and Bennett2020). For Tax Credits this entailed payment of child-related elements to the main carer while for Universal Credit, the lead parent’s work requirement is reduced according to the age of the youngest child. Revealing the unspoken enduring commitment in benefit policy to (gendered) specialisation, there is no possibility of care-related adjustment to the work-requirements of the parent who is not nominated lead carer status, either among couples or separated parents (regardless of amount of contact). Universal Credit thus hinders equal sharing of responsibility for both earning and parenting among low-income parents, and irrespective of gender-neutral terminology the effect is as gendering as Beveridge’s explicit male breadwinner/female carer model. Despite the rhetoric of non-interference in private matters and families’ right to choose how to organise unpaid work, in practice UK policy intervenes and regulates low-income family life quite willingly, through its commitment to a patriarchal family model.

The lone parent

Social policy and politics have long wavered between treating lone parents as workers or as carers (Lewis, Reference Lewis1998). In the wake of attachment theory’s rising popularity, the post-war social security system treated lone mothers primarily as carers. They were eligible for benefits without job seeking requirements, albeit differentiating between widows allowance and the lower (means-tested) income support for divorced or unmarried mothers (Gray, Reference Gray2001; Lewis, Reference Lewis1998; Meulders-Klein, Reference Meulders-Klein1996; Song, Reference Song1996). In fact, resident lone fathers have historically been less of a policy and activist concern. This silence reflects a perception of them as a less ‘problematic’ people category (more likely to be in paid work thus claiming less social security benefits; Ermisch and Wright, Reference Ermisch and Wright1995). With about nine in ten (resident) lone parents being mothers, the term is often used synonymously with ‘single mother’ (also conflating parenthood status and legal marital status), and the female-headed household is equated with the ‘lone parent family’. By extension, ‘family’ in this context means common residence, and the (non-resident) father, whether he plays an active role or not, is conceptualised as ‘absent’ from the family (although since the 1990s not excused from financial provision). However, the lack of concern about lone fathers also highlights the explicit linking of men with breadwinning, even when they are (sole) caregivers. Lone fathers on benefits were until 1975 required to look for work (Song, Reference Song1996).

Subsequently, the benefit system’s formally neutral position on lone parents’ employment has been eroded since the early 1990s, marking a clear conceptual shift toward viewing lone parents as (potential) workers rather than as primarily carers (Lewis, Reference Lewis1992, Reference Lewis1998), or perhaps more accurately to viewing their families as ‘workless households’ (Knijn et al., Reference Knijn, Martin and Millar2007). With growing political concern about ‘welfare dependency’ during the 1980s, in addition to mandating paternal financial provision, the Child Support Act (1991) also included some part-time work incentives. Further, activation policies have since the turn of the century imposed increasing, and increasingly punitive, job-seeking requirements on lone parents receiving benefits (Millar, Reference Millar2019; Millar and Ridge, Reference Millar and Ridge2018). As these requirements only apply to poor lone parents on benefits (and more recently also low-income coupled mothers on Universal Credit), policy is now effectively positioning (full-time) childrearing and care as a privilege open only to some families with a sufficiently high-earning (likely male and co-resident) breadwinner.

Yet, perhaps counter-intuitively, the male breadwinner family model has also underpinned Britain’s policy approach to lone parents throughout, as the household-level means-test typically results in withdrawal of the lone parents’ benefits on re-partnering. Not only does this assume income-sharing to an extent that may not reflect reality in newly formed relationships where children from previous partnerships are involved (Griffiths, Reference Griffiths2017; Lewis, Reference Lewis1998; Millar, Reference Millar2008), but it again reveals that policy constitutes mothers as first and foremost provided for by a man’s income. Only in the absence of a man (or his earnings) will the state provide benefits and employment support to her. The subjectification effects of these policies for lone mothers, but particularly poorer, more likely to be working class and/or young mothers, and in receipt of benefits, are clear. While held responsible for their children’s care and wellbeing, instead of recognition of their additional financial and support needs they have been increasingly positioned as ‘problematic’ due to their deviance from the ‘norm’ of being attached to a male breadwinner.

The father

Britain’s policy concern with fathers has been strongly and persistently focused on financial provision. Prior to 1973, married fathers had sole legal guardianship of their children, while unmarried fathers had few automatic legal rights or responsibilities (Lewis, Reference Lewis and Hobson2002b; Meulders-Klein, Reference Meulders-Klein1996). Not until the early 2000s did being named on the birth certificate become sufficient for conferring automatic parental responsibility to unmarried fathers (among different-sex parents). However, since the 1989 Children Act, both parents have retained their parental responsibility for their child(ren) following divorce or separation, shortly followed by the Child Support Act firmly asserting that all biological fathers be required to support their children financially, irrespective of past or present living arrangements or marital status. Further, as noted above, through joint assessment for (subsequently re-partnered) lone mothers’ benefit eligibility, the social security system also expects a man to provide for the family he lives with, whether or not children in that household are related to him. In effect, both biological and social fathers are positioned as financial providers. Despite the capacity to pay maintenance being potentially limited for fathers on a low income, especially if also supporting co-resident children, and possibly affecting the affordability of maintaining father-child contact (Lewis, Reference Lewis1998, Reference Lewis and Hobson2002b; Millar, Reference Millar1994), the ministerial response to this critique in the early 1990s unequivocally prioritised financial maintenance over contact (Lewis, Reference Lewis and Hobson2002b). Thus, clearly demonstrating a lack of policy commitment to paternal caregiving, the policy reproduces traditional gender roles of the father as breadwinner and the mother as carer, whether living together or apart (Millar, Reference Millar1994).

From the 1991 Child Support Act to the current benefit system, policy also has an inflexible conceptualisation of the living arrangements of families with separated parents. For the purposes of Child Benefit, child elements of Tax Credits, and Universal Credit, child(ren) can be registered as living with one parent only (irrespective of actual arrangements). Although social housing allocation operates on a local level with possible variations in the treatment of ‘non-resident’ parents as either a single adult or a family unit (Harding and Newnham, Reference Harding and Newnham2015), housing benefit treats a parent whose children are not registered as living with them as a single individual, thus restricting the affordability of housing suitable for shared care and overnight stays. That active parenting and caregiving across households cannot be accommodated within the benefit system again reinforces the nuclear family ideal, with the classing implication of paternal contact and caregiving being positioned as a privilege.

The policy ambivalence on fathers as carers is also evident in the lack of policy on father-involvement in childrearing aimed at fathers residing with their children. Policy and politicians have long considered family leave and pay policies barriers to economic competitiveness of businesses that employ (potential) parents (Browne, Reference Browne2013), rather than valuing and rewarding the care and time parents devote to reproductive labour. Thus, the two weeks of statutory paternity leave introduced for fathers (and second parents in the case of same-sex couples and adoption) in 2003, is paid at a low flat rate. From 2011, Additional Paternity Leave enabled mothers to transfer up to half of the maternity leave entitlement to the father/second parent, conditional on her return to paid work. However, neither Additional Paternity Leave, nor its replacement Shared Parental Leave from 2015, incentivised take-up by fathers. The government impact assessment report on Additional Paternity Leave suggests its introduction was a symbolic gesture rather than a genuine attempt to support more involved fathering. Projecting low take-up at 4-8% of eligible fathers and noting that among those the ‘time taken is unlikely to be greater than 13 weeks but could be considerably less’ (BIS, 2010, p. 4), these assumptions were justified based on evidence from other countries showing low take-up without father-specific entitlement quotas. Despite repeated calls for reform, at the time of writing, findings and recommendations from a 2018 government evaluation of Shared Parental Leave have yet to be published. It is also notable that the leave is only available to couple fathers; another institutionalised limitation of non-resident fathers’ ability to share care.