Advancing Universalism in Neoliberal Times? Basic Income,Workfare and the Politics of Conditionality

Introduction

Neoliberalism is typically understood as promoting individualism and competition, at least amongst workers, thus fostering precarity and inequality. Via processes of liberalization, markets advance while the state contracts. However, critics increasingly question this view. In practice, ‘actually existing neoliberalism’ often involves an extension of the state, in terms of spending, regulatory scope and coercive powers, in ways that enforce market norms (Wacquant, 2012). Workfare is an important example of how an extension of coercive state powers shifts risk from governments and corporations onto workers and households, reinforcing precarity.

Workfare has developed unevenly (Peck, 2001). The most pernicious forms are generally associated with the liberal welfare states of the Anglosphere. The resilience of earlier distinctions between social democratic and liberal forms of capitalism across processes of liberalization suggests potential in pursuing updated social democratic strategies (Thelen, 2014). Alternatively, others have focused on a novel strategy of universal basic income as a specific response to labour market precarity generated by workfare and ‘bullshit’ jobs (Bregman, 2016).

Universalism, fostering social solidarity rather than the differentiated individualism of market competition, is at the core of both social democracy and basic income. Yet, the application of universalism is quite different in the two approaches. Where basic income envisages the same payment to all as a mechanism to decentre paid work, social democracy typically seeks to build social insurance for specific social risks, reinforcing the ongoing importance of paid work. Indeed, welfare state theorists have been sceptical of flat-rate benefit systems, which are susceptible to being residualized.

The diversity of experiences of liberalization is a growing theme of scholarship on neoliberalism (Cahill and Konings, 2017). Comparative welfare state scholars have observed not only a lack of convergence, but the ongoing salience of existing welfare institutions in shaping experiences of liberalization (Wren, 2013). Thus, apparently common processes, such as marketization or financialization, have substantively different consequences. A related geographic literature has identified the importance of variegation within neoliberalism, moving away from ‘property-based essentialism’ (Block, 2019), which emphasizes shifts in ownership structure, to focus on how politics, culture and history shape liberalization.

This paper builds on these literatures to investigate the variegated politics of universalism within neoliberal reform of benefit systems. It takes Australia as a case study, where benefit structures were significantly restructured as the economy was liberalized in ways that prefigured more recent reforms, like Britain’s Universal Credit (Whiteford and Millar, 2019). It compares reforms to family benefits and unemployment benefits. While both payments appear to follow a common template of liberalization, producing flat-rate, means-tested benefits, I distinguish two distinct trajectories. Spending on family payments expanded significantly, remains broadly unconditional (beyond income-testing), and recipients are far less stigmatized, while unemployment benefits have stagnated and become part of a pernicious workfare system.

The paper begins with the literature on neoliberalism, focusing on theories of ‘actually existing neoliberalism’. I use this account to highlight variegation in processes of liberalization within the welfare state. In particular, I highlight examples of hybridity, where reform creates new policy models that seem to combine redistribution and market exchange in new ways. Using this concept of hybridity, I explore the Australian case study. I argue family benefits were reshaped by social movement pressure intersecting with an increasingly pro-competition and technocratic state. This produced a new, hybrid, pattern of means-testing, I call affluence-testing, challenging the dualism between fiscal and social welfare. Alternatively, unemployment benefits succumbed to a divisive ‘wedge politics’. While these differences reflect possibilities for advancing a politics of universalism, they also highlight the importance of institutional legacies, which may limit these strategies in more genuinely social democratic contexts.

Neoliberalism, Workfare and Universalism

Neoliberalism has become a central, albeit contested, concept in sociological accounts of contemporary political economic transformations (Cahill and Konings, 2017). The range of definitions and applications of the term does create conceptual challenges (Venugopal, 2015). Yet, as Jessop argues, there remains utility in identifying the broad shift in the organization of states and markets from the 1980s as the outcome of a political project. This project challenged post-War Keynesianism by reasserting ‘exchange-value’ over ‘use-value’ (2018: 347), through processes like liberalization, deregulation, privatization and marketization.

The outcomes of this project, however, sit uncomfortably with some early understandings of the term. Rather than involving a simple expansion of markets at the expense of states, ‘actually existing neoliberalism’ is spatially and institutionally embedded, often involving an extension of state power to enforce market norms (Wacquant, 2012). And rather than leading to convergence, as some anticipated, important national differences remain. Both observations have led to an institutionalist turn in analysis of neoliberalism (Cahill and Koning, 2017: 11) and an emphasis on neoliberalism (Brenner and Theodore, 2002) or liberalization (Thelen, 2014), as a process, rather than an outcome, of change.

Workfare is an important example of both the extension of state power to enforce market norms and of the variegation of liberalization. Developing alongside the service transition and post-Fordism, it centres ‘activation’ and labour market flexibility (Jessop, 2018). Activation, however, has been embraced in distinctive ways. While Scandinavian social democracies have tended to emphasize active labour market programmes and ‘flexicurity’, more liberal Anglophone countries have emphasized conditionality (Thelen, 2014). A more geographically infused literature has highlighted diversity even within these categories, suggesting a greater degree of variegation across space (Peck and Theodore, 2007).

As Block observes, work on the variegation of capitalism and neoliberalism can be understood as a distinct wave of a common intellectual project to understand ‘obvious and substantial differences in the way that different developed market societies (are) organised’ (2019: 1168). Initially comparative welfare state theory, particularly Gosta Esping-Andersen’s (1990) ‘three worlds of welfare capitalism’, formed the primary framework for this analysis. Within this tradition more recent scholarship has noted how processes of liberalization reproduce and transform existing national differences, rather than producing policy convergence (Wren, 2013). For example, while marketization has been applied across the OECD, the markets that are produced vary in how they are designed and function and in whose interests they advance (Gingrich, 2011). Such variety is not limited to cross-national experience, but emerges across policy domains within countries (Davidson, 2009).

Common across the literature on neoliberalism as a political project is a focus on changes in the distribution of power and risk. Despite the diversity of national experiences, liberalization coincided with rising inequality within nations, declining trade union power and rapidly rising incomes at the very top of the distribution. These changes reflect a shift in risk from states and corporations onto individuals and households (Hacker, 2006). Workfare represents one of the most acute instances of risk shifting, reinforcing other trends in industrial relations and macroeconomic management to produce an increasingly insecure and inadequate supply of paid work. For some, these trends fundamentally change the social structure of capitalist societies, creating a ‘precariat’ of insecure and poorly paid workers at the margins of highly profitable economies (Standing, 2011).

In response to the rise of insecure and unrewarding labour, many have called for a universal basic income. Guy Standing suggests a universal basic income as a key component of his precariat charter (2014). Likewise, Rutger Bregman has argued for a basic income, alongside a shorter working week, as a response to precarious employment, workfare and ‘bullshit jobs’ (2016). The proposal has a Utopian spirit: by helping to decouple subsistence from wage labour, many argue, basic income can reshape fundamental power imbalances, giving workers (Wright, 2011) and women (Pateman, 2004) more voice and reconciling environmental sustainability and human development (Van Parijs, 1992: 29–32).

However, proposals for basic income have enjoyed little practical policy success and remain controversial. Alongside advocates of Utopian egalitarianism, basic income has also received support from key advocates of liberalization. Milton Friedman (1968), for example, was a key proponent of a version of the scheme when it briefly received mainstream political attention in the United States. His model did not provide a uniform payment to all, but sought to achieve a similar goal through a form of integration of tax and welfare systems known as a Negative Income Tax (NIT), a model similar to affluence-testing. De Wispelaere and Noguera (2012) suggest it is precisely the diversity of support for basic income that undercuts its policy appeal, making it difficult to build any viable real-world coalition for reform.

A particular obstacle to reform is the unease with which traditional social democratic actors, trade unions and social democratic parties, view basic income (Vanderborght, 2006). This reticence echoes scepticism within welfare state scholarship, which highlights competing conceptions of ‘universalism’. Universalism was central to Esping-Andersen’s account of the emergence and success of social democratic welfare regimes. Social democracies fostered solidarity through systems of universal social insurance, to which everyone contributed and from which everyone received benefits. He contrasted universalism to targeting, where welfare resources were focused on those most in need. Counter-intuitively, by targeting welfare spending to the poor, the state achieved less redistribution overall because few voters supported welfare spending (Korpi and Palme, 1998).

Social insurance models were designed to deal with specific social risks, typically those faced by industrial workers, such as unemployment, injury and old age. To be effective, insurance systems needed to maintain a workers’ income, by providing benefits roughly equivalent to their wage. Thus, insurance benefits are only paid in specific circumstances and are linked to previous labour market status. In contrast, basic income is paid to all at a flat-rate likely to be significantly lower than average wages.

In practice, flat-rate payments, like those envisaged through basic income, are associated with the least generous liberal welfare regimes. Esping-Andersen argued these benefits reinforce the primacy of the market in allocating incomes and life-chances, residualizing welfare as a safety net (1990: 26). He rejected the ‘universalism’ of flat-rate benefit schemes because in practice such schemes encouraged affluent groups to create their own systems of social insurance, either through entitlements restricted to high-status workers (occupational welfare) or through tax benefits for private welfare spending (fiscal welfare). Rather than building solidarity, this entrenched ‘dualism’ between social welfare systems used by the poor and other forms of welfare accessed by the well-off.

Yet, the design and impact of flat-rate benefit schemes differ significantly. While aiding commodification (Lavinas, 2015), it is difficult to equate the anti-poverty programmes of Latin America to highly punitive workfare. Even within countries variation occurs. In New Zealand flat-rate benefits for working age populations have explicit work requirements, while flat-rate benefits for older people remain genuinely universal. Just as marketization of services and financialization of debt have been common features of welfare state restructuring internationally, but can in practice reflect different politics and interests, it is likely that changes in benefit structures also follow diverse trajectories.

Examining differentiation within processes of neoliberalization often highlights hybridity (Peck, 2012). Policies explicitly framed as extending market competition regularly expand bureaucracies and state regulation. For example, many countries combined the reintroduction of student fees in higher education with publicly provided income-contingent loans, simultaneously extending commodification while potentially socializing risk. Such hybridity produces new axis of contestation over the extent of commodification, with national loan schemes facilitating access and distributing costs differently (Bryant and Spies-Butcher, 2018). As Jamie Peck (2012: 121) argues, ‘significant opportunities beckon, if markets – in all their hybrid, variegated, and heterogeneous forms – are seriously problematized as objects of study’.

Understanding the differences in how benefit systems have been restructured, and how the politics of universalism and dualism have been mobilized, promises to help us identify strategies for change in what can otherwise appear an impervious model of technocratic neoliberal governance. Identifying variegation and hybridity involves exploring how similar reforms develop differently. To do this, the paper takes a case study approach, focusing on Australia, where flat-rate benefits have historically been associated with greater universalism and inclusion (Castles, 1985) and where neoliberal restructuring has produced very different models. In particular it focuses on reform of family benefits, which have increased significantly and remain largely free of workfare-like conditionality, and unemployment benefits, which have been restructured into a particularly punitive workfare regime (Mcdonald and Marston, 2005).

Flat-rate Payments in Antipodean Wage Earner Universalism

Australia and its Antipodean neighbour New Zealand are most commonly classified as liberal welfare states. This categorization not only reflects Esping-Andersen’s initial assessment, but also more recent analysis of neoliberal restructuring (Jessop, 2018: 4). However, at least prior to the neoliberal period the Antipodes were the subject of academic dissention. Francis Castles argued the countries represented an alternative model of welfare organization where strong forms of migration, labour and capital market regulation substituted for some of the features of more expansive social spending and taxation (Castles, 1994). Arbitration became a central feature of 20th-century politics that marked the region out from other British settler colonies (Fenna, 2012).

Arbitration complicates any reading of Antipodean welfare as it clearly applied the social principles of welfare to employment conditions. The courts established a social benchmark for wage setting, which entrenched both a notion of social adequacy and a normative definition of the nuclear family through a ‘family wage’, i.e. a wage sufficient to support a husband, wife and two children (Nolan, 2003). From mid-century Australia and New Zealand collectively adopted successful commitments to full employment (Nolan, 2003: 74), universalizing the wage earner condition for able-bodied white men. Over time arbitration was used to advance new social principles, which might ordinarily be thought the domain of social legislation, including paid care (Whitehouse, 2004). Given the near universal model of arbitration, occupational welfare effectively followed the principles of social insurance to protect against social risks.

Flat-rate benefits funded from general revenue emerged as the other most distinctive feature of the Antipodean model (Fenna, 2012). Starting with aged pensions strongly targeted to need, which initially entailed other forms of conditionality, over time payments became less conditional. The system of secure, well-paid work acted to reinforce the payments model. For working age men, the unemployment benefit was usually a short-term measure, and while technically tied to job search requirements, in practice, conditionality was barely enforced. It thus acted in a similar way to a guaranteed minimum income.

The highly standardized employment model made the payment effectively categorical, rather than means-tested; all unemployed men received it, and once employed the payment was replaced with a higher full-time wage (Castles, 1985). Arbitration was later extended to support a system of near universal home ownership, which acted in concert with flat-rate pensions to protect workers’ living standards in retirement (Castles, 1998). Castles argued arbitration created a ‘wage earners’ welfare state’, distinct from liberal residualism. However, having been built around white male bread-winners, it was a distinctly ‘weak’ understanding of universalism (Bryson, 1992). Alongside arbitration, social policy was also supported by extensive tax concessions for healthcare, family support and pensions, many of which expanded with neoliberalism to produce a ‘dual welfare state’ of targeted social welfare for the poor and opaque fiscal welfare directed to the well-off (Stebbing and Spies-Butcher, 2010).

Taken together, these features made the wage earner model one of the most vulnerable to the process of economic restructuring (see Fenna, 2012). Strong regulation of trade, finance and migration was harder to maintain in the face of the global shift away from Keynesianism, especially for a small, culturally isolated and capital importing economy. Racialized migration and the strong gendering of family and work roles were likewise coming under attack from new social movements. Briefly, Australia flirted with a genuine social democratic turn, but by the 1980s Labor governments instead oversaw a radical process of market restructuring.

Unlike the United Kingdom and the United States, the Antipodes were distinctive in experiencing the most intensive period of neoliberal reform under Labor governments. Australia’s unions and Labor Party helped ‘build’ neoliberalism (Humphrys and Cahill, 2017), supported by a bureaucracy increasingly dominated by committed market advocates (Pusey, 1991). Thus, neoliberal policies were enacted by a government that explicitly rejected the anti-statist ideology of the New Right (Cahill, 2014: 44–52), but wholeheartedly embraced the need for greater competition and fiscal constraint (Brenton, 2016). These forces, I argue, created incentives for movements to work within a pro-competition state, fostering technocratic political contestation in some areas, while giving way to more overtly punitive reforms in others.

Renovating Flat-rate Universalism

A more inclusive model of flat-rate benefit provision for families with children, known as affluence-testing, was maintained and even expanded alongside liberalization, free of the work conditions placed on similar schemes in other liberal states. Expansion was driven by the women’s movement, which like its counterparts internationally, enjoyed a surge of energy in the 1970s and 1980s. However, to a much greater extent than elsewhere, the Australian women’s movement pursued a state-orientated strategy focused on achieving policy goals through the bureaucracy. The ‘femocrat’ strategy, as it became known, paralleled the broader Accord process (1983–1996), which encouraged movements to engage with the Labor Government. Simultaneously, the Australian state experienced a neoliberal ascendency, as neoliberal bureaucrats, called ‘economic rationalists’ in Australia, took control of policy via the central economic agencies (Pusey, 1991).

The term ‘economic rationalism’ reflects Australia’s more technocratic and less overtly anti-statist reform project and created the conditions for an unusual interplay of what Marian Sawer calls ‘femocrats’ and ‘econorats’ (1996). Affluence-testing, I argue, can be understood as the result of the interplay of these political forces and existing policy architecture. In reflecting both the normative demand to address social needs and the economic imperatives of competition, the policy integrated policy-logics usually identified with different components of Australia’s ‘dual welfare state’ producing a form of policy hybridity. Perhaps unsurprisingly, the femocrat strategy also proved more vulnerable to co-option, becoming dependent on state support and then receding in the face of (conservative) government cuts from the late 1990s (Sawer, 2007). Economic rationalists maintained their institutional dominance, yet the underlying policy architecture of ‘affluence-testing’ has proven broadly resilient.

A significant literature explores the role of feminist policy makers in reforming social spending away from the breadwinner model (see Cass and Brennan, 2003; O’Connor et al., 1999; Sawer, 1996). Australia led bureaucratic developments in women’s policy as femocrats maintained a more central position and more direct executive access than elsewhere (Sawer, 1996), fostering a particular kind of technocratic reform. While the women’s movement was diverse, one important strategy involved feminist activists taking on roles in newly formed sections of the bureaucracy explicitly aimed at advancing gender equality. These activists, dubbed ‘femocrats’, retained links to the movement outside the state, while also pursuing policy goals within it. The strategy was always controversial amongst activists and partly reflected the inability of the movement to pursue more independent and aggressive tactics (Sawer, 1993: 15). However, it did facilitate a genuine connection between movement action and state mechanisms.

The combination of movement pressure and fiscal austerity saw family payments expanded in novel ways. From the 1980s an array of individual payments consolidated around two models I describe as affluence-tested and means-tested. This involved three distinct steps. First, feminists established gender equality on the political agenda and successfully persuaded the Labor Government that reform would aid its electoral appeal. Unsurprisingly, for most of the 20th-century Labor received greater support from (breadwinner) men than from women. That began to shift internationally from the 1980s, as women moved to the political left. Sawer argues the women’s movement became adroit at exploiting the previous ‘gender gap’ to win policy concessions (Sawer, 1996: 10). Funding for childcare, family payments and single parenting payments were tied to this political claim making, which culminated in Labor committing to end child poverty. While later viewed as hyperbolic, over the next 15 years Australia’s child poverty rate fell faster than in any other OECD country (Whiteford et al., 2011: 91).

Second, femocrats mobilized conventional economic arguments to their cause. Consistent with economists’ preference to flatten and broaden the tax base, they argued new social spending should be financed by closing tax concessions and redirecting revenue (Cass and Brennan, 2003; O’Connor et al., 1999). Wage earner welfare included income tax deductions for those with dependent children and income earners with a dependent spouse. These social tax expenditures (STEs) provided much larger gains to high-income households and created significant work disincentives for secondary (primarily women) earners. Policy reform effectively converted these concessions into direct spending, and in the process radically redesigned government support. Once accounted for as spending, it became untenable to maintain the highly inegalitarian distributional structure of STEs. The change also shifted distribution from the primary earner through taxation to the primary carer through payments. Ingalill Montanari (2000) estimates that the substantial increase in family spending achieved by 1990 only marginally exceeded the previous combination of spending and tax expenditures provided in 1975.

Finally, economic rationalists reconfigured these resources within social spending to more strongly target low-income households by limiting access from higher income households. Layering of various targeted benefits over time had left a complex system of payments. As payments were consolidated through the reform period economic rationalists successfully means-tested the universal element of family benefits to fund feminist demands to increase rates and expand access to more tightly means-tested payments. These changes effectively achieved the primary aim of expanding spending for low- and middle-income families within the constraints of fiscal austerity (see Whiteford et al., 2011: 86–88). While targeted, 90 per cent of families received the previously universal benefits (Cass, 1990). This distinguished the approach from traditional means-testing, which narrowly restricts benefits to the poor. Instead, ‘affluence-testing’ ensures the majority of people in a group are eligible, and instead restricts access to the affluent. Recent changes have tightened means-testing, but largely as a mechanism to expand funding to others (Wilson et al., 2013).

The conversion of STEs and affluence-testing of universal benefits followed a common logic, consistent with economic rationalist analysis, but also supportive of feminist aims, integrating social welfare benefits with the tax system. Given the government’s strong commitment to fiscal restraint, both tactics framed policy change as restructuring rather than increasing government distribution. Feminists equated tax concessions to social spending. Alternatively, economic rationalist attempts to limit universality effectively equated means-tests to income taxation, an equivalence captured and more broadly advanced through the concept of an ‘effective marginal tax rate’ (EMTR). ¹ Both reflect a rational choice understanding of action as the outcome of utility maximization. Actors focus on net gains or losses, rather than the mechanisms that produce those results. The same logic underpins Milton Friedman’s (1968) proposals for a NIT version of basic income.

Distinguishing between the mechanisms for delivering social benefits continues to matter for other reasons. It is unsurprising that feminists drove the first type of change, from STEs to spending, and economic rationalists the second. Highlighting the role of STEs was central to challenging a breadwinner model that treated the resources gained through markets as privileged compared to those gained through states, and thus privileged earners over carers. By making STEs visible feminists subjected them to politics, rendering their distributional consequences unsustainable and shifting power within the household (see Cass and Brennan, 2003). Alternatively, equating means-tests to tax rates reflected the fiscal politics of the state. If the same outcome had been achieved by taxing back the family benefits of higher income households then this would appear as an increase in taxation, expanding government. Constructed as a means-test on the same households it appears as a decrease in spending, shrinking government. Of course, they are not identical. A tax rate would continue to apply up the entire distribution, meaning very high-income households pay more in tax than they receive in payments. Means-testing, even affluence-testing, cuts out once the payment is exhausted, reducing the redistributive effect and changing incentives so the highest EMTRs apply to middle-income earners (see Harding and Polette, 1995).

Ending or Transforming Dualism?

Changes in family payments effectively broke down previous forms of welfare dualism, by ending fiscal welfare and ensuring social welfare was extended to most families. While children have often been considered more ‘deserving’ than the adult unemployed, Australia’s more inclusive payment structure is not necessarily reflected elsewhere. In many liberal countries extending family support has come via tax credits. As credits in the tax system, the benefit can only be accessed by those paying income tax, and thus effectively requires paid work. Some systems also establish explicit minimum work requirements (Meyer, 2010; St John and Dale, 2012).

In Australia, significant tax reform in 2000 converted family payments into tax rebates. ² This potentially shifted support from carers to earners and created new disincentives for many women workers (Cass and Brennan, 2003). However, it did not re-attach work requirements and those on low incomes continued to receive the entitlement as a cash benefit. Over time, accessing entitlements via the tax system proved difficult and unpopular. Given the choice of receiving a fortnightly payment or an annual tax refund, almost all now opt for the payment (Department of Social Security, 2016). The attempt to re-centre the tax system was also notable in its failure to re-establish the distributional structure of STEs. Even as a tax refund, entitlements were effectively flat-rate and governed by means-tests. Indeed, family support was explicitly used to compensate low- and middle-income households for regressive increases in consumption taxation (Wilson et al., 2013: 633). The distributional changes that accompanied the conversion of STEs into spending, it appears, are asymmetric and difficult to reverse.

Other changes to welfare payments entrenched new dualisms. Across the 1980s and 1990s the payment system was consolidated, reflecting both the principles of efficiency and gender neutrality. A number of explicitly gendered payments, like widows’ and wives’ pensions, were phased out, and payments associated with care were redefined in gender neutral terms (Mitchell, 1998). The consolidation produced two broad categories of payments, pensions and benefits. Pensions typically went to social groups deemed to have a specific need, which also reduced expectations of paid work. This included those with disabilities, single parents and the aged. Under the wage earner model there was no expectation these groups would work. However, increasingly many of those within these categories wanted access to work, which sat comfortably alongside economic imperatives to increase labour force participation. Benefits typically went to those expected to work under the older wage earner model, particularly the unemployed.

The payments differed in structure, conditionality and level, reflecting different political and social expectations. The structure of all payments began to respond to changes in the labour market and growing awareness of incentives. Initially virtually all of these payments (or their predecessors) applied strict means-tests, in some cases cutting payments entirely for those earning any market income. Means-tests were relaxed across the board, but much more so for pensions than benefits. This reflected differential norms. Those on pensions were only expected to look for part-time work, complicating the implicit distinction between those reliant on market versus state income. EMTRs were potentially much more important for part-time workers, while those on benefits were expected to find full-time work.

The focus on EMTRs, as in family benefits, reflects a growing acceptance of the need to combine market and non-market income, eroding traditional welfare dualism. Pension payments were also higher, reflecting an acceptance that, for many, pensions would form a long-term income, while benefits continued to be set at rates that were only effective for short periods of frictional unemployment. Similarly, pensions had weaker forms of conditionality. Even so, the targeted nature of the transfer system continues to create problems of high EMTRs for a range of low- and middle-income earners (Harding et al., 2009). Pensions followed a model closer to affluence-tested family benefits (see Whiteford and Millar, 2019). In both cases the majority of the potential recipient population receive some benefit, which is somewhat responsive through means-testing to marginal losses of income, reflecting risks of insecure and under-employment. Alternatively, support for the unemployed languished, largely unreformed, becoming ‘means-tested’ and residual despite the radical unwinding of the arbitration and full employment foundations of wage earner welfare. Similarly, in New Zealand the aged pension was entrenched as a genuinely universal payment while virtually every other payment became more conditional.

Universalism (and Dualism) in the Workfare State

Where affluence-testing fostered solidarity, mitigating political attacks, conditionality expanded for those on means-tested payments. Just as affluence-testing advanced by blurring the boundaries between tax concessions, payments and benefit withdrawal rates, so conditionality advanced by shifting the boundary between benefit-like payments and pension-like payments. As the two payment systems reflect different institutional logics, having the two systems operate in parallel facilitates forms of dualism. The boundaries between the universal norms of affluence-testing and the application of workfare to means-testing have shifted, bringing parents of younger children, those with less severe disabilities (Mendes, 2009: 105–106) and the ‘young’ old into the benefit logic of workfare.

This erosion of unconditional income support is reinforced by the advance of conditionality for those deemed least deserving. Wilson and Turnbull argue this reflects a logic of ‘wedge politics’, which is a ‘calculated political tactic’ that focuses on divisive social issues, targeting social groups that ‘attract resentment or antipathy in the wider electorate’ to gain advantage (2001: 386). Welfare conditionality is initially targeted towards visible and marginalized sub-populations, later expanding to broader populations of the poor. Wilson and Turnbull initially noted the tactic applying to waiting periods for benefit applicants, which first applied only to recent migrants (Wilson and Turnbull, 2001: 396). However, the same logic has since been used to introduce ‘work for the dole’ (first to the young unemployed) ³ and quarantining of benefit payments (first to Indigenous recipients). ⁴

Wedge politics also appears asymmetric, although in the opposite political direction to affluence-testing. Once new forms of conditionality are established, conditionality is often extended but rarely unwound. In the case of income quarantining the initial introduction was so overtly racist that it required the suspension of the Race Discrimination Act, a remarkable development in a notionally liberal democracy (Altman and Hinkson, 2007). Yet, when an incoming Labor Government, opposed to the Act’s suspension, was confronted with this situation it opted to extend quarantining to non-Indigenous communities rather than to wind back conditionality. The logic of conditionality and the need to prove deservingness are thus advanced in two related processes, by creating new conditionality on the most marginalized and extending existing conditionality to new social groups. Both processes rely on social differentiation to erode universalism.

Alongside the rise of workfare, however, has been an equally notable defence of the most universal benefits. Both sides of politics have expanded family benefits and the aged pension – the two most clearly affluence-tested payments. Labor has prioritized lifting base rates, while the Coalition has focused on loosening means-testing, but both have seen additional, egalitarian spending on these payment schemes. Alternatively, both have neglected the means-tested unemployment benefit, Newstart, despite a stakeholder consensus to increase it. Unlike the wedge politics of workfare, the inclusive design of affluence-testing creates a larger supportive constituency and political incentives for expansion.

These experiences reinforce Esping-Andersen’s general observation, and paradox of redistribution thesis, that universalism generates more fiscal resources for equality than targeting. Yet, the findings suggest universalism might be possible with flat-rate benefits. Australian affluence-testing appears to follow principles that not only resemble a basic income, but also what Theda Skocpol (1991) described as ‘targeting within universalism’, where policy design follows a universal principle, but allocates greater resources in practice to those in greatest need. Van Lancker and Van Mechelen have since found such a principle might promote equality even more strongly than traditional social insurance (2015).

Affluence-testing, Universalism and Basic Income

Australia’s experience of benefit reform reflects predictable processes of liberalization. Payment systems were ‘simplified’ and ‘streamlined’. Withdrawal rates were adjusted to better incentivize work. And in all cases benefits remained or became subject to some form of income-testing, or ‘targeting’. Yet the social outcomes of these reforms, in terms of precarity and social stigma, and the ongoing political dynamics, vary significantly. Family benefits remain politically popular, are received by most families with children and are subject to little if any stigma. Unemployment benefits are stingy, conditional and demoralizing.

Welfare state scholarship has long been sceptical of flat-rate benefit systems. Analysis of neoliberal restructuring seems to confirm this suspicion, as liberal welfare states appear to have experienced more extreme processes of liberalization, as evidenced by Australian workfare. Yet, Australia’s experience of benefit reform suggests greater variegation, even within the liberal world. These differences challenge conventional forms of welfare analysis, as different reform trajectories involve apparently similar technical reforms. Just as neoliberal restructuring can advance commodification by extending state power, so hybridity potentially mitigates commodification in otherwise marketized reforms.

The example of family benefits suggests an alternative political strategy, which mobilizes a form of universalism to expand social protection along more egalitarian lines. Crucially, that strategy involved traditional forms of movement politics, through organizing and mobilizing women as a political constituency and through making normative claims about the need to end poverty and gender-based discrimination. This political pressure then intersected with the neoliberal politics of market competition and fiscal austerity. The result was a process of conversion, which integrated and reorganized fiscal welfare into the system of social welfare. The result was a policy hybrid, affluence-testing, which reflected the welfare system’s focus on targeting and the tax system’s focus on incentives.

Interestingly, Australia’s experience seems to prefigure models now emerging in the basic income literature. Recent rational choice analysis concludes the most likely outcome of efforts to implement a universal basic income in representative democracies is something very similar to affluence-testing – a payment available to most, but not the rich (Cremer and Roeder, 2015). Likewise, the focus on EMTRs parallels discussion of ‘net’ distributive, incentive and fiscal effects in the basic income literature (see Martinelli, 2019: 4).

The success of affluence-testing partly reflects changes in the normative foundations of solidarity. As the gendered nature of the breadwinner model came under attack, so too, to some extent, did the highly gendered model of standardized full-time employment central to the initial wage earner compromise. A new universalism emerged that sought a degree of flexibility in the form of paid work, focusing attention on the logic of incentives through EMTRs. The expansion of part-time work was a conscious strategy of femocrats, justified in similar economic terms. Importantly, it positioned benefit provision as supporting paid work, not decoupling income from work. However, changes to the labour market not only created diversity, but also radical casualization and underemployment (see Wilson and Ebert, 2013), while the model of childcare initially developed by femocrats along social democratic lines has been significantly eroded by marketization (Brennan, 2007). To the extent affluence-testing is a success, it must be tempered by these significant failures, highlighting the limitations of pursuing income support divorced from other attempts to help households mitigate risk.

Extending affluence-testing to cover all residents potentially offers an avenue for defusing the politics of workfare (Spies-Butcher and Henderson, 2019). Yet, the same institutionalist analysis that highlights the potential for more inclusive forms of benefit reform also suggests limitations. The basic income-like structure of affluence-testing was built on the foundations of Australia’s wage earner model. It is potentially less appropriate, and less desirable in other contexts. Understandably, the traditional allies of welfare state development, trade unions and social democratic parties, have been slow to advance calls for a basic income (Klitgaard, 2007: 187; Vanderborght, 2006). The Ghent system linking income-based social insurance to union membership is hard to reconcile with flat-rate benefits (Scruggs, 2002), and its erosion would be a serious challenge to union density and, ultimately, equity.

Finland’s recent basic income trial potentially reinforces the broader institutional point. Finland embraced economic reform and competition policy more actively than other Scandinavian countries (Kantola and Kananen, 2013). While partly foreshadowing a new universalism, the Finnish trial also reflects the erosion of existing universalism. Halmetoja et al. outline how Finnish welfare reform gradually shifted away from the insurance model common elsewhere in the region to one more compatible with flat-rate payments. Payments were individualized, paid through the same mechanism as social insurance and funded out of general revenue (2019: 324–325). The shift, they argue, reflects the rise of unemployment and casualization following the economic downturn of the 1990s. The trial was championed by the Centre Party and framed around poverty reduction and work incentives. It was limited to the long-term unemployed and efforts to extend the trial beyond two years were rejected, suggesting the vulnerability of narrow experimental models and basic income trials (see Widerquist, 2018).

Rather than framing affluence-testing, or basic income, as a substitute for traditional social democracy, the Australian experience highlights the potential for contestation within liberalization. As Jessop (2018) has suggested, the transformation of the welfare state from its post-War Keynesian form can take different routes, emphasizing Schumpeterian entrepreneurialism or Riccardo’s race to the bottom. Affluence-testing suggests efforts towards universalism remain possible across worlds. By interrogating the diversity of liberalization processes we may identify forms of hybridity that can reduce precarity. However, this depends on whether reforms are driven by social demands and political mobilization, as they were around child poverty through the femocrat experience, or whether they act to extend contestability via price, as Adkins and Ylostalo (2018) fear the Finnish trials suggest. It is useful to remember that the Keynesian institutions of the post-War order themselves took time to develop, did so alongside new technocratic expertise in policy making and statecraft, and always represented a compromise between the needs of capital and the growing power of labour movements. Perhaps basic income, in some form, may play a similar role for advancing universalism in new economic circumstances.