It is tempting to think, as many do, that people are either in employment or ‘on welfare’. That is too simplistic. Those who are in employment are paid for their labour time, but they, like everyone else, also benefit from much of the spending undertaken by the state. Many workers, despite their status as workers, also benefit specifically from the welfare state. On the other side, many welfare recipients perform work and work-like activities in order to receive their government payments. The categories of work and welfare are intertwined. More of one does not necessarily mean less of the other.

A large part of the complication stems from the profound economic and demographic changes that have unfolded over the last few decades (Armingeon and Bonoli 2006; Bonoli and Natali 2013; Hausermann 2010; Jordan and Drakeford 2012; OECD 2018; Pierson 2006; Sarfati and Bonoli 2002). Consider the now long-term absence of full employment and the growth of underemployment, due in part to ‘de-industrialisation’ and the associated decline of the trade union movement. Associated with that there has been a proliferation of the ‘working poor’ in many countries. There have been increases in vulnerable and ‘atypical’ employment, including the rise of ‘gig work’ and the broader phenomenon of casualisation. On the positive side there is an ongoing increase in women’s labour market participation. Yet changing fertility rates in what are ageing societies offer up challenges to policymakers, including questions of migration amidst calls to address ‘skills gaps’. There has been increasing need for workers, employers and social security systems to address work–family life conflicts. Family composition has been changing, with increases in sole-parenthood and one-person households. There is also a growing need to recognise care work throughout the life course, particularly as life expectancy increases.

Many of these phenomena interrelate in complex ways, but they all challenge researchers to consider the meaning of ‘work’; how ‘work’ can differ from ‘employment’; what welfare actually means; who receives welfare; at what stages of life work is more or less important; and which categories of people—whether workers or not—are more reliant upon the welfare state.

Consider also that in many countries a growing proportion of welfare state beneficiaries have been subjected to what is called ‘conditionality’, as part of the ‘welfare-to-work’ policy agenda (e.g. Considine et al. 2015; Dwyer 2019; Watts and Fitzpatrick 2018; Wright 2016). As part of conditionality, those who receive government payments need to regularly meet strict eligibility criteria and often perform actual work or activities that resemble work in order to receive monetary assistance. The difference between officially unemployed people and wage and salary earners in terms of the expectation to be ‘productive’ is shrinking. The unemployed are compelled to be more demonstrably active in job-searching in ways that mirror employees’ need to demonstrate enhanced conventional efficiencies in their workplace. Of the continuing differences between employees and those not in paid employment, however, the most striking is that the unemployed face a social stigma attached to their perceived welfare dependency.

Finally, technological change also complicates attempts to classify people as either workers or beneficiaries of the welfare state. The growth of artificial intelligence and robotics is raising questions on how the nature of work is changing and will change as the labour market and the workforce evolve (e.g. Baird et al. 2018; OECD 2019). Who will do what work? What are the gender implications and the ramifications for other disadvantaged members of the labour market? What will be the roles of the welfare state for people who are continually in and out of paid work? Some analysts argue that people will do different work from what they do today, and in roles that often cannot be predicted or named now. Some have argued that in a world of artificial intelligence and robots, a state-guaranteed minimum income, or what is often called a ‘basic income’, will be inevitable as labour market opportunities shift (Downes and Lansley 2018). Though reliable prediction or even projection is extremely difficult, it is likely that work and welfare will increasingly need to be considered by scholars in the one analytical frame, rather than as separate spheres to be analysed by different disciplinary communities of scholars.

There are concepts which sit in and between different disciplines that help us. An important one in this book is ‘social protection’.

Social Protection

In the everyday life of people in capitalist societies, employment and welfare lie at the centre of what Karl Polanyi (1944) called social protection. Social protection represents the totality of public policies and institutions designed to shield individuals from social harm. In his Great Transformation, Polanyi (1944) contended that the institutions of social protection were a prerequisite to market society’s continuous self-recreation. Capitalism, he argued, was governed by a ‘double-movement’, and the interaction of ‘two organizing principles’:

The one was the principle of economicliberalism, aiming at the establishment of a self-regulating market, relying on the support of the trading classes, and using largely laissez-faire and free trade as its methods; the other was the principle of social protection aiming at the conservation of man and nature as well as productive organization, relying on the varying support of those most immediately affected by the deleterious action of the market – primarily, but not exclusively, the working and the landed classes – and using protective legislation, restrictive associations, and other instruments of intervention as its methods. (Polanyi 1944: 132)

Social protection defends workers through minimum employment standards, and it defends welfare state beneficiaries through state-provided income assistance and services. In order to clarify, however, Polanyi (1944: 72) expanded beyond these two institutional arenas. This was necessary because he was writing during World War II, and so before the construction of the post-War welfare state. He argued that land, labour and money were all bought and sold in markets as if they were commodities, but in fact their status as commodities was ‘fictitious’. They were not constructed as products for sale by the forces of nature. Labour was not produced, being merely a name given to a human activity. Marx (1867) had argued similarly and much earlier in relation to ‘labour power’, which was treated as a commodity under capitalism, by capitalists. Polanyi also argued that land was not produced, though it was a factor of production. Similarly, money was merely a token of purchasing power, having come into being as an outcome of monetary institutions. In short, the self-regulating market ‘commodified’ labour, land and money.

Yet, consistent with the double-movement, this commodification process had to have its limitations. Polanyi recognised that the market could not be let loose entirely and indefinitely, because this would be injurious to society as a whole, not only to the working classes, but to the capitalist classes who whole-heartedly supported and relied upon market recreation and maintenance. Unrestrained commodification would have disastrous market consequences, including the exploitation of labour to the point where it would be unproductive. Labour, after all, is only as efficient as the individual who performs it is able to be. As a result of extreme social dislocation, humans would die from poverty, vice, perversion and crime. Resulting from environmentally irresponsible production, land would be robbed of its beauty and its economic usefulness. Due to extreme money market volatility, purchasing power would be subject to radical shifts between shortfall and oversupply. In short, ‘no society could stand the effects of such a system of crude fictions even for the shortest stretch of time unless its human and natural substance as well as its business organization was protected against the ravages of this satanic mill’ (Polanyi 1944: 73).

The contents of social protection are thus crucial to the ‘de-commodification’ of humans, whether worker or not. This includes protections in the labour market in the form of minimum labour standards. Minima typically apply to standards on wages, workplace and occupational health and safety, hours of work, employment protection, and paid and unpaid leave. Social protection also calls for the consideration of policies which constitute attempts to equitably redistribute labour market rewards, ordinarily established by the operation of trade unions. Social protection within social policy includes all of the ‘de-commodifying’ aspects of welfare states. This includes income poverty alleviation and income maintenance measures, which form the basis of traditional social security systems; welfare services in health, housing, education and employment; and human and community services in the public, semi-public and private spheres.

Social protection’s two main branches are employment relations and social policy.

Employment Relations

It is important to analyse protective institutions in the labour market, but it is equally important to understand how those institutions are formed. The processes which determine labour standards form the core of the field of employment relations are treated in this book as interchangeable with the older and more established concept of ‘industrial relations’ (Bray et al. 2018). The term industrial relations is believed to have been used for the first time in 1885 by British parliamentarian Lord Thomas Brassey, in a speech delivered to the Co-operative Movement’s Industrial Remuneration Conference at Oldham (Morris 1987: 532). In the United States it was first used in 1912, when (then) President Taft established the Commission on Industrial Relations (Kaufman 1993: 3, 200), though another source cites first usage in the context of an industrial strike in 1894 (Spates 1944: 6).

Several perspectives are helpful in understanding the term. Beatrice and Sidney Webb, who are widely seen as founders of the academic field of industrial relations (Kaufman 2014), contributed the pioneering research on the interaction of collective bargaining and trade unionism, and on how these contribute to the determination of minimum labour standards. In social protection terms their perspective was best laid out in the highly influential volumes, Industrial Democracy (Webb and Webb 1897) andHistory of Trade Unionism (Webb and Webb 1894). The former work was the most influential, defining the field in terms of how workers encroach upon the free will of capitalist employers through legally defensible, state-guaranteed and irreducible minimum and above-minimum standards in wages and other working conditions.

Before Industrial Democracy, Marx (1849: 17) had focused on the importance of the broader political economy that governs employment relations, which he referred to as the ‘economic relations which constitute the material foundation of the present class struggles’. For Marx, the relations that underpinned production shaped politics and labour market institutions. As the field of industrial relations developed, the reverse perspective became dominant. That is, institutions would shape the relations, and thus would determine employment conditions. Definitions of industrial relations have varied, some focusing squarely on the dynamics of the ‘employment relationship’ (Keenoy 1985; Keenoy and Kelly 1996). Older conceptions had emphasised the formation of ‘rules’ in the employment relationship, and ‘job regulation’ (Flanders 1965). Dunlop (1958) took the ‘systems’ that sit above the employment relationship as the core; where systems were made up of the state, employers, employees, and the institutional representatives of each. Still other traditional focal points included the importance, and the orientations and ‘types’ of trade unions (Hoxie 1924), industrial conflict, and unions’ long-term struggles and strategies (Hyman 1975). With working-class struggle as the underpinning, Hyman (1975: 12) defined industrial relations in neo-Marxist terms. For him it reflected ‘the processes of control over work relations; and among these processes, those involving collective worker organisation and action are of particular concern’.

In recent decades, the field has concentrated more on the workplace itself, a focus which is seen as necessary in the context of the decentralisation of employment relations, the increased importance of enterprise- and workplace-level negotiations, and the individualisation of relations between employee and employer (Howell 2019). It is in that context that the term ‘employment relations’ has gained greater usage (Giannikas 2004). This book uses employment relations rather than industrial relations though, as stated earlier, the two are used interchangeably and industrial relations are referred to where more historically appropriate. Using employment relations avoids the unhelpful connotation that the subject is overly focused on industrial conflict, or on collectivist social relations to the exclusion of more meso- and micro-level dealings between employees and employers (Bray et al. 2018).

At the same time, the book borrows in its various sections from the other ‘industrial relations’ conceptions already discussed. Thus, employment relations are taken to refer generically to the institutional and political factors, and policies, which shape wages and other conditions of work. This borrows from the Webbs (1897), who defined the concept in terms of how workers encroach upon the free will of employers through legally enforceable minimum standards. It borrows also from Hyman’s (1975) conception that employment relations involves continuous struggle for control over the terms of the employment relationship. It is a study in social, economic and political dynamics, the processes being inextricably tied up in broader institutional arrangements. Borrowing also from Dunlop (1958), employment relations is referred to in some places as representing a ‘system’. Based on Flanders (1965), employment relations are also about employment regulation and the formulation of laws and formal and informal ‘rules’. The book also demonstrates that some rules and laws are protective of the employee, though equally they can be anti-worker, or they may be intended purely to maximise efficiencies and thus based on employer preferences (Kochan et al. 1986). The primary means to determine working conditions is the process of industrial and political negotiation and contest between the primary protagonists—employees and their representatives, employers or their associations, and various organs of the state.

Despite the fact that the employment relations concept adopted here portrays the domain as vitally linked to social protection, the subject of employment relations by itself has relatively little to say about social protection beyond the labour market. For that, an understanding of social policy is required.

Social Policy

Contributions to the understanding of how work and employment relate to welfare have come more from the field of social policy than from employment relations. There are three main, somewhat overlapping, historical stages in intellectual development. All of them are relevant to the remaining chapters of the book.

Integration: Before the Welfare State

The kinds of social policiescommonly associated with the welfare state were first introduced in Bismarckian Germany in 1883, taking the form of social insurance schemes covering health, accident, old age and disability (Baldwin 1990: Chapter 1). The foundational English-speaking literature from this period, however, was predominantly British, though the United States, Australia and New Zealand also feature. Britain lagged behind Germany, Australia and New Zealand, but intellectually the period between the late nineteenth century and World War II was most fertile in Britain (Caine 1992; Lewis and Davies 1991).

In the context of rapid industrialisation and a growth in free-market ideologies, British political agitators engaged in the defence of human welfare, insisting on active state intervention to combat growing social and labour market dislocation. The analysis of social problems typically took interdisciplinary forms. Work and welfare were seen in that light, and integrated understandings of commercial employment and daily life were fostered, including for women, both in the household and in the community (Harris 1992; Rowbotham 1994; Caine 1992).

Policies on poverty alleviation through the Poor Law, labour market protections in the form of the Factory Acts, and state health and education measures formed the main subjects of what was later to be termed the field of ‘social administration’ (Walker 1981). Fabian socialists, who founded the London School of Economics (LSE) in 1895 and established the social administration tradition (Mishra 1989), conducted rigorous empirical research as a basis to stimulate public policy debate and to lobby government to improve standards relating to work, employment and community life. With this as the backdrop, Beatrice and Sidney Webb, who have often been cited as the parents of academic social policy (Williams 1989; Pinker 1971), provided the first blueprint for a comprehensive welfare state in The Prevention of Destitution (Webb and Webb 1911). This was in addition to their pioneering role in the employment relations field, as discussed earlier. Prevention was based on their two-volume Minority Report of the Royal Commission on the Poor Laws (1909a, b).

At the heart of their work was the concept of a ‘national minimum’ (on which, see also Hobhouse 1911, 1922; Hobson 1901), outlined most succinctly by Beatrice Webb in 1918 when she and Sidney were on the War Cabinet Committee on Women in Industry. Beatrice argued that the national minimum should incorporate opportunities for daily rest and recreation; considerations of sanitation and safety in both the home and the place of employment; educational opportunities for all; minima in relation to wages and the full range of employment conditions including allocations for sickness and holidays (Webb 1919: 274). Further, she argued that minima were most important for elevating the rights of women workers who toiled in ‘sweated work’. Before the 1888 House of Lords Select Committee on Sweating (1898) Beatrice argued that the typical sweated worker was a woman, attributing this mainly to the gender segregation of the labour market. Women were more likely than men to be found in the low-paid, poorly conditioned sector of the labour market (see also Cadbury and Shann 1907). In this way Webb drew a direct link between ‘reproduction and production’ (Rowbotham 1994: 25–26). Since most sweated workers were women, and since sweated work most often involved the employee taking work home after the ‘official’ workday came to an end, this would bear upon the domestic welfare roles which were most often ascribed to women: care work, the hygiene and upkeep of the home, and the nourishment of the family.

In the 1920s, interest in the gender dimension of welfare furthered the understanding of the inseparability of welfare from the world of work. However, research by this stage took on a more specific focus on the family unit in the context of worsening economic conditions which culminated in the Great Depression (Lewis and Davies 1991). In her call for the endowment of the family in England, Eleanor Rathbone (1924) argued that policy had to recognise the dual role of women as paid and unpaid workers. The family needed to be provided for, either in the wages system through family-based remuneration or through the income security system in the form of state benefits. Her method involved an examination of the prevailing British system of providing for families, as she traced its history, its effects on the distribution of wealth, national expenditure, the productivity and well-being of the workforce, and ‘on the status of women as mothers and as wage-earners’.

From the 1940s, social reformism was to undergo a change of focus as new welfare institutions were formed and new possibilities imagined as the welfare state was taking shape. With this came a focus squarely on social policy through the delivery of state-provided programmes and social services, to the partial exclusion of the labour market. Essentially there was a partial decoupling of work and welfare.

Partial Decoupling: Enter the Welfare State

In his highly influential wartime reports on unemployment, social insurance and labour exchanges in the 1940s, pioneering welfare state figure William Beveridge (1942, 1944) offered a largely welfare state–based policy framework. This is despite subsequently also ascribing an important role for non-state or ‘voluntary action’ (Beveridge 1948). In the first three decades after the War, there was a pervasive ‘welfare unitarism’ (Pinker 1993), which reflected a relative absence of work and employment and a state-centrism in social policy analysis (Walker 1981; Wilding 2009).

It was on this basis that social policy—or social administration, as it was more often referred to at the time—became an established academic field focusing on the welfare state. This was justifiable on the grounds that the welfare state had progressed to the stage where it could address the main social needs of the national populace. As Glennerster (2007: 4) argued, at the time, ‘[t]he central state apparatus … was seen as having won the war with the support of the people … [and] had done so while securing a basic food supply, clothing, emergency medical care and jobs for the whole population’.

For the most famous post-War social policy theorist in the English-speaking world, Richard Titmuss (1951/1976: 13), the focus on the state did not totally preclude consideration of employment or, to a lesser extent, unpaid work. In his most celebrated essay, ‘The Social Division of Welfare’, Titmuss’s (1956/1976) purpose was partly to demonstrate that welfare had more institutional and policy sources than many subsequent scholars recognised. In the same era, T.H. Marshall’s (1950/1963) widely cited essay on ‘Citizenship and Social Class’ situated the ‘social element’ of citizenship as the final of a three-staged evolution, after ‘civic’ citizenship, which emerged mainly in the eighteenth century, and ‘political’ citizenship, which saw light in the nineteenth. Within the civic stage, Marshall included ‘industrial citizenship’, associated with the right to work, or in Marshall’s words, ‘the right to follow the occupation of one’s choice in the place of one’s choice’, as well as the right to strike.

Titmuss’s (1956/1976) concept of ‘occupational welfare’ was one of the three institutional domains within ‘the social division of welfare’. The two other central domains were ‘social welfare’, indicated by the conventional welfare state, and ‘fiscal welfare’, which referred to welfare gained through the taxation system. Titmuss (1956/1976: 52) highlighted that reliance on occupational welfare was increasing over time and that it was being offered disproportionately to higher-income earners. Despite the fact that it served as a similar policy instrument geared towards ‘welfare’ per se—albeit one administered by state and non-state organisations alike—its effect was to ‘divide loyalties’ and to ‘nourish privilege’, even if it did promote ‘good human relations in industry’ and allowed organisations who offered it to be ‘the “good” employer’.

Finally, Titmuss (1974: 15–16) also considered domestic work and paid employment as part of the ‘essential background for the study of social policy’, which included ‘a knowledge of population changes, … the family as an institution and the position of women; social stratification …, social change and the effects of industrialisation, urbanisation and social conditions; the political structure; the work ethic and the sociology of industrial relations; minority groups …; social control, conformity and deviance …’. Yet, that these were contextual and not central to social policy is indicated by the sheer volume of Titmuss’s work on state-provided social services. This greatly overshadows his lesser known essays featuring integrated discussions of the work–welfare nexus with equal emphasis on both, the two principal examples being the essays on ‘The Position of Women’ (Titmuss 1955/1976) and ‘Industrialization and the Family’ (Titmuss 1957/1976).

Not long afterwards, social policy was partially recoupled with work and employment.

Partial Recoupling: Poverty, Gender and Comparativism

The almost unquestioned public legitimacy of the welfare state in the post-war period began to be challenged as early as the 1960s. Scholarly research started to recognise the work–welfare nexus more keenly in the context of unprecedented policy pressures on the welfare state.

First, poverty was ‘rediscovered’, and was to be found in rich countries, whether or not households were headed by a person in full-time employment. Following important post-War works by Peter Townsend (1954, 1962) and later T.H. Marshall’s (1950/1963) recognition that poverty had not been eliminated, Abel-Smith and Townsend (1965) published their report The Poor and the Poorest. By the 1970s governments not only in Britain but across the English-speaking world accepted the need to examine the problem of poverty officially through public inquiries (Townsend 1979; Henderson et al. 1970; Australia 1975; New Zealand 1972). Townsend’s (1979) Povertyin the United Kingdom was the most remarkable piece of research in relation to the work–welfare relationship. It contained in-depth examination of how poverty related to or stemmed from disability, ageing, unemployment and ‘sub-employment’, ‘working conditions’ and ‘the character of the job’, as well as sole-parenthood, low pay, employment insecurity, and unfavourable occupational health and safety conditions.

In complementary developments, feminist scholars increasingly emphasised the male-centredness of the policy assumptions underpinning social policy scholarship. In doing so they reasserted the importance of the welfare role of women’s care work and questioned the assumption of the male breadwinner household model as the sole basis for the welfare state. It was problematised that the poverty literature assumed a family financed primarily or solely by a male head who was in paid employment (Blackburn 1995). Feminist writers began to address the problem from the 1970s, highlighting the importance of women’s contributions to household welfare both through paid work in the labour market and through care work provided in the home (McIntosh 1981). In providing the first book-length critique of the traditionally male-centred approach, Elizabeth Wilson’s (1977) Women and the Welfare State analysed the development of social policy from the Victorian era to the 1970s. She argued that only through a feminist historical approach can an effective portrayal of the aggregate distribution of life opportunities be formed, and that such an approach must place at its centre the woman’s role as the ‘linchpin’ of the family.

The struggles of feminists during the late 1960s and through the 1970s made their mark on the social policy field, particularly in relation to women’s right to income security, adequate housing and accessible childcare. A literature emerged which placed work, both paid and unpaid, at or near the centre of social policyanalysis (e.g. Land 1971; Pascall 1986; Williams 1989; Wilson 1977; Lewis 1983; Lewis 1994). Hilary Land’s (1971) analysis of ‘Women, Work and Social Security’ is emblematic of the earliest of this research, arguing that the British social security system was largely based upon an outdated and discriminatory assessment of gender inequalities in the labour market. Pascall (1986) contended similarly that the welfare of women is strongly shaped by the combination of responsibilities, which for the most part they bear alone: for housework, care of children and other relatives, and over time increasingly paid market work as well. Even when they do engage in paid employment, she argued, they must face the reality that the labour market is gender-segmented. Women’s jobs were (and in many cases still are) largely segregated from those which men occupy, the former on the whole enjoying less pay, often less paid hours and lower job security and occupational welfare in the Titmuss (1956/1976) sense. With these factors in mind, Pascall argued that if it is to be understood more truly, social policy should be seen as a bridge between production and reproduction, paid work and unpaid work. In doing so, she reinvigorated the tradition established by Beatrice Webb (1919) and her contemporaries.

In a third important body of literature, the emergence of comparative social policy as a major sub-specialisation has also aided the integration of work with welfare. Like the poverty and feminist literatures, comparativism emerged in the late 1960s and the 1970s. By the 1990s comparativists began debating the welfare role of labour market institutions across countries, and how those institutions relate to and interact with the welfare state. In its formative years comparativism was predominantly state-centred in the manner of the post-War writers (Rogers et al. 1968; Kaim-Caudle 1973). It accelerated from the 1980s onwards as a means to understand transitions and transformations in welfare states, often explicitly or implicitly through national responses to globalisation (e.g. Esping-Andersen 1996; Ellison 2006; Pierson 2001, 2006).

The relationship between work and welfare more specifically has been subject to comparative analysis through the need to deal with the kinds of demographic, social and economic changes which were discussed at the outset of the chapter. Theoretically many of these are combined by political economists and political scientists under the category of ‘post-Fordist’ labour markets (Amin 1994; Jessop 1995) and ‘post-industrial’ societies (Armingeon and Bonoli 2006; Pierson 2006). The comparative political economy framework of ‘varieties of capitalism’ (Hall and Soskice 2001; Hancke 2009) represents a complementary approach. It separates developed countries into ‘liberal market economies’, which are represented by the English-speaking nations, and the ‘coordinated market economies’ of central and western Europe plus Japan. The separation point between the two is based on how firms operate within their institutional and market contexts, though the variety of capitalism approach is actor-centred and less specific to the institutional relationship between employment relations and social policy.

The role of labour market institutions in providing welfare is a major theme in ongoing research inspired by the welfare state ‘regimes’ concept, specifically in response to Esping-Andersen’s (1990) Three Worlds of Welfare Capitalism (e.g. Castles and Mitchell 1992; Goodwin et al. 1999; Lewis 1992; O’Connor et al. 1999). In identifying three developed-country welfare state regime types—‘liberal’, ‘conservative-corporatist’ and ‘social democratic’—Esping-Andersen elaborated on underdeveloped categories first proposed by Titmuss (1974: 30–31), though with different terminology. Each regime type had relatively distinct institutional characteristics, but it was important within each to look beyond ‘the traditional approach’ to social policy analysis, which mainly covered ‘income transfers and the social services, with perhaps some token mention of the housing question’. In the preferable ‘broader view’, Esping-Andersen (1990: 1–2) argued, ‘issues of employment, wages, and overall macroeconomic steering are considered integral components in the welfare state complex’.

Yet, as argued here, there is considerably more work to do if these issues are to represent an equal relationship between social policy and employment relations.

The Conundrum

In arguing that social protection is necessary for the maintenance of the market mechanism, and thus for the benefit of society as a whole, Polanyi (1944) effectively showed the importance of social protection as a fixture through time in capitalist society. The literature of social policy has demonstrated a need for understanding the interaction of work with welfare, particularly for economically vulnerable people. A conundrum is presented, however, by two phenomena that have been revealed by examining the work–welfare literatures just discussed. First, there are few conceptions of employment relations—as opposed to work and employment—which allow a detailed focus on the interaction between employment and the welfare state. Second, and more importantly, while the social policy field is relatively proficient at demonstrating the importance of employment to the broader question of welfare, it has been considerably less proficient in revealing the dynamic factors and the forces that shape the conditions of employment.

Other authors have argued similarly, though no firm body of integrative research has resulted. In articulating the case for a ‘welfare economy’ approach to social policy, as opposed to a welfare state one, Martin Rein (1981) argued that state-provided welfare is only part of the story of resource redistribution. A broader picture which includes an analysis of the private–public welfare mix is needed, particularly for comparative purposes. The welfare economy approach involved a ‘shift [in] our focus from measuring the amount of expenditure in the welfare state to an understanding of the varied institutional forms by which society carries out its welfare function’ (Rein 1981: 22). This was an important recognition.

Rein’s subsequent partnership with Lee Rainwater led to the production of a significant edited volume which argued for a ‘welfare society’ approach to social policy research. There, as that book’s title suggests, a case is made for better recognising Public/Private Interplay in Social Protection:

We believe that the main explanation [for what they see as a ‘narrow focus on the welfare state’] lies in the existence of two intellectual traditions, one conventionally associated with the study of social policy and the other concerned with the analysis of labor and industrial relations. The former focused on state action, believing that nonstate action would be perversely redistributive. The latter concentrated on industrial conflict but neglected the content of these negotiations outside of monetary wages. We believe that the traditions of social policy and industrial relations must be reintegrated if we are to better understand and act on the world in which we live. (Rein and Rainwater 1986: vii)

The chapter authors in that volume provided case studies demonstrating the need to integrate social policy and employment relations by reference to the administration of sickness benefits (Immergut 1986), pensions (O’Higgins 1986) and employee redundancy (Russig 1986).

As important as this research was, however, and as profound as the integrative language of the editors was, it did not result in a framework which conceptually integrated employment relations as the term is commonly understood. Nor did it engender a new or radically changed approach to social policy scholarship on interdisciplinary integration. Finally, rather than the concept of employment or industrial relations, Rein and Rainwater and their chapter contributors focused much more on ‘private’ means or institutions of social protection, which is closer to Titmuss’s (1956/1976) concept of occupational welfare as previously discussed.

Francis Castles (1985), whose long-standing contribution is discussed and critiqued throughout this book, provided evidence of the benefits of integrating employment relations with the welfare state. Utilising the Webbian concept of the ‘national minimum’ as discussed earlier (Webb 1919: 274), he argued that the statutory wage regulation and compulsory industrial arbitration systems that were developed in Australia and New Zealand provided the basis of a ‘unique model that might be described as the wage-earners’ welfarestate’ (Castles 1985: 103; his italics). For Castles (1989, 1996) thenational minimum in wage regulation constituted ‘social protection by other means’ (see more recently Béland 2019; Seelkopf and Starke 2019), implying that employment relations in Australia and New Zealand history provided a ‘functional alternative’ to the more orthodox welfare states seen in Western Europe and North America. His work inspired a follow-up literature in relation to Australia and to a lesser extent to New Zealand (reviewed in Ramia and Wailes 2006; Watts 1997).

As acknowledged at length here, Castles argued that employment relations institutions interact in important ways with the welfare state. However, while the dynamism of the welfare state was assumed and demonstrated, Castles treated the employment-relations half of the social protection partnership as largely static over time. This, as demonstrated in Chap. 2 onwards, resulted in important omissions which affect the comparative analysis of social protection.

Objective and Argument

The central objective of this book is to examine the comparative evolution of social protection over more than a century, focusing on the relationship between employment relations and social policy, using a two-country comparative case study. The two countries are Australia and New Zealand. The analysis starts with the 1890s, when formal social protection first took shape there—which was earlier than any in Europe or the other English-speaking countries—and traces developments to the present day. The argument of the book is that allowing both employment relations and social policy to be dynamic over time, and to be considered together as part of social protection, produces a somewhat different and more nuanced narrative on comparative policy evolution. The analysis also discusses the implications of the comparative case study for other countries as well as for the contemporary understanding of social protection.

Australia and New Zealand have often been misunderstood by both international and local observers. The importance of history has been underestimated, and as discussed here, those seeking to ‘situate’ them in an international context have missed crucial aspects of their comparative development. Some scholars have placed great faith in the proposition that they were traditionally very similar, with some assuming that the two essentially followed one model. By contrast, those focusing on the 1980s and 1990s—when major restructuring of social protection took place, and what most of the literature focuses on—often overestimate the differences as both undertook major adjustments to common global economic pressures. Finally, their re-convergence in more recent years has gone almost unaccounted for. The book also argues that a long-haul account of the comparative evolution of the two national regimes is needed for revisionist Australia–New Zealand comparison and for the understanding of other countries’ paths.

A recurrent theme through the case analysis is how the international and local policy literatures portray Australia and New Zealand as social protection regimes in relation to different periods. The periodisation guiding the case is fourfold:

  1. 1.

    The pre–World War I era, 1890 to 1914, which inaugurated a ‘national minimum’ of arbitrated employment standards, and furnished their institutional context alongside trade and immigration policies, and pioneering if residualist and often discriminatory social security measures.

  2. 2.

    The period from the end of World War I to the mid-1980s, which saw the comparatively early emergence of the welfare state, the solidification of its relationship with the employment relations system, and the subsequent expansion of social protection after World War II and up to the 1980s.

  3. 3.

    The restructuring of social protection from the mid-1980s to the mid-1990s, highlighting the markedly different relationship between employment relations and social policy that was formed in Australia as compared with New Zealand.

  4. 4.

    Developments from the mid-1990s to the current time, highlighting the re-convergence of Australia and New Zealand, as each has mainly emulated the other. The significance of the more closely intertwined relationship between employment relations and social policy is constructed as a function of home-grown but transnationally dependent trajectories in workfare. Implications are drawn for other contemporary welfare states.

The Case Study and Its Comparative Context

Social protection is dispensed through ‘regimes’, which provide forums for interaction between institutions. ‘Welfare regime-types’, a concept popularised among social policyscholars by Gosta Esping-Andersen (1990), comprise collections of social protection institutions which are similar, though never exactly the same, across several countries. Regimes also exist at the national and sub-national levels. However, instead of constructing social protection institutions in terms of ‘state, market and family’—as Esping-Andersen (1990: 26) did—the two national regimes in the case analysis are viewed through the lens of the socially protective relationship between employment relations and social policy. The purpose of the remainder of this chapter is to situate scholarly understandings of the two regimes in their international and comparative contexts, and to outline the methodology behind their comparison to each other.

Perceptions of similarity and difference between Australia and New Zealand vary, depending mainly on how many countries are in the comparative analysis, but also on the period being analysed. As might be expected, those who have sought to examine the two within the context of many-country, or ‘large-N’, studies have interpreted the two regimes as constituting basically the same social protection model. Those that have analysed Australia and New Zealand in two-country or otherwise small-N studies have often emphasised the differences. The two were similar in the late nineteenth century, but diverged markedly in the 1980s and 1990s. There is much more to the story, however, in that it depends also on the implicit definition of social protection which is adopted. With few exceptions, the small-N studies have generally been either from employment relations or from social policy, and rarely have the two perspectives been incorporated in the same analysis.

The most important exception is Francis Castles (1985), whose oft-cited book The Working Class and Welfare directly compared Australian with New Zealand social protection while emphasising similarities through a long period in time. Castles essentially constructed them as one model, captured in the concept of a ‘wage-earners’ welfare state’. He argues (Castles 1985, 1996) that the two countries’ welfare states developed earlier than, and were different to, the welfare states of Europe. In terms of substance, the central difference between these two regimes and their Western European counterparts lay in the exceptional role of minimum labour standards in Australia and New Zealand. In qualifying this difference, Castles contended that in addition to employment relations, the overall pattern of social protection relied on a combination of industry protection, restricted and selective immigration, and a ‘residual’ social security system.

That Australia and New Zealand had this four-pronged regime in common through much of their history is not in itself contested here. However, it is argued that there are two main problems with accepting the wage-earners’ welfare state account. First, as argued earlier, Castles’ account treats the employment relations system as largely static. It is barely described, and it is implicitly assumed to be hardly changing over time. Second, it overstates the historical similarities between Australia and New Zealand. The alternate account presented in this book introduces dynamism to the employment relations dimension of social protection. It also yields a different basis for comparative social protection analysis in general.

Having made the argument earlier that the international literatures of employment relations and social policy have developed somewhat separately, it is important to briefly review the state of knowledge on social protection in Australia and New Zealand specifically.

Large-N Comparisons

Examinations of ‘corporatism’ in national policy have generally placed Australia and New Zealand alongside each other. There is nothing innately wrong with doing so. Corporatist analysis involves assessments of the strength and influence of political ties and negotiations among national-level interests, principally labour, business and the state. The policy traditions of Australia and New Zealand are generally both situated in the category of either ‘low’ or ‘intermediate’ corporatism (e.g. Bruno and Sachs 1985; Calmfors and Driffill 1988; Freeman 1988; Siaroff 1999). The existence of compulsory arbitration through much of their modern histories is sited as the main reason. Beyond this, more detail is rarely offered; nor is it required for multi-country comparative studies.

Analyses of welfare regime types have for the most part placed Australia and New Zealand in the same category, as ‘liberal’ welfare states (e.g. Bambra 2012; Esping-Andersen 1990, 1996; Ferragina and Seeleib-Kaiser 2011; Karim et al. 2010; Lewis 1992; Shaver 1990, 1995; Taylor-Gooby 1991). The liberal welfare regime is characterised by highly selective and ‘residual’, as opposed to universal, state welfare schemes, allowing the market significant freedom to determine living standards and the distribution of income and life opportunities with minimal state intervention. This is in contrast to the other two main categories of welfare state: the ‘social democratic’ type, which mainly included the Scandinavian countries, and the ‘conservative-corporatist’ regimes of countries such as Germany, Italy, Austria and France.

Castles and Mitchell (1992, 1993) challenged the welfare regimes approach on the classification of Australia, New Zealand and the United Kingdom, based on their argument that these three constitute a ‘radical’ collection of regimes, with relatively low indirect but high direct taxation, low social expenditures and almost non-existent social insurance contribution systems, and redistributive labour market regulation patterns. Importantly, however, while the welfare regime types and the families of nations scholars have the relationship between work and welfare as important elements, they do not in any meaningful sense consider the importance of employment relations systems.

This suggests that more specific research on Australia and New Zealand is needed.

Small-N Comparisons

More targeted research is available. Direct-comparative or small-N analyses of Australia and New Zealand—which typically include these two countries and may include one or two more—are important because they are naturally able to provide greater institutional and policy detail on each country. Authors in the small-N category are particularly effective in revealing the factors which cause similarity and difference between the regimes of interest. However, given that such studies have emerged mainly since the late 1980s in relation to Australia and New Zealand, much of this work seeks to come to terms with the major divergence which occurred between the two regimes from the mid-1980s and the mid-1990s. There is some research on the period since and to the current time, though there is less of it.

Small-N authors delve into important differences with respect to employment relations change (Ahlquist 2011; Barry and Wailes 2004; Bray and Haworth 1993; Bray and Nielson 1996; Bray and Rasmussen 2018; Bray and Walsh 1993, 1995; Brosnan et al. 1992; Gardner 1995; Sandlant 1989; Wailes 1999; Wailes et al. 2003), social policy (Castles 1996; Castles and Pierson 1996; Castles and Shirley 1996; Deeming 2013; McClelland and St John 2005; Wilson et al. 2013), state reorganisation and change in public administration (Boston and Uhr 1996; Considine and Lewis 2003; Schwartz 1994a, b, 2000, 2010), and economic policy and economic outcomes (Ahlquist 2011; Castle and Haworth 1993; Easton and Gerritsen 1996; Goldfinch 2000; Quiggin 1998). Despite its greater importance to the current book, however, consistent with the comparative literature at large, direct Australia–New Zealand comparativism has thus far largely reproduced the divide between employment relations and social policy (Ramia and Wailes 2006). The effects of that tendency are analysed in Chaps. 2, 3, 4, 5 and 6.

Moving Beyond the ‘Wage-Earners’ Welfare State’ Model

The divide in the literature reinforces the importance of Castles, given that he is something of a lone voice in the existing social protection literature. The separate evolution of the employment relations and social policy fields is partly addressed by his ‘wage-earners’ welfare state’ model, which formed the basis of the predominant framework substantively comparing Australia and New Zealand on social protection. In this sense it is the precursor analysis to this book.

Coming to prominence three decades ago, the model has since been variously reiterated, reified and updated in many scholarly pieces comparing developments in the two countries (Castles 1990; 1993a; 1996; Castles (ed.) 1993b; Castles et al. 2006; Castles and Mitchell 1992, 1993; Castles and Pierson 1996; Castles and Shirley 1996), and in work specifically focusing on Australian developments since the mid-1980s (Castles 1987, 1988, 1989, 1994, 2001, 2002, Castles and Uhr 2007). Castles argued that there are four planks of the traditional Australian and New Zealand social protection pattern. Minimum wages, decided mainly by industrial arbitration tribunals, was the first plank, generally differentiated according to the gender of the worker, such that men were accorded higher rates than women based on the assumption that men were more often financially responsible for the family. Thus gender inequality was a legally institutionalised feature of the system. The second plank of the wage-earners’ welfare state model was industry protection, which was used as a policy tool by both Australian and New Zealand governments to entice—in the case of Australia, to compel—employers to provide comparatively high-level minimum wages.

Third, selective and restrictive immigration policies were utilised as a method to exclude workers from countries where labour was relatively poorly paid, thereby theoretically disallowing the wages of Australian and New Zealand workers from being undercut. The fourth and final plank was the system of state welfare, and social security in particular, which was overridingly selective and mainly ‘residual’, meaning that it was designed to protect only those who could not work. The welfare state as traditionally understood thus mainly picks up the social protection pieces upon the ‘failure’ of the market and the family.

Castles’ depiction of Australia and New Zealand has been either implicitly or explicitly accepted by most social policy scholars (Bryson 1992, 1994; Bryson and Verity 2009; Carney and Hanks 1994; Cass and Freeland 1992; Deeming 2013; Fenna and Tapper 2012; Jones 1990; Saunders 2006; Saunders and Deeming 2011; Wilson 2017). Someemployment relations analysts have similarly accepted it as reflecting the traditional context, or at least a context, within which the employment relations system operated; that is, to the extent that they deal with broader social protection at all (e.g. Barry and Wailes 2004; Bray and Haworth 1993; Bray and Walsh 1993, 1995; Nolan and Walsh 1994; Sandlant 1989; Treuren 2000).

Despite capturing the scholarly community, however, as an explanation of the historical patterns of social protection in the two countries the Castles model has several shortcomings. These fall into two main categories. The first is from the comparative perspective, and the second is from the historical perspective.

From the comparative viewpoint, Castles underplays the differences between Australia and New Zealand through time. Hence, an alternative or complementary comparative approach should more comprehensively set out the factors separating the two regimes while also not underestimating or underplaying the similarities. The second set of criticisms—deriving from viewing the wage-earners’ welfare state model as a historical analysis—relate mainly to its portrayal of how the key institutions underpinning social protection evolved, and the importance of their evolution for deciphering the differences between Australian and New Zealand.

In addition, however, the third problem stems from Castles’ treatment of arbitration. In an almost total failure to acknowledge its importance after being first introduced—in the 1890s in New Zealand and the early 1900s in Australia—his model assumes implicitly that arbitration, once created, remained virtually unchanged. In addition, the nature of its relationship with both state welfare and the broader social protection regime remained relatively constant. This, it is argued here, is the model’s most serious shortcoming.

Castles did not capture the significance of two key institutional departure points on arbitration between New Zealand and Australia: first, the existence of nationalminimum labour standards outside of the arbitration system in the former, and their effective absence in the latter; and second, the different institutional means by which equal pay regulations were channelled in the late 1960s and 1970s, those of Australia being introduced through arbitration and New Zealand’s being through non-arbitral legislation. As will be discussed further, these differences point to features which helped to shape, and were reflective of, the greater institutional and political commitment to arbitration in Australia. This is the great separator.

That commitment lives in varying forms to today, and it is a central ingredient in this analysis. As demonstrated in several of the chapters which follow, analysing the intricacies of the politics of arbitration is important for understanding the paths taken in both employment relations and social policy since the mid-1980s, as policymakers in New Zealand always had on offer a non-arbitral safety net to fall back on once arbitration waned in importance, as happened in both countries in the first three decades after World War II.

Relating to the second category, the historical perspective, the wage-earners’ welfare state framework failed to explicate the broader institutional configuration of employment relations. In particular, it did not pick up on the embedding of Australia’s arbitration system within a relatively rigid Constitutional framework,Footnote 1 specifically, the much more modest role for the Australian national government in directly regulating minimum labour standards. This factor allowed New Zealand more institutional and political leeway in relation to minima. This, as argued, shaped other differences through time.

The Historical-Institutionalist Approach

The book makes no particular claim to innovation in methodology, and it is more concerned with methodological application. In substantiating its argument, the book utilises long-standing and more recent developments in ‘historical institutionalism’. The analysis relies strongly, though by no means exclusively, on Steinmo, Thelen and Longstreth’s (1992) Structuring Politics, and Mahoney and Thelen’s (2010) ‘theory of gradual institutional change’. Historical institutionalism is particularly useful where the institutions in focus are evolving over time and typically located at the ‘meso-level’. Such applications are well established (e.g. Capano and Howlett 2009; Hall and Taylor 1996; Mahoney and Thelen 2010; Streeck and Thelen 2005; Thelen 1999). That is, they lie between the macro-systems of so-called old institutionalism as represented by scholars such as the Webbs (1897; 1894; 1911), and micro, individual preference-based frameworks, which are less relevant here.

Social policies and employment relations systems are typical examples of meso-level institutional arenas. They present problems of incremental historical evolution and not merely representations of recurring or intermittent ‘exogenous shocks’. The case for this kind of research was made effectively by Mahoney and Thelen (2010: 4):

We have good theories of why various kinds of basic institutional configurations – constitutions, welfare systems, and property rights arrangements [one can add employment relations systems] – come into being in certain cases and at certain times. And we have theories to explain those crucial moments when these institutional configurations are upended and replaced with fundamentally new ones. But still lacking are equally useful tools for explaining the more gradual evolution of institutions once they have been established. Constitutions, systems of social provision, [employment relations systems] and property right [sic] arrangements not only emerge and break down; they also evolve and shift in more subtle ways across time. These kinds of gradual transformations [are] all too often left out of institutionalist work.

The book prioritises the gradual transformations. On the associated question of the causes of institutional change and the need for long-haul history, the same authors argue:

In the literature on institutional change, most scholars point to exogenous shocks that bring about radical institutional configurations, overlooking shifts based on endogenous developments that unfold incrementally. Indeed, these sorts of gradual or piecemeal changes often only ‘show up’ or ‘register’ as change if we reconsider a somewhat longer time frame than is characteristic in much of the literature. Moreover, when institutions are treated as causes, scholars are too apt to assume that big and abrupt shifts in institutional forms are more important or consequential than slow and incrementally occurring changes. (Mahoney and Thelen 2010: 2)

The analysis in the book is simultaneously comparative and historical (Lange 2012; Mahoney 2004). Australia and New Zealand are chosen for the case study because their policy regimes were historically similar, as acknowledged in the comparative literature discussed earlier, but they have taken both divergent and re-convergent paths in recent decades, having undergone both slow transitions and intermittent radical changes. Traditionally they were what some comparativists—informed by John Stuart Mill’s (1882) ‘method of difference’—refer to as ‘most similar cases’ (Wailes 1999; Wailes et al. 2003). This makes their comparison compelling.

What Follows

Following the current chapter, Chap. 2 discusses the period from the 1890s to World War I, outlining and comparing the British ‘national minimum’–based social protection platforms of Australia and New Zealand as the case study countries. Chapter 3 furthers the comparative history by discussing the consolidation of the relationship between employment relations and social policy over the period from World War I to the 1940s. Chapter 4 covers the 1950s to the early 1980s.

The differences between the two countries became most marked in the 1980s and 1990s. These are captured in Chap. 5, as Australia moved into a new coupling of employment relations and social policy under a quasi-corporatist national wages and incomes ‘Accord’, while New Zealand embarked on a path of decoupling as part of an internationally infamous and radical model of neoliberalism. As outlined in Chap. 6, this picture of divergence gave way to a re-convergence form the mid-1990s to the present time, principally around the integration of employment relations with social policy. The importance of both historical and new or emerging similarities and differences is also emphasised. Finally, Chap. 7 examines the international and conceptual implications of the case study findings for other countries.