Liberalism's Difficult Relationship with the Welfare State

Abstract

This paper makes two related points. First, as liberals have started to realize that the welfare state is unable to deliver on egalitarian theories of justice, they have increasingly tried to dissociate their theories from the welfare state. Second, dissociating from the welfare state type of thinking is difficult for some liberal egalitarian theories such as John Rawls's theory of justice as his theory shares some of the same underlying thinking as found in the welfare state. For example, Rawls's understanding of universal citizenship and the difference principle resembles some of the aspects of the welfare state on how social equality and citizenship are tied to productivity and society as a venture of mutual cooperation. Consequently, liberals are caught in a difficult relationship where they can only partially move beyond the welfare state. Because of this affinity liberals should move beyond a Rawlsian framework, as Rawls's theory is difficult to completely dissociate from the welfare state.

Headnote

Abstract: This paper makes two related points. First, as liberals have started to realize that the welfare state is unable to deliver on egalitarian theories of justice, they have increasingly tried to dissociate their theories from the welfare state. Second, dissociating from the welfare state type of thinking is difficult for some liberal egalitarian theories such as John Rawls's theory of justice as his theory shares some of the same underlying thinking as found in the welfare state. For example, Rawls's understanding of universal citizenship and the difference principle resembles some of the aspects of the welfare state on how social equality and citizenship are tied to productivity and society as a venture of mutual cooperation. Consequently, liberals are caught in a difficult relationship where they can only partially move beyond the welfare state. Because of this affinity liberals should move beyond a Rawlsian framework, as Rawls's theory is difficult to completely dissociate from the welfare state.

Keywords: liberalism, the welfare state, multiculturalism, egalitarian justice

JEL Classification: D63, D30

I. INTRODUCTION

Some proponents of distributive justice have advanced the welfare state as the practical application of egalitarian liberal political theory to remedy social and economic inequalities.1 So strong has the relationship between the welfare state and liberalism become that some have labelled it welfare state liberalism (Sterba 1988). Although liberalism is a broad tradition consisting of a wide range of different theories, it is primarily the egalitarian liberal theories which have been associated with the welfare state.

Two examples are Brian Barry's Justice as Impartiality (1995) and John Rawls's A Theory of Justice (1999) and it is with this strand of liberalism this paper primarily engages. Because it has often been assumed that egalitarian liberal political theories such as Rawls's theory of justice were a defense of the welfare state, a surprising turn is that John Rawls explicitly sought to distance himself from the welfare state in his later works, as he believed that it was not sufficiently egalitarian to satisfy his two principles of justice (1999, xiv-xvi). Several others have joined Rawls' search for more egalitarian strategies, for example, Bruce Ackermann and Anne Alstott (1999) and John Roemer (1999). In addition, recently the welfare state has been under mounting political pressure, as it has proved harder to reduce inequalities and eradicate social problems through it. Thus, it is no longer clear that the welfare state is the best hope to realize distributive justice for liberals.2

Thus, there is a need for an assessment of the relationship between egalitarian liberal theories and the welfare state and that is the purpose of this paper. I argue that liberalism has a difficult relationship with the welfare state because, on the one hand, liberal ideals are more egalitarian, but on the other hand liberalism struggles with a complete dissociation because traces of the ideals underlying the welfare state can still be found in liberalism.

My argument has the following two implications. First, the welfare state struggles with realizing the egalitarian ideals of liberalism. Second, liberalism is still somehow tied to the thinking underlying the welfare state and it becomes difficult for liberals to completely move beyond this to realize their ideals of distributive justice, which, I conclude, they should do if they are serious about realizing egalitarian ideals of distributive justice.

I start my analysis with a discussion of some of the ideas underlying the welfare state that relate to liberalism. I then continue to analyze how these ideas affect John Rawls's theory of justice. I end with a discussion of the relationship between liberalism, the welfare state and the ideal of universal citizenship.

II. THE IDEAS BEHIND THE WELFARE STATE

Today the welfare state is defined and understood in a wide range of ways. I will not aim at a thorough description and discussion of the vast literature on the subject. Instead, I aim to draw attention to some features important for a comparison between the welfare state and liberalism. I begin this section by discussing different definitions of the welfare state, welfare regimes, and normative justifications for the welfare state. Then I elaborate on the idea of social equality before I argue that the welfare state is mutually advantageous for the most politically effective groups in society.

Definitions of the welfare state often emphasize that it is a response to the social and economic inequalities caused by industrial capitalism, and that the state takes responsibility for securing that no citizen's standard of living falls below a certain threshold (Gutmann 1986, 3). Although this definition points out some central features and activities, it undercommunicates the central role the welfare state played in modern societies where it is "a fundamental aspect of modern government [...] as an indispensable means of making capitalist economies socially and economically sustainable" (Garland 2016, 3). Importantly, this understanding draws attention to both how normal and indispensable the welfare state is to modern societies and that modern societies would hardly be wellfunctioning without such a state. The indispensability and normality of the welfare state are expressed through the various welfare regimes adapted to different societies' social and historical backgrounds such as the social democratic welfare state in Scandinavia, the liberal welfare state in many English-speaking countries and the corporatist-statist welfare state common in continental Europe (Esping-Andersen 1990, 26-27). What Gosta Esping-Andersen stressed in his analysis of the three welfare regimes was that they are intimately connected to the workings of capitalism and hence he called the three regimes the three worlds of welfare capitalism (1990).

Whereas Esping-Andersen analyzed the different welfare regimes prevalent in modern societies, Joseph Heath has analyzed three different normative models of the welfare state. Heath's starting point builds on the idea of an intimate relationship between the welfare state and capitalism stating that to identify "an appropriate division of labor between the public and the private sector, we need a normative standard that tells us what the state ought to do" (2011,13, original emphasis). The problem is, however, that "of the welfare state there are several different theoretical reconstructions of the normative commitments that are taken to underlie it, all of which are in tension with one another" (Heath 2011, 14). Three different models of justification are commonly employed: "the redistributive, the communitarian and the public-economic models of the welfare state" (Heath 2011, 14). The redistributive model emphasizes redistribution of resources to rectify some of the inequalities created by the market economy (14). The communitarian model argues in favor of limiting the market economy and avoid commodification of certain parts of society (14-15). The third model "regards the welfare state as playing a role essentially complementary to that of the market" (15). Together these three models cover many of the normative justifications for the welfare state.

Although all three models are plausible to some extent, Heath argues that the public-economic model is the most plausible (2011, 15). One problem with the redistributive and communitarian models is that they are unable to explain the puzzling fact that the welfare state has expanded over time regardless of whether left-leaning or right-leaning governments were in power (Heath 2011, 38). The public-economic model in contrast can make sense of this feature because:

The state taxes people in order to provide public goods. As people become wealthier, they want to spend an increasing fraction of their income gain on public goods, and so, to the extent that the state is responsive to public preferences, growth in per capita GDP will lead to an increase in state spending as a fraction of GDP. (Heath 2011, 38)

Thus, welfare state spending on public goods is not a zero-sum game but is a win-win situation based on a Pareto improvement (37-38). I will elaborate on this last point and argue that the welfare state embodies an important element of mutual advantage, and that this element is important in understanding the difficult relationship between liberalism and the welfare state.

Elements of mutual advantage can be found in some of the earliest justifications of the welfare state about the importance of reversing the social inequalities following from a capitalistic economy (Beveridge 1942; Tawney [1931] 1951; Titmuss 1976). Of these early writings, T. H. Marshall's occupy a special role. His writings contain one of the clearest expressions of why a welfare state is required and are still influential among contemporary welfare-state theorists.3 Crucial to Marshall's account is that "there is a kind of basic human equality associated with the concept of full membership of a community-or, as I should say, of citizenship which is not inconsistent with the inequalities which distinguish the various economic levels in the society" (Marshall 1992, 6). It is important to note in Marshall's notion of equality that he makes a firm distinction between social and economic equality and claims that universal citizenship rights are compatible with economic inequalities.

More controversially, Marshall tied welfare rights to participation in productive labor. This move was legitimated because "your body is part of the national capital, and must be looked after, and sickness causes a loss of national income" (Marshall 1981, 91). Welfare rights, according to Marshall's account, are thus necessary to sustain and increase the national capital. The post-war context in which Marshall's writings first appeared helps explain the strong emphasis on citizenship, the nation-state and economic growth. These views may sound alien to contemporary readers. However, contemporary welfare theorists continue to emphasize the national context and maintain that workers pay for the welfare state and argue that the expansion of the welfare state during the 1960s would have been unlikely without strong economic growth (Myles and Quadagno 2002, 36).4

For Marshall, social equality constitutes the fundamental aim of the welfare state. However, this notion of social equality is compatible with (potentially large) economic inequalities and presupposes redistribution among groups to provide welfare for the working class (and other groups). Hence, Marshall's notion of social equality is not strictly egalitarian. The significance of Marshall's concept of social equality is that it successfully pulls the social status of the working class upwards and simultaneously ties welfare rights to participation in productive labor. Marshall's concept of equality thus achieves two things: first, social status is independent of economic class, and second, welfare rights are tied to participation in productive labor. Participation in productive labor becomes a condition both for equal social status and welfare rights.5 In short: workers exchange labor for welfare rights. Hence, reciprocity connects social equality and welfare rights. As Garland (2016, 16) points out, reciprocity has played an important role in most historical societies and in this respect the welfare state is a modern continuation.

Heath's public-economic model and Marshall's emphasis on reciprocity can be combined to explain how the welfare state mutually benefits the politically effective groups in society.6 I will argue that the mutualadvantage thesis contributes to explaining how the welfare state operates and to understanding why it is a relatively stable political institution. The mutual advantage thesis has two main elements: first, that the welfare state is efficient and second, that the welfare state serves the interests of a large proportion of the population in most Western democracies.

Regarding the first condition, industrial capitalism led to rapid urbanization and in many cases widespread poverty, especially during economic recessions. Urbanization and poverty led to over-crowded housing, poor sanitation, and poor health among the growing working class. Resolving these issues was a matter of urgency and was solved through public policy, as private philanthropy and market solutions proved inadequate. In the United Kingdom, for example, these concerns led to reforms such as the Public Health Act (1875), and after the Second World War, such reforms culminated in the modern welfare state, in which the state offers services such as education, health care (NHS), unemployment benefits and pensions, among others. Public provisions of many of these services were prompted by a desire to increase national competitiveness and the need for a productive and healthy workforce (Barr 1987).

In the words of the historian Eric Hobsbawm (1994, 284), the development of the modern welfare state represented a balance that "depended on a coordination between the growth of productivity and earnings which kept profits stable". Higher productivity gave rise to higher profits, which again gave rise to higher tax revenues. The higher tax revenues made it possible to fund the welfare state without crippling profits. Higher productivity boosted efficiency, and the institutions of the welfare state were instrumental in achieving efficiency by improving the living conditions of the working class, which ultimately improved their contribution to the economy.

Concerning the second condition, that the welfare state serves the interests of a large portion of the electorate. One can find support for this feature by examining the broad coalitions that created the welfare state in the first place. Developing a welfare state required broad coalitions and political compromises to be viable. In the Nordic countries, for example, red-green coalitions amalgamated the welfare state, while in Britain the consolidation of the welfare state "came to depend fundamentally on the political alliances of the new middle class" (Esping-Andersen 1990, 31). Similar alliances developed across continental Europe, while in the USA, the middle classes "were not wooed from the market to the state" (31), and the welfare state remained residual. This perspective injects a new dimension of dynamism to the development of the welfare state and shows how its development was possible only with broad coalitions mutually benefiting the interests of a vast majority of voters.

The magnitude of the modern welfare state therefore lies in the way in which it serves the interests of a large proportion of the population. People are better off with the welfare state than without it. Furthermore, it also performs tasks not undertaken by the market or private philanthropy (that is, universal free health care, housing benefits, child benefits). Thus, there is a demand for its services. This demand does not mean that support for the welfare state is unanimous, as development has been gradual and has faced resistance from influential parts of the electorate. Mutual advantage also contributes to our understanding of why the welfare state appears relatively stable and resilient to retrenchment. Not even Thatcher's attack on the welfare state during the 1980s eradicated it in the UK. An important implication of mutual advantage element is that mutual advantage turns social conflict away from confrontation and towards cooperation. The idea is no longer that social cooperation represents a zero-sum game wherein one's loss is another's gain, but that cooperation increases productivity and "the national capital" (Marshall 1981, 91).7 Thus, the welfare state is surprisingly resilient and a relatively stable political institution, in part because it benefits large portions of the electorate.

Contrary to my argument here, Russell Hardin has argued that the welfare state (or welfare liberalism, as he labelled it) is not based on mutual advantage. His reasons are that the welfare state involves redistribution, which inevitably leads to conflict between social classes (Hardin 1999, 326) and because welfare programs "are not likely to be mutually advantageous for the most politically important groups in liberal societies. They do not serve the interests of the middle class and the politically influential entrepreneurial class" (330). I believe the broad coalitions identified by Esping-Andersen, which initially created the welfare state, undermine Hardin's argument concerning the middle class, at least in Europe. Saying that the welfare state mutually benefits the politically effective groups in society does not mean that mutual advantage is a moral justification for the welfare state. Instead, it means that mutual advantage contributes to our understanding of how the welfare state operates.8

To summarize this section, I have argued that social equality represents an important aim of the welfare state both in historical and contemporary understandings of it, and that mutual advantage contributes to our understanding of how it operates. Consequently, it seems that egalitarian concerns are not at the heart of the welfare state. Instead of relying on a strictly egalitarian foundation, the welfare state is content with reversing some of the gravest inequalities caused by industrial capitalism. This insight brings into question whether the welfare state should be associated with liberalism and is discussed in more detail in the next section.

III. RAWLS'S DISSOCIATION FROM THE WELFARE STATE

In this section, I compare Rawls's theory of justice with the ideas discussed above. In his later works, Rawls attempted to dissociate his theory from the welfare state because the welfare state emphasized ex-post redistributive schemes and allowed for potentially large economic inequalities.9 However, my comparison will show that one of Rawls's principles of justice-the difference principle-resembles to some extent the thinking underlying the welfare state and, I argue, makes it harder for Rawls to dissociate his theory from it.

Welfare-state theory is committed to social equality understood as equal citizenship rights. The first of the two principles of justice in Rawls's theory shares this concern for the importance of equal citizenship rights and states that "each person is to have an equal right to the most extensive total system of equal basic liberties compatible with a similar system of liberty for all" (1999, 266). However, Rawls notes that relying solely on equal citizenship rights is dissatisfying in a modern constitutional democracy. A flaw with modern democracies has been that wealthy and influential citizens and groups of citizens have exercised a disproportionately strong influence on democratic decision-making (Rawls 1999, xv). To remedy this flaw, Rawls' liberal egalitarianism aspires to extend beyond mere social equality.

For liberalism, the welfare state's notion of social equality is a necessary but not sufficient condition to create justice. Social equality, according to Rawls, must be supported by some kind of economic equality to fulfill what justice requires. The second principle of justice addresses economic equality and states that "social and economic inequalities are to be arranged so that they are both: (a) to the greatest benefit of the least advantaged, consistent with the just savings principle, and (b) attached to offices and positions open to all under conditions of fair equality of opportunity" (Rawls 1999, 266). The second principle implies that Rawlsian liberalism permits economic inequalities but only if they promote the interests of society's worst off and maintain equal opportunities. To determine the relationship between Rawls's theory and the welfare state, it is necessary to analyze what (a) and (b) entail. I will begin by discussing (b), the fair equality of opportunity, before discussing (a), the difference principle.

The idea of fair equality of opportunity in Rawls's theory states that from a moral point of view, natural talents, abilities, and social circumstances are arbitrary. This argument leads Rawls to suggest that social class should not impair a person's opportunity to rise in society. Rawls, therefore, held that to realize equal opportunity, the basic structure of society should ensure that "those who are at the same level of talent and ability, and have the same willingness to use them, should have the same prospects of success regardless of their initial place in the social system, that is, irrespective of the income class into which they are born" (1999, 73). Such compensation surpasses the notion of mere social equality and a minimum welfare state, instead requiring some form of economic equality to equalize opportunities. Social equality as understood in welfarestate theory is not sufficiently egalitarian to satisfy this aspect of Rawlsian liberalism, and Rawls seems correct in wanting to dissociate his theory from the welfare state.

Furthermore, the welfare state is not only less egalitarian than liberal justice but also relies on a different strategy for redistribution. The welfare state is committed to ex-post redistribution because the aim is to intervene and redress current inequalities. Therefore, Rawls is correct in observing that the welfare state "may allow large and inheritable inequities of wealth incompatible with the fair value of the political liberties" (1999, xv). Rawls's preferred alternative to the welfare state is a propertyowning democracy, which emphasizes ex-ante redistribution such as "the steady dispersal over time of the ownership of capital and resources by the laws of inheritance and bequest" (xv).10 The main difference between these two methods of redistribution is that the welfare state requires existing inequalities to be rectified, while Rawlsian justice requires that those inequalities do not emerge in the first place.

However, more must be said about the social institutions of a just society. Some of what Rawls says appears to obscure the distinction between ex-ante and ex-post distributions. According to Rawls, special branches of government are responsible for providing the social minimum and for the necessary taxation of both inheritance and income to "satisfy the difference principle" (1999, 245-246). These special branches distribute economic resources through ex-post redistribution and resemble the welfare state in their operations. These special branches and institutions appear to pull Rawls closer to the welfare state than his theory would seem thus far. In a recent paper Martin O'Neill examines the distinction between predistribution (ex-ante) and redistribution (ex-post) and argues that regarding these two types of distributions "insofar as there is an important distinction here, it has to be with regard to the aims and effects of policy, rather than in terms of the policy tools or mechanisms that are used" (O'Neill 2020, 85). Thus, relying on a firm distinction as Rawls seems to do may be more an analytic distinction rather than anything else. This underlines the difficulties confronting Rawls in dissociating from the welfare state.

Despite the similarities in both language and content between the welfare state and the institutions just described, two important differences remain. First, a Rawlsian property-owning democracy aims for the widespread ownership of capital and land. Second, Rawls is not merely concerned with the social inequalities arising from industrial capitalism and what justice requires in general, but he is also concerned with mitigating the inequalities caused by bad luck in the initial distribution of skills and talents, which goes beyond the scope of the welfare state and its main focus on social inequalities.

Concerning the difference principle (the first part (a) of the second principle of justice), this principle permits inequalities that make the most disadvantaged group better off than it would be with strict equality. I claim that this principle is more difficult to distinguish from the welfare state because the similarity between the difference principle and the welfare state is more profound than in the case just discussed. The problem with the difference principle is that it agrees with some of the mutualadvantage reasoning underlying the welfare state. The rationale behind this principle is that inequalities may be mutually advantageous both for the least advantaged and for the most advantaged because "society is interpreted as a cooperative venture for mutual advantage" (Rawls 1999, 73-74). Inequalities provide the least advantaged with more social primary goods-income, wealth, influential positions, and self-esteem than strict equality, while the most advantaged group keeps a larger share of its contribution to social cooperation (78-81). Hence, social cooperation is mutually advantageous for both the least advantaged and most advantaged. The result that follows is that social cooperation "is no longer a matter of shuffling about a fixed stock of goods" (66). When the total amount of resources in society increases, a wealthier society results.11

At this point, one might argue that this understanding of the difference principle is skewed by considering the difference principle in isolation and not concerning the first principle and Rawls's wider theory, and that therefore my interpretation might pull the difference principle too far in the direction of an economic understanding of the difference principle. I believe rather than relying on a strong economic understanding of mutual advantage and the difference principle, the latter has some built-in ambiguity regarding its interpretation, and it is this ambiguity that pulls the difference principle in the direction of the welfare state. There are some overtones of the economic thinking associated with the thought expressed by Marshall in the previous section emphasizing that citizens' bodies were part of the national capital. The difference principle agrees with this line of thinking. Rawls expresses the same ideas when he considers society to be "a cooperative venture for mutual advantage" and that society's resources are not "a fixed stock of goods," (Rawls 1999, 7374) ideas that bear a resemblance to Marshall's view. Rawls also seemingly agrees with Marshall that citizens' bodies are part of the national capital, as Rawls sees "the distribution of natural talents as in some respects a common asset" (1999, 87). Mutual advantage pushes Rawls closer to the welfare state and away from the egalitarianism of the first principle and fair equal opportunity.12 These economic overtones are underlined by Rawls's reliance on Pareto-optimality and his general premise that the persons in the original position have access to knowledge among other things about economics. Since economic inequalities are an important source of inequalities in most modern societies it makes sense to understand the difference principle also in an economic setting, although that is not the only context of the difference principle.

The overtones of economic thinking expressed by certain features of Rawls's theory are complemented by interpretations of the difference principle emphasizing solidarity, patriotism, and non-economic features (van Pārijs 2003). This line of thinking associates the difference principle with social equality and the first principle. Multiculturalists such as Will Kymlicka have expressed strong reservations with the type of thinking associated with Marshall and the understanding expressed by van Pārijs above regarding the difference principle and emphasizing solidarity and patriotism. This has sparked an intense debate concerning the relationship between the welfare state, liberalism, and universal citizenship. In the next section, I discuss the difference principle as part of this wider context where both the economic and more social features are included and discussed in considering some of the recent criticisms of universal citizenship.

More recently some theorists have tried to bridge the gap between the welfare state elements and the more radical egalitarian sides of Rawls's theory. One example is Stuart White (2015) who has argued that the welfare state must be supplemented with a scheme of Basic Capital to advance a Rawlsian property-owning democracy. Another example is Tong Zhicago (2015) who argued in a similar vein that a property-owning democracy reconnects the elements that were developed in the New Deal tradition. Combining the welfare state elements and the more egalitarian elements of Rawls's theory is a sophisticated way to overcome some of the difficulties described above. A challenge with both approaches is that the tension between the welfare state elements and the egalitarian elements is still present because the schemes hide the difficulties, rather than resolving them, between liberal egalitarianism and the welfare state.

I have now discussed some of the ideals underlying the welfare state and compared these ideals with the ideals of Rawls's liberalism. We have seen that both share a concern for social equality, but a difference is that liberalism requires not only social equality but also some type of economic equality to ensure equality of opportunity regardless of one's social background. Moreover, the difference principle seems to rely on mutual advantage and pulls Rawls in the direction of the welfare state, making it harder to dissociate.

IV. LIBERALISM, WELFARE, AND UNIVERSAL CITIZENSHIP

The main question this discussion leads up to is whether liberals such as Rawls can successfully dissociate from the welfare state. I will attempt to answer this question by connecting the criticism of universal citizenship with the ideals promoted by the difference principle and the welfare state. This approach will exemplify how the welfare state and liberal egalitarianism promote some of the same ideals and policies. I argue that these policies run counter to the egalitarian commitment of liberalism and thus I conclude that this makes it more difficult for liberals such as Rawls to dissociate completely from the welfare state.

Both the welfare state and Rawls's liberalism rely on an ideal of universal citizenship. Multicultural critics have criticized the notion of universal citizenship because they consider this ideal to be insensitive to the interests of various minority groups. Kymlicka's criticism is well known, and his critique exemplifies why a complete dissociation from the welfare state is difficult for liberals. He identifies Marshall as the intellectual originator of the traditional view of universal citizenship. Kymlicka describes Marshall's concern for health care in the following terms: "ensuring that people had health care and education was important for Marshall not just for humanitarian reasons-that is, to meet basic needs. Social rights would also help integrate previously excluded groups into a common national culture, and thereby provide a source of national unity and loyalty" (2002, 328). The group that particularly concerned Marshall was the English working class, "whose lack of education and economic resources excluded them from enjoyment of this national culture-for example, from knowing and enjoying Shakespeare, John Donne, Dickens, the King James Bible, Cromwell, the Glorious Revolution, and cricket" (Kymlicka 2002, 328). Today this ideal of universal citizenship has come under criticism from multicultural theorists such as Kymlicka because:

Many members of these groups [blacks, women, indigenous peoples, ethnic and religious minorities, gays, and lesbians] feel marginalized, not (or not only) because of their socio-economic status, but also because of their socio-cultural identity-their difference. [...] They demand these group-specific forms of citizenship either because they reject the very idea that there should be a single common national culture or because they think that the best way to include people in such a common culture is through differentiated citizenship rights. (Kymlicka 2002, 329-330)

According to Kymlicka, the traditional view of citizenship should be abandoned because it leads to a notion of social equality where a majority group enjoys a dominant role and possibly alienates and marginalizes minority groups.

The crucial point here is that the difference principle could allow for the inequalities permitted by Marshall's notion of citizenship, criticized by Kymlicka. This outcome emerges if the most disadvantaged group is better off with these inequalities than without (and there was no violation of first principle rights or the principle of fair equality of opportunity). For example, one could argue that economic arrangements more accommodating to able-bodied males would improve the well-being of the most disadvantaged group because this arrangement would increase society's productivity and efficiency. Women, old people, and (possibly) ethnic minorities would benefit from these inequalities because they would receive more social primary goods than without the inequalities. Providing ablebodied males with advantages and incentives that enable them to contribute to economic growth that can be redistributed is in line with the difference principle because the worst-off group becomes better off by implementing such schemes.

These implications are acknowledged by some of the interpretations of the difference principle. Philippe van Pārijs, for example, asserts that the difference principle:

Must not shy away from resolutely designing institutions that foster an ethos of solidarity, of work, indeed of patriotism-not of course because of the intrinsic goodness of a life inspired by such an ethos but because of its crucial instrumental value in the service of boosting the lifelong prospects of the incumbents of society's worst position, (van Pārijs 2003, 231)

This account of the difference principle is reminiscent of Marshall. The result of fostering such an ethos is precisely the promotion of what Kymlicka criticizes above and what makes it difficult for liberals to dissociate from the welfare state by stifling groups with weak ties to the labor market, especially if the ideals are too closely attached to social equality or promote a certain national culture as superior to others.

Because the difference principle justifies social and cultural inequalities if they make the worst-off group better off, it becomes difficult for Rawls to rule out such practices as unjust in principle.13 Rawls's theory contains overtones of the economic reasoning that Marshall expresses, but also an egalitarian ideal of equal opportunity. Marshall on one side and Kymlicka on the other express these two different modes of thinking within Rawls's theory. As elements in Rawls's theory are associated with both of these very different ideals, his theory produces a difficult tension. As a result, he may wish to dissociate from the welfare state, but distancing may be difficult because of the similarity between the difference principle and the welfare state.14

My conclusion is therefore that the shared reliance on mutual advantage found in both the difference principle and the welfare state makes it difficult for Rawls to completely dissociate from the welfare state. Before ending this part, I will discuss two possible objections against my conclusion. The first objection takes issue with my portrayal of the welfare state as in opposition to multiculturalism. The confrontation between Marshall and multiculturalism indicated that there is a trade-off between the welfare state's emphasis on redistribution and multiculturalism's emphasis on recognition of difference. Over the last two decades, this trade-off has been subject to debate between multiculturalists such as Will Kymlicka, Keith Banting, Richard Johnston (Banting and Kymlicka 2006; Johnston et al. 2010; Johnston et al. 2019) and liberals like David Miller (2006, 2019) as some of the protagonists.

The multiculturalist position argues that the trade-off is exaggerated and that it is not necessary to choose between multicultural policies and welfare policies. Some empirical research discussing such a trade-off argues that such a trade-off does not exist because the general trend in Western democracies is that multicultural policies are maintained and that the welfare state still enjoys widespread support (Banting and Kymlicka 2006). Hence, my presentation of the welfare state as focused on productivity and a single national identity misrepresents the modern welfare state.

In a more recent analysis, the multiculturalist position has been refined to

[cjonclude that national identity does increase support for the welfare state [...] However, we also find that identity plays a narrower role than existing theories of liberal nationalism suggest [...] This leads us to suggest an alternative theory [...] that suggests that the relationship is highly contingent, reflecting distinctive features of the history and national narratives of each country. (Johnston et al. 2010, 351)

Although the analysis above is restricted to Canada, the refined conclusion amounts to acknowledging that the trade-off is real although subject to qualification. In a more recent analysis, the trade-off is further downplayed and claims that multicultural policies are not undermining solidarity in the case of Canada (Weaver 2019; Johnston et al. 2019).

Responding to these claims, David Miller has argued that Canadian multiculturalism is more the exception than the rule as in most of Europe multiculturalism is considered a mistake (Miller 2019). Although some multicultural policies may be operating and successful, multiculturalism as an ideology is dead (2019). A multicultural ideology is defined by Miller as thinking "that cultural diversity is intrinsically valuable, that it matters greatly to people that their cultural identities should be protected, and that the state should grant equal recognition to each group culture that is substantially present on its territory" (2019, 322). An important criticism of this understanding of multiculturalism "claim that multiculturalism has had the effect of fragmenting the progressive coalition that in the period following World War II was able to tame the capitalist economy through enhanced workers' rights, redistributive taxation, and the provision of essential services on a non-market basis, creating a form of social citizenship as theorized by T. H. Marshall and others" (329).

As the discussion above suggests, the discussion on the relationship between the welfare state and multiculturalism has been ongoing for about two decades and has not been resolved, as the exchange between Kymlicka (and others) and Miller shows. Marshall, as referenced by Miller above, envisioned a shared national culture and identity as a means to create wide coalitions supporting social citizenship rights. The criticism of multiculturalism is that such broad coalitions become difficult because a shared identity and culture are rejected by multiculturalism as an ideology. My main point in emphasizing this discussion is to draw attention to the fact that this issue is far from being settled and reveals why liberal egalitarians have a difficult relationship with the welfare state and that the discussion above of social citizenship and the difference principle exemplifies this difficulty.

A second objection to the thesis advanced in this paper states that the egalitarian social-democratic welfare state found in Northern Europe and Scandinavia has approximated an egalitarian society. Some research suggests that the social-democratic welfare state is both efficient in reducing poverty and enhancing equality (Goodin et al. 1999; Ringen 2007). The social democratic welfare state is known for narrow income distribution, low levels of poverty, and extensive civil, political, and social rights. These features indicate that the social-democratic welfare state comes close to satisfying the egalitarian ambitions of Rawls's theory. Thus, the welfare state and Rawls's theory not only agree on mutual advantage but also in terms of realizing an egalitarian society. Consequently, the picture painted in this paper of a difficult relationship between the welfare state and Rawlsian liberalism is misleading when comparing Rawls with the social-democratic welfare state.

Despite the success of the social-democratic welfare state, important differences remain between this regime and Rawlsian liberalism making it difficult to equate the welfare state and liberalism. The social democratic welfare state aims to intervene in the capitalist economy: "The goal is one of redistribution. For social democrats, the point and purpose of the welfare state, narrowly conceived, is to transfer resources-goods and services, and income and wealth more generally-from the richer to the poorer members of society" (Goodin et al. 1999, 50). This passage demonstrates how the welfare state emphasizes ex-post redistribution in contrast to ex-ante redistribution in a property-owning democracy. For this reason, it is wrong to equate the welfare state and liberalism. Similarities do exist between the welfare state and liberalism, as discussed in this paper, but they are insufficient to equate the two.

V. CONCLUSIONS

One of the main differences between the welfare state and liberalism is that liberalism harbors more egalitarian ambitions than the welfare state. Schemes such as the equality of opportunity and ex-ante redistribution also distinguish liberalism from the welfare state. In contrast, the difference principle pulls liberalism toward the welfare state and makes a liberal dissociation from the welfare state difficult. Two implications follow from these conclusions. First, liberals have commonly defended the welfare state and considered it, despite its weaknesses, as an important means for realizing social justice in the real world. The welfare state can, of course, be defended for such pragmatic reasons. However, such a pragmatic defense of the welfare state runs the risk of diluting the commitment to equality because of the close relationship between capitalism and the welfare state. Second, I believe that the difficulties Rawlsian liberalism experiences in dissociating from the welfare state should urge liberals to move beyond a Rawlsian framework and re-evaluate the institutional implications of their theories if they are serious about realizing their ideal of distributive justice.

Sidebar

Harald Borgebund is Associate Professor in Political Science at Østfold University College and NLA University College (both in Norway). His research and teaching interests are liberal political theory, democratic education and democratic theory.

Contact e-mail: <[email protected]>

Footnote

1 See, for example, Barry (1973) and Clark and Gintis (1978) for some of the earlier arguments for associating the welfare state with liberalism. One more recent example is Valiler (2015) who argued that the welfare state does a better job than more egalitarian institutions.

2 See Journal of Social Philosophy 40 (3), 2009, for a special issue on Rawls's theory and a property-owning democracy. See especially Martin O'Neill's (2009) paper, 'Equality and Property-Owning Democracy', 379-396, for one of the few papers discussing the welfare state and Rawls's liberalism.

3 For contemporary discussions of Marshall see Goodin (1988), Moon (1986, 27-52), Esping-Andersen (1990), Goodin et al. (1999), Klausen and Wolfe (1997), and Kymlicka (2002). Kymlicka's account will be discussed in more detail in section four of this paper. Myles and Quadagno (2002) argues that there are important similarities between the early theorists of the welfare state and contemporary theorists.

4 See also Miller (1999) and Ringen (2007). Miller is not a traditional social-democratic welfare-state theorist such as Marshall; nevertheless, he emphasizes the nation state.

5 See White (2003) for an egalitarian justification of economic citizenship based on reciprocity.

6 This phrase is borrowed from Hardin (1999).

7 This view may sound simplistic, as in contemporary societies marked by religious, ethnic, and social diversity (among others), political and social conflicts are no longer associated solely with class struggles. I discuss this point in section four.

8 Mutual advantage can be understood in two different ways: either as a normative thesis or as an explanatory thesis. The normative thesis implies that mutually beneficial agreements are considered morally justifiable because they are mutually advantageous. See Gauthier (1986) for an example of a normative mutual-advantage theory. The explanatory thesis means that a mutually advantageous agreement is a necessary but insufficient condition for an agreement to be morally justifiable. See Hardin (1999) for an example of an explanatory mutual-advantage theory. In this paper, mutual advantage is understood as an explanatory thesis.

9 Rawls (1999: xv-xiv). See also Rawls (2001) for a discussion of the relationship between welfare-state capitalism and his theory of justice.

10 Rawls (1999, xv) leaves open whether his theory's principles are best achieved through a property-owning democracy or a liberal socialist regime. However, Rawls simultaneously states that the difference principle ought to be considered in light of a propertyowning democracy to comprehend the full force of this principle. Furthermore, in ch. 5 of A Theory of Justice, in which the institutions of the basic structure are discussed, the emphasis is on a property-owning democracy rather than other economic systems.

11 The difference principle does not require economic growth per se, but Rawls can be interpreted to say that all else being equal, a state with economic growth is superior to a situation with negative contributions from those who are better off (1999, 68). See also Rawls (2001) and Freeman (2007) for a discussion of the relationship between the difference principle and economic growth.

12 See Barry (1989) ch. 6 for an illuminating discussion of the mutual-advantage aspect of the difference principle.

13 Rawls might reply that such inequalities would violate the equality-of-opportunity principle. If that is the case, then the scope for the difference principle seems very limited, and it is not clear what role it plays in Rawls's theory.

14 The incentives discussed in relation to the application of the difference principle have tended to be purely economic, such as higher wages. However, the incentives could equally well be of a more 'cultural' character, such as those discussed in this section. According to Freeman (2003,112-113), Rawls "says a society is not required to maximize the expectations of the least advantaged measured in terms of income and wealth [...] but also their opportunities for powers and positions of office, non-basic rights and liberties, and the institutional bases of self-respect". Therefore, 'cultural' incentives can be invoked to enhance self-respect and opportunities for the most-disadvantaged members of society.

References

REFERENCES

Ackermann, Bruce, and Anne Alstott. 1999. The Stakeholder Society. New Haven: Yale University Press.

Banting, Keith, and Will Kymlicka (eds). 2006. Multiculturalism and the Welfare State - Recognition and Redistribution in Contemporary Democracies. New York: Oxford University Press.

Barr, Nicholas. 1987. The Economics of the Welfare State. London: Weidenfeld and Nicolson.

Barry, Brian. 1973. The Liberal Theory of Justice. Oxford: Oxford University Press.

Barry, Brian. 1989. Theories of Justice. California: The University of California Press.

Barry, Brian. 1995. Justice as Impartiality. Oxford: Oxford University Press.

Beveridge, William. 1942. Social Insurance and Allied Services. London: H.M.S.O.

Clark, Barry, and Gintis Herbert. 1978. "Rawlsian Justice and Economic Systems." Philosophy and Public Affairs 7 (4): 302-325

Esping-Andersen, Gosta. 1990. The Three Worlds of Welfare Capitalism. Cambridge: Polity Press.

Freeman, Samuel (ed). 2003. The Cambridge Companion to Rawls. New York: Cambridge University Press.

Freeman, Samuel. 2007. Rawls. London: Routledge.

Garland, David. 2016. A Very Short Introduction to the Welfare State. Oxford: Oxford University Press.

Gauthier, David. 1986. Morals by Agreement. Oxford: Oxford University Press.

Goodin, Robert E. 1988 Reasons for Welfare. Princeton NJ: Princeton University Press.

Goodin, Robert E., Bruce Heady, Ruud Muffels, and Henk-Jan Dirven. 1999. The Real Worlds of Welfare Capitalism. New York: Cambridge University Press.

Gutmann, Amy (ed). 1986. Democracy and the Welfare State. Princeton NJ: Princeton University Press.

Hobsbawm, Eric. 1994. The Age of Extremes. London: Abacus.

Hardin, Russell. 1999. Liberalism, Constitutionalism, and Democracy. New York: Oxford University Press.

Heath, Joseph. 2011. "Three Normative Models of the Welfare State." Public Reason 3 (2): 13-43.

Johnston, Richard, Keith Banting, Will Kymlicka, and Stuart Soroka. 2010. "National Identity and Support for the Welfare State." Canadian Journal of Political Science/Revue Canadienne de science politique 43 (2): 349-377.

Johnston, Richard, John Myles, Elizabeth Goodyear-Grant, and Will Kymlicka. 2019. "Introduction: The Progressive's Dilemma." In Federalism and the Welfare State in a Multicultural World, edited by Richard Johnston, John Myles, Elizabeth GoodyearGrant, and Will Kymlicka, 1-17. Montreal & Kingston: McGill-Queen's University Press.

Klausen, J., and A. Wolfe. 1997. "Identity Politics and the Welfare State." Social Philosophy and Policy, 14 (2): 231-255.

Kymlicka, Will. 2002. Contemporary Political Philosophy - An Introduction. Oxford: Oxford University Press.

Marshall, Т.Н. 1981. The Right to Welfare and other Essays. London: Heinemann Educational.

Marshall, Т.Н. 1992. Citizenship and Social Class. London: Pluto Press.

Miller, David. 1999. Principles of Social Justice. Cambridge MA: Harvard University Press.

Miller, David. 2006. "Multiculturalism and the Welfare State: Theoretical Reflections." In Multiculturalism and the Welfare State - Recognition and Redistribution in Contemporary Democracies, edited by K. Banting, and Will Kymlicka, 323-338. New York: Oxford University Press.

Miller, David. 2019. "The Life and Death of Multiculturalism." In Federalism and the Welfare State in a Multicultural World, edited by Elizabeth Goodyear-Grant, Richard Johnston, Will Kymlicka, and John Myles, 319-338. Montreal & Kingston: McGillQueen's University Press.

Moon, Donald. 1986. "The Moral Justification of the Welfare State." In Democracy and the Welfare State, edited by Amy Gutmann, 27-52. Princeton NJ: Princeton University Press.

Myles, John, and Jill Quadagno. 2002. "Political Theories of the Welfare State." Social Service Review 76 (1): 34-57.

O'Neill, Martin. 2009. "Liberty, 'Equality and Property-Owning Democracy'." Journal of Social Philosophy 40 (3): 379-396.

O'Neill, Martin. 2020. "Power, Predistribution, and Social Justice." Philosophy 95: 63-91.

Pārijs, Philippe van. 2003. "Difference Principles." In The Cambridge Companion to Rawls, edited by Samuel Freeman, 200-241. Cambridge: Cambridge University Press.

Rawls, John. 1999. A Theory of Justice. Oxford: Oxford University Press.

Rawls, John. 2001. Justice as Fairness - A Restatement. Edited by Erin Kelly. Cambridge, MA: Harvard University Press, Belknap.

Ringen, Stein. 2007. What Democracy is For. Princeton, NJ: Princeton University Press.

Roemer, John. 1999. "Egalitarian Strategies," Dissent. 64-74.

Sterba, James. 1988. How to Make People Just: A Practical Reconciliation of Alternative Conceptions of Justice. Totowa, NJ: Rowman and Littlefield.

Tawney, R. H. [1931)1951. Equality. London: George Allen & Unwin.

Titmuss, Richard. 1976. Essays on "The Welfare State." London: George Allen & Unwin.

Vallier, Kevin. 2015. "A Moral and Economic Critique of the New Property-owning Democrats: On Behalf of a Ralwsian Welfare State." Philosophical Studies 172 (2): 283304.

Weaver, R. Kent. 2019. "Political Institutions and the Welfare State in Canada and the United States." In Federalism and the Welfare State, edited by Elizabeth Goodyear - Grant, Richard Johnston, Will Kymlicka, and John Myles, 151-171. Montreal & Kingston: McGill-Queen's University Press.

White, Stuart. 2003. The Civic Minimum - On the Rights and Obligations of Economic Citizenship. Oxford: Oxford University Press.

White, Stuart. 2015. "Basic Capital in the Egalitarian Toolkit?" Journal of Applied Philosophy 32 (4): 417-431.

Zhichao, Tong. 2015 "Rawlsian Property-Owning Democracy: An American Historical Interpretation." American Political Thought 4 (2): 289-310.