7. Why Limitarianism?

Ingrid Robeyns

This article discusses ‘limitarianism’, which in its most general formulation is the idea that in the world as it is, no one should have more than a certain upper limit of valuable goods, in particular, income and wealth. What, if anything, does ‘limitarianism’ add to normative political philosophy?

In Section 1, I describe the context in which limitarianism has been introduced. Section 2 will provide a more detailed statement about limitarianism, including some more recent contributions to and developments in the literature. In the next two sections, I discuss egalitarianism (Section 3) and sufficientarianism (Section 4) and ask whether they can do what I envision to be the task of limitarianism. Section 5 argues that within theories of distributive justice, limitarianism is best seen as part of a pluralist account. This is illustrated by sketching the proposal of a pluralist account combining sufficientarianism, opportunity egalitarianism, and limitarianism. Section 6 concludes by pulling everything together, and will give an answer to the question of what limitarianism contributes to normative political philosophy.1

1. The Context of Limitarianism

In ‘Having Too Much’, I introduced limitarianism, which in its most general formulation is the idea that in the world as it is, as well as in the most nearby possible worlds, no-one should have more than an upper threshold of valuable goods.2 These valuable goods could be various types of scarce goods, and their distribution could be a concern at the macro or micro level. In this article, my focus will be on personal holdings of money—income and wealth in particular.

Much of the writing on material inequalities focuses on the position of the worst off, and makes a distinction between the poor or disadvantaged and those who are not poor. Economic limitarianism changes that two-tiered categorization into a three-tiered distinction between (1) those who are poor or deprived; (2) those who are not poor, but also not very rich; and (3) those who are very rich. This reflects the material basis of the classic distinction in sociology between the working class, the middle class, and the upper class. Making such a three-layered distinction makes it possible to analyse what we owe to persons in each of those three groups, and also what persons in those groups owe to others, in a way that a distinction between only the poor and the non-poor doesn’t allow us to do. In particular, since limitarianism holds that, in the present and nearby possible worlds, holding large concentrations of wealth has no overall beneficial effects, we should morally prefer limitarianism to a situation in which some have too much.3 The beneficial effects of this shift could be various, including contributing to meeting the urgent needs of others, addressing collective action problems such as the funding of effective action for climate change, or protecting democratic values.4

For political philosophers engaged in interdisciplinary research on inequalities and distributive justice, the shift in focus to the upper tail of income and wealth distribution shouldn’t be surprising. Economists have been documenting rising economic inequalities, in particular the strong concentration of wealth among the rich, for some time now.5 These data show that inequality is rising in almost all countries, because the very rich are getting even richer. Limitarianism as an idea is a contribution to the debate that there is something wrong with not just inequality in general, but with wealth concentration in particular. Limitarians believe that this can best be captured by introducing the idea of limitarianism separately, rather than understanding what is wrong with a rising concentration of wealth under the broader banner of egalitarianism.6 In that respect, it involves a symmetrical move from a concern with the elimination of poverty, which, ceteris paribus, would also reduce inequality. For the elimination of poverty too, specific reasons have been given that are not all reducible to egalitarian concerns.7

One aspect of the discussion on limitarianism is the question of what kind of theorizing it exemplifies. Limitarianism has been proposed for ‘the world as it is’.8 In the world as it is, there is hunger, destitution, and disadvantage; many people are unable to flourish; there are major collective action problems that require our collective attention, or collective goods that are endangered (first and foremost biodiversity and a stable climate); and large holdings of wealth allow their possessors to disproportionally influence politics and policy-making, to engage in forms of consumption that are highly polluting, and, in some countries, even to buy citizenship.9

I have proposed limitarianism for the present and nearby possible worlds, in which such injustices and instances of ill-being occur. Suppose we lived in a different world, in which there are no societal harms, everyone is fully flourishing, there are no injustices or cases of ill-being that can be addressed by human intervention, and concentrations of money can no longer enable corruption or the buying of political influence. In that blissful world, limitarianism would not demand any redistribution of resources.10 Similarly, limitarianism (as I proposed it) does not apply to people living in worlds that are not connected. What matters is the distribution of wealth in comparison with others in the here and now; it is a comparative notion. The other thing that matters is that people are in some way connected; it is a relational notion.11 It is not the absolute level of wealth in itself that matters independent of what others have and independent of the overall state of affairs we find ourselves in. Hence, if, after a nuclear war, everyone dies except one person, who is a billionaire, limitarianism would not judge that to be an unjust situation. Limitarianism would simply not apply.12

Thus, limitarianism, at least as I proposed it, should be understood as a contribution to problem-driven philosophy as opposed to a contribution to theory-driven philosophy.13 I introduce these labels here, as I hope they are a helpful formulation for getting the relevant distinction across. It is important to acknowledge, however, that the distinction between these two broad strands in political philosophy is a long-standing one that offers different perspectives to political philosophers regarding what they take to be their primary tasks, and, derived from this, the appropriate methods they will use and the attitudes they will take to engaging with empirical work.¹⁴

Theory-driven political philosophy engages with the theoretical work of other philosophers (or one’s own earlier work) and essentially refers to and engages with that body of work. Much of the literature on theories of distributive justice over the last few decades has been theory-driven political philosophy—political philosophers writing about how their view is different from another philosopher’s view, or how an existing view could be improved or challenged theoretically.

This discussion also has implications for my response to Robert Huseby’s argument, which I address in this article, especially in Sections 3 and 4. Huseby claims that limitarianism is in essence a combination of intrinsic egalitarianism and intrinsic sufficientarianism. In the next sections, I will deny these claims; but it is important to stress that in the work on limitarianism I have developed so far, it has been presented as a contribution to problem-driven rather than theory-driven political philosophy. Nevertheless, in Sections 3 and 4, I will go along with the theory-driven frame, since this is how I read the methodological commitments underlying Huseby’s criticism. In Section 5, I will revisit these methodological concerns and explain why I believe that limitarianism has an action-guiding force for some real-world-problems that egalitarianism and sufficientarianism on their own do not have.

2. Limitarianism Restated and Refined

Before addressing Huseby’s concerns in the next sections, I want to summarize what I have said about limitarianism in earlier work, as well as point to some other recent developments in the literature.15

Let us start with the core concepts. The limitarian view comes with four core concepts: first, the twin concepts of surplus money (or wealth) and the riches line, and second, the twin concepts of excess money and the limitarian threshold. The limitarian threshold is the line above which limitarianism claims no one should be situated: either there should be policies that make sure people are not situated above the limitarian threshold (wages regulation, caps on the lifetime inheritance level, and so on), or there should be redistribution that takes away the money above the limitarian threshold to reallocate it below the threshold or to use it for funding public goods. Excess wealth is the money a person has above the limitarian threshold and which limitarianism claims the person should not have.

The notions of the riches line and surplus wealth are special cases of the more general notions of the limitarian threshold and excess wealth, if we use the value of flourishing as the criterion to determine the limitarian threshold. The riches line is the level of wealth accumulation at which, at some point of increasing wealth, there is no additional contribution of additional wealth to one’s flourishing. Flourishing is taken to be a satiable concept, and once one arrives at a certain level of wealth, it is assumed that more money will not significantly further the value of one’s flourishing, which could either decrease or (asymptotically) stay at the same level. I have not worked out this account of flourishing in great detail so far, but to the extent that it is worked out, it is clearly an objective-list account of flourishing rather than a hedonistic or desire-fulfilment account. I have also argued that it should cover only the material side of flourishing. Moreover, it is a political or public account, in the specific sense that it is an account of the value of flourishing used for decision-making in the public realm and for the design of institutions and practices. An important assumption I am endorsing is that, on that account of the value of flourishing, the marginal contribution of money to that value declines, and it either becomes asymptotically zero or at some point becomes zero or negative. The value of flourishing, in this sense, is taken to be satiable with respect to money. Surplus wealth is all the wealth a person has above the riches line; it is wealth that cannot contribute to such an objective and public account of flourishing.16

It should be noted that in my work on limitarianism so far, no distinction has been made between the riches line and the limitarian threshold, or between excess money and surplus money. Yet discussions with critics and collaborators have made it clear that the limitarian threshold, above which we take away and reallocate money, and the riches line, being the line at which surplus wealth kicks in, need not be the same.17 Different reasons for limitarianism could point to different limitarian thresholds. For example, the limitarian threshold might be set at the level at which additional increases in personal wealth jeopardize one’s moral autonomy, or at the level at which additional wealth poses a threat to democracy. It is conceptually possible, and in practice plausible, that these different criteria will lead to different limitarian thresholds from the threshold at which the value of flourishing becomes significantly close to zero.

Dick Timmer has argued that limitarians should not assume that the marginal value of wealth becomes zero or negative, and that they could accept that there is still flourishing above the limitarian threshold, but that this is outweighed by other moral concerns.18 In my view, because the account I defended is not a first-person account of flourishing but a public or political one, this can be accounted for. The account could recognize cases in which surplus wealth could still further someone’s personal flourishing, but introducing the distinction of the political account of flourishing allows us to collectively decide that the value of that marginal contribution becomes zero. In other words, there might well be cases where flourishing itself, on that person’s own assessments, is still increasing, but the value of flourishing, as decided by the political community, is zero.

A second clarification on limitarianism relates to the recipients of redistribution. A motivation for limitarianism, as well as an implication of it, is locating those bearing the costs of redistribution more centrally in the discussion. The literature on distributive justice has been primarily recipient-oriented and has therefore paid relatively little attention to the question of who holds the duties of justice, who will pay the costs of redistributive policies, and who could be the agents of justice.19 David Miller has argued along similar lines that there is a general agreement that certain moral needs need to be met—from saving people from famines to rescuing someone in direct danger—but that political philosophers have not argued in great detail about who bears the responsibility for meeting those needs.20 Limitarianism wants to bring the cost-bearers of policies that have redistributive effects more centrally into view. As I will argue in Section 4, this is important for the argument about why limitarianism cannot be reduced to sufficientarianism.

A third point of clarification is that it does not follow from limitarianism that those who are situated below the riches line will not have to be net contributors to redistributive policies or to the funding of valuable public goods. In fact, my intuition is that the total amount of money needed to address all these ills (X) is larger than the total amount of surplus wealth (Y). Robert Huseby thinks that this is quite a bold empirical claim.21 I strongly doubt that, given that the unmet urgent needs are not only poverty and hunger, but also other forms of disadvantage and suffering (e.g. homelessness, children in need of special needs education, social exclusion, loneliness, and functional illiteracy), as well as various collective challenges, such as climate change and other ecological problems. There are currently very few, if any, countries that are meeting all these needs and collective challenges. Taken together, the problems are vast. In any case, my assumption that X>Y explains part of the intuitions I had that prompted limitarianism: if the resources required to meet these unmet urgent needs are so vast that X>Y, then we should start by addressing those needs by using the money that is not used for people’s flourishing, and if the aggregate surplus money is not enough to address all these injustices, unmet urgent needs, and societal challenges, then we can call upon the middle classes to also contribute to addressing them.

Fourth, so far I have defended two arguments for limitarianism, although I have stressed that I do not think these two arguments are necessarily exhaustive in terms of what could ground limitarianism.22 The first argument is the unmet urgent needs argument. As I indicated above, unmet urgent needs are related not just to poverty, but also to forms of disadvantage, isolation, and stigmatization, and to mental ill-health, as well as to collective action problems and public goods that are not sufficiently addressed, such as restoring biodiversity or climate-adaptation policies. The unmet urgent needs argument states that if there are interventions (whether by the state or other agents of change) that can mitigate unmet urgent needs and that require financial resources, the surplus money should be used to meet those needs. The other argument is the democratic argument, which states that surplus money is a threat to political equality and that, on the assumption that it is not possible to build institutional walls between the spheres of politics and the sphere of money, we should put limits on how much money people can have.

Fifth, limitarianism is a partial account of justice, which can be combined with different views of what justice requires below the threshold. Clearly, the idea that limitarians would not care about what happens below the threshold is very implausible, especially for those endorsing the unmet urgent needs argument. But there are several different principles that they can endorse below the limitarian threshold. They could, for example, additionally commit to prioritarianism, sufficientarianism, and egalitarianism. And all those principles could be responsibility-sensitive or not; moreover, they could be outcome- or opportunity-based principles. Adding limitarianism to this menu of principles gives us the tools to develop accounts of distributive justice that combine several of these principles rather than defending just one. While such pluralist accounts of distributive justice might perhaps be seen by some as less elegant, they empower philosophical thinking on issues concerning distributive morality. Moreover, pluralist accounts are especially important for action guidance in the real world, since we have reasons to let different distributive rules apply to persons with different levels of wealth.23 I will return to this claim in Section 5, where I suggest what such a pluralist account of distributive justice could look like. But, first, I will explain why limitarianism cannot be reduced to either sufficientarianism or egalitarianism.

3. Egalitarianism and Limitarianism

If we have egalitarianism and sufficientarianism, do we still need limitarianism? According to Robert Huseby, limitarianism is best seen as a combination of instrumental egalitarianism and sufficientarianism, and therefore it is not limitarian as such.24

Let us first look at egalitarianism. In essence, limitarianism cannot be reduced to egalitarianism, because, although there are different understandings of egalitarianism, for each of the plausible understandings there are fatal objections. Egalitarianism can, and does, mean several different things; it is a concept that has to be formulated into a more specific notion before it can be compared with limitarianism, since otherwise we run the risk of making a comparison between limitarianism and a ‘moving target’. I will first show this by responding to Huseby’s argument that the democratic argument for limitarianism is essentially an argument for equality, and then provide a further argument concerning how limitarianism and egalitarianism are distinct.

A. Does Political Equality Provide Support for Limitarianism?

Huseby argues that limitarianism is neither necessary nor sufficient for protecting the value of political equality. I agree that it is not sufficient and have never claimed otherwise. Yet limitarianism not being sufficient for political equality should not bother limitarians, since limitarianism is explicitly stated to be a partial principle of justice. Clearly, there are also non-distributive societal changes that are needed to protect political equality, such as citizens being educated and properly informed, as well as a particular distribution of wealth and other goods below the limitarian threshold. Thus, protecting political equality requires an array of measures.25

According to Huseby, the danger to the value of political equality is not that people have surplus wealth; rather, inequality (in financial terms) is the root problem. I agree that large inequalities are what undermine political equality, and I think it is correct that this might give us reasons to adopt a relative limitarian threshold rather than an absolute one, as in the case of the riches line.26 While the argument from unmet urgent needs leads us to adopt the riches line as the corresponding limitarian threshold, the democratic argument might lead to another limitarian threshold. But that doesn’t make limitarianism superfluous; rather, it requires a modification. Wealth limits are still a means to limit the size of the inegalitarian gap. It is relevant, too, to point out that there is something special about surplus money for democratic purposes, which is that the opportunity cost in terms of flourishing for those who spend it on political influencing (thereby undermining political equality) is zero. Those without surplus wealth who spend money on political influencing are paying an opportunity cost in terms of their own material flourishing, and, moreover, are extremely limited in how much they can spend because they are not super-rich. The democracy-undermining effects of their gifts will be very limited. Meanwhile, given that the fortunes of the super-rich are on a scale reaching many billions of dollars, they can spend massive amounts of money at no real opportunity cost to themselves.

B. Is Egalitarianism the Superior Alternative?

Based on his analysis of the distribution of economic means and the value of political equality, Huseby argues that egalitarianism is the superior alternative to limitarianism. He writes:

the problem is really caused by a discrepancy in the holding of economic means between individuals. In light of this, the most reasonable strategy is to address the discrepancy as such. To do so, one can start at the top end, at the bottom end, or both. Robeyns chooses to start at the top end, without offering arguments for why we should not start at the bottom end … or, most plausibly, why we should not start at both ends simultaneously.27

In response, at least three things should be said. First, this quote ignores the fact that limitarianism has not been introduced as a replacement for any of the other patterns, but rather as a perspective that is currently missing in the philosophical debates on distributive justice.28 The aim is to complement, not to replace. Second, to the extent that the value we want to protect is political equality, starting at the bottom end of the distribution will not solve much. Either people who are poor will use the increase in income to improve their low standard of living or they will be able to make very small donations to political campaigns. In other words, the result of reducing inequality by lifting the worst off up is not the same as the result of limiting what the best off have. Both strategies may be needed for some goals, but for enhancing political equality, the second strategy is more effective. Limitarianism is especially powerful and useful for drawing our attention to the negative effects on certain values that are particularly caused by a concentration of wealth. Third, to the extent that we are concerned with all values that are affected by the distribution of money, it follows from my argument that we should start at both ends simultaneously, as the argument from unmet urgent needs proposes that money should be moved from the top of the distribution to the bottom of the distribution, or that those at the top of the distribution should pay the cost of collective action strategies.

Yet according to Huseby, there is an available alternative that he regards as superior and which will reduce the problem to a greater extent. That alternative is egalitarianism. However, I do not think this claim is true. Egalitarianism does not make limitarianism superfluous, because egalitarianism is underspecified.

For one thing, egalitarianism could be specified as an outcome-based view or as an opportunity-based view. As an opportunity-based view, egalitarianism runs the risk either of being epistemically too demanding to be implementable in the real world or of leading to forms of disrespect, for example because its epistemic requirements lead to shameful revelations.29 Opportunity-based egalitarianism might also allow for inequalities in outcome that have the bad and unjust effects that defenders of sufficientarianism and limitarianism have pointed out. Hence, limitarianism could easily have different implications from egalitarianism when the latter is understood as equality of opportunity.

However, perhaps the most charitable interpretation of Huseby’s arguments is not opportunity egalitarianism, but outcome egalitarianism. Does outcome-based egalitarianism make limitarianism superfluous?

Outcome-based egalitarianism faces many well-known objections. The criticism of outcome-based views led, on the one hand, to opportunity-based views and responsibility-sensitive egalitarian theories and, on the other hand, to defences of relational egalitarianism and sufficientarianism with a low sufficiency threshold. It seems utterly implausible to believe, as a view on distributive justice in the real world, that no inequalities in outcomes could be justified at all. Clearly, people who lead more frugal lifestyles and spend less will save more and will justifiably have more wealth. Working more or doing unpleasant, dangerous, stressful, or otherwise very demanding work can all justify some inequalities in wages and thus in outcomes.30 For those who reject strict outcome egalitarianism, limitarianism allows limits to be put on inequalities in outcomes, while retaining opportunity egalitarianism for other parts of the distribution. In particular, what I find an intuitively very plausible view is having low-level outcome-based sufficientarianism and a limitarian threshold combined with equality of opportunity for those who are situated between the sufficiency and limitarian thresholds. I will briefly return to such a hybrid view in Section 5.

Limitarianism in financial resources is an outcome-based notion. As a consequence, a distributive view that includes limitarianism could be either more demanding or less demanding than an egalitarian view. It could be more demanding when compared with opportunity-based egalitarianism, and less demanding compared with outcome-based egalitarianism. Limitarianism in itself allows for inequalities under the limitarian threshold that an outcome-based egalitarian view does not permit. This makes limitarianism less demanding than such an egalitarian view. On the other hand, if a limitarian account sets the limitarian threshold relatively low, then it might not allow for certain holdings of material wealth that an equality of opportunity view allows for, depending on the level of the riches line and depending on the specification of the opportunity view.

In sum, I agree with Huseby that egalitarianism, as an outcome-based notion, would reduce many problems, including political inequalities, that limitarianism aims to address to a far greater extent than limitarianism. But as an opportunity-based notion, it might not. Moreover, as an outcome-based notion, there are many reasons not to endorse egalitarianism. Limitarianism is one piece of a more comprehensive view that allows us to say that some outcome inequalities may be justified, but that even if these inequalities are justified, the discrepancy cannot become too big.

C. Tensions Between the Two Arguments for Limitarianism

In his discussion of limitarianism, Huseby observes potential tensions between the democratic argument (and the corresponding limitarian threshold) and the argument from unmet urgent needs (and its corresponding limitarian threshold, the riches line). I agree that this was a weakness of my original statement of limitarianism. And these tensions might become more important if we endorse more reasons for limitarian thresholds, depending on whether each reason justifies a different limitarian threshold, and depending on the consequences for other parts of the wealth distribution. Working out what this implies will require more than I can do here, but let me briefly respond to two points.

The first point is that in some contexts, limitarianism might be overdetermined in the sense that when the argument from unmet urgent needs is put into practice, it requires all surplus money, making the democratic argument unnecessary. I agree that this might happen. But while I read Huseby as thinking this is a weakness of my arguments, I think overdetermined moral responsibilities should be welcomed. Problem-driven political philosophy tries to have value for the world as it is, and overdetermined moral claims are easier to put into practice and therefore politically more robust: as long as all those involved in collective decision-making can endorse one of the reasons for limitarianism, they will find each other agreeing to implement the institutional change in question, despite not agreeing on what the single right reason is for doing so.31 To the extent that we want political philosophy to be problem-driven and not merely theory-driven, a principle being overdetermined is a desideratum, not a bug.

Second, as I indicated in Section 2, I agree that the two arguments might well lead to different limitarian thresholds. This insight will require limitarianism to work out how to deal with multiple limitarian thresholds, but I don’t think it is a reason to reject limitarianism. Rather, the analysis should prompt egalitarians, sufficientarians, limitarians, and others to embrace the agenda of developing pluralist, hybrid theories, as will be argued in Section 5 But first I turn to the critique that limitarianism is reducible to sufficientarianism.

4. Sufficientarianism and Limitarianism

Is limitarianism reducible to sufficientarianism? As with egalitarianism, there are many versions of sufficientarianism—for example, different versions can differ in their scope, their metric, the level of the threshold, and whether they are a single-threshold or a multiple-threshold view.32 The comparison between sufficientarianism and limitarianism is not straightforward for at least two reasons. The first reason, which Huseby mentioned in another recent article, is that the version I developed ‘is explicitly non-ideal and resource-oriented, which makes a direct comparison between the two a bit difficult’.33 In addition, although the sufficiency thresholds are often not spelled out in the sufficientarian literature, they range from a threshold at the poverty line, or perhaps at the mean quality of life among the population, to a much higher threshold akin to a fully good life. If we are interested in comparisons of substantive positions and not merely in the structures of a certain concept, we must have a sufficiently clear sense of what, substantively, we are comparing. Otherwise, comparing sufficientarianism and limitarianism will require limitarianism to be compared with a moving target.

My own view is in line with those who think that sufficientarianism captures something regarding obligations of justice to the genuinely badly off: that is, those who are suffering in some important sense, not in a marginal sense. The higher the sufficientarian threshold is, the more sufficientarianism is prone to critiques of overdemandingness and leading to recommendations that create unfairness. There is also another reason why the sufficientarian threshold should be low, which applies to the sufficientarians who support the thesis that once everyone has secured enough, no distributive criteria apply to benefits.34 Shields has argued that these sufficientarians cannot set the threshold high, since it would deprive the view of any distinctive guidance.35 One might construct views that one labels ‘sufficientarian’ and that consist of other distributive principles in the range below the threshold or thresholds, such as prioritarianism; but then we are talking about hybrid views, and Shields is right that the distinctive guidance they offer is not a sufficientarian one. Although, as I will argue in Section 5, I endorse advancing hybrid accounts of distributive justice, for the task of asking whether limitarianism boils down to sufficientarianism, we must compare it with a ‘pure’ and plausible sufficientarian account, not with those that are de facto hybrid views.

Recall that sufficientarianism entails two theses: ‘[t]he positive thesis stresses the importance of people living above a certain threshold, free from deprivation. The negative thesis denies the relevance of certain additional distributive requirements.’36 If sufficientarianism entails only one threshold, and we do not collapse the sufficiency threshold and the limitarian threshold, then sufficientarianism cannot do the work that limitarianism does. For those who believe that the category of ‘the poor’ or ‘the deprived’ is relevant to normative thinking, limitarianism has been introduced to make a distinction between three groups: the deprived; those who are not deprived, but are not fully flourishing; and those who have more resources than they need for full flourishing. It has also been introduced to put more focus on those who have obligations or responsibilities regarding financial support.

Table 1 illustrates that the work limitarianism does, with its explicit focus on who should pay for the costs, cannot be reduced to single-threshold sufficientarianism. In the initial situation, the lowest-level group is deprived, since it needs 20 units of resources to cross the poverty line, or the sufficiency threshold, but only has 10. Where will those 10 units be taken from? That is where limitarianism comes in. Suppose that with 50 units of resources, one is flourishing fully; all additional resources represent surplus wealth. State A is sufficientarian, but not limitarian; State B is sufficientarian and limitarian, and State C is egalitarian. Sufficientarianism, at least to the extent that such a view does not have particular additional principles above the sufficientarian threshold, would be satisfied with a move to A, but limitarianism would not.37

Table 1. Sufficientarian redistributions with and without limitarianism

One might respond that what really matters is the comparison at a more principled level, and that we should look at which values are foundational. But there, too, limitarianism cannot be reduced to sufficientarianism. Let us grant, for the sake of argument, that sufficientarianism can be grounded on an impersonal values, such as ‘the impersonal disvalue of individuals having insufficiently good lives’.38 It is correct that the argument from unmet urgent needs is grounded in the same disvalue. But that is not the only impersonal value that grounds this argument; the other two values focus on the contributors.

The first is the disvalue of waste: since by definition surplus money cannot contribute to the flourishing of the super-rich, it is wasteful to let them spend their money on private yachts if that money can meet urgent needs of the deprived. The second is the principle of financial capacity. In his discussion of distributive responsibilities, Miller discusses the principle of capacity, which states that those who are most able to address the needs or the problem have a greater or first responsibility for doing so. In the present context, we might split this principle of capacity into two parts: those who have the financial capacity to deliver the money that is needed to solve the relevant problems might not be the same people as those who have the agentic capacity to set up and execute the acts if and when they have the necessary financial resources. Limitarianism is also based on the principle of financial capacity: those who have the strongest financial shoulders should be the first to contribute.

What about the comparison with a multi-threshold sufficientarian view? Perhaps, one might argue, a two-threshold sufficientarian view that sets the second sufficiency threshold as equal to the level of full human flourishing makes limitarianism otiose? Such a two-level threshold, whereby there is a lower threshold at the poverty or deprivation level and a second at the limitarian level of a fully flourishing life, seems to be the view that, qua structures, brings sufficientarianism and limitarianism most closely together.

Yet even here the similarity is deceiving. Sufficientarianism states that justice requires everyone to meet those two thresholds, and sufficientarianism in itself doesn’t care about what happens above the thresholds.39 Limitarianism also entails a threshold, but focuses on the people situated above the threshold, claiming that we have reasons to not want people to be situated above that threshold—but it does not say anything about what is needed between the low sufficiency threshold and the higher sufficiency threshold. Indeed, Volacu and Dumitru capture it accurately:

Robeyns is seeking to turn the sufficiency view on its head. Sufficientarians are interested in the beneficiaries of redistributive processes, specifying a threshold of relevancy and attempting to bring as many individuals as possible to that threshold or close to that threshold. Limitarians, by contrast, are primarily interested in the duty-bearers of redistributive processes.40

Sufficiency views are about why everyone should be on or above one or more sufficientarian thresholds; limitarian views are about why everyone should be below a limitarian threshold. Sufficientarian views are about securing that everyone has enough of what matters; limitarian views are about nobody having too many scarce resources. Both are threshold views, but they are asking fundamentally different questions. Sufficiency views are an answer, or a partial answer, to the question ‘What is owed to people?’. Limitarian views are a partial answer to the question ‘What do those above the limitarian threshold owe to others?’. And in answering the latter question, limitarians not only refer to the values of equality, or sufficiency, but can also refer to a plurality of values that need protecting by constraining excess wealth accumulation.

A sufficientarian might still object that, qua structure, limitarianism is a two-level version of sufficientarianism. I would want to resist this conclusion for at least three reasons. First, as I mentioned at the beginning of this section, I side with those who think the strong intuition of sufficientarianism is to focus on meeting a set of basic needs that we want to fulfil for all. Two-level sufficientarianism with additional principles applying to the ranges between thresholds can no doubt be worked out in interesting theories, but they are pluralist or hybrid theories that add additional elements to what is genuinely sufficientarian.

Second, keeping the focus of sufficiency on unconditionally meeting basic needs is especially important for philosophy aiming to address real-world problems, since the poverty line (and addressing disadvantage) is clearly recognised as a separate concern. There are separate societal challenges that arise from the issues caused by wealth concentration. Sufficientarian and limitarian thresholds are indeed all thresholds, but the underlying empirical reality and the problems of poverty or deprivation and that of excess wealth are vastly different.41 Philosophers lose action-guiding power by calling hybrid theories ‘sufficientarian theories’.

Third, conceptually there remains the question of separating out the bearers of the costs from the recipients of the increased benefits. This is especially important for the questions about the distributive ethics of providing public goods—questions that are virtually absent from the sufficientarian literature.

In the end, rather than arguing over whether limitarianism can be reduced to multiple-threshold sufficientarianism, I think more is to be gained from exploring the value of combining those views, to which I now turn.

5. The Role of Limitarianism in Pluralist Theories of Distributive Justice

Limitarianism is explicitly a partial distributive justice rule; it can be combined with rules or patterns below the threshold in order to become a full account of justice. It adds an explicit focus on the bearers of the costs—those from whom the resources for redistribution will be taken or those who will have fewer resources (compared with the pre-limitarian situation) due to measures such as maximum income legislation or a cap on inheritance. But something similar arguably holds for sufficientarianism. For example, Shields has recently argued that sufficientarianism is much better able to respond to the objection that it is implausibly indifferent to what happens above the threshold if it becomes part of a pluralist view that combines sufficiency with prioritarianism, egalitarianism, and/or utilitarianism.42 Shields rightly traces pluralist accounts to John Rawls, and Paula Casal also discusses pluralist accounts in her detailed critique of sufficientarianism.43

In my view, limitarianism can provide sufficientarians with some tools that sufficientarians need to make their account more plausible, and the reverse is equally true. If the sufficientarian threshold is low, limitarianism can help us make a choice between moving everyone above that low threshold by taking resources from those who have more than they need for fully flourishing lives or from those who are not deprived, yet are not fully flourishing, as the numerical example in Table 1 illustrated. If there is a second, higher sufficientarian threshold that is lower than the limitarian threshold, exactly the same questions can be asked, and limitarianism can play exactly the same role in developing an account of distributive justice. If, however, the higher sufficientarian threshold lies at the level of the limitarian threshold, we have a theory that is essentially egalitarian in aspiration.

Sufficiency is essentially recipient-oriented: it deals with what we want people to have. Limitarianism tries, on the one hand, to avoid the harms of excess wealth, and, on the other hand, it is contribution-oriented: it is concerned with who should contribute to the redistributive policies and interventions aimed at collective action problems, and how much they should contribute. Combining one or more sufficientarian thresholds with limitarian thresholds allows sufficientarians to say something about where the resources for the goals of sufficiency will be taken from.

Pluralist accounts of distributive justice can be pluralist regarding the distributive rules that apply to different parts of the distribution, or they can be pluralist in relation to the distributive rules that apply to different metrics of justice. They are also very likely to be pluralist with respect to the underlying values that the proposed patterns will aim to protect. One pluralist view that, in my view, requires further examination is having an account of distributive justice that has three tiers.44 The first tier would be a low-level sufficientarian threshold in terms of basic functionings and capabilities, which makes sure everyone, independent of differential needs, can live a dignified life without suffering poverty or social exclusion. For this first tier, the values of sufficiency and human dignity would be the most important, and values of personal responsibility and efficiency would not play a role. The second tier would impose a limit on financial means (income and especially wealth) so as to protect a range of values, including political equality, non-waste, and non-domination. The third tier would sit between the sufficiency threshold and the limitarian threshold and an incentives-compatible account of equality of opportunity would be its most important value. While I do not have the space here to work out such an account, it is pluralist in terms of its metrics, distributive rules, and values. Given that distributive justice affects multiple values, and that those values conflict in some contexts, a pluralist account of distributive justice as sketched here allows us to give lexical priority to some values in some parts of the distribution.45

This very brief sketch no doubt leaves many questions unanswered, but I hope it can serve to illustrate the point that by introducing limitarianism to our theorizing of distributive justice, we are making theories of distributive justice more potent in spelling out what we want from them for the world as it is and for nearby possible worlds.

6. What Can Limitarianism Contribute?

In the previous sections, I argued against the view that limitarianism can be reduced to a combination of sufficiency and egalitarianism. Starting from the theory-driven terms of the debate that Huseby set, I have argued that limitarianism has things to offer theories of justice that we cannot find either in a defensible form of egalitarianism or in plausible forms of sufficientarianism (to the extent that they are sufficientarian and not in so far as they are hybrid). In this closing section, I want to return to some comments made earlier and draw some threads together.

First, when judging whether a proposed view is distinct, problem-driven philosophy would ask us to engage with all its arguments taken together. Even if there is a better view for reason 1 and another better view for reason 2, it might be the case that those who are concerned with both reason 1 and reason 2 should be endorsing a third view that best addresses these two reasons. I think this general methodological argument applies to how one should judge limitarianism. Neither of the existing views captures a distinct worry that the limitarian view tries to address, which is that some (in the middle classes) would have costs imposed on them for redistributive or public goods purposes while others could keep surplus wealth which does not improve their flourishing. It does not follow from the fact that some limitarian arguments are shared by other views that limitarianism cannot have its own set of grounding values and arguments.

There is a further point to be made. Suppose it were the case, as Huseby believes, that limitarianism adds nothing to egalitarianism or sufficientarianism. It would then have to be true that we could fully understand and appreciate the normative concerns with surplus wealth from the perspective of equality and sufficiency. But the whole point about introducing a focus on the upper side of the distribution is to enable us to ask and investigate what the distinct reasons are for worrying about extreme wealth. Excess wealth creates worries that concern all of the following elements: not asking enough about who will pay for the costs of redistribution; the undermining of democratic values by those who can do so at no significant cost to themselves; a radical waste of resources; power imbalances; the loss of moral autonomy; domination and the undermining of human dignity; and easy funding solutions to collective action problems not being seized. These normative concerns cannot be expressed with the same precision and clarity when relying on the notions of egalitarianism and sufficientarianism.

My final point concerns how I see the playing field on which limitarianism should be judged. Limitarianism was first introduced as ‘a view of distributive justice’ or ‘a partial account of distributive justice’.46 I still hold that to be true, but I have also come to see more clearly that political philosophers have different views on what they take to be a view of distributive justice—and my take on this is presumably broader than those of many others. Theory-driven political philosophers are especially interested in what the reasons for a certain normative claim are and whether those reasons are theoretically distinct. Problem-driven political philosophers are less interested in whether, at a theoretical level, one of those reasons can be reduced to another reason. Instead, they seek reasons that have a distinct contribution to make to addressing real-world-problems. I believe this is the case for limitarianism, and I interpret the political interest in limitarianism to support this belief.

Answering the question about what limitarianism contributes cannot be reduced to asking whether at a theoretical level limitarianism is reducible to egalitarianism and sufficientarianism. There are likely to be people who endorse the position that egalitarianism is overdemanding, but that there should be limits on inequality by protecting the vulnerable and by limiting excess wealth, while allowing inequalities between those two extremes. There will also be those who care less about distributive justice in itself and think that the absence of domination, both in politics and in the social realm, is the most important value to protect, and who endorse limitarianism on those grounds. And there will be those who have moderate welfarist inclinations and believe that concentration of wealth where it only leads to the possession of status goods and luxuries is wasteful in a society in which so many urgent needs are unmet.

All these groups, and others, can agree on the political desirability and justification of limits to wealth concentration; for them it is important to agree about having a reason for acting to limit wealth concentration (but each can have another reason). They don’t need to know whether there is, in a narrow philosophical sense, something ‘distinctive’ about limitarianism.47 Given reasonable value pluralism, it is likely that not all citizens will have reasons to endorse outcome egalitarianism; but if all citizens have reasons, albeit different ones, to endorse limitarianism, then that is a major strength for the limitarian view. If an important role of political philosophy is to address real-world problems, then in situations such as this one, limitarianism will be more action-guiding, since everyone can sign up to the limitarian proposals, but not everyone would sign up to egalitarian or high-threshold sufficientarian views.

The question of what the value and limits of limitarianism are cannot be answered, therefore, without first asking what we want from a theory of justice.48 Many different answers can be given to that question. Of course, problem-driven philosophers also want there to be sound reasons to endorse a normative claim; they have as much an interest as theory-driven political philosophers that their normative analyses be sound. But in addition, they want limitarianism to be judged by the answer to the question whether it can help to address problems in the world as it is and in nearby possible worlds, either by it being developed into guiding ideals or by shifting the public discourse in a different way. Because there are distinct reasons to endorse limitarianism, and because it does not commit us to the normatively implausible parts of egalitarianism, limitarianism does have a role to play in real-world political morality.

Acknowledgement

For very helpful comments on earlier versions, I am grateful to David Axelsen, Constanze Binder, Bob Goodin, Tammy Harel Ben-Shahar, Robert Huseby, Elena Icardi, Tim Meijers, Lasse Nielsen, Dick Timmer, Alex Volacu, Nicholas Vrousalis, and an anonymous referee. For furthering my understanding of limitarianism, I also thank Colin Hickey, Chris Neuhäuser, Debra Satz, and Liam Shields. Finally, I thank the participants at seminars and workshops in Utrecht (January 2019), Bucharest (June 2019), Dortmund (November 2019), the LSE (May 2021), and Rotterdam (November 2021). Research funding was provided by the European Union’s Horizon 2020 Research and Innovation Programme, ERC Consolidator Grant No. 726153. This chapter was first published in The Journal of Political Philosophy, 2022, 30(2), pp. 249–270.

References

Facundo, Alvaredo, Atkinson, Anthony B., Piketty, Thomas & Saez, Emmanuel. 2013. The top 1 percent in international and historical perspective. Journal of Economic Perspectives, 27, 3−20. https://doi.org/10.1257/jep.27.3.3

Atkinson, Anthony B. & Piketty, Thomas. 2007. Top Incomes over the 20th Century. Oxford: Oxford University Press.

Axelsen, David & Nielsen, Lasse. 2021. Why we should worry about extreme wealth. MS: University of Essex.

Brouwer, Huub & van der Deijl, Willem. 2021. More onerous work deserves higher pay. In Anders Örtenblad (Ed.), Debating Equal Pay for All (pp. 55−98). Cham: Palgrave Macmillan.

Casal, Paula. 2007. Why sufficiency is not enough. Ethics, 117, 296−336. https://doi.org/10.1086/510692

Goodin, Robert E. 1982. Political Theory and Public Policy. Chicago: University of Chicago Press.

Gosseries, Axel. 2011. Sufficientarianism. In E. Craig (Ed.) Routledge Encyclopedia of Philosophy, https://www.rep.routledge.com/articles/thematic/sufficientarianism/v-1>. https://doi.org/10.4324/9780415249126-S112-1

Green, Fergus & Brandstedt, Eric. 2021. Engaged climate ethics. Journal of Political Philosophy, 29, 539−563. https://doi.org/10.1111/jopp.12237

Harel Ben Shahar, Tammy. 2019. Limitarianism and relative thresholds. SSRN Working paper http://dx.doi.org/10.2139/ssrn.3404687.

Herlitz, Anders. 2019. The indispensability of sufficientarianism. Critical Review of International Social and Political Philosophy, 22, 929−942. https://doi.org/10.1080/13698230.2018.1479817

Hickey, Colin, Meijers, Tim, Robeyns, Ingrid, & Timmer, Dick. 2021. The agents of justice. Philosophy Compass, 16, e12770. https://doi.org/10.1111/phc3.12770

Huseby, Robert. 2019. Sufficientarianism. In William R. Thompson (Ed.). Oxford Research Encyclopedia of Politics. Oxford University Press. https://doi.org/10.1093/acrefore/9780190228637.013.1382.

Huseby, Robert. 2020. Sufficientarianism and the threshold question. Journal of Ethics, 24, 207−223. https://doi.org/10.1007/s10892-020-09321-7.

Huseby, Robert. 2022. The limits of limitarianism. Journal of Political Philosophy, 3, 230–248. https://doi.org/10.1111/jopp.12274.

Miller, David. 2001. Distributing responsibilities. Journal of Political Philosophy, 9, 453−471. https://doi.org/10.1111/1467-9760.00136

Mulligan, Thomas. 2021. Equal pay for all: an idea whose time has not, and will not, come. in Anders Örtenblad (Ed.), Debating Equal Pay for All (pp. 21−35). Cham: Palgrave Macmillan.

Neuhäuser, Christian. 2018. Reichtum als Moralisches Problem. Frankfurt: Suhrkamp.

O’Neill, Onora. 2001. Agents of justice. Metaphilosophy, 32, 180−195. https://doi.org/10.1111/1467-9973.00181

Piketty, Thomas. 2014. Capital in the 21^st Century. Cambridge, MA: Harvard University Press.

Rawls, John. 1999. A Theory of Justice, rev. edn. Cambridge, MA: Harvard University Press.

Robeyns, Ingrid. 2012. Are transcendental theories of justice redundant? Journal of Economic Methodology, 19, 159−163.

Robeyns, Ingrid. 2017. Having too much. In Jack Knight & Melissa Schwartzberg (Eds.), Wealth - Yearbook of the American Society for Political and Legal Philosophy (pp. 1–44). New York: New York University Press.

Robeyns, Ingrid, Buskens,Vincent, van de Rijt, Arnout, Vergeldt, Nina & van der Lippe, Tanja. 2021. How rich is too rich? Measuring the riches line. Social Indicators Research, 154, 115−143.

Rothstein, Bo. 1998. Just Institutions Matter. Cambridge: Cambridge University Press.

Sen, Amartya. 2006. What do we want from a theory of justice? Journal of Philosophy, 103, 215−238.

Sen, Amartya. 2009. The Idea of Justice. London: Allen Lane.

Shields, Liam. 2012. The prospects for sufficientarianism. Utilitas, 24, 101−117. https://doi.org/10.1017/S0953820811000392

Shields, Liam. 2020. Sufficientarianism. Philosophy Compass, 15, e12704. https://doi.org/10.1111/phc3.12704

Sunstein, Cass R. 1998. Practice reason and incompletely theorized agreements. Current Legal Problems, 55, 267−298. https://doi.org/10.1093/clp/51.1.267

Timmer, Dick. 2019. Defending the democratic argument for limitarianism. Philosophy, 47, 1331−1339. https://doi.org/10.1007/s11406-018-0030-6

Timmer, Dick. 2021a. Thresholds in distributive justice. Utilitas, 33, 422−441. https://doi.org/10.1017/S0953820821000194

Timmer, Dick. 2021b. Limitarianism: pattern, principle, or presumption? Journal of Applied Philosophy, 38, 760–773. https://doi.org/10.1111/japp.12502

Timmer, Dick. 2021c. Thresholds and Limits in Theories of Distributive Justice. PhD dissertation, Utrecht University.

Valentini, Laura. 2012. Ideal vs. non-ideal theory: a conceptual map. Philosophy Compass, 7, 654−664. https://doi.org/10.1111/j.1747-9991.2012.00500.x

Volacu, Alexandru & Dumitru, Adelin Costin. 2018. Assessing non-intrinsic limitarianism. Philosophia, 47, 249−264. https://doi.org/10.1007/s11406-018-9966-9

Wolff, Jonathan. 1998. Fairness, respect, and the egalitarian ethos. Philosophy and Public Affairs 27, 97−122. https://doi.org/10.1111/j.1088-4963.1998.tb00063.x

Wolff, Jonathan. 2015. Political philosophy and the real world of the welfare state. Journal of Applied Philosophy, 32, 360−372. https://doi.org/10.1111/japp.12125

Wolff, Jonathan. 2018. Methods in philosophy and public policy: applied philosophy versus engaged philosophy. In Annabel Lever & Andrei Poama (Eds.), Routledge Handbook of Ethics and Public Policy (pp. 13−24). London: Routledge. https://doi.org/10.4324/9781315461731-2

Zwarthoed, Danielle. 2019. Autonomy-based reasons for limitarianism. Ethical Theory and Moral Practice, 21, 1181−1204. https://doi.org/10.1007/s10677-018-9958-7