1 Introduction

Let us start with these facts. There is staggering economic inequality both within and between countries. One significant aspect of the growth of economic inequality is the concentration of wealth in the hands of a small group of very rich people. The rich are getting much richer (Oxfam 2022). There is also no disputing that the political will to decrease economic inequality by targeting the rich is limited to say the least. In fact, economic inequalities are actively sustained through a politics that defends wealth (Winters 2017, 160–165). It has become a received truth that the personal wealth of the rich must surely be deserved or, if not deserved, then somehow useful for the economy. The current credo is that economic justice is to be achieved mainly through economic growth, not redistribution, as testified for instance by Goals 1 and 8 of the 2030 Agenda for sustainable development.

Recent research in economics and social science turns a critical eye to the increases of economic inequality and the political defence of wealth (Dorling 2015; Milanovic 2018; Picketty 2015; Stiglitz 2012; Wilkinson and Pickett 2010). Within human rights theory, there is an emerging critical debate over the field’s lack of attention to economic inequality and the threat to equal political rights that comes with inequalities of economic power (Moyn 2019; Brinks et al. 2019).

One recent philosophical contribution to these debates is Ingrid Robeyns’ “economic limitarianism” (Robeyns 2017, 2019, 2022): the notion that there should be an upper limit to wealth or income. Limitarianism is a mirror image of the sufficiency approach, familiar from human development and rights theory, as well as from philosophy. It similarly operates with a moral idea of economic thresholds. The sufficiency approach focuses on a minimally acceptable level of basic provision below which no one should be allowed to fall (Frankfurt 1987; Nussbaum 2013). Where that level should be set will be dependent upon moral reasoning. Limitarianism makes the case for a maximum level, above which no one should be allowed to rise. Economic resources above the maximum, or “riches line”, count as “surplus money” to which no moral claim can be made.Footnote 1 There is such a thing as having too much, hence there should be a cap on wealth.

I share Robeyns’ position that extreme wealth is offensive and I share her concerns about the damaging effects that vast economic inequality has on democracy. Despite these sympathies, I argue that if you are concerned about economic inequality, you have reason to be wary of limitarianism. My current ambition is quite limited. I do not discuss whether inequality is bad in itself or not, nor do I go into any details about various accounts of distributive justice. I think Robeyns wants her own position to be independent of such debates. A valid reason for focusing only on the very rich is to diffuse the objection that economic redistribution is potentially unfair and needs careful moral grounding: surely we have no reason to feel sorry for the billionaires. I restrict myself here to showing that Robeyns’ economic limitarianism fails on its own premises; it cannot do what it sets out to do.

In the next section, I give a brief outline of economic limitarianism in Robeyns’ version of it. In Sect. 3, I argue that limitarianism does not hold up as theory, not even as the partial theory of justice it claims to be. The normative reasons invoked in its favour fail to support it. I focus on the facts that Robeyns’ wealth threshold is set in terms of human flourishing, not in terms of how well or badly people fare in relation to each other, and that it is a “local” not a “systemic” principle of justice. A crucial problem is that Robeyns’ wealth threshold disconnects the question of having too much from the economic system that generates inequalities.

Robeyns wants limitarianism to be problem-driven – not theory-driven – philosophy and to be judged by how well it addresses problems in the world as it is (2022). Problem-driven philosophy aims to be action guiding and policy relevant, and so needs to be informed by empirical research. In the final section, I take Robeyns at her word and discuss limitarianism in relation to existing research on attitudes to economic inequality. My contention is that limitarianism lacks purchase against attitudes that condone economic inequality. In conclusion, if you agree that economic inequality and concentration of wealth are unjust and if you see a value in problem-driven political philosophy, then you still lack sufficient reason to find limitarianism convincing.

2 Limitarianism, Flourishing, and the Wealth Threshold

In her latest statement of limitarianism, Robeyns makes explicit a distinction that has implicitly been there all along, between what I call limitarianism simpliciter and the limitarianism that she defends (2022). Limitarianism simpliciter includes two ideas. The first is evaluative – a moral intuition – and the second is normative or action guiding: there is such a thing as having too much money and a threshold should be introduced to prevent anyone from rising above it. Limitarianism simpliciter does not decide on the criterion or principle for settling what “too much” is and on what ground such a judgement should be made. People who share the intuition can do so for different moral, political, or pragmatic reasons.

Robeyns’ own account is not limitarianism simpliciter. It is important to bear in mind that she does not only argue for a limit on income or wealth. She argues for a particular moral reason to find a limit on income or wealth plausible: the good of human flourishing. I wish to emphasize that it is this morally specified limitarianism that I am concerned with here.Footnote 2

The notion of flourishing is familiar from Martha Nussbaum’s version of the capability approach (2013), which incorporates a sufficiency threshold below which a person counts as unjustly deprived of necessary conditions for human flourishing. Robeyns uses flourishing to theorize about an upper threshold, above which a person’s flourishing, to the extent it depends upon economic resources is, we might say, saturated (Robeyns 2017). Economic resources held above that level count as surplus money and no claim to it can be made (2022, 5), at least not in a world where there are urgent unmet needs.

Being rich is the name given to that full or saturated level of materially dependent flourishing. Importantly, flourishing is objective in the sense that it is not a matter of experienced personal satisfaction. The calculation is not a utilitarian one. A person can feel perfectly content and satisfied without being rich, and vice versa.Footnote 3 When a Russian oligarch – to take a current example – senses the pinch of sanctions, he might feel gravely deprived but if his losses leave him above the riches line, he has no valid complaint.

So, the riches line is independent of individual desires but still socially constructed. What counts as full quality of life will be dependent upon shared social attitudes that vary over time and place. A mundane example is that coffee used to be a luxury item but is now an everyday consumption staple. Robeyns stresses that deliberating over what counts as full flourishing will always be a contextual and discursive exercise (2017, 26) and any settled line will be provisional and subject to renegotiation. This is perfectly reasonable. Indeed, it can scarcely be otherwise. But a relevant question for Robeyns’ limitarianism is whether the riches line can ever be non-arbitrary, even within a deliberative context. If it cannot, it is hard to see how it can function as a principle of justice.

3 Does Limitarianism Work as a Partial Account of Justice?

In this section, I discuss whether limitarianism holds up as the partial account of economic justice that it claims to be. By assessing the wealth-threshold, the role of flourishing in the theory, and the power that comes with wealth, I find that the arguments offered for limitarianism indeed provide strong reasons for redistribution from the rich, but not to favour limitarianism over alternative redistributive schemes.

3.1 The Wealth Threshold

What kind of exercise is it to negotiate a Robeyns-style wealth threshold? The first thing to note is that it is a riches line, not a best-off line (Robeyns 2017). Limitarianism is set on identifying those who are rich, not those who are best off compared to others. Limitarianism theorizes from the intuition that it is morally offensive if people have much more money than they themselves can conceivably benefit from, not that it is morally offensive for some people to have much more than others. It does not within itself contain a moral position on the overall distributive pattern in a given society, only on the cut-off-point or threshold between rich and the rest – the non-rich – a category covering everything from poverty to comfortable affluence just short of whatever is regarded as rich.Footnote 4

Tammy Harel Ben Shahar has argued that the purposes of limitarianism would be better served by a relative or comparative riches line.Footnote 5 She suggests a ratio account, similar in structure to the Ten Times Rule proposed by Sam Pizzigati in 1992. The idea is to not set a definite upper wealth-limit, but instead to limit the ratio between the economic resources of the worst and the best off.

Pizzigati’s Ten Times Rule and other accounts of a maximum wage are concerned with the wage structure, not – like limitarianism – with economic resources generally. (Remember that Robeyns does not distinguish between income and wealth.) Maximum wage accounts build on the familiarity of the minimum wage, and suggests a maximum set in a certain ratio to the minimum wage. Pizzigati proposes a maximum wage at ten times the minimum wage and that income above that be taxed at 100%. Harel Ben Shahar, like Robeyns, focuses on economic resources in general, not wages, and she does not settle on any specific ratio.

Harel Ben Shahar thinks that a relative (or comparative) ratio account would serve limitarianism better since it captures the limitarian intuition that unlimited wealth is morally repugnant and wasteful, while solving the problem of negative incentive. The worry is that the incentive structure throughout the economy might be negatively affected if people are not allowed to keep their money over a certain wealth level.Footnote 6 Ratio accounts avoid this problem since there is no absolute cap. The rich are allowed to become ever richer as long as their gain is also a gain for those at the bottom (Ramsay 2005). But for a ratio model to still be a version of limitarianism, one does need to trust it to actually limit wealth accumulation – remember that the moral intuition underpinning limitarianism is that extreme wealth is morally offensive. A defender of ratios might say, first, that accumulation of massive wealth would, on their account, be impossible in practice since you cannot become super-rich without exploiting the poor for your own benefit, and second, that any accumulated wealth within the ratio is not wasteful since the condition for allowing it is that it raises the level of the poor. One can easily make a connection here to the moral idea behind Rawls’ difference principle: economic inequality is not unfair provided that it is to the benefit of the worst off (Rawls 1971, II:13).

Robeyns is not likely to be swayed by this. She has explicitly rejected accounts that are relative in this comparative sense, regardless of whether they use ratios, percentages of shares, or any other model. The two reasons she invokes for this rejection are, however, quite weak (Robeyns 2017). A relative account, she says, would have to accept that if the general level of material welfare changes in a society, so that everyone becomes either worse or better off, the number of rich people would stay the same (2017, 16). But that is just a restatement of how relative thresholds work; it is not a counter-argument. A defender of relative thresholds would simply agree. Her second argument is that even in a situation of general dire need, there will be some who have more than others – a bit more food perhaps, or some tool that others do not have. They are comparatively better off yet still seriously deprived, so it would be counterintuitive to call them rich. But again, a defender of relative thresholds would simply agree, adding that relative accounts do not depend upon notions like “being rich”. Robeyns’ objections are not counter-arguments, they are intuitive aversions to calculating economic injustice in terms of who is better or worse off than others, rather than in who is rich by some standard and who is not.

I now proceed from the implications of Robeyns’ commitment to non-relative thresholds and show that her approach cannot be sustained, at least not given the reasons she herself states in favour of limitarianism. I use Elizabeth Anderson’s distinction between local and systemic principles of justice to show that Robeyns’ limitarianism is self-defeating.

3.2 Flourishing as Metric or Goal

Elizabeth Anderson makes a useful distinction between two classes of justice principles: local and systemic (Anderson 2017, 72; compare Elster 1992; for use in an egalitarian approach to human rights, see Halldenius 2022). A local principle of justice focuses directly on allocation of goods to or from particular people. The pertinent question is whether a person’s set of goods is rightly theirs. Relevant reasons for assessing whether it is are to do with facts about that person or how they came to have this set of goods. A familiar example is Robert Nozick’s libertarian defence of property rights, according to which a person has an absolute right to their set of goods as long as it was acquired through voluntary transactions without fraud. Another familiar local principle is meritocracy: a person’s set of goods is rightly theirs if it reflects their hard work, valuable contributions to society, or similar grounds for merit. A necessary assumption is that answers to such questions – how a set of holdings came about – are at least in principle available and morally significant. On local principles, assessing whether a situation is fair or not is a matter of tallying things up, one person (or possibly household) at a time. If each has what the local principle grants as their due, then there is justice regardless of what the distributive pattern looks like overall.

On a systemic principle of justice, the object of economic justice is the overall distributive pattern and its expected consequences. Local principles have a role, but are subordinated to systemic principles. For instance, wages may be negotiated in terms of particular employees’ personal merits or efforts – a local principle – but only within the bounds of a wage structure that on the whole has beneficial consequences in terms of some agreed goal – a systemic principle. Rawls’ difference principle is systemic. The difference principle does not say anything about who should have how much of what and why; it is purely systemic, saying only that the basic structure of society should be organized such that the distributive pattern it generates is as good as possible for the worst off. Even if a wage inequality can be explained in local terms, say by differences in personal merit, the inequality is unfair if it does not support the systems-consequence of benefitting the worst off. How is this relevant for assessing Robeyns’ limitarianism as a contribution to theories of economic justice?

The limitarian threshold, with a riches line set in terms of a person’s flourishing is, on Anderson’s distinction, a local principle. It is a variation of the form “to each, according to his P” (Anderson 2017, 72; “P” can be merit, prudence, need, or any other morally pertinent feature). For Robeyns, the form is not “to each” but “let each keep”, and P is “full flourishing” translated into money. Contrary to a principle of personal merit or libertarian property rights, this principle does not consider how a person’s resources came about, only what values they create in that person’s life.

It should now be noted that Robeyns talks of flourishing as a value, but the work flourishing does in the theory is mainly as a metric. It features in the limitarian deliberation over where the cut-off point should be. The cut-off point is moral in the limited sense that money held above it has no moral weight and is therefore available for redistribution, but full flourishing is not an overall social goal to be promoted generally throughout the economy. In other words, there is no normative limitarian principle saying that the economy should be organized in such a way that it promotes flourishing among the population as a whole. It is merely a metric for settling when someone’s resources no longer add value for them. To the extent that flourishing is a principle of justice at all here, it is local, not systemic.

The arguments that Robeyns gives in favour of limitarianism also tell us that flourishing functions as a metric and not as a social value or goal. The problem with wealth, says Robeyns, is not rich people per se, but “the effects of the situation of extreme wealth on society” (2017, 258). This is a systemic way of reasoning and it makes no reference to flourishing. It instead focuses on two other things: political equality and urgent needs. As we will see, these systemic concerns sit uneasily together with the flourishing metric and, crucially, one can be fully convinced by the force of these concerns without finding any reason to support limitarianism over alternative redistributive schemes with the same goals.

One of Robeyns’ arguments proceeds from the fact of urgent unmet needs. Poverty, deprivation, and collective action problems – like the climate crisis – exist in the world and combating them requires funds. Given the limitarian position that wealth held above the cut-off point is in excess (it is wasteful as judged by the metric of full flourishing) and has no moral weight, such wealth can and should be redistributed to support measures against these urgent problems (ibid.).

The structure of the argument is that there are no moral reasons against redistribution from the rich, as long as the rich remain in a situation of full flourishing also after their excess wealth has been taken. The wealth that they retain will still be considerable but not wasteful in terms of the metric. But this is peculiar. Anyone who is convinced of the urgency of poverty and the climate crisis as rectifiable problems of survival and justice, has reason to support redistributive taxation to fund climate action and poverty eradication. But that conviction does not make a case for limitarianism over alternative redistributive models like progressive taxation or relative ratios. Remember that the focus in this section is on limitarianism as a partial theory of justice, not as practical politics. I doubt that any limitarian believes that combating climate change and poverty is less morally important than letting the affluent keep their wealth as long as they continue to flourish by it, but that seems to be the implication of limitarianism as a principle of economic justice. The argument about urgent unmet needs supports redistribution in general, but not the limitarian way of doing it.

We saw before that on systemic accounts of justice, overall judgements of justice are made on the level of the system of production and distribution (Anderson 2017). The favoured systems-consequences (like eradication of poverty, realization of equal opportunities, or sustainable development) condition local principles of justice (like personal merit or freedom of transaction), which are subordinate to them. If the workings of the local principle do not sustain the systemic goal, the local principle will be circumscribed. But limitarianism turns this relationship on its head. Its local principle (“let each keep their wealth up to the point of full flourishing”) has priority over its systemic goals, so the urgent systemic problems can be addressed and funded only as long as the local principle allows it.

Here are two possible counter-arguments to what I have just said. One is that I have underestimated the strength that limitarians give to the systemic goals. Limitarians might say that they support the wealth threshold only on the assumption that capping wealth at that level actually is the best way to raise revenue for meeting urgent needs. If that turned out to be wrong, they would support whichever redistributive scheme is more efficient. But if that were the case, then limitarianism is just one option among many in pragmatic calculations of optimal tax policy and no significant contribution to justice theory at all. Another possible counter-argument is that I neglect the fact that limitarianism is only a partial account, and that limitarians are free to combine it with other justice principles, say to tackle inequalities below the riches line. But compensating the inadequacies of limitarianism with the virtues of whatever principles you combine it with, is not really a point in favour of limitarianism. Robeyns emphasizes that “introducing the idea of limitarianism separately” (Robeyns 2022, 2, emphasis in original) is the best way to capture what is wrong with wealth concentration; consequently, it also should withstand being scrutinized separately.

3.3 The Power of the Rich

The other reason that Robeyns invokes in support of limitarianism – in addition to meeting urgent needs – is that substantial economic inequalities “undermine the value of democracy and the ideal of political equality in particular” (Robeyns 2017, 254). This points to the crucial insight that economic inequality is not only about who has how much of what. It is about power. The power of the rich is corruptive of democracy in a sense that Robeyns rightly contemplates: the wealthy can set the agenda of political decision-making. Through campaign financing, media control, and lobbying they can change perceptions and make their interests look true and reasonable. They can exert political pressure by threatening to move their money elsewhere. But let me emphasize that the power of the rich is also strictly economic: those who have nothing to sell but themselves are dependent on the arbitrary will of rich employers who might or might not hire them at minimum wage. As we recently noted, it is difficult to become very rich without exploiting the poor.

The fact that unequal economic power poses a threat to political equality and democracy is a powerful argument, but just as the argument from urgent needs one can be fully convinced by it without finding limitarianism to be the best way of dealing with it. Robeyns focuses, again, on the very rich and on what they do with their money. But as substantial research shows (for an overview and analysis of data, see Schäfer and Schwandler, 2019), the situation is much more complex. For instance, voter turnout decreases as economic inequality increases. Voter turnout is always lower among the poor – they do not feel included and they know that things are stacked against them anyway – but turnout decreases over all income groups when inequality increases (Schäfer and Schwandler 2019, 405).

Importantly, the fact that the interests of the more affluent are favoured by political decision makers is to do with comparative affluence, not a binary cut-off point. This is evident in normal politics and crisis management alike, as we see from the design of support-schemes during the Covid-pandemic. To give a current example: in the wake of the petrol price hike, the Swedish government gave a one-off compensation payment to all car owners, while those who depend on public transport did not get anything, even though the cost of public transport had increased more than the cost of driving. In liberal democracies where the middle class is big and vocal, with economic aspirations and high expectations, politics gear towards their interests – particularly in an election year – but none of them would be above Robeyns’ riches line, so limitarianism, consequently, has nothing to say about it.Footnote 7

The conclusion is the same as with the argument from urgent need: even if you agree with this reason to worry over economic inequality, you still have no ground to prefer limitarianism to other redistributive schemes, like progressive taxation in combination with general and publicly funded social services. Political equality is a systems outcome and so should set the conditions for local principles of justice (Anderson 2017). But the principle of the riches line is to let each keep their wealth up to the point of full flourishing. Its cut-off point is set in terms of the full flourishing of the wealthy, not in terms of what would benefit the less favoured, nor what would strengthen democracy. Any alignment of this principle with political equality would be co-incidental.Footnote 8 The point is that the political equality argument – like the argument from urgent needs – cannot settle the issue in favour of limitarianism over redistributive schemes that aim to decrease economic inequalities structurally and throughout the economy. Consequently, the reasons invoked in favour of limitarianism support other redistributive principles equally well, or better.

4 Does Limitarianism Work as a Guide to Policy and Economic Decision-Making?

A possible objection to the arguments I have made so far is that they do not matter for what limitarianism ultimately wants to be: a practical guide to decision making. As I noted in the introduction, Robeyns wants limitarianism to be judged by how well it can address practical problems in the world as it is (2022, 4 & 18–20). Even if it turns out that limitarianism is not theoretically distinctive or does not hold up as a well-structured account of justice, it could still make politically useful critical points about inequality. We need to start somewhere. If limitarianism could shift public discourse such that people in general became more aware of the ills of wealth concentration, and policy makers more willing to enforce redistribution from the very wealthy to the economically most vulnerable, then that is a good place to start. In practical terms, limitarianism does not aim to gain support for economic equality in general or for high sufficiency levels, since there can be reasonable disagreement about such things (Robeyns 2022, 19). Instead, by disregarding inequalities over which there can be reasonable disagreement and concentrate only on what should be obviously offensive to anyone – extreme wealth in a world where there is also dire poverty – wide support can be gained for making the very rich contribute more.

This is indeed a fine ambition. We might differ on how much to tax the “merely” affluent, the limitarian will say, but surely we all agree that those who can afford every luxury and still have plenty left “to do extraordinary things” (Robeyns et al. 2021, 130) cannot complain if we say enough is enough. But this way of reasoning is likely to instead counteract support for redistribution from the wealthy.

Contrary to theoretical idealizations, practical policy suggestions need to be backed up by empirical findings about how public attitudes are formed and how discourse is legitimated. Limitarianism proceeds from the assumption that disregarding lesser inequalities, and concentrate only on the super wealthy, will make it easier to get public support for redistribution, at least redistribution from those who are very rich, but this assumption is shaky at best. Counter-intuitively, we will see that the wider people believe the inequality gap to be, the more willing they are to accept it and to morally justify it. By focusing on the very wealthy, limitarianism might well serve to shift the acceptance of inequality upwards.

In the next sub-section, I discuss what “problem-driven philosophy” is and critically assess Robeyns’ and her team’s attempt to empirically test people’s attitudes towards wealth. After that, I assess limitarianism as problem-driven philosophy in a wider empirical context, focusing on studies on the so-called adjustment hypothesis and on the strength of meritocratic beliefs. People’s moral attitudes to economic inequality tend to be shaped by what they know or believe about existing inequalities. According to the adjustment hypothesis, the more inequality there is, or is believed to be, the more accepting people are of inequalities (Trump 2017). Attitudes adapt. But limitarianism relies on people widely sharing the moral attitude that the more wealth is concentrated in the hands of a few, the more legitimate it is to enforce redistribution from the rich. Limitarianism depends on the adjustment hypothesis being false. I conclude that limitarianism is insufficiently supported as guide to practical policy. It might even inadvertently legitimate great inequalities in wealth.

4.1 Problem-Driven Philosophy and the Robeyns Survey

Robeyns claims that limitarianism is “problem-driven”, not “theory-driven”, political philosophy (2022, 2f), which means that it aims to provide useful knowledge that can help solve problems rather than make fine-grained conceptual distinctions. This is a variation of the more familiar debate about ideal and non-ideal theories of justice (Rawls 1971, for an overview, see Valentini 2012). An ideal theory of justice provides an account of what a just society would look like, and it makes or allows idealizing assumptions about people and society. For instance, John Rawls assumed full compliance with the principles of justice. A case for ideal theory is that philosophy plays an important role in challenging collective lack of imagination and intellectual courage through thought-provoking suggestions without practical constraints. If philosophers don’t dare to think beyond the seemingly feasible, who will?

We can accept this in full while also agreeing that there is value in philosophical accounts of justice that are designed to be action guiding, that proceed from the world with all its flaws, and are empirically informed about what people are actually like, how attitudes are shaped, and how institutions work.

Robeyns acknowledges that problem-driven philosophers need to engage with empirical research. How tough a requirement that is will be a matter of debate. I proceed from the position that for a philosophical theory to provide useful knowledge for addressing practical problems in the world it needs to be supported by empirical research which itself is methodologically sound and bears scrutiny. We have already seen – in Sect. 3 – that limitarianism seems out of sync with how economic inequalities operate politically: empirical studies of how the power of money impacts political participation tells in favour of systemic approaches to decreasing inequalities throughout the economy, not a local principle of flourishing. A cut-off point in terms of flourishing carries no information about when or how the power of someone’s money gives them political advantages. I will now consider the central moral idea of limitarianism – the “riches line” or “limitarian threshold” – in relation to empirical findings about formations of attitudes to economic inequality. I start with a survey conducted by Robeyns herself and her team.

Robeyns notes that where the riches line is will vary between contexts and should be negotiated in deliberative processes: the riches line operates with a metric for flourishing which is objective – it does not simply reflect the subjective preferences of the rich – but the acceptance of it, as well as deliberations over where it should be drawn, will be a matter of attitudes. How people think about issues of wealth and inequality is therefore a crucial empirical factor for limitarians to consider. Remember that limitarianism depends on both the moral intuition that excess wealth is without moral value and that the moral value of wealth is tied to flourishing, not to things like personal merit. If that double intuition is not widely shared, the limitarian deliberation will not get off the ground.

In recognition of this quandary and in order to check the prevalence of limitarian attitudes, Robeyns and her colleagues conducted their own survey (Robeyns et al. 2021). They tested a representative sample of Dutch respondents on two things: if they were willing to draw a riches line (referred to as the evaluative question), and if they would support the enforcement of limits to wealth and income (the normative question). The results, as interpreted by the team, were that a majority of the respondents did identify a riches line but very few expressed support for a cap on income (12% in favour) or on total wealth (5% in favour). A lingering question was how this discrepancy could be explained: if wealth above a certain level is evaluated as having zero moral weight, why does that not translate into a normative position that wealth (or income) should be capped at that level? Let me briefly comment on both these results.

In the evaluative part of the survey, the respondents were asked to place example families along a grade scale from having “just enough to get by” to being “extremely rich”; the latter meant being able to afford every luxury and still have plenty of money left (2020, 130). The description of that category was “No one needs that much luxury”. A respondent classifying an example family as “extremely rich” was counted as accepting a riches line in the limitarian sense, including the limitarian idea that excess money is morally worthless. The interpretation, then, was that respondents evaluated the situations presented to them in the same way as the research team did: by using flourishing as criterion for the moral worth of holdings. But that does not follow. Anyone would agree that Jeff Bezos is extremely rich and has way more money than anyone needs, but it is possible to agree on that without also agreeing that his wealth is morally useless. That latter attitude depends on one’s favoured criterion for when wealth is legitimate. A limitarian thinks that flourishing is that criterion, but we know that meritocratic attitudes are very strong, even when they are unfounded. In fact, the normative part of Robeyns’ survey confirms that, as we will now see.

Robeyns’ respondents were asked to express their judgement on abstract statements on “major income inequality” and people who are “extremely rich”. A substantial majority found nothing wrong with either (Robeyns et al. 2021, 126). An even bigger majority said no to imposing an upper limit on income and wealth. How can this be explained? Crucially, respondents were asked to state their attitudes to extreme wealth, only as long as the rich earned their money “fairly according to the rules and do not evade tax”, there was no discrimination in the labour market, and everyone had access to high-quality education. In other words, respondents were asked whether they would accept inegalitarian outcomes in a counter-factual meritocratic society with full compliance, which is exactly the kind of idealization that problem-driven or non-ideal philosophy is not supposed to use. The respondents were not asked for their opinion on inequalities resulting from discrimination, exploitation, and tax evasion, which is how the rich get rich in the real world. (Anyone in doubt is recommended to consult The International Consortium of Investigative Journalist (ICIJ) 2021, https://www.icij.org/investigations/pandora-papers/). The normative part of Robeyns’ survey does not support (but neither does it disprove) people using flourishing as a moral criterion. But it does seem to lend support to what is already known from other studies: that people’s moral attitudes to wealth are significantly shaped by ideas about personal merit (even when there are few or no empirical reasons to think that wealth is explained by personal accomplishments or effort, see e.g. Sandel 2020) and that the tendency to justify wealth with meritocratic ideas is particularly prevalent in societies with high levels of inequality.

In what follows, I will relate the empirical viability of limitarianism to studies on meritocratic beliefs and the adaptability of attitudes towards inequality. As practical guide to economic policy, limitarianism could conceivably serve as a kind of foot in the door: by focusing only on the very rich and counting on wide support for the illegitimacy of their wealth, it aims to normalize the idea that those who have more money than they need are rightly expected to also contribute more, thus lending support for economic redistribution from the haves to the have-nots. But I will show that we have scant reason to think that it works like that.

4.2 Adjusted Attitudes and the Trust in Meritocracy

The potential of limitarianism as a guide to policy and economic decision-making depends on the intuition that people are more prone to object to inequality the more there is of it but, as we will see, the reverse seems to be true: acceptance adjusts upwards when inequality increases.

In a study into the acceptance of economic inequality (2017), Kris-Stella Trump builds on several previous studies that all have confirmed that “economic inequality does not lead to demands for redistribution” (Trump 2017, 931; Alesina and Glaeser 2004; Kenworthy and McCall  2008; Luttig 2013, and McCall 2013). These findings form the basis of the adjustment hypothesis: “income inequality affect perceptions of legitimacy in income differences” (Trump 2017, 934). People who believe that income inequality is high, tend to be more accepting of inequality than people who believe that inequality is low. And when people who believe that income inequality is low are told that it is higher than they thought, their acceptance shifts upwards.

The adjustment hypothesis is summarized thus: “Individuals who believe that income inequality in their society is high also tend to think that higher income inequality is legitimate” (Trump 2017, 932). There is a plausible explanation for this, which tallies with what we know about social cognition: persons are motivated to avoid the discomfort of concluding that their social system is unfair. This is referred to as “system justification motivation” (2017, 933) and the relevant point of it for our purposes is that people are inclined to think of the society they live in as fair. This adjustment is all the more likely if there is a “culturally acceptable reason” (ibid.) readily at hand for rationalizing why inequalities are not objectionable, like the idea that the rich justly deserve their wealth.

A study by Jonathan Mijs confirms that a widespread belief in meritocracy works as just such a culturally acceptable reason for accepting inequalities. Meritocracy here means that inequalities are due to personal accomplishments and hard work, not structural factors or luck (Mijs 2021). Mijs also shows that this belief and – by implication – this legitimation of inequalities is stronger the more inequality there is. In a society where economic inequalities are large, people are less prone to explain inequalities in structural terms and more prone to explain them in terms of personal accomplishments. One reason is that in vastly unequal societies, where people in different economic situations live isolated from each other, they are more likely to believe that people earn whatever situation they are in (Mijs 2021, 11). In settings where people intermingle across different economic levels, believes in structural reasons for inequality are more prevalent. Cultural narratives of the personal success of the rich and the personal failures of the poor (the rich are rich because of hard work and grit; the poor are poor because of bad choices and lack of dedication) are more easily sustained when people in different economic circumstances live as strangers to each other’s realities.

These studies together find that people tend to be more accepting of economic inequalities the more inequality they believe there is, that there is a justificatory bias towards the fairness and legitimacy of one’s own society, and that meritocratic beliefs that, however erroneously, justify inequalities are stronger in more inegalitarian, socio-economically stratified societies. This is relevant, and problematic, for limitarianism in several ways. In fact, the results of the Robeyns survey are compatible with these findings. Given the design of the questions, their respondents’ normative acceptance of large inequalities and aversion to a cap on wealth, were premised on structural causes and corruption being taken out of the equation, while personal factors like merit were left in. Just as in Mijs’ study, Robeyns’ respondents accepted inequalities as long as personal merit was available as explanation for them.

This ready acceptance of merit as explanation is a kind of cultural code that functions to keep up the image of one’s own society as fair, which is convenient for the economically powerful. Needless to say, people are not hard-wired into thinking in these terms. There is political and pragmatic opposition to inequality, as well as waves of public rallying against wealth concentration, like the protests in Chile and Ecuador in 2019 and the Occupy movement (McCleave Maharawal 2013; Palacio Ludeña and Díaz Pabón 2020). The important point, which Trump also emphasizes, is that given the political power of the affluent and the general tendency for people to adjust, it is opposition to large inequalities, not acceptance of them, that needs to be explained (Trump 2017, 931f). This is a crucial insight for anyone who wants to shift discourse in favour of redistribution.

The limitarian assumption is that when inequality increases and people are adequately informed about just how rich the richest in their societies are, demands for redistribution of wealth will be the default reaction; concentrating only on the very wealthy will make it easier to get public support for redistribution. But that assumption is unwarranted, and as long as one sticks to it, one will find no reason to investigate just how opposition to inequality can be activated and sustained.

So, can limitarianism shift public discourse and shape actions and institutions to support redistribution from the rich to the poor (2022, 4, 20)? In an interview to CBS News, Robeyns said “I think you can have a fully flourishing life and do all the things you want with, well, perhaps not with $10 million, but then with $20 million. I don’t think you need a billion” (CBS News, January 2022). It is hard to see how attitudes capable of countering the adjustment process can be activated if the discussion we are to be having concerns how many millions short of a billion the rich need in order to flourish fully.

In the previous section, I showed that limitarianism fails as a partial theory of justice. Even so, one might have thought it could be useful for eliciting support for redistributive economic policy, but we have now seen that we lack reason to believe that it has such use.

5 Conclusion

I have argued that Robeyns’ limitarianism fails on its own premises, both as a partial theory of justice and as a practical guide to economic policy. The arguments invoked in its favour as theory – political equality and meeting urgent needs – give no reason to prefer it over alternative redistributive accounts, even if they are fully accepted. Limitarianism has no resources within itself that can be mobilized for or against setting any particular level of permissible wealth. The reasonableness of any riches line will depend on attitudes about inequalities and wealth but attitudes tend to adjust to rising levels of inequality.

If intuitions about “having too much” are the moral ground we are supposed to stand upon in our critique of economic inequality, our account will be vulnerable to the fact that attitudes are more accepting of economic inequality, the more unequal a society is. Focusing exclusively on the very rich and not on the economic system that produces inequalities, limitarianism lacks the resources to explain how resistance to such adjustment could be activated.

It is easy to understand the temptation to side-step the complicated and contested world of less dramatic inequalities and concentrate only on the morally purer case of the super-rich, but it comes with the price of also side-stepping the realities in which most people’s experiences are formed. For a philosophy of justice to be truly problem-driven and capable of delivering solutions, it needs an empirically solid and unblinkered understanding of what the problems are.