Progressive taxation schemes and redistributive social policies, as well as public services such as those discussed already in the fields of education, healthcare, housing, sport and recreation are of course indispensable tools to combat poverty. However, they also present certain limitations. They depend on economic growth, as measured by an increase of the country’s GDP per capita. In addition to the ecological barriers that such a strategy now faces, this may create an incentive to create a ‘business-friendly’ investment climate and to lower the regulatory and tax burdens on corporations, as a means to stimulate wealth creation, with potentially exclusionary impacts: it is a strategy, in other terms, that may encourage the development of an extractive and exclusive economy, when what is needed is a regenerative and inclusive economy. Second, tax-and-transfer approaches may be politically unsustainable, at least where social protection takes the form of welfare policies narrowly targeted to benefit the poor, rather than the middle class: the result of targeting thus conceived is that the median voter may not support such policies, insofar as they are perceived as (and presented by populist politicians as) taking from the rich and deserving groups of the population to provide aid to ‘undeserving’ poor. It is this finding that is at the heart of the ‘paradox of redistribution’ highlighted by Korpi and Palme already a generation ago (Korpi and Palme, 1998).

It is therefore vital, in addition to designing progressive taxation schemes and to strengthening social protection and access to essential services, to move towards a more inclusive form of economy, one that ensures real opportunities for all. This is not just a slogan. It means designing a model of development that takes seriously the duty of the state to ensure access to employment. It may require providing social support to young adults, particularly from low-income backgrounds, to compensate for the obstacles they face and reduce the perpetuation of advantage and of disadvantage. It means, finally, taking seriously the prohibition of discrimination on grounds of poverty, in order to combat effectively all forms of povertyism. We consider these tools in turn.

A jobs-rich model of development: making the right to work a reality

Given the negative impact on children’s well-being and development of growing up in households with low resources, a key element in breaking the cycle of poverty must be developing a range of policies to enhance access to adequate income and resources for these households. Improving access to employment for adults from low-income backgrounds therefore has an essential role to play in combating child poverty.

How to achieve this? The focus since the 1990s has been on supply-side factors, rather than on the demand side of the employment market; on the individual’s incentives to work, rather than on the structural factors that society can influence. The emphasis on ensuring that ‘work pays’ belongs to that approach, and it is of course essential that in-work poverty be addressed, if it is to provide a remedy against the perpetuation of poverty (see Peña-Casas et al, 2019). Similarly, lifelong learning and training policies can support parents’ access to the labour market by ensuring that they acquire the ‘right’ qualifications: those that the market demands. Finally, the obstacles to employment that result, especially for women, from family responsibilities, can be addressed by promoting family-friendly working conditions (such as parental leave, workplace support and flexible working arrangements); by ensuring access by disadvantaged groups to affordable quality childcare; and by the promotion of gender equality in the labour market and in family responsibilities.

All these tools matter and should be strengthened further. Another approach, however, complementary to the policies that focus on the individual, is to impose on governments a duty to ensure that each individual able and willing to work will be provided with a job, paid at least a living wage and offering working conditions consistent with the requirements of a decent job.

This would go beyond the right to work as it is currently recognised in international human rights law, most notably in article 6 of the International Covenant on Economic, Social and Cultural Rights. Until now, this right has not been interpreted as requiring from states that they provide jobs to the long-term unemployed. Instead, it has been read more narrowly, as imposing on states obligations of means, rather than obligations of result. As expressed in the ILO Employment Policy Convention, 1964 (No 122), states have a duty to adopt ‘an active policy designed to promote full, productive and freely chosen employment’. In other terms, states must endeavour to promote employment, by appropriate macroeconomic and fiscal policies; but they are excused if they fail to guarantee a job for all. It is now time to ask whether the interpretation should be revised.

The proponents of the ‘Job Guarantee’ idea, in particular within the Levy Economics Institute at Bard College (Tcherneva, 2020), take as their departure point that unemployment has huge costs both to the individual and to society. For those affected, unemployment is not only a loss of income, significantly increasing the risk of poverty (Sen 1997). It also results in a depletion of skills and in a loss of confidence for individuals. It exposes to discrimination: employers are reluctant to hire long-term unemployed candidates, since they tend to interpret the fact that a person has been unemployed for long periods of time as betraying a lack of motivation (Van Belle et al, 2018).

In addition to its impact on the individual, unemployment imposes a huge cost on society. It represents a waste of talent, when so many societal needs are still unmet. In advanced economies that provide unemployment benefits or social aid, unemployment has to be compensated for with public funds, to provide basic income security to the individual. In addition to those direct costs, unemployment has a range of indirect costs (Watts and Mitchell, 2000). It is also correlated with poor health and depression, and with higher crime rates (Raphael and Winter-Ebner, 2001). The failure to address unemployment may compound the impacts of an economic crisis, since the decrease in demand may create the conditions for a longer recession. In local communities where layoffs take place, indirect job destruction spreads like a disease to surrounding areas (Tcherneva, 2019).

In contrast to the general acceptance of mass unemployment as a permanent feature of the economy (and, for some, as a convenient means to justify paying workers less), the ‘Job Guarantee’ recognises that each individual has a right to work (Mitchell, 1998). Its basic premise is that governments, including in particular local government, or private entities with the support of public funding, will provide a decent job at acceptable conditions (including a fair remuneration) to each working-age individual who seeks to work, if that individual is unable to find employment at acceptable conditions elsewhere.

A society that provides a Job Guarantee recognises that each individual has a valuable contribution to make to social progress. It is a society that recognises the worthiness of each of its members, and that refuses the idea that some are redundant. It is also a society that acknowledges the distinction between employment that is created in response to demand, as expressed by the market, and employment that is created in response to societal needs, including needs that the market cannot satisfy because no household or enterprise is willing and able to pay.

Jobs that benefit the local community or society as a whole, rather than only the specific employer, will typically be economically viable only if supported through the public purse. This is the case for jobs that establish or maintain ‘commons’, such as ecosystems or projects that are accessible to all, from collective vegetable gardens to community-led energy cooperatives, public housing projects and initiatives in the sharing economy; jobs that contribute to the circular economy, encouraging the repair, reuse and recycling of consumer items; and jobs that provide care and support to groups of the population that cannot afford to pay for such support themselves – including older persons, people with disabilities or low-income households. The type of jobs generally proposed in the context of the Job Guarantee are often not prioritised or seen as financially viable by the private sector, which means that most of these jobs do not compete with private sector jobs.

Guaranteeing the right to work through a Job Guarantee can be costly (since the jobs created will be financed by the state). Yet, the immediate fiscal implications should be balanced against the enormous benefits, both for the individual and for society, of combating long-term unemployment. In addition to the savings involved – since unemployment benefits or social aid to the job-seekers can be suspended – the workers provided with employment will spend as consumers and contribute to social insurance schemes. The services they will provide will benefit society, meeting needs that, in the absence of a solvent demand, markets currently cannot respond to, and can contribute to environmental sustainability. This will also give meaning to their work, bringing it closer to what one scholar has described as ‘socially capability-enhancing work’, in contrast to dominant approaches to labour that value it only to the extent that it is so-called ‘productive’ work, that is, work valued by the market (Bueno, 2022; Veltman, 2016).

An additional benefit of introducing a Job Guarantee is that it will serve to de-commodify labour – the immediate implication being that, in all the branches of the economy, including in the profit-driven sectors, the bargaining position of workers will be strengthened, since they have the fall-back option of seeking decent employment paid for by the public purse. In 1919, article 427 of the Treaty of Versailles stated that ‘labour should not be regarded merely as a commodity or article of commerce’, and this principle was again enunciated in 1944, in the ILO Declaration of Philadelphia. It is this idea that the Job Guarantee will help to realise.

Two fears have been expressed in the debate concerning the introduction of a Job Guarantee through public employment schemes. First, it is said, the introduction of the Job Guarantee could be used as a convenient pretext for making social protection conditional on accepting a job, provided such a job is considered ‘suitable’. This is a legitimate concern. There exists a strong political pressure to increase work requirements within existing social protection support, a trend sometimes described as the ‘activation’ of social protection (De Schutter, 2015). In countries where social protection is weak and fiscal space limited, it also may be politically easier to provide income security by establishing or expanding public employment programmes, rather than by expanding other forms of (unconditional) social protection.

The shift ‘from welfare to workfare’ is not inevitable, however. Participation in the scheme should be on a strictly voluntary basis, not as a condition for receiving other kinds of support. The introduction of a Job Guarantee scheme could be paired with a requirement of non-retrogression in the provision of unconditional social protection, to avoid a slide towards workfare. It could also be combined with initiatives to better value care work, performed within households or communities, often without remuneration or even formal recognition.

Second, some actors, unions in particular, have expressed the fear that public employment programmes could lead to lowering public sector wages, and to weakening the bargaining position of public sector unions, since such programmes would lead to the creation of low-paying jobs to perform the same duties as traditional public sector jobs: public administrations may be tempted to downsize certain services and to outsource them to the Job Guarantee scheme. Yet, this too is not inevitable. Such a consequence would follow only if the wages paid in the Job Guarantee scheme were lower than market wages – for instance if they corresponded to no more than the statutory minimum wage. However, leaving aside the fact that in many developing countries many informal jobs are in fact paid at wages below the minimum wage, the Job Guarantee scheme can be designed to provide better wages and working conditions, thus raising the bar across the employment market and strengthening the bargaining power of workers across the economy (Atkinson, 2015, p 144).

A basic income for young adults

A truly inclusive society, one that is designed to ensure real equality of opportunities, may also rely on the tool of a universal basic income (UBI). While a range of proposals have been made to implement this idea, either to complement classic welfare programmes (Van Parijs and Vanderborght, 2017) or, in a libertarian-conservative version of a ‘negative income tax’, to replace it (Friedman, 1968), our preferred scenario is that of a UBI introduced to provide all young adults, between the end of secondary education and age 25, with a monthly stipend, on an unconditional basis. This would be providing children from a low-income background with a third chance, where interventions in early childhood and attempts to provide inclusive education remain insufficient. Providing a UBI to support all young adults as they enter into their adult lives may be the most realistic option, both politically and economically: the universal nature of the benefit ensures that it may gain support from large parts of the electorate; and the burden on public budgets would remain bearable, whereas a UBI provided throughout the life cycle, or to all the adult population, could result in difficult trade-offs as other public expenses may have to be cut.

Contrary to a widespread assumption that income support provided unconditionally may discourage work, the provision of a UBI in such a form may in fact favour labour market integration. A study of a randomised cash transfer in Uganda showed that most youth (who normally would not have access to credit in the absence of aid) invested the transfer in vocational skills and tools, leading to significant increases in cash earnings (almost 50% relative to the control group): the real annual return on capital was 35% on average (Blattman et al, 2014). Similarly, an experiment with the introduction of a UBI scheme in rural Kenya shows the benefits of even a relatively modest guaranteed and unconditional income scheme (of the equivalent of US$0.75/day) for improved food security, mental and physical health, and entrepreneurship – protecting households from having to sell productive assets in times of crisis and encouraging them to invest in productive investment (Haushofer and Shapiro, 2016). Studies of UBI schemes in rich countries show either no negative impact on employment or only a marginal impact (a 10% income increase induced by an unconditional cash transfer decreasing labour supply by about 1%), but significant improvements in health and educational outcomes, especially among the most disadvantaged youths (Marinescu, 2018).

By definition, due to its universal nature, UBI for the youth is not stigmatising, and the risks associated with targeting in means-tested programmes are avoided. In most countries, such schemes could be financed by increasing taxes on inheritance. This would also be a coherent way of tackling the growth of wealth inequalities. In OECD countries, the inheritances and gifts reported by the wealthiest households (top 20%) are close to 50 times higher than those reported by the poorest households (bottom 20%), which illustrates the important role of inheritance in perpetuating and even reinforcing inequalities – since wealth inequalities lead to increased income inequalities. Yet only 24 out of 37 OECD countries tax inheritance, estate or gifts across generations, and the levies are typically very low, accounting for only 0.5% of total tax revenues on average for the 24 countries concerned. There are important differences between countries, of course: in the US, only 0.2% of estates are subject to inheritance taxes, and parents may transfer up to US$11 million to their children exempt of taxes, while the figures are 48% and US$17,000 for the Region of Brussels-Capital in Belgium (OECD, 2021a). Overall, however, in combination with the introduction of a UBI scheme for young adults to allow them to pursue their education or to start a small business, taxing inheritance or increasing progressivity in inheritance taxes may be an administratively easy and essentially painless way to break the cycle of poverty. (Thomas Piketty makes a very interesting point in relation to inheritance tax. He suggests that the ‘ideal tax system’ is ‘a compromise between the incentive logic (which favours a tax on the capital stock) and an insurance logic (which favours a tax on the revenue stream stemming from capital)’. He argues that ‘the unpredictability of the return on capital explains … why it is more efficient to tax heirs not once and for all, at the moment of inheritance … but throughout their lives, via taxes based on both capital income and the value of the capital stock. In other words, all three types of tax—on inheritance, income, and capital—play useful and complementary roles’ [Piketty, 2014, p 527].)

The prohibition of discrimination on grounds of socioeconomic disadvantage

Given the extent to which discrimination causes and perpetuates poverty and social exclusion (see Chapter Two, in section ‘Discrimination’), strengthening the protection against discrimination on grounds of poverty (or socioeconomic disadvantage) should be part of any anti-poverty programme. Indeed, under international human rights law, states already have an obligation to address such a form of discrimination. Article 2(2) of the International Covenant on Economic, Social and Cultural Rights mentions ‘social origin’ and ‘property’ among the prohibited grounds of discrimination. The Committee on Economic, Social and Cultural Rights has reiterated that people ‘must not be arbitrarily treated on account of belonging to a certain economic or social group or strata within society’ (UN Committee on Economic, Social and Cultural Rights, 2009, para 35), and it insists that such grounds should be included in the anti-discrimination framework adopted by the states parties to the Covenant.

At its core, the idea is simple enough: people in poverty cannot be treated adversely simply because they are poor; in principle, their underprivileged socioeconomic situation cannot be allowed to result in a reduced ability to enjoy human rights. Yet, discrimination against individuals or groups of individuals on grounds of socioeconomic disadvantage remains widespread, and largely unpunished. The introduction of an explicit protection from discrimination on grounds of social condition in particular would not only have a strong symbolic value, sending a clear message to policy makers in particular that people may not be treated less favourably because they are poor. It would also have institutional consequences, in particular by allowing Equality Bodies (independent institutions tasked with preventing and addressing discrimination and promoting equality) to contribute more effectively to the fight against poverty, alongside the specific contribution other human rights mechanisms can make.

This remains a largely unfinished task. While some progress has been achieved on paper (in the legislative and regulatory framework), the protection from discrimination on grounds of social status is in practice, at best, highly uneven (Thornton, 2018); there are in fact few examples of this ground being used effectively. The European Network of Equality Bodies notes that relying on such a ground of discrimination raises specific challenges, both because ‘the meaning, situation and character of the socio-economic status ground is not understood by many people in the society and even within the equality body’ and because, in the large majority of cases, discrimination on grounds of social condition (or socioeconomic status) is combined with discrimination on other grounds (particularly sex, race or ethnic origin, disability or age), requiring specific methodologies to be developed to address multiple discrimination appropriately (Equinet, 2010). However, as the Equality Bodies also note, the inclusion of such a ground in anti-discrimination legislation has an essential role to play, not least since poverty is often an obstacle to the filing of discrimination claims, a barrier that the explicit reference to social condition (or socioeconomic status) in the mandate of Equality Bodies could help to overcome – since it would provide a clear encouragement to people in poverty to use the tool of filing individual complaints to challenge discriminatory practices.

The potential role of the prohibition of discrimination on grounds of ‘social condition’ could be maximised by relying on an expanded notion of the concept of ‘social origin’ which appears in Article 2(2) of the International Covenant on Economic, Social and Cultural Rights. The Committee on Economic, Social and Cultural Rights (CESCR) understands this expression to refer to the ‘social and economic situation when living in poverty or being homeless’ (UN Committee on Economic, Social and Cultural Rights, 2009, para 35). As noted by Angelo Capuano, however, this definition may be unnecessarily restrictive, and thus potentially irrelevant in many contexts where discriminatory treatment is based. First, the ‘social status’ of a person ‘is reflected more by prestige and esteem rather than merely property status, wealth or economic status’ (Capuano, 2017, p 105). Moreover, ‘the criteria which the CESCR seems to use to give content to the concept of “social status” – property status, caste, and economic and social status such as homelessness and poverty – are not likely to be bases upon which an employer will commonly have the opportunity to discriminate’ (Capuano, 2017, p 106). Instead, he suggests, discrimination most frequently occurs on the basis of family relationships, of the schools the person has attended or of childhood circumstances: these are instances of ‘ascribed status’, rather than ‘achieved status’, and it is these instances in particular that, in such a context, one should be devoting more attention to (Capuano, 2017, pp 109–110).

Two implications follow. A first implication is that, in the understanding of ‘disadvantaged socio-economic condition’ or simply ‘poverty’ (whatever the anti-discrimination framework provides for), attention should be paid to both objective and subjective factors – in other terms, to both the circumstances a person faces and the stereotypes or prejudice associated with such circumstances. Guidance may be found in the interpretation provided to the prohibition of discrimination based on the term ‘social condition’, which appears in Article 10 of the Charter of Human Rights and Freedoms of the Canadian Province of Québec:

The definition of ‘social condition’ contains an objective component. A person’s standing in society is often determined by his or her occupation, income or education level, or family background. It also has a subjective component, associated with the perceptions that are drawn from these various objective points of reference. A plaintiff need not prove that all of these factors influenced the decision to exclude. It will, however, be necessary to show that as a result of one or more of these factors, the plaintiff can be regarded as part of a socially identifiable group and that it is in this context that the discrimination occurred. (Comm. des droits de la personne v. Gauthier [1993], 19 C.H.R.R.D/312 [English summary])

A second implication is that, in addition to direct discrimination on grounds of socioeconomic disadvantage or poverty, indirect discrimination should be prohibited, where decisions made on other grounds, including on apparently neutral grounds, disproportionately affect people in poverty. Homelessness, for instance, as a proxy for poverty, should not be allowed to lead to discrimination (Farha, 2015, para 39). Employers should not be allowed to reject job applicants based on where they live (in poor neighbourhoods) or on the reputation of the schools the candidate attended (disproportionately attended by pupils from disadvantaged backgrounds). Landlords should not be allowed to refuse to rent an apartment to a lessee who relies on social aid. Schools should not be allowed to penalise students who cannot buy teaching materials or who lack access to the internet. Beyond the diversity of such instances of discrimination, what unites them is that people in poverty are penalised for being poor, adding to the disadvantage that stems from lack of income alone.

Legal tools can be designed to address these situations in their variety. In Ireland for instance, the Equality (Miscellaneous Provisions) Bill 2021 (which is still pending adoption at the time of writing) defines having a socioeconomic disadvantage as being member of a

socially or geographically identifiable group that suffers from such disadvantage resulting from one or more of the following circumstances: (a) poverty, (b) source of income, (c) illiteracy, (d) level of education, (e) address, type of housing or homelessness, (f) employment status, (g) social or regional accent, or from any other similar circumstance. (Houses of the Oireachtas [Ireland], 2021)

In South Africa, the Promotion of Equality and Prevention of Unfair Discrimination Act (implementing Section 9 of the Constitution) contains a ‘Directive Principle’ that requires giving special consideration to the inclusion of, inter alia, ‘socio-economic status’ in the list of prohibited grounds: this expression is defined as the ‘social or economic condition or perceived condition of a person who is disadvantaged by poverty, low employment status or lack of or low-level educational qualifications’. The formulae may be different, but the central idea remains the same: it should not be allowed to circumvent the prohibition to discriminate on grounds of poverty by using proxies such as a lack of a fixed address, a long-term unemployed status or reliance on social aid, whether such characteristics are relied upon consciously or unconsciously by decision makers.

Public entities, moreover, should not be allowed to make policy decisions or decide regulatory reforms without inquiring into the impacts on people in poverty and ensuring that their decisions do not worsen inequalities. We detail this positive duty further below.1

The three roles of equality and non-discrimination in combating poverty

The prohibition of discrimination on grounds of social origin or property extends to any action or omission that disproportionately affects members of a particular group, in the absence of a reasonable and objective justification. (Such form of discrimination is referred to as de facto discrimination, to distinguish it from discrimination that is de jure, in other terms, that relies explicitly on poverty or socioeconomic condition as a characteristic on which adverse treatment is based.) There is both a negative and a positive face to this. First, regulatory or policy measures that are neutral on their face may be considered discriminatory if they do not take into account the disparate impacts they may have on certain groups of the population, defined for instance on the basis of ‘property’, or income levels. In South Africa for instance, a Western Cape Equality Court considered that the drastic difference in resources allocated by the South African Police Services to impoverished, predominantly Black communities in comparison to wealthier, White communities amounted to discrimination on the basis of race and poverty, the latter being an ‘analogous unlisted ground’ on which a claim of discrimination can be based because it ‘adversely affects the equal enjoyment of a person’s rights and freedom in a serious manner that is comparable to discrimination on a listed ground’ (Western Cape High Court, Social Justice Coalition v Minister of Police, 2018 ZAWCHC 181, para 65).

It is this idea that is at the heart of a ‘positive duty’ to consider the impacts on poverty in law and policy making, as imposed for instance in Scotland since April 2018 as part of the Equality Act 2010: this duty, referred to as the ‘Fairer Scotland Duty’, imposes on a number of public bodies in Scotland to ‘actively consider (“pay due regard” to) how they can reduce inequalities of outcome caused by socio-economic disadvantage, when making strategic decisions’.2 In practice, this means that decisions such as where to locate a school or a hospital, or how to develop a neighbourhood, should be made with the involvement of the local community, and aim at adopting a decision that will reduce, rather than increase, the exclusionary impacts of lack of income – thus contributing to a more inclusive, less divided society. Diana Skelton, a volunteer for ATD Fourth World, described to one of us3 how low-income families expressed their concerns after plans were announced for the Monklands University Hospital in North Lanarkshire to move: these families were living in the vicinity of the existing hospital, and they feared that they might not be able to travel to the new location to seek treatment. Thanks to the consultations that took place, a compromise could be found, providing for some primary healthcare services to remain in Monklands.

Human rights impact assessments serve to alert policy makers to the impacts on the human rights of the poor of the policies they design and implement. Human rights impact assessments, it should be emphasised, are distinct from other types of assessments, including social impact assessments or sustainability impact assessments, with which they present certain similarities. The specificity of human rights impact assessments is that they examine the intended and unintended impacts of policy measures on the ability of the states parties to these agreements to respect, protect and fulfil the human rights of people living in poverty. They therefore should be based explicitly on the normative content of human rights, as clarified by the judicial and non-judicial bodies that are tasked with monitoring compliance with human rights obligations. References in impact assessments to development goals or to poverty are therefore not a substitute for a reference to the normative components of human rights.

The requirement of non-discrimination against the poor is especially important where states face an economic or financial crisis and adopt fiscal consolidation (so-called ‘austerity’) programmes in order to reassure their creditors as to the health of their public budgets (Bohoslavsky, 2018). In his Letter of 16 May 2012 to the States Parties to the International Covenant on Economic, Social and Cultural Rights on austerity measures, the Chairperson of the CESCR – the expert body tasked with supervising implementation of the covenant – emphasised that fiscal consolidation policies ‘must not be discriminatory and must comprise all possible measures, including tax measures, to support social transfers to mitigate inequalities that can grow in times of crisis and to ensure that the rights of the disadvantaged and marginalised individuals and groups are not disproportionately affected’. In a statement adopted in 2016 by the same Committee, it noted, by reference to Article 2, para 2 of the Covenant, that ‘Low-income families, especially with children, and the workers with the lowest qualifications are disproportionately affected by measures such as loss of jobs, freezing of the minimum wage and cutbacks in social assistance benefits, potentially resulting in discrimination on grounds of social origin or property’ (UN Committee on Economic, Social and Cultural Rights, 2016, para 2). The 2012 Guiding Principles on Extreme Poverty and Human Rights also emphasise this duty, noting that:

Given the disproportionate and devastating effect of economic and financial crises on groups most vulnerable to poverty, States must be particularly careful to ensure that crisis recovery measures, including cuts in public expenditure, do not deny or infringe those groups’ human rights. Measures must be comprehensive and non-discriminatory. They must ensure sustainable finance for social protection systems to mitigate inequalities and to make certain that the rights of disadvantaged and marginalised individuals and groups are not disproportionately affected. (OHCHR, 2012, para 54)

The Guiding Principles on human rights impact assessments of economic reforms presented in 2018 by the Independent Expert on foreign debt and human rights provide further guidance as to how the human rights impacts of fiscal consolidation programmes, as they are adopted following an economic crisis leading to an increase in the sovereign debt and thus additional borrowing by states, should be conducted (Bohoslavsky, 2018).

The equality requirement goes beyond this negative duty, however. In order to prevent discriminatory results, states may have to provide positively for differential treatment benefiting certain categories of the population facing systemic disadvantage. In cases of entrenched discrimination, states may be under an obligation to adopt special measures to attenuate or suppress conditions that perpetuate discrimination. Such measures are legitimate under human rights law to the extent that they represent reasonable, objective and proportionate means to redress de facto discrimination and are discontinued when substantive equality has been sustainably achieved. Courts have sometimes considered with suspicion differential treatment benefiting certain underprivileged groups defined by their ethnicity or gender – a suspicion which reflects adherence to a formal understanding of equality, resulting in a restrictive reading of non-discrimination law as forbidding the ‘sin’ of discrimination rather than as a tool to remedy injustices that have their source in society-wide mechanisms of exclusion (Sullivan, 1986). But courts have been far more open to affirmative action measures taken to improve the situation of those who are economically deprived, since socioeconomic condition is not a suspect ground: in fact, such measures are at the very heart of the construction of welfare states, the main purpose of which is to provide support to those who are excluded by the mechanisms of the market.4

States therefore should dedicate greater resources to improve the condition of groups who face systemic discrimination (UN Committee on Economic, Social and Cultural Rights, 2009, para 39). They also should move up the causality chain, to tackle the underlying causes of social exclusion. Indeed, once it is recognised that ‘a great deal of poverty originates from discriminatory practices – both overt and covert’, it follows that poverty-reduction strategies will be fully effective only if they also address ‘the socio-cultural and political-legal institutions which sustain the structures of discrimination’, and eliminate ‘the laws and institutions which foster discrimination against specific individuals and groups’ (OHCHR, 2005, para 21). The fight against inequalities, based in particular on social condition, should concern not only the sphere of economic, social and cultural rights but also the sphere of civil and political rights, since inequalities in access to political influence and socioeconomic inequalities are mutually reinforcing (Alston, 2015, para 21).

Finally, in order to properly assess the contribution of prohibiting discrimination on grounds of socioeconomic disadvantage to breaking the cycles perpetuating poverty, the discrimination faced by disadvantaged individuals and households should be seen for what it is: a form of systemic discrimination which affects a range of areas including health, education, housing and employment.

Addressing discrimination on grounds of socioeconomic disadvantage is therefore ineffective if limited to one sphere alone. For instance, ensuring that employers do not discriminate on grounds of poverty will have a limited impact if disadvantaged individuals continue to face obstacles in having access to quality education, or live in poor neighbourhoods distant from the place of work; supporting schools with a high proportion of disadvantaged pupils may not make a significant difference to these pupils if residential segregation remains unchallenged, so that these pupils remain concentrated in certain schools; and neither combating discrimination in employment nor in education will suffice if health inequalities persist, lowering workers’ productivity and academic achievement.

This also points to the limits of an approach to equality of opportunities which relies on classic understandings of ‘merit’. In fact, societal improvements pursued in the name of ‘meritocracy’, including the use of a classic anti-discrimination framework simply prohibiting discrimination but without including class-based affirmative action, could be counter-productive. As emphasised by the Harvard political philosopher Michael Sandel, the more a society promotes meritocracy by insisting on equality of opportunity at the starting line, the more it risks justifying whatever inequalities follow as the legitimate result of inevitable differences in talent between individuals, combined with the hard work, persistence and discipline of those who succeed (Sandel, 2020). It is not mere coincidence that the belief in meritocracy is entertained in particular by the wealthiest groups within the population (Roex et al, 2019): high-income earners have an interest in believing (and in making believe) that people in poverty are less ‘meritorious’ and more deserving of their socioeconomic situation, thereby justifying inequality on the basis of unequal merit (Redmond et al, 2002; Heiserman and Simpson, 2017). This is equivalent to blaming people in poverty for being poor, triggering feelings of shame among the poor (Walker, 2014, pp 132–156). Pushed to its limit, ‘meritocracy’ as an ideology, despite its popularity in public discourse, thus both may reduce empathy towards affected groups and make inequality look like an inevitable and, to some extent, even desirable phenomenon – a means to incentivise people to achieve more. Social psychologists have shown how ideological frames could distort our attentiveness to inequality, and thus our willingness as a society to address it (Waldfogel et al, 2021): this selectiveness towards inequalities (denounced as a scandal where it is based on ethnicity or sex, but seen as legitimate and even desirable where it is seen as attributable to the individual’s choices and attitudes) may be one high price to pay for the adherence to a meritocratic ideal of society.

But the meritocratic ideal is problematic at an even more basic level. Reliance on ‘meritocracy’ is entirely inappropriate, indeed, where disadvantaged individuals have not been given fair opportunities to acquire certain qualifications or to have their experiential competences formally recognised. Instead, affirmative action policies are essential to break the vicious cycles that result from the systemic nature of the discrimination faced by people in poverty. Whereas preferential treatment is well established as regards the allocation of goods or services that compensate for poverty or social exclusion, as in means-tested social protection schemes or in the award of scholarships to help overcome financial barriers to education, it is less common and more heavily contested where it is seen to challenge the mainstream narrative about ‘deservingness’, as in access to employment or to the most coveted schools or universities. Yet, affirmative action is especially needed in such fields, if real equality of opportunities is to be achieved (De Schutter, 2022a, paras 37–40).

Examples abound. Israel successfully designed a form of class-based affirmative action to access the country’s most prestigious universities since the mid-2000s, which determines socioeconomic disadvantage on the basis not only of financial status but also of neighbourhood and high school attended, family socioeconomic status (including parental education and family size) and ‘individual and/or family adverse circumstances’ (Alon and Malamud, 2014). In India, while the Constitution includes various anti-discrimination provisions and bans the practice of ‘untouchability’ (Art 17), it also states that special measures may be adopted ‘for the advancement of any socially and educationally backward classes of citizens’, as a means to reduce social inequalities for members of these groups (Art 15, (4) and (5)). This mainly takes the form of reserved seats in public offices and educational institutions (both public and private), as well as job reservations in the public sector, for the castes and tribes mentioned in Articles 341 and 342. In addition however, Article 16(4) of the Constitution allows for ‘the reservation of appointments or posts in favour of any backward class of citizens which, in the opinion of the State, is not adequately represented in the services under the State’: consistent with this constitutional mandate, the Central Educational Institutions (Reservations in Admissions) Amendment Bill stipulates that 27% of seats are reserved for ‘Other Backward Classes’ in publicly funded higher education institutions, a policy which led to significant improvement in the socioeconomic diversity in universities (Basant and Sen, 2020).

Affirmative action is in principle acceptable under international law (Bossuyt, 2002). Human rights bodies recognise that it may be required to combat systemic discrimination, and legislation implementing affirmative action programmes occasionally frames such programmes not as a derogation from the principle of equal treatment, but instead as a consequence of that principle. Domestic courts have correctly taken the view that such policies are not a derogation to the principle of non-discrimination, but rather should be seen as implementing the mandate to ensure effective equality, in particular for low-income groups. In Society for Un-aided Private Schools of Rajasthan v Union of India, the Indian Supreme Court upheld a requirement imposed on private unaided schools under Section 12(1)(c) of the 2009 Right to Education Act to fill 25% of the seats in Class I with children from weaker and disadvantaged groups, taking into account that the Act sought to remove ‘financial and psychological barriers which a child belonging to the weaker section and disadvantaged group has to face while seeking admission’, and that this objective could justify reasonable restrictions to the economic freedoms of educational establishments. In Kenya, the High Court allowed a government policy providing more opportunities in national schools to students from public institutions as opposed to students from private institutions: it found that this measure was aimed at achieving substantive equality by reducing the inequality gap between the rich and the poor and was consistent with Article 27(6) of the Kenyan Constitution, which commits the State to give full effect to the realisation of the right to equality and freedom from discrimination by taking legislative and other measures, including affirmative action programmes and policies designed to redress any disadvantage suffered by individuals or groups because of past discrimination.

Affirmative action contributes to increased diversity in different sectors and levels of the professional sphere, providing role models to adolescents and young adults from underprivileged backgrounds. We have also seen how such promotion of diversity can breed greater tolerance and understanding between social groups. Greater diversity also results in decisions made in institutions being better informed by the lived experiences of people in poverty, reducing the risk of indirect (including unconscious) discrimination; and the services provided by such institutions will be more attentive to the specific circumstances of low-income people. Beyond its role as a social-engineering tool, affirmative action recognises the specific obstacles that people in poverty face owing to the persistence of povertyism, thus questioning the mainstream narrative about society distributing outcomes on the basis of ‘merit’. This is not merely of symbolic value: it sends a powerful message across society, contradicting the mainstream view about success or failure being attributable to the individual rather than to society’s being insufficiently inclusive.

2

See Fairer Scotland Duty: Interim Guidance For Public Bodies (Scottish Government, March 2018), p 5.

3

At a meeting attended in London by Olivier De Schutter, in November 2019.

4

This should be nuanced, since not all welfare states seek to achieve equality: some models aim only at protecting individuals from extreme deprivation, without setting wealth redistribution as an objective in its own right. See Gosta Esping-Andersen (1990).