Editorial
We are most grateful to the Political Quarterly for permission to reprint ‘The Many Faces of Universal Basic Income’, by Jürgen de Wispelaere and Lindsay Stirton. This substantial article offers us “a better appreciation of the wealth of proposals falling under the rubric of universal basic income, and the potential diversity of arrangements that exist at the level of concrete design and implementation.”
It raises a question about that rubric, debated before in this newsletter and also on our website: the question of terminology. Since the first Basic Income European Network congress in 1986 the terms ‘Basic Income’ and ‘Citizen’s Income’ have referred to an unconditional payment to every individual as a right of citizenship, paid automatically, and nonwithdrawable. The article employs the term ‘Basic Income’ to denote a family of schemes, the schemes characterised by a diversity of conditionalities, withdrawabilities and coverage. The question is: Should ‘Basic Income’ and ‘Citizen’s Income’ be used to denote only schemes in which nonwithdrawable payments are made to every individual unconditionally, or should the meaning of the terms be stretched to cover other income maintenance programmes sharing one or more of the characteristics of a Citizen’s Income ?
Readers’ comments are always welcome.
The many faces of Universal Basic Income
by Jürgen de Wispelaere and Lindsay Stirton
During the past decades the debate on the desirability and feasibility of universal basic income has reached maturity. Associated with a growing number of scholars, social activists, public advocacy groups and political parties, basic income is no longer perceived as yet another crackpot idea of the radical left. Indeed, it is increasingly accepted that basic income advocates have something valuable to contribute to the debate on welfare reform and employment regulation. But with maturity comes the need to rethink the ideal of a universal basic income. As the debate expands, the standard definition of basic income as an income granted by right to each individual, without means test or work requirement, may no longer capture the diversity of policies advanced within the basic income community and beyond.
This article contributes a first step to this enterprise by charting the many faces of universal basic income. Our starting point is the belief that successfully implementing a universal basic income crucially depends on our being able to match the design features of a particular scheme with the surrounding policy context or administrative environment, which differs extensively from one country to another. This, in turn, requires a better appreciation of the wealth of proposals falling under the rubric of universal basic income, and the potential diversity of arrangements that exist at the level of concrete design and implementation. It is towards this latter task that this article is specifically directed.
The devil in the detail
For many scholars and practitioners, basic income constitutes a distinctive social paradigm within contemporary welfare theory, leading to vigorous normative arguments and ideological disputes between its proponents and adversaries. This is not to say that there is no substantial disagreement pertaining to form or content of the normative principles underlying universalism amongst its principal advocates. Neither does it imply that we must buy into ideological cleavages to find arguments for or against basic income. In fact, one of the intriguing aspects of basic income is precisely its capacity to secure support across the ideological spectrum.1
Moving from social philosophy to policy it becomes apparent that the paradigm of universalism does not constitute a single identifiable policy, but represents a myriad of social support schemes that differ substantially along a range of policy dimensions. Basic income supporters readily acknowledge that there exists ample choice of which policy to pursue within a broadly universalist approach. Consequently, the debate has now moved from defending universalism writ large to a dispute within the basic income community itself over the preferred form of basic income. The result is substantial disagreement at the level of ideal-type policies: some scholars favour a negative income tax scheme, others advocate an unconditional basic income or a participation income, and still others believe stakeholder or basic capital grants are superior, and so on.2
Furthermore, at the level of fine-grained design and implementation, apparently similar proposals are even further differentiated along dimensions that are characteristically not captured in ideal-type analysis. An additional concern is that universal schemes that are substantially similar in design may still end up producing widely divergent outcomes because of different interaction effects with policies already in place. It is a mistake to assume that a universal basic income would operate in something resembling an institutional vacuum. To the extent that fine-tuned distinctions also produce distinctive outcomes, both normative and empirically driven research ought to take differential design features seriously. In the next section we discern seven principal dimensions along which concrete basic income proposals can be differentiated.
Dimensions of basic income
Universality
Universality refers to the extent of the population that is covered by a given policy. Typically, universal policies are open to all, while more selective measures single out a subset of the population as beneficiaries. One category of subjects often excluded from even the most universal schemes is non-citizens (however defined), while more selective measures discriminate even further to select eligible individuals or groups from the broader population. Selectiveness immediately invokes debate regarding the principles and mechanisms employed to decide on eligibility. In practice this implies building in some level of conditionality, discussed further below.
The distinction between universal and selective measures, however, is often overstated on ideological grounds. To begin with, the label ‘universalist’ is misleading in cases where policies are universal in some respects but selective in others. Most policies in contemporary welfare regimes appear to fit this mixed-bag category. In addition, a strict divide between universal and selective measures is easily blurred in practice. Circumstances typically introduce selective effects in an otherwise universal policy; conversely, selective measures may well combine to mimic the effects of a universal policy.
Basic income advocates often favour an incremental approach to instituting a full basic income. One way in which this could be done is to have basic income type policies in a specific domain-child benefit, basic pension or sabbatical accounts-which are then gradually expanded or ‘universalised’ over time.3 Here too we must be wary about attaching too much importance to the label and ignoring what happens on the ground. In what follows we review various ways in which universal basic income schemes can be more or less universal, as well as other salient dimensions in which concrete proposals can be differentiated.
Individuality
Individuality refers to the standard unit at which a policy is directed. Welfare policy schemes basically face a choice of administering their services either directly to individuals or indirectly through a household unit. Basic income is routinely advocated as a form of income support that caters to individuals. However, some advocates, mainly for reasons of goodness-of-fit with traditional approaches in welfare policy, have been willing to compromise on this and recommend instituting a basic income targeted to households. The dimension of individuality therefore does not lose its relevance for policy purposes.
In the case of households many questions arise with respect to the appropriate definition of a household and its internal composition. Tony Atkinson distinguishes four types of household units, each with its own delineation and composition problems: ‘households’ based on common residence; ‘spending units’ based on common spending patterns; ‘family units’ defined by blood ties or marriage; and finally the ‘inner family’ defined in terms of a sustained dependence relationship.
Switching from one household base to another in policy design has been shown to imply up to a quarter of reductions in the measurement of people on low income.4 Measuring the actual effects of welfare policies is clearly sensitive to choice of household type. Similar observations can be made regarding the administrative challenges associated with targeting policies to households. These reasons often lead policy-makers to favour a more individualised approach.
From a normative point of view, the chief sources of concern are the often arbitrary discrimination of life-style choices associated with household-based policies; unacceptable inequalities between single-income and double-income households and between single persons and double-income households (sometimes leading to perverse redistributive effects from the poor to the well-off); and the fact that non-individualised rights often generate employment traps or trap partners into a dependency relation.5 While conservative political factions are often keen to use welfare policies to strengthen the traditional nuclear family unit, the increasing variation of living arrangements within and across generations suggests this argument may have outlived its usefulness. Although much attention has recently been devoted to the effects of individualising benefits on intra-householder power relations, the research remains surprisingly ambivalent about its implications for basic income.6
Conditionality
Conditionality implies the extent of conditions built into a policy that may restrict a person’s eligibility for a service. Most welfare policies come with different types of conditions attached that recipients need to satisfy to gain or maintain eligibility. Basic income is of course distinctive precisely in that it is purportedly unconditional or, failing that, at the very least only employs conditions that do not violate the programme’s inclusiveness. An example of a conditionality requirement that ostensibly does not affect inclusiveness is Tony Atkinson’s well-known proposal for a participation income.7
To understand better the dimension of conditionality a number of distinctions need to be kept in mind. Conditionality refers in the first instance to formal criteria of eligibility that either imply a set of characteristics necessary to acquire eligibility status or, alternatively, impose certain behavioural constraints to retain eligibility (ex ante and ex post conditionality, respectively). In addition to these two main types we can discern hidden or implied forms of conditionality: a universal basic income can become more conditional because of the interplay with external contingencies, which may result in the policy effectively treating recipients differently within a formally uniform framework. Suppose we institute a fully unconditional basic income at a variable level related to a macroeconomic performance indicator such as GDP or employment rates. The level of the grant decreases when more people opt out of formal employment or if productivity decreases below a certain threshold indicator, which serves as a ‘soft incentive’ to push people back into work. While such a scheme does not have any formal conditions attached to it, it nevertheless institutes a set of incentives to contribute towards maintaining a certain level of production or employment.
Next, conditions can be strict or weak depending on whether they are ‘set in stone’ or there is a significant measure of bureaucratic discretion in assessing when a claimant has satisfied a requirement. Bureaucratic discretion invites a measure of arbitrariness, and may induce welfare administrators to engage in behaviour that violates professional standards. Interestingly, welfare workers often oppose discretion and prefer a system that rigorously outlines their duties precisely because they want to minimise the risk of unprofessional conduct.8 In addition, bureaucratic discretion may boost administrative error, particularly when rules change rapidly and become increasingly complex.9 At the same time, basic income research should be aware of the literature in public administration and administrative law that points to the limits of administration ‘by rule and rote’ and of the appropriate uses of discretion.
Finally, conditions can also be narrow or broad depending on whether they result in more or less exclusive policies-that is, policies that capture a larger subset of the population. The Earned Income Tax Credit, for instance, only applies to those in work, whereas a participation income is meant to encompass a broader range of activities, and hence a broader range of target beneficiaries. This of course raises the precarious problem of who ends up making the decision to value certain social activities by including them in the participation requirement. This is not a moot point: conditions are often introduced within a universal basic income for political reasons, because decision makers believe there will not be sufficient political support for unconditional measures-although occasionally economic grounds are also put forward as arguments in favour of some conditionality. On the other hand, increased target efficiency associated with improved take-up rates is often cited as the strength of unconditional measures. In practice, the choice of a basic income scheme and its level of conditionality will depend in large part on which constraint we believe to be the stronger.
Uniformity
Uniformity is the extent to which all those who are eligible receive a similar level of benefit. Universal basic income schemes can deviate from this strict interpretation in at least two ways. First, we may decide to allocate different levels of transfer to different types of recipients, thus imposing a form of ex ante conditionality within the scheme. A familiar example is the use of age to differentiate the allocation of grants to children, adults of working age and pensioners. By making good strategic use of a distinction that is already embedded in existing welfare systems, basic income proponents have a better chance of bringing basic income in via the back door. Even noted opponents of unconditional basic income, such as Gøsta Esping-Andersen, favour universal child benefits and basic pensions, effectively endorsing a basic income ideal for a subset of the population. Differentiating uniformity provides a handy tool for policy design and advocacy.
Contingencies also affect the uniformity of basic income. Imagine, for instance, a universal scheme that is formally uniform but with the value to its recipients fluctuating in line with a set of external circumstances, such as the regional variation in cost of living. It is a matter of some discussion whether a basic income should remain uniform, as argued by Philippe Van Parijs, or instead regional price differences should provide a legitimate departure from the uniformity rule.10 Of course, policy-makers may well decide to use the differential value of the grant to actively influence certain behavioural traits: like taxes, grants may end up serving multiple purposes that need to be balanced. In principle, then, both uniformity and differentiation are consistent with most forms of basic income. Of course, one should keep in mind that at the margin a heavily differentiated scheme may no longer satisfy the key requirement of universality, blurring the line between ‘differentiation’ and ‘selectivity’.
Frequency/Duration
Until recently, the dimensions of frequency and duration were somewhat neglected within universal basic income schemes. But at the end of the 1990s, a real cleavage emerged between universal basic income proposals that provide a regular income stream, as in unconditional basic income or participation income, and schemes where beneficiaries receive a one-off payment, constituting a capital stock as in stakeholder or capital grant proposals.11 With respect to income streams, a further relevant distinction should be made with respect to the timing of regular instalments. It does make a difference whether a recipient receives the grant on a weekly, monthly or even yearly basis. Shorter intervals often draw support from those who emphasise basic security, whereas advocates of equal opportunity, suspicious of any form of paternalism, typically favour longer intervals. Of course, timing may simply be determined by the surrounding ad-ministrative time frame: until recently, wages were commonly paid in weekly instalments in the UK and Ireland as compared to the majority of European countries which employed a monthly pay system. Having basic income ‘piggy-back’ on whatever system is in operation at any given time often makes good administrative sense.
While the distinction between streams and stocks informs much of the current debate, the distinction is prone to overstatement. Under the right circumstances income streams can be converted into stocks and vice versa, though it remains unclear whether such condit-ions are currently present in even the most advanced welfare regimes. In addition, many of the basic capital approaches seem to have some in-built mechanisms of ensuring that the entire grant is not wasted on so-called ‘stakeblowing’ activities. Once we take this expansion into account, the distinction between income and capital grant schemes diminishes.12
A final consideration concerns the duration aspect of basic income. Putting a time-limit upon receipt of assistance is a measure common to most selective income support policies, but could conceivably be used to render universal basic income socially and politically acceptable. A recent proposal by Stuart White argues in favour of introducing a temporary basic income scheme to combat exploitation and free-riding.13 Limiting the receipt of basic income to, say, a total of five years may deflect free-riding by recipients who would otherwise take advantage of the scheme, or at the very least render its overall effect less socially damaging. In addition to these normative considerations, a time-limited basic income policy may also reflect practical considerations, such as fitting neatly with other policies that make up the institutional background of that particular welfare regime (such as child benefit or universal pension provisions).
Modality
Modality refers to the particular shape that a universal transfer takes. When debating basic income we commonly think about cash transfers, but certain forms of in-kind transfers (for example, food coupons, education or travel vouchers, housing benefits) should not be dismissed out of hand. The defining feature of a universal basic income scheme is not the distinction between cash or in-kind transfers as such, but rather whether social assistance takes the form of public or private goods. Universally distributed private in-kind measures such as education vouchers may be con-sidered part of a universal basic income, as opposed to strict public goods such as road infrastructure. Having said this, there are many good reasons why most universal basic income schemes will rely on cash transfers, but in principle at least part of a basic income or capital grant could be transferred in kind.
Few researchers seem to appreciate fully how many distinctive forms cash transfers can take. Consider, for instance, the difference between schemes that deliver the grant by postal cheque, in the form of a debit card with automatic top-up, or as a refundable tax credit. Each of these forms has benefits and drawbacks that need to be carefully considered at the level of design and implementation. One important administrative factor is the level of integration, the ease with which a given scheme operates within the existing administrat-ive environment. Highly integrated grants make use of existing tax-and-transfer mechanisms only, whereas weakly integrated versions may require additional, often costly, administrative measures. The latter must of course be offset by the corresponding advantages: for example, higher levels of take-up of schemes that oper-ate independently of other, more selective, welfare benefits.
Adequacy
A final dimension of universal basic income schemes relates to the capacity to satisfy recipients’ basic needs. Strictly speaking, a universal basic income need not be fixed at subsistence level: it can conceivably both exceed as well as fall short of what is commonly considered adequate in a given society.14
Some scholars have made the case for a partial as opposed to a full basic income, and most cognate universal schemes can be varied along this dimension. The key distinction here is between partial proposals that allow for basic income to be complemented by other types of cash or in-kind assistance, and proposals where the partial basic income becomes the sole means of social assistance. Not surprisingly, ideological positions differ considerably as to which form is most desirable. Neo-conservatives like James Buchanan, Milton Friedman or Charles Murray have all at times endorsed a welfare state that adopts a single universal scheme for social assistance, provided we simultane-ously cut all other types of state intervention. Socialists and social democrats, on the other hand, oppose such proposals and insist that a partial basic income should always be complemented by other forms of social assistance. With respect to one-off grants, things are slightly different: in this case the goal is not primarily income security but rather improving one’s stock of personal assets. Typical examples of how to use a basic capital include investing in education or using the grant as start-up capital for a small business.15
Focusing for the moment on income stream versions of universal basic income, the adequacy dimension raises a number of problems. There is first the familiar problem of defining and measuring the level of subsistence at any given time or place. There exists an immense literature on various ways of delineating basic needs in contemporary welfare societies. While most of this literature accepts that there is something arbitrary about determining a uniform level of subsistence across society, we should nevertheless appreciate that even arbitrary benchmarks often serve a useful purpose in social policy. This is not merely a matter of debating the proper criteria but also of finding ways to properly assess how differential contingencies affect people’s lives.
A related concern is whether the level of a universal grant needs to be fixed at all. Perhaps we should take a dynamic approach by making the level of the grant periodically revisable, which could be done in two ways. First, the level of the grant can be periodically revised by Parliament or an independent commission, taking into account reflections of economic perform-ance, political will and so on. A second possibility is to link the level of the grant automatically to some macro-economic indicators, as is already the case in some European countries. This indexation approach has the clear advantage of minimising direct political interference, but it remains a question whether this is in all cases a good thing. Two further advantages of the dynamic approach, whichever form it takes, are that the resulting level of grant reflects the overall state of the economy as well as being able to respond much quicker to dynamic changes in behaviour that will inevitably occur with the introduction of a basic income. A potential downside of such a dynamic approach, however, is perhaps that it does not instil the same degree of basic security as a fixed level grant achieves, which in turn may have adverse effects at the personal as well as the social level.
From ‘adversarial’ to ‘fuzzy’ policy design
The history of basic income is replete with deep-seated divisions between advocates and adversaries. Today, both camps still regard basic income broadly as a radical departure from welfare policy rooted in the Bismarckian or Beveridgean traditions. However, as mentioned before, this adversarial approach to policy design is increasingly inadequate because welfare policy defies simple normative and ideological opposition. The point applies to almost any kind of policy, but universal basic income is a case in point.
Basic income may end up serving a number of goals, not all of which are compatible or even desirable. More significantly, what Brian Barry has labelled the principled argument for basic income can only justify its broad contours, leaving detailed features, such as those discussed, undecided.16 This raises a serious political problem for basic income advocates, who remain unsure which political forces to court. Each political faction on the ideological spectrum seems divided as to whether they should fully endorse basic income and, even in cases where they do, which partic-ular variant to sponsor. And even if all of this could be resolved, as we argued, practicalities associated with the implementation of basic income repeatedly interfere with neatly drawn theoretical categories, rendering a principled, adversarial approach fruitless. Taking a less antagonistic approach, we propose that the mature stage of basic income debate would be more suited to the sort of ‘fuzzy’ policy design that features prominently in recent policy discourse.17
This has a number of potential advantages. First, it takes seriously the idea that policies do not simply follow from a prescriptive statement of desirable goals. It is often suggested that policy design has to contend with second-best solutions because of economic, political or administrative feasibility constraints. But this assumes that a preferred policy can be determined independently from these background constraints, which in our view does not make much sense. Because policy must necessarily fit a number of contingencies, it is inherently pragmatic and compromising in nature (even if one agrees that any policy must start from a normative argument about desirable social goals). Fuzzy design clearly welcomes the idea of basic income as a family of concrete proposals, which can be better fitted to the background circumstances at hand.
Following from this, we recommend that basic income design take a bottom-up approach where circumstances dictate both constraints and opportunities, and good policy requires that policy-makers experiment with different ways of solving practical problems and achieving certain goals. The idea of basic income having many faces sits well with the rejection of ‘one size fits all’ approaches in public policy analysis. From a fuzzy perspective it is perfectly congruent to advoc-ate basic income in principle, while recognising that the detailed outlook of any actual proposal will be largely determined by the specific constellation of goals, constraints and opportunities. One interesting implication in the context of an emerging social Europe is that different states or welfare regimes do not need to endorse precisely the same policy in order to deliver universal welfare. Recent years have witnessed a massive literature on the difficulties of obtaining positive coordination on social measures across EU member states.18 The variety within the basic income ideal should be appreciated as one of its key advantages, allowing it to match a wide range of political limitations.
Relinquishing antagonism to ‘fuzzy congruence’, then, has important implications for the comparative analysis of universal welfare policies. Acknowledging the many faces of basic income allows for a specific comparative approach to basic income research which need not focus on programme specifics, but instead allows for evaluating policy outcomes. To policy-makers it matters less whether a basic income is fully unconditional or incorporates a weak participation requirement, as long as both score roughly equally well on desirable goals such as combating poverty, increasing equal access to employment, supporting a variety of life-styles, etc. And even where programmes score unevenly, a comparative approach might provide good indications why this is either not desirable-maybe different countries rate competing social goals differently-or perhaps not feasible. After all, different economic, political or administrative background conditions entail different possibilities for policy implementation.
Summary
The main lesson of this article can be summarized as follows: there is no such thing as a preferred basic income scheme independent of the overall institutional and policy context. The debate concerning the best possible basic income design can only generate productive results when carried out within the rich institutional environment of case studies. It is part of the policy-making balancing act that what works here may not work over there, and what seems a good idea now might become counterproductive or obsolete at a later time. This insight should not lead to despair; instead, the crafty policy designer should wholeheartedly embrace it. ‘Fuzzy’ policy design paints a world in which policy reform explicitly acknowledges the many faces of basic income, and uses this feature as its main strength to further the cause of social justice across Europe and beyond.
Acknowledgements
This article is part of a wider project on the administration of universal grants. We are grateful to John Cameron, Norma Henderson, Malcolm Torry and particularly José Noguera for comments on a previous draft.
The Citizen’s Income Trust is grateful to the Political Quarterly for permission to reprint this article. It originally appeared in The Political Quarterly, vol.75, no.3, July-September 2004, pp.266-274. The Political Quarterly is published by Blackwell, and further details can be found on www.blackwellpublishing.com/journals/poqu.
Notes
1 Philippe Van Parijs, ‘Competing justifications of basic income’, in Arguing for Basic Income, London/New York, Verso, 1992. Brian Barry has suggested the capacity to transcend ideological cleavages is a particular strength of basic income, but unfortunately so far it has not generated sufficient critical mass to influence agenda-setting or coalition-building strategies in advanced industrial countries. See Barry, ‘Surfer’s saviours?’, Citizen’s Income Bulletin, no. 22, 1996, pp. 2-4.
2 Keith Dowding, Jurgen De Wispelaere and Stuart White (eds), The Ethics of Stakeholding, Basingstoke, Macmillan, 2003.
3 Philippe Van Parijs, ‘Reply’, in Philippe Van Parijs, Joshua Cohen and Joel Rogers, What’s Wrong with a Free Lunch?, Boston, Beacon Press, 2001, pp. 124-7.
4 Anthony Atkinson, Poverty in Europe, Oxford, Blackwell, 1998, pp. 34-5.
5 Maria Jepsen and Danielle Meulders, ‘The individualization of rights in social protection systems’, in Hedva Sarfati and Giuliano Bonoli (eds), Labour Market and Social Protection Reforms in International Perspective, Aldershot, Ashgate, 2002, pp. 108-10.
6 Compare, for instance, Tony Fitzpatrick, ‘Feminism and basic income’, in Freedom and Security, Basingstoke, Macmillan, 1999; Ingrid Robyens, ‘Hush money or emancipation fee? A gender analysis of basic income’, in Robert van der Veen and Loek Groot (eds), Basic Income on the Agenda, Amsterdam, University of Amsterdam Press, 2000; Alisa McKay and Jo Van Every, ‘Gender, family, and income maintenance. A feminist case for citizens’ basic income’, Social Politics, vol. 7, 2000, pp. 266-84; Anne Allstott, ‘Good for women’, in What’s Wrong with a Free Lunch?
7 Anthony Atkinson, ‘The Case for a Participation Income’, The Political Quarterly, vol. 67, no. 1, 1996, pp. 67-70.
8 See Mary Jo Bane and David Ellwood, Welfare Realities: From Rhetoric to Reform, Cambridge, Mass., Harvard University Press, 1994.
9 Atkinson, Poverty in Europe, pp. 131-3, discusses errors in administration due to imperfect information. For a sustained discussion of formal and informal discretion in the welfare state see Robert Goodin, Reasons for Welfare: The Political Theory of the Welfare State, Princeton, Princeton University Press, 1988; for a socio-legal view of discretion in welfare administration see Jerry Mashaw, Bureaucratic Discretion, New Haven, Yale University Press, 1983, and, more generally, Keith Hawkins (ed), The Uses of Discretion, Oxford, Clarendon Press, 1992.
10 The implications of regional differences in purchasing power are in fact barely discussed with respect to universal basic income; but see Philippe Van Parijs, Real Freedom for All, Oxford, Clarendon Press, 1995, pp. 37-8 and 244, note 8.
11 For a comparative analysis of income and capital approaches see Keith Dowding, Jurgen De Wispelaere and Stuart White, The Ethics of Stakeholding, as well as a recent issue of Politics and Society, vol. 32, no.1, 2004. Although in some proposals the actual delivery of a one-off payment will still occur in a few large instalments, this remains substantially different from a life-long regular income stream.
12 See in particular Robert van der Veen, ‘Assessing the unconditional stake’, in The Ethics of Stakeholding.
13 Stuart White, ‘Fair reciprocity and basic income’, in Andrew Williams and Andrew Reeves, Real Libertarianism Assessed, Basingstoke, Macmillan, 2003; ‘Freedom, reciprocity, and time-limited citizens’ income’, in The Ethics of Stakeholding. White’s proposal is very similar to the sabbatical grant idea proposed by Claus Offe.
14 See Van Parijs, Real Freedom for All, pp. 30-57, for a discussion of ‘the highest sustainable basic income’. Theodore Marmor rightly distinguishes between ‘aggregate’ and ‘individual’ standards of adequacy. See Marmor, ‘On comparing income maintenance alternatives’, American Political Science Review, vol. 65, 1971, pp. 86-7.
15 Bruce Ackerman and Anne Alstott, The Stakeholder Society, New Haven, Yale University Press, 1999; Julian LeGrand and David Nissan, ‘A capital idea: helping the young to help themselves’, in The Ethics of Stakeholding.
16 Brian Barry, ‘Real freedom and basic income’, Journal of Political Philosophy, vol. 5, 1997, pp. 274-86.
17 B. Guy Peters, ‘Policy reform: is uniformity the answer?’, The Political Quarterly, vol. 74, no. 4, 2003, pp. 421-8.
18 Fritz Scharpf, Governing in Europe, Oxford, Oxford University Press, 1999.