There are many reasons for preferring a “universal basic income” to the present pattern of social supports. Some are moral, others pragmatic. The pragmatics may be philosophically less showy, but they might also be politically more powerful. Here I shall supplement Philippe Van Parijs’s useful round-up by offering another pragmatic reason for shifting to a universal basic income, and then suggest how that proposal might be politically saleable, even in seemingly hostile environments.

The Destandardization of Social Life

Public policymakers are long accustomed to responding to social need “categorically.” Public assistance is offered under headings that supposedly correspond to the standard sorts of “social risks” that can characteristically lead to hardship: old age, sickness or injury, death or desertion of the household’s breadwinner, unemployment, and so on. Those same categories also incidentally separate the “deserving poor” from the “undeserving,” whose hardships are ostensibly due to their own recklessness or fecklessness–but I’ll leave that to one side for now. My present concern is with the limits of categoricalism, not as a form of social moralism, but as an instrument of social policy.1

Categorical approaches to the relief of social distress work fine only so long as people’s lives and needs follow broadly predictable patterns. In late-Victorian England, Rowntree and Booth found that they did. The traditional sort of poverty that they uncovered basically tracked the life cycle: people were poor when they were young, when their own families were young, and again when they became too old to work.2With industrialization came other sorts of standard social misfortunes associated with the production process (occupational sickness and injury) and the economic cycle (unemployment).

Where social risks were relatively standard across the population, so too could be the social response. Governments instituted categorical benefits of one sort or another (child benefit, family allowances, old-age pensions) to address standard risks of poverty at various points in the life cycle. Governments mandated contributory social insurance schemes of one sort or another (unemployment, disability, old age, death, or desertion) to deal with cases in which industrial workers and their families were without market income. Social safety nets were still required to catch anyone whose needs were in some way peculiar, and not covered by any of those standard schemes. But those were supposed to be, and largely were, residual: small-scale mopping-up exercises, with the vast majority of cases being covered by one of those other standard programs of social relief.

In that world of “standard problems and standardized responses,” public programs were propped up by other equally standard, non-public “pillars of social security.” Assuming standard family relations (one partner for life) and standard employment relations (full-time, full-year employment with the same employer for life), people could ordinarily draw on family resources and occupational benefits to tide them over. That made them more “self-reliant,” which in the context of this debate meant merely less reliant upon state support.3

That is “the world we have lost.” Maybe it was never completely true anywhere. But certainly it is not true now for most people in most places across the OECD. The standard employment relation is no more. Most of us, we are told, ought to expect to have two or three careers across the course of our lives–and probably just as many partners (and fractured families). And that means that we can no longer rely on either of those non-state pillars of social security. It also makes it much harder for public programs to process more than a few claimants through any small set of standard social categories. If people’s needs are increasingly non-standardized, so too must be the social response.

One way of doing that is to give more discretion to social workers and rely on them to be attuned to each person’s unique circumstances. But that of course would represent a return to the old poor laws. It would remake welfare-right claimants into abject supplicants for the substantially discretionary favors of public and private charity.

To my mind, a much more attractive response to the destandardization of social life would be to abandon conditionality and categoricalism altogether. Let’s give up trying to second-guess how people are going to lead their lives and crafting categorical responses to the problems they might encounter. Instead, simply give them the money and let them get on with it.

Participation-conditioned Income

That is an admittedly pragmatic argument for universal basic income. But the main arguments against universal basic income are essentially pragmatic as well. Philippe Van Parijs has effectively addressed the main economic ripostes encountered by such proposals. But an even more telling argument against universal basic income is political: whatever the abstract merits of the case, you will simply never be able to sell people politically on the idea that people should get “something for nothing,” when everyone else in the community is working for a living.4 Maybe, as Van Parijs says, surfers should be fed, but there is little political support for feeding them so well as to keep up the physically demanding sport of surfing at public expense.

This point is pragmatic in the first instance, principled in the second. There is no political constituency precisely because it offends people’s principles (their sense of fair play) that someone should get something for nothing when others have to work hard for it. To some extent those principles are internalized by welfare recipients themselves, whose self-esteem is heightened if they are able to “give back something to society” rather than just being seen, by themselves as much as anyone else, as leeches on the commonweal. Whether or not those sentiments are well grounded, they are indisputably real and politically powerful. And that makes it politically impossible to enact a basic income in any universal and unconditional form.

Thus the most politically saleable form of basic income, for now and the foreseeable future, may be a “participation income.”5 Under that scheme, everyone would draw a basic income on condition that they perform some socially useful labor. They can satisfy that condition by working in the paid labor market. But equally well, they can satisfy it by caring for young, old, or disabled members of the community, by participating in community service or environmental projects, or through some other activity.

Obviously, participation income is conditional in a way that universal basic income is not. Equally obviously, everything depends on what counts as “socially useful labor” and who gets to decide whether any given person has done enough. At its worst, such a scheme could be as oppressive as the most tyrannical labor exchange. At its best, it could well allow people credit (and cash) for doing all sorts of good things that they want to do and perhaps would be doing anyway. In any case, participation income is pragmatically as close as many of us suppose we will ever be able to get, politically, to universal basic income.

In the present political climate, how can even participation income be enacted?

Start by recalling that some of the most progressive developments have come under the most unlikely regimes. Bismarck was a deeply conservative politician who nonetheless gave us the welfare state in its modern form. Nixon was a deeply ambiguous politician who came close to giving us a negative income tax.

Consider, more specifically, the history of the enormously generous Dutch “social minimum.” Roughly speaking, that program pays any family whose income falls below half the median national wage money enough to get them above that threshold, which is roughly what cross-national researchers would deem the “poverty line.”6 One might imagine that that program was the brainchild of left-wing politicians. But, as it happens, the social minimum was the bequest of a series of traditionally conservative “confessional” coalitions in the 1960s. Not only were its instigators conservatives, the arguments they gave for it were couched in traditionalist terms. They wanted, above all else, to give social recognition to the important unpaid contributions made by stay-at-home wives and mothers in the traditional social order.

Might the same sorts of conservative impulses that led to this Dutch “backing into progressive social policies” be marshaled in support of basic income in its participation-tested form? Recall, first, that certain sorts of social benefits have been participation-tested for years. From their earliest days, unemployment benefit programs have always contained requirements that recipients be “actively looking for work.” Advocates of many of the more reactionary forms of welfare reform have recently urged, with considerable success, that that principle be extended. “Workfare” (or “active labor market policy” or “mutual obligation”) requirements have thus been imposed on a wide range of social assistance.

Here’s the twist. Suppose workfare requirements are taken to imply an “activity test” that can be satisfied in any of many ways. Central among them are, of course, activities directly related to paid labor markets: working in paid labor, looking for such work, training for such work. But also invariably included–either as exemptions from the “activity test” or as ways of satisfying it–are provisions for those who are themselves unable to work (by reason of age or disability) and provisions for those who are unable to work because of their responsibilities caring for young children. Those provisions, in turn, are increasingly expanded to include caring for not only pre-school-aged children but also disabled relatives. Those, in turn, are expanded to include provision for “voluntary” (which is to say, unpaid) participation in approved community-service or environmental-protection projects.7

At the end of the day, we may well find punitive and draconian workfare schemes being thereby transformed, in effect, into state salaries for socially useful labor of many (if not quite all) forms. Maybe unpaid labor in households without very young or disabled members might not be included. But all manner of other community work, which had previously been unpaid, might suddenly attract, in effect, a salary from the state.

All we then have to do is persuade people to apply for it. That is to say, we just have to persuade them that “the dole” is not just for people who find themselves out of paid work, but also for people doing socially useful unpaid work. Once that gestalt shift is effected, workfare will have become a first approximation to a participation income. Once again we truly will have backed into progressive social policy.



1 These points are elaborated in my “Crumbling Pillars: Social Security Futures,” Political Quarterly 71 (April 2000): 144-50, which builds in turn on my “Toward a Minimally Presumptuous Social Welfare Policy,” in Arguing for Basic Income, ed. Philippe Van Parijs (London: Verso, 1992), pp. 195-214.

2 B. Seebohm Rowntree, Poverty: A Study of Town Life, 2nd ed. (London: Macmillan, 1901); Charles Booth, Pauperism and the Endowment of Old Age (London: Macmillan, 1892).

3 I argue this point in “Social Welfare as a Collective Social Responsibility,” in Social Welfare and Individual Responsibility: For and Against, David Schmidtz and Robert E. Goodin, eds. (New York: Cambridge University Press, 1998), pp. 97-194.

4 What finally did in Aid to Families with Dependent Children in the United States and cognate programs around the world was arguably increased female labor force participation. So long as the ordinary expectation was that women of school-aged children would stay at home with them if they were economically able, a case could be made for public assistance to allow other women to do so as well. But once middle-class mothers were typically “working for a living” in the paid labor market, “welfare mothers” were increasingly expected to do the same.

5 Such a proposal has been endorsed by the UK Labour Party’s Commission on Social Justice. See Social Justice: Strategies for National Renewal (London: Vintage/Random House, 1994). Elaboration is provided by Tony Atkinson, a member of that commission, in his “The Case for a Participation Income,” Political Quarterly 67 (Jan/Mar, 1996): 67-70.

6 “Roughly,” because exact payments depend on household structure, and so does the “poverty line”; but it is the flavor of the program rather than the details that matter here.

7 In Australia, for example, the unemployment benefit (“Newstart Allowance” or “Youth Allowance”) is subject to a “mutual obligation” activity test which can be met through “voluntary work 24 hours each fortnight for at least 8 of 13 fortnights” for 18-20 year olds, or 30 hours for people over 21. The agency responsible for administering that scheme says, on its web page that “the choices of voluntary work are almost limitless: helping people with disabilities; helping people who are frail; home care; shopping; administration; community welfare; translating; providing companionship and support; working with community-based organisations or clubs.” For further information, see