In relatively short order, universal basic income (UBI) has transformed from what was little more than a glorified thought experiment into a concrete policy option, and discussion in the media has mushroomed accordingly. Debate has likewise intensified on the Left, taking on a sometimes productive, sometimes acrimonious, tenor. The reasons for the latter are obvious, but when productive, the discussion has proceeded as a debate among those who share a set of moral commitments but disagree on questions of strategy or analysis. In the case of UBI, an abstract policy measure with no history of genuine implementation, it is natural to see a good number of different intuitions, crosscutting hypotheses, and wide-ranging concerns about unintended consequences. Indeed, the debate on the Left may ultimately rest on empirical outcomes. Will UBI in fact improve people’s lives? Will it facilitate wider and deeper transformations? Or is it just a neoliberal mirage?
This essay surveys the debate surrounding UBI that has emerged in recent years, focusing on the main objections from the Left. This entails analyzing the gamut of possible empirical effects, from the impact on wages and labor force participation to gender and collective action. The debate on these empirical questions, it must be said, is decidedly unresolved. As with any major social transformation, the impact of offering substantial cash transfers to all might generate outcomes that are impossible to foresee. To claim otherwise — that we have a clear-eyed understanding of the full set of consequences — would be foolhardy. With this caveat registered, it should be said that we luckily do know something about the impact of UBI, and drawing on the available evidence we are able to say something meaningful about its consequences in multiple spheres of social life.
The concept of a universal basic income refers to a monthly cash income paid to each member of society without regard to income from other sources and with no strings attached.1 There is no precise level of payment built into the definition. Proposals on the order of $14,000 per person — a number exceeding the official poverty line for single individuals in the US ($12,000) and totaling about one-quarter of US GDP — are often seen as somewhere between modest and substantial. Proposals that are more generous tend to hover around $18,000 or $20,000 per person. I have the number $14,000 in mind as more or less the minimum payment level required to achieve the normative objectives discussed herein; in particular, this sum is meant as the lower-bound threshold that affords people an above-poverty fallback position, providing everyone with a measure of freedom from work, and therefore, of power at work.
Understanding basic income requires a consideration of its likely empirical consequences, as well as clarifying the underlying normative agenda. In some cases, there are pragmatic empirical tests that any normative vision must pass for it to be realized; in others, the normative arguments can hold their own whatever the empirical consequences. All things considered, including some ambiguities discussed below, there is a powerful socialist case for basic income. This essay shows that the scheme, were it sufficiently generous and universal, would help realize the moral vision socialists ought to hold. It is worth getting back to basics, so to speak, to make some sense of this debate.
The Ends of Social Policy
One of the constitutive aspects of left politics is that the policies advocated for are not mere ends in themselves, but rather instruments for realizing a broad set of normative commitments that envision how the world ought to be. Sometimes the Left, mistakenly in my view, evades these lofty commitments because they are far removed from the grind of political struggle or because moral argument is seen as the domain of liberal and conservative politics. But this position has never been persuasive. To evaluate policies and politics we have to commit to a moral vision, even if it is somewhat hazily characterized as a future defined by human flourishing and real, substantive freedom.
When it comes to the impact of actual policies, it is useful to distinguish ameliorative from emancipatory reforms. Ameliorative reforms, like traditional welfare policies, are valuable because they provide direct material benefits and improve people’s lives, which is a normative end in itself. If a political vision loses sight of life-improving reforms, it will be abandoned by poor and working people; they would rightly see that vision as callous to their needs. Still, it is difficult for left political operatives to get overly excited about purely ameliorative reforms. While they make people’s lives less painful, such policies do not, by definition, help to mobilize people or expand their power. The concept of an emancipatory reform, on the other hand, refers to some social policy that may ameliorate a particular deprivation but does so in a way that pushes us closer to an underlying moral vision. These are policies that tip the balance of power and strengthen the position of poor and working people when facing off against bosses, spouses, and other powerful individuals in their lives.
The main reason UBI ought to be a part of a left normative vision is because it facilitates exit from relations of exploitation and domination — the power of exit has ameliorative as well as emancipatory significance, as I will show. The foundational Marxist objection to the structure of capitalist labor markets is that they are superficially free but substantively unfree. Dispossessed of the means of production, and therefore of subsistence, workers can happily choose between capitalists, but are ultimately forced to choose one. This is what Marx termed “double freedom”: our freedom to be exploited by the employer of our choosing is coupled with the freedom to remain hungry should we choose none. For those who object to the compulsory nature of the capitalist labor market, basic income is appealing because it ensures that people not only have the abstract right to freedom, but the material resources to make freedom a lived reality. It gives people the power to say no — to abusive employers, unpleasant work, or patriarchal domination in the home.
People often use that power. In the case of the Canadian Mincome experiment from the late 1970s, some participants did in fact take up their newfound ability to quit. In the town of Dauphin, Manitoba, a three-year guaranteed annual income led to an 11 percentage point drop in labor force participation.2 Across the five major guaranteed annual income experiments previously conducted in the US and Canada, there was a wide range of average labor supply reductions for men and women, from a low of nearly zero in some cases to a high of about 30 percent.3 The guaranteed annual income is not identical to the UBI; the former phases out above a certain income threshold, reducing its universality and, to an extent, its desirability. However, even this version touches a wide swath of the population: a high guarantee level and a low phaseout rate will run deep into the middle class. It moreover makes the option of work withdrawal universally available and allows for a good amount of inference about a fully universalistic model. As discussed below, I also found evidence suggesting that in the Mincome case, the guaranteed income reduced domestic violence. In providing people with a decent fallback position, such a policy affects underlying power relations and changes the background conditions under which negotiation takes place, both at work and at home.
But there is a stronger point about emancipatory reforms to be made here: as a social policy, basic income can pave the way toward broader social transformations. In particular, UBI can help set in motion a dynamic process that empowers people to struggle to build a better society. It achieves this in two ways: the power of exit, noted above, and the institutionalization of solidarity. The former allows poor and working people a better footing to bargain from, instigating broader and more far-reaching gains; the latter, by redrawing the social boundaries carved by categorical welfare states and reducing the appeal of “defection” from collective action, improves the odds that they do so collectively rather than individually. At bottom, the vision of basic income is attractive because of its dual function as an ameliorative and emancipatory policy measure.
In this hopeful depiction, basic income thus articulates both an economic alternative and a theory of social change. There is a concern, however, that social change happens not when people are given exit options, but when circumstances lock them into unavoidable interaction, when the lack of alternatives leaves collaboration and struggle as the only viable option. To be sure, it is sometimes argued that the Left ought not to allow people an exit option; that is, if we aspire to build power and mobilize people, we ought to encourage “voice” over “exit.”4 As an empirical matter, this argument cannot be dismissed.
Indeed, there is always a chance that giving people the freedom and capacity to do what they want might mean that they do things we would prefer them not to do. Perhaps basic income would be emancipatory for individuals, but inadvertently fragment us as collectivities. After all, some might choose to withdraw from the social world entirely.
To the contrary, however, there is good reason to believe that it is the possibility of exit that facilitates voice. If a stable flow of cash gives you the power to threaten to leave a marriage or a job — that is, if your threat of exit has real credibility — you are in a better position to speak your mind. In what follows I attempt to make this case, though I advance basic income as a desirable reform even if it fails this empirical test. Differently put, basic income can provide resources to facilitate collective action, as will be explored below, but it does so without precluding more solitary escape routes. This position ought to be seen as perfectly consistent with a socialist ethics: we wish to nurture collective action by fostering its conditions of possibility in a positive sense — not through the active obstruction of alternative pathways, and not by leaving collective action as the only path to individual survival.
Basic income thus both enhances people’s negative freedom from coercion and their positive freedom to do what they want. There are few on the Left who would disagree with these principles. Do we wish, for example, to block a Walmart worker from quitting her job if she so desires? If we are in favor of basic human autonomy, the answer is no. The answer ought to be no, even if my argument about the positive relationship between basic income and collective action fails to persuade — even if collective action is nourished only when people are locked inside conflictual relationships. Rapunzel might survive best in her tower, but that would scarcely convince her of its value. There is a real sense in which left opposition to basic income’s underlying principle entails advocating some degree of coercion. This might be philosophically defensible, but it does not square with a commitment to decoupling Marx’s double freedom, nor with deep socialist commitments to expanding the domain of human autonomy. We return to these core philosophical issues after taking stock of an array of normative and empirical questions, and addressing the major left criticisms of basic income.
Neoliberal in Practice?
Left objections to unconditionally giving people money have proliferated of late, sparked no doubt by UBI’s strange bedfellows on the Right. Some of these objections are highly pertinent and have pushed the debate in positive directions; others are less persuasive. With basic income on the policy agenda in a number of countries around the world, it is necessary to appreciate the broader context of the discussion.
The first and most important objection has been stressed recently by John Clarke of the Ontario Coalition Against Poverty, among others: given the constellation of forces and the political commitments of many proponents, odds are that basic income, if implemented, will come in a neoliberal guise, dishing out meager payments and accompanied by severe austerity measures.5 Indeed, like every social policy, basic income could be implemented in neoliberal fashion, and over the past four decades there has been no shortage of such regressive proposals in Canada and the US.
This is a legitimate worry, and it is in the implementation of the policy where the strange bedfellows problem will be resolved, one way or another. The rogue’s gallery of right-wing supporters, from Milton Friedman to Charles Murray, is often unambiguous in its desire to use basic income as a knife to eviscerate the expensive insides of the welfare state. To different degrees, recent support within elite tech-chauvinist circles, from Peter Thiel to Mark Zuckerberg, might be similarly understood. How on earth could Marxists form a political alliance with the boy-king of Silicon Valley? Perhaps some elites see basic income as a pragmatic means to avoid the radicalization of a population that has seen little improvement in living standards in recent years, but others envision a Trojan horse designed to raid the citadels of Social Security, Medicare, and education spending.
If basic income is little more than a dressed-up neoliberal policy, there is no doubt: it ought to be resisted. But why not work towards a better version of basic income? There are vastly different visions for what a basic income would look like, and a small basic income deployed in a libertarian fashion to replace the welfare state is not just different from a generous version built into the existing welfare state, but it is actively rooted in the opposite philosophical vision. Where the former is designed to reduce the tax burden on the rich and avoid supposedly paternalistic social policies, the latter is designed to negate the coercive nature of the capitalist labor market and empower popular forces. Quantitative changes in generosity induce qualitative changes in result. There are qualitatively different varieties of basic income, and it is entirely possible that in the contemporary political context an undesirable vision becomes reality. But no political vision, it must be said, can escape the uncertainty built into the passage from theory to practice.
An instructive comparison here is the call for guaranteed work. If a jobs guarantee were implemented in the contemporary context, it is easy to imagine a version that is far from liberatory, where the jobs would be backbreaking and the breaks would be few. Political scientist Adam Przeworski argued against this unsavory but plausible vision of a work guarantee: “Making people toil unnecessarily, just so they can be paid something without others complaining and so they will not hang around with nothing to do, is to substitute one deprivation for another.”6 This is not to claim that a progressive vision of a jobs guarantee is unimaginable; to the contrary, a workable scheme of that sort has a great deal of potential, and if implemented successfully would be a vast improvement on the current configuration of social policies. But the forces that might sabotage a basic income would operate similarly in the case of a jobs guarantee.7 There is, moreover, a well-known historical example of an ugly implementation of the jobs guarantee; it was called the workhouse. For centuries the old English workhouse tied public assistance benefits to toil and operated on the principle of “less eligibility,” a doctrine ensuring that workhouse conditions be made worse than those outside in order to deter its use. It might also be noted that highly suspicious proponents were attracted to this system of poor relief. For example, Jeremy Bentham advocated for the workhouse because it was a “mill to grind rogues honest and idle men industrious.”8
The problem, therefore, is a general one. As a rule, the Left opposes meager unemployment insurance and embraces generous unemployment insurance. Bad health care policy is bad, and good health care policy is good. Left strategy has always involved fighting to improve those policies, and any model of the world suggesting that decent unemployment insurance or good health care is won through struggle would apply equally to UBI. Criticism of the abstract idea thus ought to be distinguished from criticism of its concrete implementation — this talking point ought to be old hat for socialists, at least those old enough to remember unsavory implementations of their dearest ideas. As with all social policy measures, a basic income might be implemented in an appalling way. Should we therefore reject the idea out of hand? As an argument against the impulse to loosen the compulsory nature of capitalist labor markets, this line of reasoning is hardly sustainable.
A related critique is that basic income is a bloodless, technocratic social policy — many UBI advocates seem to imagine that once the appropriate legislation is passed, the job is done. They imagine a policy that gets wonkishly imposed, outside the context of social struggles, as if policy and power exist in separate worlds. But the critique here is primarily of those advocates, not the idea itself. Indeed, if basic income is abandoned to the technocrats we will be sure to get a tepid or even regressive set of social policies; a desirable, radical version will find many opponents, in particular employers, and will require massive popular mobilization. But it is strange to believe this problem is unique to basic income.
Neoliberal Even in Theory?
Apart from the anxieties about right-wing politicians implementing their preferred version of basic income, there are a number of criticisms of even a generous and truly universal basic income. This essay evaluates a range of empirical arguments concerning gender, capitalism, and collective action, but in this section I zero in on two oft-made normative arguments: (1) that we ought to expand the public provision of key services before we consider income maintenance; and (2) that we ought not have a basic income because we have an obligation to work, contribute to the community, and not live off of the productive labor of others.
To begin, some argue that money earmarked for a UBI should instead be spent decommodifying important services like housing, childcare, transportation, and more. This objection to basic income, first made by the economist Barbara Bergmann, is a powerful one, but in the end is not persuasive.9
The issue is sometimes framed in the following way: If you had one additional dollar to spend, where should it be spent first? The services-over-income argument is perhaps most powerful in the form of a utilitarian hypothesis. Using one marginal dollar of additional tax revenue to expand existing public transportation or health systems, or to provide new kinds of public services, might improve people’s lives more than offering people the equivalent dollar in cash.10 Perhaps it would more effectively extend the average lifespan or improve people’s subjective well-being. This is an unanswered empirical question, but if true it would be hard to ignore. Framing the question in narrowly economistic terms, however, posits a false choice between decommodifying labor power and decommodifying services — as if both cannot be pursued at once. In a rich, productive society we ought to be able to afford both a basic income and high-quality public goods. Were popular forces powerful enough to make progress on one, they very well might be powerful enough to make progress on the other.
But granting the framing, the calculus still makes two mistakes. First, it ignores the goal of real freedom as a non-instrumental moral objective. On the grounds of freedom — in particular, the positive freedom to decide on the activities we want to pursue and how to spend our days — it is worth defending a strategy that directly and forcefully erodes workers’ background condition of market dependence. That is, it is a good thing to be able to quit your Walmart job whatever the long-term consequences. Second, the Bergmann argument ignores the process whereby reducing labor market coercion and providing a genuine fallback better positions people to achieve broader goals.
There is, of course, a degree of symmetry between the exit option provided by basic income on the one hand and a comprehensive set of publicly provided goods and services on the other. However, I believe the expansion of freedom and power is weaker in the latter case. As Offe and Wiesenthal emphasize in a well-known essay, the needs and preferences of poor and working-class people are deeply heterogeneous — the needs of a young man living in a small rural town, a single mother in a large city center, and an older suburban couple are inescapably diverse.11 On these grounds, money, a highly fungible good, can better satisfy diverse needs and subjective preferences than even a fairly comprehensive suite of specific goods and services.12 This means that basic income would more effectively reduce the costs of being fired and better create an alternative to the labor market for a wide swath of society; by more effectively constructing a fallback position it would better expand workers’ leverage at work.
If we must choose between expanding the public provision of services and providing a basic income, and we go with the former, we should be clear on the meaning of this choice. It implies that we prefer a system where people remain somewhat more dependent on the labor market for survival, that we prefer to retain, in all likelihood, Marx’s double freedom. By contrast, a basic income insists that it is important to decommodify not only a range of goods and services, but labor power itself.13 It says that taking coercion out of the labor market and abolishing what the labor movement once called “wage slavery” may ultimately be more liberatory than taking a broad spectrum of commodities off the market. It says that we ought to have the positive freedom to spend our time as we wish. Rather than improving our ability to get to work, UBI provides the means to avoid it if we need to.
There is another side to the services-over-income objection. John Clarke argues that even in the best case, giving people money will foster a consumerist society. Labor power might be decommodified, but if everything else must be bought, we will end up spending all our time as “customers in an unjust society.”14
It is worth making two points in response. First, a world with an open market in most goods but no compulsory capitalist labor market in fact might be a decent transitional vision of market socialism. The injustices of capitalism have much more to do with the coercive nature of the labor market than the existence of markets for consumption items. Indeed, the anti-consumerist argument misidentifies the sources of injustice in capitalism. The goods market is not so much a bad in itself; the problem is rather that people have insufficient purchasing power to make effective demand correspond to actual want and need.15 A more egalitarian distribution of purchasing power would help bring the neoclassical fantasy that market demand equals need into alignment with reality.16
Second, it seems perfectly reasonable to expect a basic income to make for a much less consumerist life. As noted above, the Dauphin experiment generated a nontrivial drop in labor force participation. For some people, basic income might also mean exiting the paid labor force, collecting a lower income, and thus having less, not more, to spend. It is often hoped and hypothesized that socially valuable activities would be encouraged if people’s basic needs were secured outside the labor market. Moreover, the virtue of basic income is its potential to expand people’s leisure activities. We can turn to data from the urban portion of the Mincome experiment — a randomized controlled trial based in Winnipeg conducted in conjunction with the Dauphin portion of the experiment — to analyze this very question. Mincome inquired into the day-to-day activities of basic income recipients who left the labor force; relative to controls, the intervention led to growth in a range of socially valuably activities, including care work and education (see Table 1). The intervention also led to growth in the portion of men and women reporting that they were not working simply because they “did not want to work.” In a free society, this decision ought to be available to the poor as well as the rich.
Treatment effects for survey question, “What is the main reason you were not working?”*
NOTE: “Treatment effects” refer to the isolated effect of the experiment, or the “difference-in-difference”. The difference-in-difference subtract the baseline/study period change in the control group from the baseline/study period change in the treatment group. For example, the treatment effect of the experiment for answering “education” is 2.6 percentage points. In this case, the percent of control subjects reporting that they were not working due to education increased from 4.6% at the baseline to 5.7% during the experiment, and the percent of Mincome treatment subjects reporting the same answers increased from 4% at the baseline to 7.7% during the study period, leaving the full treatment effect at 2.6 percentage points. See also D. Calnitsky, Latner, J., & Forget, E. 2017. Working Paper. Life after work: The impact of basic income on non-employment activities. Available upon request.
|Job/work conditions [“Labor dispute”; “No jobs available”; “Available wages too low”]||5.9|
|Family [“Wanted to take care of family”; “Child care too expensive”; “Pregnancy”]||3.9|
|“Did not want to work”||4.0|
|Education [“In school”; “In job training”]||2.6|
|“Ill or disabled”||-3.8|
|* Data from randomized controlled trial, Winnipeg sample, Mincome experiment; right column is percentage point change|
While it’s possible that some people would spend more of their newly freed-up time shopping — not to mention doing experimental theater and rollerblading, the most underrated of all basic income critiques — they also might spend free time with others, pursue social and political projects, undertake care work, or engage in a wide array of other, non-consumption-related activities.
Before moving on, it is worth making note of a second normative critique of basic income, one that stretches from Rosa Luxemburg to Jon Elster, and which is anchored in a good deal of liberal and left political theory: that we do not have a right to live off someone else’s earnings.17 Rather, the argument goes, we have a moral obligation to contribute to the community, and therefore, to work. In part, this is what led Tony Atkinson to propose a “participation income” in place of basic income: the participation income would provide a stream of income conditional on participation in some socially valuable activity, be it inside or outside of the formal labor market.18
Here I see two issues worth contemplating. First, from the viewpoint of socialist freedom, there is every reason to believe that rather than equalizing work levels and incomes we ought to provide people a choice between higher income and more leisure. This is consistent with what G. A. Cohen has named “socialist equality of opportunity.”19 In such a world, inequalities in income and leisure reflect nothing more than differences in personal tastes for income and work — that is, differences consistent with socialist justice. For Cohen, a society where each person has roughly equal work/wage bundles is inferior to one allowing for a choice between varying bundles of income and leisure. Basic income goes some way to allow for individuals who might choose a basic-income/maximal-leisure bundle or a high-income/minimal-leisure bundle. I return to this question of socialism and freedom in the conclusion.
Second, the normative argument that people do not have a right to live off another’s earnings — and by implication that only those who work shall eat, that only those engaged in productive labor ought to be compensated — is unacceptably libertarian in its underlying theory of remuneration. The theory ignores the non-attributability of outputs to production inputs: production is a deeply interdependent activity and, particularly in a world of non-constant returns to scale, the abstract process of linking one person’s productive effort to their ultimate compensation is always an ambiguous exercise. This means that the very concept of an individual’s appropriate earnings is ill-defined. But even more importantly, the principle suggesting that we ought not live off the labor effort of others gives far too much weight to current productive labor — that is, the labor of living workers rather than the whole history of work — as the driving force of current output. As Herbert Simon has argued, high levels of individual productivity in rich societies are, for the most part, consequences of the brute luck of being born into a rich society.20 High incomes and high productivity are attributable less to current labor effort and more to past labor effort, and all members of society ought to benefit from the work of prior generations and the overall wealth and development of society. For the current generation, this means that through no contribution of our own we have been endowed with highly developed technologies, infrastructure, language, and culture, and this gives current income, in large part, a morally arbitrary character. This is, therefore, a powerful reason to redistribute a good amount of it to people whether they work or not.
Basic Income and Capitalism
A separate set of critiques from the Left concerns the unintended consequences of basic income on the labor market, employer behavior, and capitalism more broadly. These arguments are often framed in terms of the apparent limits of capitalism and the subterranean economic forces that compromise progressive social transformation. As a general rule, arguments taking the form “a decent basic income is impossible under capitalism” should be treated with the same suspicion we have for claims about capitalism’s fundamental incompatibility with a decent welfare state. History has shown capitalism to be a highly flexible system; what was once said to be impossible under capitalism is later said to be an essential feature of its legitimation. In such arguments it is pro forma to allude to some deep and unmovable economic (rather than political) impasse, but the idea that a decent basic income is impossible under capitalism boils down to the claim that real reform of capitalism is impossible.
However, one very real feasibility constraint concerns labor market participation: if basic income pulls most of the workforce out of the labor market, the scheme’s ultimate revenue source will dry up. Yet, as noted above, experimental evidence suggests that basic income payments hovering at half of median family income induce some labor market withdrawal, but not catastrophic levels. To my mind, this result is more or less desirable: no work reduction would mean no expansion of freedom and no lessening of toil, but extreme work reduction in the short-run risks unraveling the program. Contrary to common opinion, basic income should not be understood by itself as a post-work utopia: indeed, if most everyone dropped out of work, there would be no revenue to fund the scheme. The wager is that even though work would be a choice rather than an economic necessity, people would for the most part continue to find work attractive, albeit less so; poorly remunerated jobs would be bid up (itself a process that makes work more appealing, partially compensating for exits elsewhere), and workplaces characterized by the worst forms of domination would be less sustainable.
A further prediction made by David Purdy is that workers who reduce labor hours or exit from the labor market will make it easier for underemployed or unemployed workers to find work.21 If it is indeed the case that employers require replacement hires for exiting workers — and it should be said that there exists no evidence for or against this hypothesis because of data limitations — this particular mechanism predicts not so much an increase or decrease, but rather a redistribution of available work. There are reasons, therefore, to expect increases in labor market participation in some cases, even if the scheme generates net declines.
Nevertheless, despite evidence to the contrary, it may be the case that basic income does drain most workers out of the labor market. Or perhaps these perverse effects would eventually materialize with a massive basic income. If so, the non-sustainability argument has force, and it means that there is some Goldilocks level of basic income, above which people drop out in droves. My own estimation is that if some such level exists, it is considerably higher than the figures posed above: none of the evidence from a wide range of benefit levels in the various experiments comes close to inducing a collapse in the labor market. Because of the benefits of added income, the inherent appeal of work, and its potentially growing attractiveness due to changing power relations, it seems to me that an increasingly generous basic income will face other sustainability problems long before some mass exodus of labor buckles the economy.
This brings us to a second feasibility constraint: a decent basic income might be impossible under capitalism due to capital flight. In this story, high taxes or high wages will lead capitalists to disinvest, thereby undermining the revenues required to fund an expensive basic income. How narrow are the bounds of a progressive welfare state within the context of capitalism? Would basic income provoke debilitating levels of capital flight, thereby depleting the tax base necessary to fund the scheme?22 Although UBI is undeniably expensive, this criticism is overstated. One way to think about the problem is as follows: At the highest level of abstraction it is clear that a country like the United States is far from the threshold where tax revenue as a percentage of GDP reaches its theoretical limit inside a more or less capitalist economy. If the lower bound for this theoretical upper limit is the Danish level of about 51 percent, the United States, at about 26 percent, can afford to double its spending. On abstract feasibility grounds, there is plenty of room to grow the portion of resources that we devote to public purposes before the Marxist theory of the state kicks in to insist on a hard limit to left policymaking within capitalism.23 This counterargument obscures many important details — for example, the types of tax instruments used can significantly impact the likelihood of capital flight — but it is worth recalling that the threat of capital flight is often just that: a threat.24 If higher social spending is forcibly imposed on them, there is good reason to believe most capitalists would accept it, albeit unhappily, rather than abandon their firms.
Even if there is good reason to believe that the threat of paralyzing capital flight is itself far off, it may still be ultimately fatal at some threshold. At this point, however, it is likely that social and political conditions also begin to change. Indeed, as basic income grows — because of rising expectations, the program’s rising popularity, and an increasingly empowered populace — there will be a greater need to find new funding by directly taxing capital through a range of mechanisms. Perhaps funding schemes that heavily tax capital are avoided at first because of the sensitivity of investment, but eventually it becomes an unavoidable revenue stream, thereby exacerbating the capital flight threat. One solution that may become viable for political leaders is — in fits and starts, and in specific industries — a program to socialize various means of production. The initial hazard posed by capital flight may thus become an opportunity. This will help solve the underlying economic problem of abating private capital’s need for profits, while also serving as a fresh source of funding. For example, John Roemer’s coupon socialism model is essentially a basic-income-like dividend funded by the universal ownership of all capital assets.25 This story is of course highly speculative, but as a sketch of the transition to socialism it seems about as plausible a way to surmount the capital flight problem as any. Call it the basic income road to socialism.
To conclude this section, it is worth making note of a final, more pointed economic critique of basic income; namely, that the policy is nothing more than an employer subsidy. One version of the argument goes like this: there is a subsistence wage out in the world that is historically determined, but more or less fixed, and if the state can be made to cover some of that wage, employers will happily pay less of it.26 Apart from resting on an unsustainably functionalist argument about wage setting, the inner logic is absent. Wage declines do not happen magically — they have to be imposed. But when workers have an exit option, a bargaining chip, wages are likely to go up rather than down.27 Indeed, in the case of Mincome, we can observe this very effect: relative to businesses in control towns, basic income forced Dauphin businesses to raise wage offers in order to better attract workers who now had a decent alternative.28
The argument goes further. Even a small but unconditional basic income would not be an employer subsidy. To clarify, take a seemingly similar case: The US Earned Income Tax Credit is an employer subsidy, but not because of some functionalist mechanism about subsistence wages; it is an employer subsidy because it is an income transfer that is conditional on work and therefore increases the labor supply, which pulls wages down.29 In contrast, a small unconditional basic income would, in a small way, raise the reservation wage of labor, just as food stamps, in a small way, raise the reservation wage of labor and lower work hours — they allow people to be just a bit pickier.30Provided a basic income policy is not conditional on work, even a modest version added to the current welfare state would make it marginally easier to say no to bosses because it offers a modicum of an alternative.
It is important to put the employer subsidy position — a truly classic case of Albert O. Hirschman’s perversity thesis — to bed because, first, there is no evidence to support it, and second, it forecloses on the otherwise reasonable strategy that views a small basic income as a way station to a large one.
Basic Income, Collective Action, and Solidarity
If the above argument about wage growth is correct, coupled with public support, an insufficient-but-unconditional basic income presents a viable path to a more generous one. While I make this case below, it is first worth laying out an argument leveled against the potential impact of basic income on solidarity: UBI will not only dramatically increase the tax burden on some and redistribute a good amount to others, but it will do so in a way that is immediately socially recognizable as a transfer; unlike, say, health care and housing, transferring actual cash from one party to another is conspicuous. As a consequence, it is easy to imagine a vulnerable group being publicly accused of laziness and dependence. Is it possible that the net-contributors to the program will strongly differentiate themselves from and even begrudge the net-recipients?
In response, it is useful to distinguish between different types of income transfer programs. For example, unlike a negative income tax, where some people — those below a given threshold — collect payments and others do not, the universal basic income makes everyone a recipient. The calculation of the net impact of a UBI is far less conspicuous than under a negative income tax where you either physically receive payments or don’t. The UBI computation requires comparing the amount you receive to the portion of your tax contribution allocated to the program. Post-tax-and-transfer winners and losers are far less visible, even if the two schemes achieve the identical post-tax-and-transfer income distribution. It is also worth mentioning that family allowances — in Canada, France, and the UK — are (or were) near-universal cash transfer programs and among the most popular social policies in those countries. Indeed, there are plenty of cash transfers that are robust and popular. Those that are, as I discuss below, tend to avoid distinctions between the “deserving” and “undeserving” poor, and thus escape the cycle of stigma and victim-blaming that so many social assistance programs are vulnerable to.31
By contrast, traditional welfare policies suffer from inherent limits to political mobilizing: they impact only a small, poor, and marginalized group, and are consistently at the top of the list of the most unpopular social policies. Because so few people are touched by welfare policies targeted at the poorest populations, organizing benefit increases is always an uphill battle and requires disproportionate reliance on moral arguments, rather than material ones. It is for the same reason that such policies are uniquely vulnerable to austerity. Yet even a weak basic income could touch a broad array of people and help build a solid constituency to support its continual growth and expansion. As more people are folded into a program, two things happen. First, quality improves. And second, it becomes a political third rail. Programs with benefits dispersed broadly across diverse social layers tend to become highly popular and can start to be seen as a civic right, making for ratchet effects where gains become irreversible.
Indeed, this popularity effect is clear from the qualitative commentary from Mincome participants in Dauphin. Mincome helped blur the usual lines of demarcation between the deserving and undeserving poor. For many, welfare was viewed in moralistic terms; it was a signal of a tarnished moral character and consistently too humiliating for most to consider joining. Mincome, however, was viewed as a neutral, pragmatic program, and its widespread availability meant it was not interpreted as a system for “other” people. People took casual and positive attitudes toward Mincome and participated because they simply “needed money,” while the vast majority despised welfare because, among other things, it was for the “needy and bums.” They often distinguished their own Mincome receipt — which was based simply on needing cash in an economy with precarious employment opportunities — from the circumstances of welfare receipt, which were caused by recipients’ moral failings. Even Mincomers with strong ethics of self-reliance or negative attitudes toward government assistance felt able to collect Mincome payments without a sense of contradiction.32
There is thus a powerful argument that the universalism of UBI would facilitate solidarity that is otherwise obstructed in a highly fragmented and categorical welfare state marked by deep tensions between low-wage workers, unemployed workers, and social assistance recipients. Similar life experiences are critical in facilitating communication and solidarity (for Marx, it was the similarity of life inside the walls of the factory that galvanized solidarity). At minimum, even if a UBI does not actively nurture solidarity, breaking down the categorical nature of social provisioning may reduce the barriers to alliances across otherwise separated groups of poor and working people.
However, there are other aspects to consider when thinking through the impact of basic income on collective action and solidarity. Indeed, it may be the case that the overall impact of basic income on solidarity is somewhat indeterminate, with certain forces facilitating it, and others running against the grain. Although we have seen that the impact on wages is likely to be favorable, what can we say about the manner in which those wage gains are made? Put differently, if wage increases can be won through individual or collective strategies, how might UBI play out in this respect? The basic fact of an exit option might mean that individuals use their newfound powers to bargain on their own, not collectively. It might allow them, moreover, to opt out entirely. After all, basic income increases workers’ bargaining power with their bosses, but it also increases their power with respect to their unions. Offering people alternatives to economic dependence on employers also means alternatives to economic dependence on collective solutions.33
The optimistic view proposes that basic income would for the most part facilitate collective action. It is sometimes suggested that a UBI could operate as an inexhaustible strike fund; indeed, the National Association of Manufacturers (NAM) was the first to recognize this in their Congressional testimony on Nixon’s Family Assistance Plan, a guaranteed income that nearly passed in Congress in 1970. The business group was willing to support the plan as long as it was a significantly watered-down variant of the original, fairly radical and work-unconditional proposal. In congressional hearings the NAM insisted that they would support the program only “if the basic allowance is a realistic minimum, and if the earnings disregard provides a true incentive to work and advancement and if the work requirement is strong.” Finally, they expressed concern about the link between the guaranteed income and labor upheaval: “We suggest that anyone directly involved in a labor dispute should be ineligible for benefits under the family assistance plan.”34 This worry on their part seems perfectly reasonable.
In this vision a UBI aids collective action because it provides the positive resources to facilitate it. Moreover, the policy would reduce the temptation to “defect” from collective action. Desperate workers, individuals with few alternatives, would be less inclined to scab if they had another decent survival option. However, while basic income provides the positive sustenance for collective action, it weakens the negative motivations that spur it on. Much collective action happens because workers have no alternative but to struggle in conjunction with others. Basic income eliminates the external condition of starvation, the condition that forces collective action on people as the only viable path to advancement. Thus, while it undermines the push factor, it strengthens the pull factor by providing the material support that makes collective action more likely to occur and succeed.
It is of course perfectly reasonable to imagine that basic income might empower people both as individuals and collective actors, facilitating both solitary and collective struggles against powerful social actors. From the perspective of socialist freedom, this approach to collective action strikes me as desirable. Moreover, as noted in Table 1, survey data on why people were not in the labor force during the Mincome experiment reveals some evidence suggesting that people acted individually and some suggesting collective action. I noted above that survey data showed that care work and education were cited, but the strongest reason for not working was related to dissatisfaction with job or work conditions. One can see answers relating both to workplace struggles and opting out in the data — another common answer, as indicated above, was “did not want to work.”
But what if, contrary to my arguments, universalistic income maintenance ultimately hampers solidarity? If basic income enhances some of the positive reasons for collective action and undermines some of the negative reasons, the net effect could still be negative. It may turn out that the only way to nurture solidarity is to leave workers with no exit option, and no alternative to collective action. Perhaps free (or freer) people will not choose solidaristic strategies and prefer to go it alone. Should we decide that it is then preferable to maintain an external starvation constraint in order to better ensure group solidarity? Even in this limiting case, it would be strange for the Left to argue in favor of economic dependence on the capitalist class. Certainly the freedom tradition in socialism would find little in the way of argument to justify an instrumental case against current autonomy in the anticipation of greater autonomy in the faraway future. The intuition that suggests workers ought not have a basic income because they might behave in ways we don’t like is the same intuition recommending that the Garden of Eden ought to be destroyed should it one day be discovered on Earth. A place like Eden, where our subsistence needs can be met by plucking fruit from the trees, where we can make ends meet on our own, might corrupt our other-regarding impulses. But that would be a bad argument against Eden. The issue is best conceived as a socialist wager: we hope and hypothesize that free people would prefer cooperative, collective action, but if they don’t then they don’t. That sad counterfactual is an insufficient reason to limit their freedom.
Basic Income and Gender
Among the open questions concerning the empirical consequences of basic income, the issue of gender is sometimes seen as the most ambiguous. Before interrogating the evidence on this matter it is worth recalling the 1970s Marxist-feminist campaign for “wages for housework,” a social movement (and demand) with much affinity to basic income, as demonstrated by Kathi Weeks.35 Wages for housework was in part a real demand for remuneration for valuable economic activity, and in part an attempt to socially recognize the unpaid care work done disproportionately by women. It was meant to make visible labor that was otherwise invisible. The demand itself was straightforward: Women do valuable and productive, but unremunerated, domestic work and they ought to be paid for it.36There is a “social factory” that is largely invisible but facilitates the very existence of the industrial factory insofar as the former partially produces (or “reproduces”) the human inputs for the latter.
However, even the chief proponents in the movement were hesitant to commit to the normative demand as a concrete social policy. Ellen Malos noted that it was “not clear whether campaigners for wages for housework really want what they are asking for.”37 As an earnest normative demand it was a nonstarter. Few feminists could get on board with a scheme that is dangerously essentialist and at bottom a categorical social policy only available to women — or women who do housework. As it was designed, it would fortify a highly gender-inegalitarian division of labor — indeed, male housework was sometimes regarded as scab labor in the milieu of the time. Moreover, the perspective views the assignment of housework to women as more or less appropriate. The wages for housework demand might render the housework done by women visible, and recognize it as socially valuable, but it also naturalizes it, and buttresses a gendered division of labor. For these reasons, wages for housework, taken as a genuine attempt to reorganize social life and envision a just system of remuneration, was indefensible.
In Weeks’s overview of the debate, she draws a straight line from wages for housework to basic income, arguing that the latter better achieves the underlying objectives of the former. Weeks writes that wages for housework proponents sought a “measure of independence”: a certain level of autonomy — and power that flows from it — was the underlying objective, and wages for housework was the means for achieving it. The problem was that it was a categorical social policy that poorly realizes its own core normative vision. For Weeks, “[p]recisely because it does not address its potential recipients as gendered members of families, the demand for basic income is arguably better able to serve as a feminist perspective and provocation.”38 Unlike wages for housework, basic income comes without the strings of actual housework: for that reason it better undermines economic dependence and better realizes the twin goals of autonomy and power.
From a Marxist viewpoint, one of the central conditions undermining autonomy and facilitating exploitation in the labor market is the double freedom discussed above. There is a clear parallel here with the historic conditions underwriting women’s subordination to their husbands. In a traditional marriage, without access to external means of subsistence, women remain economically dependent on male breadwinners. As a consequence, their power both inside and outside the context of marriage is constrained.
If double freedom is a stylized fact of capitalism, from the Marxist-feminist lens, what happens then when a social policy breaks the second half — the freedom to starve — of that dictum? The Marxist hypothesis is that power relations between workers and employers will be transformed. The corresponding Marxist-feminist problematic centers on the ways that social policy weakens or entrenches women’s dependence on their husbands. Basic income operates as an outside option that can modify the internal dynamics of marriages. If you have a viable exit option, your power inside a marriage can improve. If you have no outside options, you are more likely to remain a subordinate partner.
These issues were debated in the context of the American guaranteed income experiments. The debates played out in the pages of the American Journal of Sociology, and they were framed in an exceedingly narrow fashion — would the guaranteed income undermine “marital stability”? — but the implications for women’s power and autonomy lurked in the background. Some evidence appeared to show that women would leave their husbands because they could make do without them (this was termed the “independence effect”) and some evidence seemed to show that extra income would improve marital stability (the “income effect”).39 The debate generated an immense controversy on empirical and methodological grounds, but a weakness of equal importance was that the core questions were undertheorized. At no point did researchers attempt to investigate the ways that an outside option would impact the power relations internal to marriages.
It was seldom acknowledged that if some marriages dissolved, perhaps they were bad or abusive marriages, formed and sustained in the context of limited alternatives. Likewise, if some marriages were stabilized — as others found — then perhaps it was because the guaranteed income ameliorated underlying financial stressors. There are, however, further hypotheses that were ignored. Rather than simply making exits more likely, basic income may impact the balance of power and decision-making within relationships by making the threat of exit credible. It may also mean that relationships prone to large inequalities in power were less likely to form and solidify. It may be hypothesized further that these changes in the positional power of women, their expanded capacity to realize their demands, have broader effects, including possible reductions in the risk of violence. This view shifts attention from the dissolution of marriage to changes in the power relations interior to them, from actual exit to the threat of exit, and poses a further empirical hypothesis: basic income could increase the bargaining power of wives vis-à-vis husbands and thereby reduce the risk of violence by making credible the threat of exit. In the Dauphin case, I find some preliminary evidence of a decline in domestic violence, and several mechanisms — actual exits from marriage such that exposure to potential violence declines, changing power relations due to the availability of the threat of exit, and a decreased risk of violence due to reduced financial stress — may have all played a role.
However, if the impact on power and autonomy is a net positive, what are we to make of the potentially negative implications for women? It is often argued that a universal basic income would disproportionately reduce female labor market participation and entrench a gendered division of labor. This itself might have implications for the reduced power of women inside relationships. Indeed, the experimental evidence from the 1970s shows that women reduced their labor supply a good deal more than men did. Would a contemporary implemented UBI have the same disproportionate effects?
While it may still be the case that women would reduce work more than men, it is highly unlikely that the effect will be as disproportionate as it was in the 1970s. With a far narrower gender wage gap, many women today will find the opportunity costs of work withdrawal to be too high, and thus decide, like most men, to continue to work. Even so, it is still possible that women would see a somewhat greater impact than men on this front, generating some negative empirical outcomes, including the entrenchment of a gendered division of labor. One response would be to say that while this might be true, on balance — and especially considering the evidence on power, autonomy, and violence — a UBI would have net gender-egalitarian consequences. A second response would be to admit that some outcomes might be negative, and like any social policy measure with unintended negative effects, it ought to be countered by other supplemental policies that bolster a more gender-egalitarian division of labor. A third response would emphasize the limits of the old strategy of replacing domination by husbands with domination by bosses. Such a substitution may have once had appeal under certain circumstances, but weakening economic dependence as such is preferable. However one falls on this issue, what has to be asked is whether these empirical and theoretical ambiguities should prompt us to surrender the freedom to quit. Again: do we wish to disallow a Walmart worker from quitting her job if she so desires?
Basic Income and the Socialist Project
With right-wing variants of basic income on the table, it is natural to see a flurry of left criticism. Nonetheless, it should be recognized that the UBI concept dovetails with a normative vision that has deep roots on the Left. The fundamental goals of the socialist left have long been fixed on emancipation, self-realization, and the satisfaction — and even expansion — of human needs. As Adam Przeworski writes, “Socialism was not a movement for full employment but for the abolition of wage slavery … it was not a movement for equality but for freedom.”40 It was only when those goals appeared closed off by political and economic circumstances that we narrowed our horizons and settled for a productivist alternative, characterized by more rather than less work. Having found it unworkable in the medium-run to eradicate exploitation and alienation, socialists set out to universalize them.
Socialism lost something in the reorientation from a vision defined by the abolition of the wage relation to one that fastens us all to it. A generous basic income defined by a genuine exit option from the labor market ultimately has a real affinity with the socialist project. The moral question at the fore is whether or not we wish to retain the coercive and compulsory quality of the capitalist labor market. Fighting for an exit option ought to be a priority because, first, it gives people the power to confront their bosses or their spouses — the possibility of exit facilitates voice — and second, because it gives people real freedom to enact their life plans, unencumbered by the dull compulsion of economic relations.
Expanding people’s real freedom and eroding the background condition of market dependence are core features of basic income and at best secondary goals in the jobs-and-services strategy. The objective is to free workers not only from a given capitalist, but also from capitalists as a class. This is why a generous and truly universal basic income ought to be a plank in any broad socialist agenda. That said, social order is often secured through some measure of coercion, and forms of social organization that strive to reduce coercion and expand people’s freedom always run the risk of dysfunctionality. This is a danger baked into the socialist project. Likewise, the risk that basic income gets co-opted and turned in on itself is a problem facing any abstract policy proposal. It is a danger inherent in the move from theory to practice, and it presents itself whenever an idea narrows in on reality. Whether or not that happens is ultimately up to us.