The Architecture of a Basic Income

Introduction

The notion of a universal basic income, or UBI, has captivated academics, entrepreneurs, policymakers, and ordinary citizens in recent years. Across the globe, countries ranging from Brazil to Finland, the Netherlands, Italy, Kenya, Uganda, and Canada are conducting or have recently concluded pilot studies of a UBI.1 Here in the United States, the city of Stockton, California, has partnered with a nonprofit organization to give checks of $500 a month—no strings attached—to several dozen families;2 a mayoral task force in Chicago has recommended a similar pilot program that would pay one thousand people $1,000 a month;3 and the Hawaii legislature has launched a working group to evaluate a UBI in the Aloha State.4 A tech-funded nonprofit has hatched a plan to give $1,000 a month to one thousand families in two as-yet-unannounced states,5 with the stated goal of taking a “first step toward defining a new social contract for the 21st century.”6 And prominent voices from across the ideological spectrum have taken up the UBI idea as well: libertarian Charles Murray,7 Facebook cofounder and Obama campaign strategist Chris Hughes,8 and former labor leader Andy Stern9 all have offered proposals for nationwide programs that resemble a UBI.10 Former President Obama floated the idea of a universal income in a July 2018 speech in South Africa;11 Hillary Clinton now says she seriously considered adding a basic income plank to her 2016 presidential campaign platform;12 and entrepreneur Andrew Yang—who made a UBI the primary plank of his bid for the 2020 Democratic presidential nod—attracted more support in fundraising and polls than many well-established politicians, including several sitting senators.13

All the while, the mechanics of how a nationwide UBI would work have drawn relatively little attention. Most experiments, for example, are designed to test the effects of unrestricted cash transfers on health, educational attainment, workforce participation, and subjective measures of well-being among a small population—not to gauge whether such programs are scalable to a country of 330 million. Those who have offered concrete nationwide UBI proposals have skipped over details that could determine the success or failure of a UBI—such as precisely how it would phase out, whether benefits would be assignable, and which agency would administer what would likely be the largest program of periodic cash transfers in human history. Some insights can be gleaned from academic studies in the late 1960s and early 1970s regarding the design of a “negative income tax,”14 but technological transformations and several decades’ worth of social science research render much of that work in need of updating.

To be sure, even the most optimistic advocates for a UBI will acknowledge that nationwide implementation lies years—if not decades—ahead. And accordingly, one might argue that hashing out the nitty-gritty programmatic specifications of a UBI puts the cart before the horse. But as we seek to show below, these design details in many cases are the horse. What might seem like technical aspects of a UBI (for example, whether it is paid weekly, monthly, annually, or once in a lifetime; whether an individual’s future stream of UBI payments can be posted as collateral for a loan; and whether a spouse’s income is factored into the calculation of a phaseout) turn out to be essential elements that affect whether a future UBI will live up to the high expectations that supporters have set.

Moreover, while we hazard no predictions as to the imminence of a UBI, we note that if and when a UBI comes, it could come fast. A predecessor to the UBI—the negative income tax—went from academic idea to presidential policy priority in seven short years in the 1960s.15 In April 1970, the House of Representatives voted overwhelmingly to adopt a bill providing a minimum income of $500 per parent and $300 per child for families in the United States—the equivalent of more than $10,000 for a family of four in today’s dollars.16 And while the bill ultimately failed in the Senate, the very notion that such a scheme could command bipartisan support in one house of Congress would have seemed virtually unimaginable only a few years earlier. The path from political implausibility to national policy is sometimes quite short.

If and when such support does coalesce into law, the implementation of a UBI may prove to be the most important domestic policy endeavor of our time. If it succeeds, it could mean the end of extreme poverty—at least in advanced industrialized economies. Yet if it does not deliver on its promise to improve the lives of recipients or if it causes massive departures from the workforce that undermine economic growth, then the failure of a UBI could cast a long shadow over cash-assistance efforts for decades to come. In sum, the stakes involved in the implementation of a UBI are potentially enormous, making attention to implementation details all the more essential.

This Article seeks to advance the conversation among academics and policymakers about UBI implementation. Our prior work has focused on the philosophical foundations of a UBI;17 here, we build up from those foundations to identify the practical building blocks of a large-scale cash transfer program. Specifically, we focus on six building blocks of a UBI. By “building blocks,” we refer to core components of any UBI that will shape policy outcomes significantly. To preview, our six18 building blocks are:

• Size: How much will the basic income be?
• Eligibility: To whom will the basic income extend, and what eligibility criteria (if any) must potential recipients satisfy?
• Uniformity: Will payments vary based on geographic location, household size, or wealth?
• Assignability: Can individuals transfer their right to future payments (for example, as collateral for a loan)?
• Payment Mechanism: How will the actual cash transfer to UBI recipients work? How often will payments be made? Which agency or agencies will be tasked with implementing the program? and
• Funding Mechanism: Will a UBI be funded by the elimination of other social welfare programs (and if so, which ones?), by additional taxes, or by a combination of both?

Our analysis draws from studies of the existing tax-and-transfer system and from ethical and political theory. In most instances, our analysis yields a set of specific recommendations for policymakers. For example:

• Size: While no number is magic, a basic income of $500 per person per month would bring an end to extreme poverty in the United States and would be feasible given the current size of the US economy;
• Eligibility: A comprehensive basic income should cover citizens and lawful permanent residents of all ages;
• Uniformity: Marriage penalties, asset tests, and geographic differences in payment amounts are all best avoided;
• Assignability: Adult UBI recipients should be able, within limits, to use future payments as collateral for short- and medium-term loans—either from private lenders or from the government;
• Payment Mechanism: Payments should be made via direct deposit on a biweekly basis and delivered through the Social Security Administration; and
• Funding Mechanism: The cost of a UBI should be offset by, at minimum, the consolidation of some or all current cash and near-cash transfer programs as well as a surtax layered on top of the existing tax structure and administered by the Internal Revenue Service (IRS).

Importantly, the building blocks of a UBI do not necessarily determine its outward face. By this, we mean that economically identical programs can be described in very different ways without altering any of the essential features. For example, a basic income for all individuals funded by an additional income tax can—without any substantive change—be recast as a “negative income tax” that only provides benefits to individuals or households below a certain income threshold. Likewise, a monthly or annual UBI that can be posted as collateral for a loan also can be redescribed as a once-in-a-lifetime “stakeholder grant”—again without much more than a cosmetic change. Of course, packaging does matter to the public perception of a UBI, and we consider reasons why some characterizations of the program may prove more popular than others. A key contribution of this Article is to sort the building blocks of a UBI out from the cosmetic components, thereby clarifying which elements of a UBI shape implementation and which ones affect only the external appearance.

We approach these questions from a distinct disciplinary perspective—as tax law scholars—and our own views as to how the building blocks of a UBI ought to be assembled are colored by our observations of the federal income tax system in operation. The federal income tax serves as a model for a UBI in many respects but certainly not in all. The definition—or rather, definitions19 —of income in the Internal Revenue Code provide a starting point for any calculation of the surtaxes that would be used to finance a UBI. Experiences with the earned income tax credit (EITC) over the past four decades and with the child tax credit over the past twenty years yield insights regarding the implementation of cash-transfer programs. Challenges related to bubble rates and marriage penalties—familiar from the income tax context—loom over the design of any UBI scheme. And the IRS stands as one, but not the only, agency that might have the capabilities and expertise to administer a program of a UBI’s size and scope.

Our analysis also is informed by our engagement with ethical and political theory. How the building blocks of a UBI should be assembled will depend in large part on the philosophical foundations upon which a UBI rests. A variety of philosophical approaches—including welfarism, resource egalitarianism, and libertarianism—potentially lend support to the idea of a UBI, but translating the abstract idea of a UBI into a concrete proposal exposes the junctures at which these various approaches diverge. These differences go beyond the questions of size, eligibility, and funding mechanisms: for example, welfarists and libertarians may find that they arrive at very different answers to the question of whether rights to future UBI payments should be assignable. While we believe that a successful UBI movement will need to adopt an open-tent attitude toward supporters from different philosophical perspectives, thinking through the design details of a UBI will shed some light on whether this intellectually divided house can stand.

The Article proceeds as follows. Part I provides a brief overview of the UBI idea, distinguishes it from traditional welfare programs, and explains the essential similarities among UBI proposals with different labels. Part II bores down into the foundations of a UBI with the aim of identifying the ethical and philosophical premises upon which a UBI rests. Part III canvasses the considerations that inform the design of a UBI and then provides policymakers with a set of specific recommendations influenced by both tax policy and ethical theory. Part IV looks toward the outward appearance of a UBI: How should the program be presented to the public? We weigh the political advantages and disadvantages of various framings (UBI vs. negative income tax; phaseout vs. no phaseout), and we explain how the outward framing of a UBI can be separated from its inner workings. We conclude by considering the ways in which the UBI debate can inform the design of antipoverty programs in the here and now, before support for a UBI achieves critical mass (if it ever does). A UBI, we argue, is not all-or-nothing, and many of the innovations discussed here can be repurposed to improve upon existing antipoverty and cash-transfer programs.

I. The Basics of a UBI

A universal basic income, or UBI, refers to a program that ensures that all members of a polity have access to at least a minimum sum of money.20 At bottom, it is a method for distributing resources among individuals within a given society. It could—but need not—replace existing welfare programs. It also could—but also need not—be cast as a response to automation-induced job losses. What distinguishes a UBI from other transfers is that it is unconditional and unrestricted: that is, receipt of a UBI is not contingent upon a particular characteristic (for example, age or disability) or a particular activity (for example, work), and benefits are not restricted to any particular use.

Unconditional and unrestricted cash-transfer programs go by various names, such as a “basic income guarantee,” a “guaranteed income,” a “citizens income,” a “negative income tax,” and a “social dividend.”21 Although these names may reflect different philosophies or objectives, they do not change the program’s core features or dictate distributional outcomes. Any given name or frame, however, may affect the level of political support. For ease of exposition, we use the term “universal basic income,” or “UBI,” in Parts I–III and then discuss alternative framings in Part IV.

A. UBI vs. the Welfare State

A UBI resembles existing welfare programs insofar as it entails the redistribution of resources from higher-income to lower-income individuals and households. This is true even if a UBI is characterized as being “universal” in the sense of being distributed to all individuals regardless of need.22 This is because virtually every UBI proposal envisions that the UBI will be funded, implicitly or explicitly, with additional revenues from taxes that increase with income. Higher-income individuals will pay more in additional taxes than they receive from the UBI; lower-income individuals will pay less.23

But a UBI differs from existing welfare programs in several key respects. First, a UBI is unconditional in the sense that it is not contingent upon an individual showing of deservedness. Recipients need not work nor demonstrate that they are unable to participate in the labor force. In this respect, a UBI contrasts with most current programs. For example, the federal Supplemental Security Income (SSI) program requires non-elderly recipients to demonstrate that they are blind or disabled.24 The Supplemental Nutrition Assistance Program (SNAP) requires able-bodied adults without dependents to work or to participate in training or education for at least eighty hours per month in order to continue to receive benefits.25 Temporary Assistance for Needy Families (TANF) is available only to pregnant women and individuals who care for minor children.26 A UBI, at least in its purest form, would not be conditional upon disability, work effort, volunteering, caregiving, the presence of dependents, pregnancy, or any other individual factor. Also, a truly unconditional UBI—unlike many current programs27 —would not exclude beneficiaries because of drug use or past criminal activity.

Second, a UBI—unlike many current welfare programs—is delivered via unrestricted cash or cash-like payments. This contrasts with programs that provide in-kind benefits (such as the National School Lunch Program) and programs that provide vouchers for specific goods and services (such as SNAP and Section 8 rental assistance). A UBI—again, at least in its purest form—also does not restrict the range of goods and services that recipients can purchase. By contrast, the TANF program generally delivers aid through debit-like “electronic benefit transfer” cards,28 and federal law mandates that states take steps to prevent beneficiaries from using those cards at any liquor store, gambling establishment, or strip club.29 Many states impose additional restrictions; Kansas, for example, prohibits the use of electronic benefit cards for tobacco, tattoos, and body piercings, and at sporting events, jewelry stores, nail salons, movie theaters, and swimming pools.30 A true UBI would contain no such restrictions on use.

A final feature of a UBI is that it comes in the form of regularly recurring periodic payments. The payment period could be weekly, monthly, annually, or any other stretch. Most current welfare programs have a similar periodic-payment feature, delivering benefits monthly (for example, SNAP and SSI) or annually (for example, the EITC). A UBI differs, however, from the idea—rooted in the writing of Thomas Paine31 and popularized in the legal academy by Professors Bruce Ackerman and Anne Alstott32 —of providing individuals with large lump-sum grants upon entering adulthood. As we discuss in Part III.D, the difference between periodic payments and lump-sum grants may be largely cosmetic if individuals can assign their future stream of periodic payments as collateral for a lump-sum loan. Without free assignability, however, the choice between periodic payments and a lump-sum grant has important implications for the operation of—and outcomes from—a comprehensive cash-transfer scheme.

In theory, a UBI can act as a substitute for the existing welfare system or as a complement to it. There is much to be said for both approaches, and we review the main arguments in the substitute-versus-complement debate as part of our discussion of funding mechanisms in Part III.F. Many of the key design details of a UBI—such as how and how frequently payments would be made, whether future benefits would be assignable, and who would administer it—do not depend upon the answer to the substitute-versus-complement question. Others—such as size and eligibility—almost certainly do.

B. UBI vs. Negative Income Tax

While the differences between a UBI and the existing welfare state are significant, the differences between a UBI and a so-called negative income tax are not. Moreover, some of the distinctions drawn between different UBI variants—for example, a UBI with a phaseout and a UBI that never phases out—turn out to be alternative ways of accounting for the same flow of funds.

To illustrate: Imagine a UBI of $6,000 per year ($500 per month) that extends to all individuals regardless of income, coupled with an income tax of 20 percent on all non-UBI income. Thus:

• A, who earns $0, receives a UBI of $6,000 and pays no income tax, for a net transfer from the government of $6,000;
• B, who earns $30,000, receives a UBI of $6,000 and pays income tax of $6,000, for a net transfer to/from the government of zero; and
• C, who earns $60,000, receives a UBI of $6,000 and pays income tax of $12,000, for a net transfer to the government of $6,000.

Now imagine that instead of the above-described system, we implement a UBI of $6,000 that phases out at a 20 percent rate starting with one’s first dollar of non-UBI income, and that we couple that with a tax of 20 percent on non-UBI income over $30,000. Thus:

• A, who earns $0, receives a UBI of $6,000 and pays no income tax, for a net transfer from the government of $6,000;
• B, who earns $30,000, receives no UBI (because her UBI has fully phased out) and pays no income tax, for a net transfer to/from the government of zero; and
• C, who earns $60,000, receives no UBI (again, on account of the phaseout) and pays income tax of $6,000, for a net transfer to the government of $6,000.

The first example involved a UBI with no formal phaseout; the second UBI phased out explicitly. But as should be evident, the first and second arrangements are identical in economic terms.

Now imagine that instead of a UBI, we implement an arrangement involving a negative income tax. Specifically, the income tax is 20 percent × (income − $30,000), with individuals whose tax is negative (that is, whose pre-tax income is less than $30,000) receiving a payment from the government of 20 percent times the difference between $30,000 and their pre-tax income. Thus:

• A, who earns $0, faces an income tax of 20 percent × ($0 − $30,000), which is to say, receives a net transfer from the government of $6,000;
• B, who earns $30,000, owes nothing and is owed nothing; and
• C, who earns $60,000, pays a tax of 20 percent × ($60,000 − $30,000) = $6,000.

As should be apparent, this third regime is the exact same in economic terms—resulting in the exact same net transfers between individuals and the government—as the first and second. That is, the same arrangement can be described as a UBI with no phaseout, or as a UBI with a phaseout, or a negative income tax.33 The various framings may affect the political popularity of a UBI proposal, but they do nothing to change its essential features.

II. The Foundations of a UBI

Our discussion so far has focused on the basic workings of a UBI. In this Part, we shift from questions of mechanics to questions of morality and ethics. We present a brief survey of the normative foundations for a UBI in philosophy and political theory. These justificatory arguments will—as we illustrate in Part III—prove significant to the design of a basic income program.

The argument for a UBI boils down to three premises: (1) society should redistribute from the rich to the poor; (2) redistribution should come (largely or exclusively) in the form of unrestricted cash transfers; and (3) those cash transfers should be unconditional—that is, available to all regardless of age, disability, work history, or other individual characteristics. These arguments, in turn, derive varying levels of support from at least three main ethical theories: welfarism, egalitarianism, and libertarianism. We focus on these three theories because of their predominance in debates over tax-and-transfer policy in the United States and abroad.34 We note, though, that these three perspectives do not exhaust the possible philosophical justifications for a UBI. One author has offered an argument for a UBI rooted in the capabilities approach associated with Professors Martha Nussbaum and Amartya Sen;35 others have defended a UBI on communitarian grounds.36 Our focus on welfarism, egalitarianism, and libertarianism ought not be interpreted as an implicit judgment regarding alternative ethical theories or on their amenability to a UBI.

Before proceeding, we note that none of the arguments for a UBI outlined below imply any claim about automation and the future of work. Some advocates of a UBI, including labor leader Andy Stern,37 Tesla founder Elon Musk,38 and former presidential candidate Andrew Yang,39 emphasize the potential role of a UBI as a safety net in the event that automation leads to large-scale job losses. Many UBI skeptics seize upon that argument as a reason not to back a UBI. For example, former vice-presidential candidate Senator Tim Kaine told a reporter in October 2017 that “[t]he premise of UBI is that work will be destroyed so much that you will need something like that [to replace it].” He added: “I don’t completely accept that that’s the way things are going.”40 Professor Jason Furman, who chaired the Council of Economic Advisers under President Obama, has attacked the premise that automation will lead to large-scale unemployment and criticized a UBI as “giving up on work and giving up on people.”41

As will become apparent, one can support a UBI without believing that work will wither away. For example, one need not place faith in forecasts of widespread job losses in order to believe that some individuals will be locked out of labor markets (for example, on account of undiagnosed disability) and that a UBI serves as a normatively attractive means of guaranteeing a minimum level of well-being and autonomy for all members of society. One also can oppose a UBI while crediting the most pessimistic predictions about future job losses. An argument along these lines might posit that targeted in-kind transfers (for example, Medicaid and food stamps for the unemployed), along with job retraining, represent more effective uses of resources than a UBI.

This is not to say that technological change is irrelevant to the UBI debate. For example, insofar as technological change supersizes the rewards for certain skills and reduces the demand for manual labor, it may contribute to widening inequality and thereby strengthen the case for redistribution.42 But even if the case for redistribution is undisputed, the case for a UBI still depends upon claims about the scope of redistribution (universal rather than conditional) and the form of redistribution (unrestricted cash transfers rather than restricted transfers or transfers in kind). Accordingly, the coming of automation-induced unemployment is neither a necessary nor a sufficient element of the case for a UBI. For reasons explained at greater length elsewhere, we are skeptical of automation-based arguments for a UBI.43 For the purposes of this Article, we adopt an attitude of agnosticism toward claims about a coming “robot apocalypse,”44 and we note instances in which automation-induced unemployment would be especially relevant to a UBI’s design.

A. The Case for Redistribution

1. Welfarism.

One possible case for a UBI proceeds from welfarism, which is—at least within the academic fields of economics and law—the dominant approach to the normative analysis of taxes and transfers.45 Under welfarism, the proper role of political institutions is to maximize some conception of “welfare,” “utility,” “well-being,” or “happiness.”46 Arguments that redistribution increases overall welfare generally rely on the premise that the marginal utility of income or wealth is declining.47 According to this view, taking a dollar from a rich person decreases her utility less than it increases the utility of the poor person who receives that dollar, thereby increasing overall utility.48

This premise of diminishing marginal utility of income draws support from intuition as well as empirical evidence. Intuitively, it seems almost axiomatic that an extra $500 per person per month would mean a great deal to a family in poverty but would be unnoticeable in the life of Jeff Bezos (who as of this writing was the world’s richest person).49 Empirically, cross-national happiness studies find that additional income leads to a larger increase in self-reported happiness among citizens of poorer nations than among citizens of wealthier ones.50 A similar phenomenon emerges from within-nation studies conducted in the United States: while rich people tend to be happier than their lower-income counterparts, the relationship between income and happiness ebbs as income rises.51 The more money one already has, the less an additional dollar contributes to well-being.

Although welfarists tend to support redistribution,52 welfarism does not translate into total egalitarianism. If, at the extreme, a country imposed a 100 percent tax on income above the median, then presumably no one (or virtually no one) would earn above the median amount, and gross domestic product would decline precipitously. Most welfarists would thus agree that the welfare gains generated by redistributing from low-marginal-utility to high-marginal-utility individuals (that is, from the rich to the poor) must be balanced against the welfare losses generated by the disincentive effects of high income taxes. A rich literature on optimal taxation seeks to identify the mix of taxes and transfers that best balances these two effects.53 We do not seek to reproduce that literature here, except to note that the optimal tax literature will ultimately have much to say about the size of a UBI.

2. Resource egalitarianism.

Resource egalitarianism is an increasingly prominent alternative to welfarism in tax-policy discussions.54 While welfarists generally believe that the ex post transfer of income or wealth from the rich to the poor enhances overall utility, resource egalitarians—in the words of Professor Alstott—“tend to emphasize ex ante equality” in the distribution of wealth and opportunities.55 Resource egalitarians further argue that the financial circumstances of one’s birth are arbitrary and should not limit one’s ability to develop skills and talents. To that end, resource egalitarianism supports ex ante redistribution in order to level the playing field.56

Two key aspects of resource egalitarianism are a belief in “neutrality” and the so-called choice-chance principle.57 “Neutrality,” according to Alstott, entails the view that “society should not endorse any particular way of life or underwrite any particular tastes or vision of the good, because each person is a moral agent whose life plan is worthy of equal respect.”58 The choice-chance principle, according to Alstott, “holds that distributions of society’s resources among individuals ought to reflect individual choices . . . but not ‘pure’ bad luck.”59 Because of the choice-chance principle, resource egalitarianism is often referred to as “luck egalitarianism.”60 As Professor Samuel Scheffler explains:

This theory has different variants, but the central idea is common to all of these variants. The core idea is that inequalities in the advantages that people enjoy are acceptable if they derive from the choices that people have voluntarily made, but that inequalities deriving from unchosen features of people’s circumstances are unjust.61

In theory, this principle could be implemented by dividing initial resources equally among individuals, so that the later distributions reflect individual choice and not the chance circumstances of birth.62 Professors Ackerman and Alstott’s stakeholder proposal reflects this ideal by attempting to limit the role that chance plays in the financial resources available to one upon entering adulthood.63 The choice-chance distinction also can be implemented by redistributing ex post to those in unfortunate circumstances due to luck (but not to those in unfortunate circumstances due to choice). Perhaps A was born into a financially disadvantaged family in a poor neighborhood that lacks decent schools, and B’s business goes bankrupt due to unforeseen political circumstances beyond his control and for which there is no insurance market, while C grows up amidst grand wealth but squanders his inheritance. The choice-chance principle suggests that A and B—but not C—merit redistribution. Here, resource egalitarianism and utilitarianism potentially diverge: inequalities that arise from different individual choices do not justify transfers from the rich to the poor according to the resource egalitarian view, even if those transfers would increase overall utility.

Not surprisingly, one criticism of luck egalitarianism is that distinguishing between choice and chance is a fraught exercise.64 If a smoker contracts lung cancer, is that a function of choice or chance?65 What about an individual who contracts skin cancer after failing to wear long sleeves or apply sunscreen in young adulthood? Critics of luck egalitarianism argue that implementation of the choice-chance principle ultimately requires “moralizing judgments” of individual choices and prioritizes some people’s judgments over others. The result, critics contend, is a profoundly anti-egalitarian regime that ranks individuals as more or less deserving of state assistance based on the subjective beliefs about deservedness held by the majority or by those in power.66

The philosopher Professor Elizabeth Anderson has argued for a version of egalitarianism that, in her view, avoids some of the pitfalls of the choice-chance principle.67 She posits that the “proper . . . aim of egalitarian justice is not to eliminate the impact of brute luck from human affairs” or “to ensure that everyone gets what they morally deserve,” but instead to “end oppression” and “to create a community in which people stand in relations of equality to others.”68 On this view, each individual—regardless of her choices—should have “effective access to enough resources to avoid being oppressed by others and to function as an equal in civil society.”69 This includes access to adequate nutrition, shelter, clothing, medical care, and education, along with a range of other civil, economic, and political rights.70

We will have more to say about the implications of Anderson’s view in Parts II.B and II.C. The key point for present purposes is that egalitarians of all stripes accept (and indeed, embrace) a program of robust redistribution from rich to poor, either ex ante or ex post. The more difficult questions for egalitarians will be what form such redistribution should take (cash vs. in-kind) and whether it should be conditional upon age, work, disability, or other individual characteristics.

3. Libertarianism.

Perhaps surprisingly, some of the most outspoken proponents of a UBI—such as Charles Murray and Milton Friedman—have grounded their support in libertarian theory.71 We say that this is somewhat surprising—or rather, may seem somewhat surprising to many readers—because libertarianism is often thought to be particularly hostile to the first premise underlying the pro-UBI argument: that society should redistribute resources from the rich to the poor.72 Elsewhere, we have argued at length73 that many (but not all) strands of libertarianism can accommodate a system of progressive resource transfers and may even require some amount of redistribution.74 We briefly recapitulate those arguments here.

By libertarianism, we refer to a set of ethical and political theories sharing a belief that individuals have a right to be free from coercion, that free markets with well-defined property rights are preferable to government command-and-control arrangements, and that the results of market interactions presumptively should remain undisturbed.75 Underlying all of these principles, according to philosopher Professor Eric Mack, is a belief in “the separate value of each person’s life and well-being.”76 This principle potentially poses a challenge for libertarians who believe that property holders have a right to protect their holdings by force (or who believe that the state should provide such protection): How can a property holder—or the state—justify the application of force against an individual who, at no fault of his own, finds himself in such dire circumstances that he must steal or trespass to find life-saving food or shelter? Requiring such an individual to sacrifice his or her life to honor the property holder’s rights would be to deny the former’s value as a separate person. Mack therefore concludes that libertarianism mandates some system for ensuring that no member of society finds himself or herself in a position that requires such self-sacrifice. The provision of a minimal income to all members of society serves that purpose.77

Mack’s argument offers one (self-described) libertarian approach to justifying redistribution, but not the only one. A second argument, advanced by philosopher Professor Matt Zwolinski and further developed by the two of us,78 takes as its starting point the “Lockean proviso” at the heart of much of libertarian thought. The Lockean proviso states that individuals can justly acquire property rights by mixing their labor with unowned natural resources so long as “enough, and as good” is left for others.79 The problem for libertarians in a world of scarce natural resources is that private appropriation might not satisfy this condition. When, for example, X claims desirable beachfront property for herself, Y loses access to that same land, and while Y and Z might snap up the adjoining lots, that still leaves A, B, and C out of luck.

How, then, can libertarians who accept the Lockean proviso as an ethical constraint ever justify the appropriation of natural resources? One response is to say that a rising tide lifts all boats—for example, A, B, and C may be rendered better off by X, Y, and Z’s appropriation of beachfront property because X, Y, and Z will build hotels and resorts that then provide A, B, and C with additional opportunities for employment and leisure. But if A can’t find work and can’t afford a room or a meal or a spa treatment at any of the new beachfront properties, then A would seem to have a strong argument that “enough and as good” have not been left for him.

And here is the role for redistribution. By imposing some (perhaps small) tax on X, Y, and Z while redistributing the revenues to A, B, and C, we can potentially ensure that A, B, and C are compensated for their loss of access to the ocean. Whether that tax ought to fall on property, income, or some other base—and how much the tax and the corresponding transfer should be—are questions that lie beyond the scope of this brief discussion. And, of course, embedded within this entire argument are a number of contestable empirical premises (for example, that A, B, and C really are better off with the transfer than with unimpeded access to the beach). The point here is simply to illustrate that in a world of scarce resources, some redistribution may be required to satisfy the Lockean proviso, after which X, Y, and Z can (according to the libertarian view) justly dispose of their property as they wish.

A number of other arguments for redistribution emerge from the branch of libertarianism known as classical liberalism. One such argument, fleshed out by Professors Gerald Gaus and Loren Lomasky, is based on consent: In a just society that treats all individuals as free and equal, any coercion (such as in the form of a state) must be justified.80 A just state is therefore one that each member concludes is better than none. Gaus thus argues that “[i]f the system of property rights is to be publicly justified, it must be the case that everyone has reason to accept it and no one has reason to reject it.”81 He concludes: “Since some people inevitably are left out of the general abundance of modern economies, it seems that the [Lockean] Proviso would require that they be provided with sufficient property so that they too are, manifestly, not disadvantaged by a system of private property rights.”82

A second argument arises from classical liberalism’s acceptance of a role for the state as a producer of public goods.83 Redistribution is plausibly a public good for at least two reasons. One is that redistribution may reduce crime and thus serve a similar function to state-provided police services.84 Another is that poverty—in the words of economist Mark Pauly—“offends the esthetic and moral sensibilities” of individuals who encounter it,85 and thus redistribution to alleviate poverty might be justified on the same basis that classical liberals justify expenditures on street cleaning and public parks.

As discussed below, these various rationales for redistribution will have divergent implications as to the amount of redistribution that might be required or desired. We defer discussion of that issue to Part III.A. Before that, we turn toward the second premise underlying the case for a UBI: that redistribution ought to take the form of unrestricted cash transfers.

B. The Case for Cash Transfers

Assuming that redistribution to the poor is justified, why should that redistribution occur through unrestricted cash transfers? Why not instead transfer benefits in kind, or through vouchers for specific goods and services? As above, we consider these questions from welfarist, egalitarian, and libertarian perspectives, explaining how each approach might justify the cash-transfer aspect of a UBI.

1. Welfarism.

While the welfarist case for redistribution is relatively straightforward, the welfarist claim that redistribution should take the form of unrestricted cash transfers is more tentative. The challenge for welfarist proponents of a UBI is to prove that unrestricted cash transfers increase utility by more than a well-designed and equally expensive program of redistribution in-kind. Intuition and empirical evidence can bring us some of the way toward validating this proposition, but not all of the way.

Start with the intuition. Most of us would prefer to receive one dollar in cash rather than either one dollar’s worth of goods in kind or a one-dollar voucher to purchase particular goods and services. Imagine that you have the choice between a dollar in cash, a one-dollar candy bar, and a one-dollar voucher allowing you to purchase certain kinds of food in a grocery store (for example, fruits and vegetables but not soda). With a dollar in cash, you can choose whichever of the candy bar, fruits and vegetables, and soda that you desire the most. Just as importantly, you can also walk across the street and spend the dollar at a record store if you prefer.86 For this reason, most of us prefer to receive our salaries in monetary form rather than as bundles of goods and services selected by our employers (except perhaps when tax incentives encourage us to choose the bundle of goods and services over the cash).87

Indeed, imagining a world in which individuals are paid in fruit-and-vegetable vouchers rather than cash serves to illustrate the inefficiency of restricted transfers. The individual who prefers soda over fruits and vegetables might use her vouchers to buy bananas, sell her bananas on the secondary market for cash, and spend that cash on soda. While ultimately the individual can buy the soda she desires, this series of exchanges generates unnecessary transaction costs that deplete the value of the transfer. Anecdotal evidence suggests that some SNAP recipients already engage in these sorts of transactions to translate their benefits into cash that they then use to purchase goods and services that SNAP does not cover.88 From a welfarist perspective, these transaction costs are pure waste that society should strive to avoid.

Yet intuition and anecdote get the welfarist only so far. The welfarist presumably cares not only about short-run utility but about long-run utility: Perhaps today we would prefer the dollar in cash so that we can buy a soda, but years from now we might for health reasons prefer that our choice had been restricted to fruits and vegetables. And while some recipients of restricted or in-kind transfers might engage in secondary market transactions to convert their benefits into cash, others will stick with what they are given. Moreover, the welfarist cares not only about the welfare of the redistribution recipient but also the welfare of others. For example, the recipient might choose to spend a dollar on soda, but the welfarist cares as well about the negative externalities from the obesity epidemic. Externality concerns apply with even greater force when the transfer goes to an adult who cares for children.89

At this point, the welfarist argument turns from intuition to empirical evidence. And existing evidence from cash-transfer programs suggests that such transfers increase measurable outcomes associated with welfare in the long term. Studies in domestic contexts suggest that cash transfers are associated with improvements in infant health, increased test scores, additional years of schooling, and a lower rate of arrests among young adults.90 More evidence is available from studies in the developing world, which show that cash transfers lead to reductions in child hunger and increased consumption of vitamin- and protein-rich foods such as fruits, vegetables, and meat.91 Importantly, these studies also suggest that recipients do not increase spending on goods associated with lower welfare, such as alcohol, cigarettes, or gambling.92 Instead, beneficiaries in the developing world generally either make capital investments (such as acquiring livestock or replacing a thatched roof with an iron roof) or increase spending on education, food, and medical care.93

These studies strongly suggest that unrestricted cash transfers can be welfare enhancing. They do not, however, establish that such transfers enhance welfare more than equally expensive programs of restricted transfers enhance welfare. Frustratingly, most UBI pilot experiments also do not include a separate treatment group receiving in-kind or restricted transfers.94 For this reason, the welfarist case for unrestricted cash transfers remains uncertain. These studies and experiments should help assuage skeptics who wonder whether recipients will waste cash transfers on frivolous expenditures, but they do not establish the superiority of cash transfers over other forms of redistribution.

2. Resource egalitarianism.

For resource egalitarians such as Alstott, the argument that redistribution should take the form of unrestricted cash transfers is much more straightforward. The neutrality principle suggests that society should not compel individuals to purchase particular baskets of goods or services; if individuals choose to spend their basic income at, say, liquor stores, gambling establishments, or strip clubs, then who are we to say that their version of the good life is wrong? And since individuals who enjoy the benefit of inherited wealth already have the opportunity to choose their own consumption basket, equality-of-opportunity principles suggest that others ought to have the same option. Otherwise, some individuals enjoy greater consumption choices than others solely due to the luck of being born into wealthier families. To be sure, resource egalitarians might be concerned about negative externalities arising from consumption decisions, but those negative externalities can be addressed through regulations or regulatory taxes that affect all individuals alike—heirs and heiresses as well as redistribution recipients.

The case for cash rather than in-kind transfers is not uniformly accepted by self-described egalitarians, however. Consider Anderson’s critique:

The preference for income rather than in-kind transfers . . . gives no special priority to freedom from disease over the freedom to idle: freedom is freedom. As an account of what we owe to one another, that seems misguided. . . . We owe each other the rights, institutions, social norms, public goods, and private resources that people need to avoid oppression (social exclusion, violence, exploitation, and so forth) and to exercise the capabilities necessary for functioning as equal citizens in a democratic state. From a social point of view, then, we should grant higher priority to securing certain goods, such as education, over others, such as surfing opportunities, even if some individuals prefer surfing to schooling.95

The schooling-versus-surfing distinction is, to our ears, a rhetorically powerful one: surely education is more essential for functioning as an equal citizen than wave riding is. Yet egalitarian proponents of cash transfers can make two compelling counterarguments to Anderson’s critique. First, many of the decisions involved in crafting an in-kind transfer scheme entail much more contestable value judgments than the prioritization of schooling over surfing. For example, should subsidies be available for tuition at an unaccredited vocational training school? If we rule out surfing, what about gym membership? If we allow a specific amount as a housing allowance, then what about an adult who would prefer to crash on a friend’s couch for a couple of months while saving up so she can start a small business or afford college? More generally, many of the personal financial decisions that low-income individuals must make do not have obvious right or wrong answers. The government might say, for example, that tuition subsidies should be available only for attendance at an accredited institution, that gym membership should remain an unsubsidized recreational activity rather than a subsidized component of personal health, and that individuals should not skimp on decent housing in order to save up for other purposes. It is not so clear, though, that these judgments are any “better” (in the sense of facilitating one’s functioning as an equal citizen) than the decisions that individuals would make if left to their own devices.

A second resource-egalitarian response to the Anderson critique of cash transfers is that empowering individuals to make their own personal financial decisions is itself part of what it means to respect them as equal citizens. Let’s say that the individual who chooses to attend the unaccredited beauty school would have been better off financially if he had spent the money on community college courses; even so, giving the individual the ability to make the decision between beauty school and community college allows him to more meaningfully determine his own destiny. Equal citizenship, on this view, entails not only the opportunity to participate in collective governance but also the opportunity to engage in self-governance. Cash transfers facilitate self-governance to an extent that restricted or in-kind transfers cannot.

3. Libertarianism.

The case for cash rather than restricted or in-kind transfers is most straightforward for libertarians. First, libertarians who are skeptical of state capacity will, we think, be especially doubtful that the government can “choose better” than individuals can. Second, and similarly to resource egalitarians, we expect that libertarians will prefer cash-transfer programs that show respect for individual autonomy over in-kind or restricted transfer schemes that seek to dictate an individual’s consumption basket.96 They will react negatively to the paternalism inherent in in-kind transfer programs that dictate how much individuals can allocate to nutrition versus shelter versus other priorities.

This is not to say that the libertarian case for cash rather than in-kind transfers is open and shut. One counterargument—and a counterargument that may appeal to some libertarians—is that in-kind benefits serve as a screening mechanism that distinguishes the truly needy from others. For example, if offered a UBI of $500 per month, individuals of all ability levels might gladly accept. If offered soup kitchen meals and a bed in a homeless shelter that cost $500 per month to provide, an able-bodied adult who is fully capable of finding a job might decline. The lower quality of goods and services provided in kind—combined with the stigma associated with acceptance of such benefits—may reduce take-up among those not in dire straits and, as a result, reduce the cost of the transfer program or leave more resources available for the very neediest.97

We address this screening argument at greater length elsewhere.98 Here, we briefly recapitulate two potential rebuttals. First, the claim that in-kind benefits screen out “non-needy” individuals is highly uncertain. Insofar as the provision of aid in kind rather than in cash makes some individuals more likely to take up benefits than others, there is no guarantee that those who are screened out are the ones we might want to screen out. Soup kitchen meals might screen out the gourmands who have no particular need for state aid, or they might screen out the individuals whose caregiving obligations make them unable to take their meals at a particular time and location. Homeless shelter beds might screen out the surfers whose low incomes are a function of a strong preference for leisure, or they might screen out the assault victims who assign particular value to privacy as a result of past trauma.99 Second, even for libertarians, minimizing the cost of transfer programs is not the only goal. Whatever screening benefits come from in-kind transfers must be weighed against the loss of individual autonomy that comes with them. The as-yet-unproven screening benefits from in-kind transfers would have to be quite powerful to overcome the libertarian case for redistribution in cash.

C. The Case for Unconditional Transfers

The third and final component of the case for a UBI is the claim that redistributive cash transfers should be unconditional—not limited to individuals of a particular age (for example, the young and the elderly), or to individuals with a demonstrable disability, or to individuals who participate in the workforce. This is probably the most controversial component of the case for a UBI and the point on which welfarism and resource egalitarianism are most equivocal in their support. (Libertarians may find the claim less problematic.) Again, we consider each theory in turn, highlighting the ways in which lack of conditionality can be justified and evaluating potential counterarguments.

1. Welfarism.

Conditionality generates winners and losers. The winners are individuals who meet the conditions for aid (for example, by participating in the workforce or demonstrating disability). The losers are, of course, those who do not. The game is arguably zero-sum insofar as there is a fixed pie of resources available for redistribution: more recipients means less for each one. For welfarists, the trade-off between the scope of benefits (that is, reaching more people by removing conditions) and the size of benefits (that is, increasing the transfer per beneficiary) is one of the greatest challenges in the design of a cash-transfer regime.

Conditionality would be more attractive if we knew who was capable of work and who was faultlessly in need. What we do know, by contrast, is that understandings of disability and deservedness are ever evolving. Before the 1960s, for example, few in the United States understood alcoholism to be a disease.100 By the 1990s, alcoholism and drug abuse were considered to be disabilities that potentially qualified an individual for benefits under the Social Security Disability Insurance (DI) and Supplemental Security Income (SSI) programs—though Congress reversed course in 1996 and barred the Social Security Administration (SSA) from considering alcoholism and drug addiction as factors qualifying individuals for benefits.101 Prior to the 1980s, individuals suffering from mental illnesses such as major depressive disorder struggled to qualify for DI and SSI benefits;102 today, more than a third of DI beneficiaries and more than six in ten nonelderly SSI beneficiaries fall into the “mental disorders” diagnostic category.103 Individuals unable to work on account of substance abuse disorders or depression might have been branded as “at fault” for their circumstances in an earlier era. Awareness of our own epistemic limitations should make us think twice before setting conditions that exclude individuals from aid today; it may be our own lack of knowledge—or lack of sympathy—that causes us to conclude that they are to blame for their circumstances.

Aside from the problem of defining disability in the abstract, any conditional cash-transfer program that denies benefits to able-bodied nonworkers encounters the additional challenge of applying the definition of disability in individual cases. Internal and external audits of SSA disability determinations find high error rates, with studies suggesting that anywhere from 20 percent to 60 percent of disability benefit rejections are “false negatives.”104 Whatever the precise figure, welfarists must weigh the inevitability of error in a conditional transfer program against the putative upside of limiting aid to those who are “truly” in need.

The same welfarist logic that justifies redistribution in the first place would suggest that these error costs are asymmetric: providing nothing to disabled individuals who are erroneously classified as “work capable” is worse than providing somewhat less to disabled individuals who are correctly classified as such. This, again, is a function of the diminishing marginal utility of income: the first dollar matters more than the thousandth and the millionth. The welfarist who is worried about false negatives in disability determinations might well conclude that ensuring a subsistence-level income for all individuals is more important than trying to channel somewhat more in benefits to a subset of disabled individuals by excluding others from aid.

The trade-off between scope and size is not the only factor that a welfarist might consider when wrestling with the conditionality question. Proponents of making cash transfers conditional upon workforce participation or a demonstration of disability also worry about the effect of a UBI on labor supply. A UBI—so the argument goes—may lead to widespread exit from the workforce and thus to a reduction in total resources and overall welfare.105

There are two ways in which a UBI may affect labor supply: an “income effect” and a “substitution effect.” The term “income effect” refers to the effect on labor supply of making someone richer or poorer. If an individual becomes richer and therefore can meet more of her material needs or desires, then she may allocate more time to leisure; conversely, if she becomes poorer, then she may allocate more time to labor. The term “substitution effect” refers to the effect on labor supply of changing the terms of the labor-leisure trade-off. Labor becomes more attractive relative to leisure when after-tax wage rates rise (and, conversely, less attractive when after-tax wage rates fall).

A UBI financed through income or consumption taxes will increase the income of individuals at the low end of the income ladder (who will receive more in benefits than they pay in additional taxes) and will reduce the income of individuals at the high end (who will pay more in additional taxes than they get back). We would expect the income effect on labor supply to be negative for low-income individuals (who now have more income) and positive for high-income individuals (who now have less). The substitution effect on labor supply of higher tax rates is negative across the board.

Past experiences with cash-transfer programs shed some light on the magnitude of these effects. One well-examined program involved members of the Eastern Band of Cherokee Indians in western North Carolina who began to receive annual disbursements of casino profits starting in 1996.106 The disbursements amounted to approximately $4,000 per year for each adult tribal member. A study comparing households with children that did and did not receive the casino-profits payments found no economically or statistically significant effect on labor supply for mothers or fathers.107 Another program—the Alaska Permanent Fund Dividend—has used the state’s substantial oil revenues to pay approximately $2,000 per resident per year since 1982.108 A study comparing Alaska to other states with roughly similar demographics and industry compositions again found that introduction of the dividend had no significant effect on overall employment, labor-force participation, or hours worked.109 Note, though, that in both cases the cash-transfer program would have affected labor supply only through the income effect—there was no substitution effect, as tax rates (and after-tax wage rates) did not change. An inflow of casino profits or oil revenues is about as close as one can get to money dropping like manna from the heavens. The Eastern Cherokee and Alaska natural experiments thus suggest that the income effects on labor supply of a cash-transfer program in the range of $2,000 to $4,000 per person are small or nonexistent, though they tell us nothing about the substitution effect.

Studying the substitution effect of a UBI is somewhat more complicated. Much is likely to depend on the rate of the tax (or tax plus phaseout) used to fund a UBI—the higher the rate, the larger the labor-supply effect. Much will also hinge on whether the UBI supplements or replaces existing cash-transfer programs. Some of those programs entail very high implicit tax rates. For example, a nonblind individual who earns more than $1,220 per month in 2019 will lose DI benefits, which can mean a loss of up to $771 per month for a single individual.110 The marginal tax rate on the 1,221st dollar is thus as high as 77,100 percent. Replacing DI with a UBI and a less cliff-like marginal tax rate schedule may increase labor supply among former DI beneficiaries.

Finally, it is not obvious whether the labor-supply effects of a UBI—even if negative—should be considered a bug of a basic income or a feature. The aim of a “sufficient social income,” wrote the late Austrian-French philosopher André Gorz, “is not to force the recipients to accept any kind of work on any terms whatsoever, but to free them from the constraints of the labor market.”111 A UBI, Gorz emphasizes, potentially will allow individuals to opt out of low-paying and otherwise unrewarding market labor and instead to devote their time to their families, their communities, or to other ennobling pursuits. While Gorz does not frame his argument in welfarist terms, the welfarist could make a similar point: The objective of welfarism is to maximize well-being, not to maximize output. Some amount of labor-market leakage may be tolerable and even desirable if it allows individuals to live more fulfilling lives.

Work requirements for able-bodied adults are not the only sorts of conditions that might be attached to a cash-transfer program. Conditional cash transfer (CCT) programs in low- and middle-income countries often require that beneficiaries—as a string attached to aid—ensure that children attend school, receive medical and nutrition examinations, maintain vaccinations, or satisfy other education and health requirements. While several of these studies have yielded encouraging results,112 relatively few studies have compared CCTs to unconditional cash transfers. Thus, it is difficult to know which aspect of the CCT program—the conditionality or the cash—is driving positive outcomes. In the few studies that compare conditional and unconditional cash-transfer arrangements, results are ambiguous. For example, one study targeting families with adolescent girls in Malawi compared cash transfers conditional upon school attendance with unconditional cash transfers. Girls whose families received any cash transfer (conditional or unconditional) were less likely to drop out of school than girls whose families received no transfer, and the effect was strongest in the conditional-cash-transfer arm of the study, but only girls whose families received an unconditional cash transfer saw a reduction in teenage pregnancy.113 Comparisons of conditional and unconditional cash-transfer programs in other low-income countries have arrived at similarly mixed results.114 Not only are the lessons from low-income countries uncertain, but those lessons also are difficult to generalize to higher-income countries such as the United States where baseline rates of school attendance are higher and teenage pregnancy rates are lower.115

Unfortunately, ongoing and proposed UBI experiments will do little to adjudicate the debate between unconditional and conditional cash transfers. For example, the Stockton program—$500 a month to several dozen families with no strings attached116 —will not tell us whether total welfare would be higher if instead those several dozen families received $500 with attached strings (for example, a work requirement or a requirement that children remain in school or receive health and nutritional examinations). Y Combinator’s proposal similarly includes no comparison group subject to a cash-transfer arrangement conditional upon work or other criteria.117 The “Family Rewards” experiment in New York and Memphis, billed as “the first test of a comprehensive CCT in the United States,”118 likewise included no unconditional cash-transfer comparator.119 Welfarists might rightly wish for further data on the relative performance of conditional and unconditional cash-transfer programs before embracing one or the other wholeheartedly. In the meantime, though, there is no compelling evidence that cash transfers conditional upon educational and health criteria consistently yield larger welfare improvements than equivalent unconditional cash transfers.120 In the absence of such evidence, the fact that conditional cash transfers entail higher administrative costs supplies an argument—though a tentative one—for favoring unconditional programs on welfarist grounds.

2. Resource egalitarianism.

Resource egalitarianism generates a strong case for unconditional rather than conditional cash transfers, though counterarguments rooted in egalitarian principles remain. Recall that resource egalitarians generally believe that society should strive to eliminate inequalities due to chance and then should adopt an attitude of neutrality toward individuals’ free choices. Work requirements and other behavior-dependent conditions on the receipt of cash transfers run up against this neutrality principle. If A and B are both initially entitled to a cash transfer so as to equalize the initial distribution of society’s resources, but then A chooses to work while B chooses not to, why should B thereby lose her entitlement to the transfer? According to Alstott, “voluntary unemployment in a well-functioning market is not a matter for concern” for resource egalitarians so long as “[e]very individual is free to decide whether to work for the wage she commands, or to use her time elsewhere if the wage seems too low.”121 And if the individual chooses to use her time elsewhere, why should society strive to penalize her for that decision?

One possible answer comes from resource egalitarians who believe that disabilities should be taken into account. As Anderson writes, “disabled people typically require more resources to achieve equivalent freedoms—to move around, to get access to information, and so forth—than those who are not disabled.”122 Most luck egalitarians similarly account for disabilities on the theory that such individuals require more resources in order to have the same life choices open to them as others.123 And holding total cost constant, a disabled individual would receive a larger benefit under a cash-transfer regime that is conditional upon a demonstration of disability than under an unconditional cash-transfer program. This reasoning might suggest that cash transfers should be available only to those who can demonstrate disability or who for other reasons require more resources. But as for the latter, Anderson notes that “[p]eople who engage in unpaid dependent care work also require more resources to achieve equivalent freedoms.”124 Thus, even the egalitarian argument for conditionality would not seem to justify a requirement that recipients be engaged in formal-economy labor.

There are at least two responses that a resource egalitarian committed to unconditional cash transfers might make to those who would account for disabilities—one a counterargument, the other a concession. The counterargument is that denying benefits to work-capable dependent-less adults is defensible only if society is itself capable of identifying disability. The fact that federal law still requires the SSA to deny that addiction is a disability should call into question our faith in our country’s capacity to implement a system of accurate and compassionate disability determinations.125 This skepticism is deepened by the fact that various federal agencies and courts cannot arrive at a consensus as to which medical professionals are qualified to diagnose disability.126 The concession is that the case for a UBI need not imply the elimination of additional benefits for disabled individuals and/or individuals who care for dependents. As we discuss in Part III.F, the combined cost of DI and SSI programs is sufficiently low in comparison to the total potential cost of a national UBI that maintaining or eliminating those programs will not dramatically affect a UBI’s financial feasibility.127

3. Libertarianism.

The case for unconditional rather than conditional cash transfers is perhaps easiest to make on libertarian grounds. First, cash transfers show respect for individual autonomy, whereas in-kind or restricted transfers that seek to dictate consumption decisions limit individuals’ ability to control important aspects of their own lives.128 In-kind and restricted cash transfers exemplify the sort of paternalistic government-imposed value judgments that libertarians abhor, just as resource egalitarians generally eschew such judgments under the neutrality principle. Second, while most libertarian arguments would in theory render transfers conditional upon a demonstration of work or inability to work,129 the difficulties in implementing such a condition run up against the deep-seated libertarian skepticism regarding the state’s ability to carry out social programs. Do we want armies of government bureaucrats deciding who is and isn’t “deserving” of a cash payment? For some of us, this scenario is unexceptional (and more or less describes the current DI and SSI programs). For committed libertarians, though, this intrusiveness is anathema. Likewise, the inevitability of error in these determinations means that a conditional safety net would not be tight enough to absolutize and legitimize property rights. In order to ensure that a UBI is therefore serving the purpose that libertarian theories envision, it will probably have to be unconditional.130

As a general matter, we expect that libertarians will be amenable to the claim that if transfers are to occur at all, they should be unconditional and unrestricted—and more skeptical of the claim that any redistribution (regardless of form) is consistent with libertarian principles. In this respect, the libertarian case for a UBI is the opposite of the welfarist case for a UBI: whereas welfarists are generally comfortable with the notion of redistribution but less convinced that such redistribution should come in the form of unconditional and unrestricted transfers, libertarians are generally more skeptical of redistribution but easily convinced that—if there is to be redistribution—it ought to be UBI-like in character.

III. Building Blocks for a UBI

Having highlighted why one might pursue a UBI, we now turn to the heart of this Article: How should policymakers design such a program? More specifically, we identify and scrutinize the design decisions that determine a UBI’s ultimate shape: size, eligibility, uniformity, assignability, payment mechanism, and funding mechanism. Our approach is informed both by the philosophical literature on the foundations for a UBI and by research in economics and law on the design of tax-and-transfer policies.

A. Size: $500 a Month as a Starting Point

In discussions about the size of a UBI in the United States, two focal points have emerged: $500 per month (the figure suggested by Facebook cofounder Chris Hughes131 and used in the Stockton experiment132 ) and $1,000 per month (the figure proposed by a Chicago mayoral task force,133 labor leader Andy Stern,134 entrepreneur and former 2020 presidential candidate Andrew Yang,135 and basic-income activist Scott Santens,136 among others).137 The $500-a-month figure happens to correspond, almost precisely, with the US Census Bureau’s threshold for “deep poverty” for a single individual.138 The Census Bureau estimates that more than 18 million Americans—including nearly 6 million children—currently live in “deep poverty”;139 a $500-a-month UBI would bring deep poverty (almost)140 to an end. The $1,000-a-month figure corresponds almost precisely with the Census Bureau’s overall poverty threshold for single individuals; more than 40 million Americans—nearly an eighth of the country’s population—lives below the poverty threshold today.141

There is, to be sure, nothing magical about the deep poverty and poverty thresholds. The poverty level is set at three times the cost of a “minimum food diet,” updated to reflect inflation since 1963,142 and the deep poverty level is simply half that. Those figures do not account for geographic differences in the cost of living or for person-to-person differences in health-care and dependent-care expenses.143 A full-time college student with cash income of less than $12,000 per year might live quite comfortably and yet technically be below the poverty level. A single adult who pays out of pocket to maintain an ailing parent at a long-term care facility may fall well above the poverty level and yet still live a hand-to-mouth existence.

Calculations of poverty status also account for only some income sources. The definition of money income used to calculate poverty status includes Social Security’s Old-Age and Survivors Insurance (OASI), DI, and SSI benefits, workers’ compensation, and unemployment compensation, but excludes noncash benefits such as SNAP and housing subsidies as well as refundable tax credits such as the EITC and the child tax credit.144 Replacing noncash benefits and refundable tax credits with a cash UBI of the same value would make it appear as though millions of Americans had been lifted out of poverty without necessarily improving their material well-being.145 Consider a single individual with no dependents and no cash income. The maximum Section 8 voucher for a single person in New York City as of January 2018 is $1,590,146 and the maximum monthly SNAP allotment for a one-person household in the lower forty-eight states for the October 2017–September 2018 year is $192.147 Replacing Section 8 and SNAP with a $500 per month UBI might make it look like this individual is no longer in deep poverty, because his countable income rises from $0 to $500. But if he loses Section 8 and SNAP benefits along the way, then this person might be substantially worse off than before.

While we therefore think it is important not to accord talismanic significance to poverty thresholds, we believe that a basic income of $500 per person per month—approximately the threshold for deep poverty for a single individual—is a sensible starting point for discussions of a UBI.148 We should note that since our proposed UBI would include children, our $500-per-person-per-month suggestion would yield the same household benefit for a single parent with one child, or for a two-parent family with two children, as proposals by Murray, Yang, Stern, and others,149 who would set a UBI at approximately $1,000 per person per month but limit eligibility to adults. Our proposal, however, better protects single-parent families with children against the possibility that a UBI would leave them worse off compared to the current system. We discuss the inclusion of children at further length in Part III.B.

Why $500? First, a UBI of $500 per person per month would make a measurable difference in the lives of millions of Americans. Indeed, studies in the United States have shown that unconditional cash transfers of even less than that amount can yield significant effects. A study of the Eastern Cherokee cash-transfer arrangement noted above found that an annual transfer of $4,000 per adult ($333 per month) was associated with an extra year of educational attainment by age twenty-one for children in families receiving the transfer as well as a reduction in crime rates.150 A study of the Alaska Permanent Fund Dividend—again, approximately $2,000 per year—substantially reduced the likelihood that mothers would give birth to low-birth-weight babies.151

Second, a basic income of $500 per person per month—$6,000 per year—would be financially feasible given the current size of the US economy. In a country with approximately 314 million citizens and lawful permanent residents and a gross domestic product of approximately $20.49 trillion,152 a basic income of $500 per person per month translates to a cost of roughly 9 percent of GDP, or about 7 percent after the various expenditure offsets discussed in Part III.F are taken into account.153 To put that in perspective, US general government spending as a percentage of GDP is now around 38 percent—or about 11 percentage points below the level in Norway and 12 percentage points below the level in Sweden.154 We could thus afford a basic income of $500 per person per month while keeping our government spending-to-GDP ratio below Nordic levels. By contrast, a basic income of $1,000 per person per month would vault us almost to the top of the government spending-to-GDP ratio rankings, barely behind Finland (57 percent) and France (57 percent), and ahead of Denmark (55 percent).155 In the medium term, a basic income of $1,000 per person per month would be difficult to sustain unless the United States were willing to become one of the most heavily taxed nations, if not the most heavily taxed nation, in the world.

Third, setting a basic income at $500 per person per month (or thereabouts) should allay the concerns of those who worry that a UBI will lead to a mass exodus from the workforce. It strains credulity to think that a substantial number of working-age American adults will leave their jobs to live on an income roughly equivalent to a $3-an-hour full-time wage.156 While this observation might seem to be in tension with the first, it is not. According to a Federal Reserve survey, four in ten US adults do not have the cash to cover a $400 emergency expense without borrowing from friends or family or running up a credit card balance.157 For them, $500 per month would not mean a life of leisure, but it could make the difference between, say, repairing a broken-down car and losing a job due to lack of transportation.

But while a basic income of $500 per month should not stoke the fears of those who think that a UBI might lead to mass indolence, it also will not live up to the hopes of those who see a UBI as providing a safety net in an automation age.158 Average weekly earnings for US manufacturing workers as of February 2018 were about $900;159 a basic income of $500 per month will only slightly soften the blow if that manufacturing worker is replaced by a robot. Again, we hazard no prediction as to whether wide-scale technologically induced unemployment is on the horizon.160 We do believe, though, that the size of a basic income is a question best approached with a clear-eyed view of a UBI’s possibilities and its limits.

Ultimately, the sizing of a UBI entails a number of difficult trade-offs and value judgments. These include the trade-off between expanding unconditional cash assistance and maintaining existing welfare state programs; the trade-off between more redistribution of wealth and greater deadweight loss from explicit or implicit income-based taxes; the trade-off between the interests of the present and those of future generations that will bear the burden of deficit spending; and value judgments about how much redistribution is ethically justified. We suggest $500 per person per month as a starting point but not as an end to the discussion of amount. The size of any UBI inevitably will depend upon both the fiscal capacity and political will of the relevant jurisdiction; as fiscal capacity and political support (hopefully) grow, the UBI may increase in tandem. Size could be—and likely would be—adjusted on an ongoing basis, whereas other building blocks of a UBI may prove more difficult to refashion year to year. Accordingly, we shift our attention to those other elements while acknowledging that size will remain a significant decision point and source of disagreement.

B. Eligibility: All Citizens and Lawful Permanent Residents Should Qualify

Aside from size, a key aspect of any UBI is a specification of eligibility criteria. Just how “universal” should a universal basic income be? Proposals differ in their eligibility criteria in at least four respects: (i) whether they require workforce participation of some form; (ii) whether they apply to the elderly; (iii) whether they apply to children; and (iv) whether they apply to noncitizens. As explained below, we believe that (i) no work requirements should be imposed; (ii) seniors should be eligible; (iii) children should be eligible; and (iv) a UBI should extend to lawful permanent residents in addition to US citizens.

Start with workforce participation. President Richard Nixon’s Family Assistance Plan (FAP) would have required all able-bodied adults to pass a “work test” in order to receive cash aid. A woman whose youngest child was under the age of six could have satisfied that test by caring for the child. Most other adults would have been required—as a condition for receiving the FAP—to register with a state employment agency and to accept training or employment opportunities that the agency deemed suitable.161 The FAP also would have been limited to adults with children under eighteen (or under twenty-one if the child was in school); childless adults and adults with grown children would have been excluded entirely.162

Most other UBI proposals define eligibility to include all working-age adults,163 but some do not. Chris Hughes’s proposal—which he acknowledges is not a truly “universal” UBI—tracks the FAP in significant respects: adults (of any gender) would be eligible if they cared for a child under age six, but others would have to work, attend school, or enroll in vocational training.164 Dylan Matthews, a writer for the online publication Vox and a prominent voice in UBI debates, has suggested linking a UBI to participation in national service: individuals who serve in the military or participate in AmeriCorps- or Peace Corps–like programs for one to two years would become eligible for a lifetime UBI.165 An advantage of a basic income program along the lines suggested by Hughes and Matthews is that it might tap into widely shared views regarding deservingness: a basic income would not simply amount to getting paid for doing nothing. The disadvantages of work-linked transfer programs are detailed above—including the economic distortions stemming from subsidies, the administrative costs associated with disability determinations and work verification, and the autonomy objections rooted in resource egalitarian and libertarian thought.

Regardless of how a UBI treats working-age adults (and we think that any program that calls itself “universal” must include generous eligibility criteria), designers also must consider how to treat the old and the young. The existence of Social Security’s OASI program complicates the treatment of the elderly. In light of OASI, we can anticipate at least two potential arguments for excluding senior citizens—one principled and one pragmatic. The principled argument is that a UBI would largely be duplicative of the basic income that OASI already provides: allowing seniors to continue receiving OASI while phasing out other welfare programs unfairly advantages seniors over other current beneficiaries of transfer programs. The pragmatic argument is that excluding the 52.7 million Americans who receive OASI benefits from a UBI would reduce the cost of a new basic income program significantly.166 Notably, senior citizens are excluded from the existing cash-transfer program most similar to a UBI—the federal EITC—unless they have a qualifying child in their household,167 presumably because policymakers view the EITC to be duplicative of OASI.

We find the arguments for automatically excluding all senior citizens from a UBI to be unpersuasive. Social Security has not brought the problem of elder poverty to an end. The sixty-five-and-older demographic has the lowest poverty rate of any age group, but still, 9.3 percent of senior citizens fell below the poverty line in 2016.168 And while the average monthly Social Security benefit for retired workers and their dependents as of December 2017—$1,404—would put an individual above the poverty line,169 some beneficiaries receive substantially less than that, and approximately 4 percent of senior citizens will never receive Social Security benefits.170 Excluding senior citizens whose OASI benefits are less than the UBI amount would seem arbitrary and would be difficult to justify.

One potential solution—suggested by Murray—is to give senior citizens an either-or choice between OASI benefits and a UBI.171 While that approach would surely cut costs, it also could leave some seniors substantially worse off than under the status quo. The most recent data indicates that approximately 4.8 million individuals over age sixty-five receive SNAP benefits;172 about 2.4 million receive SSI benefits;173 and approximately 500,000 live in Section 8–subsidized units.174 Eliminating these programs and replacing them with a UBI that many low-income senior citizens choose not to take because it is less than their OASI benefit could leave some seniors worse off than under the status quo.175 Moreover, requiring senior citizens to choose between OASI and a UBI would effectively impose a much higher marginal tax rate on OASI benefits than other income sources—an outcome that will grate at seniors who see OASI benefits as something that they have “paid for” already.176 We suspect that many of those same individuals would view a choice between OASI benefits and a UBI as an unfair choice between something that they have earned and something that others receive automatically.

Another option would be to include senior citizens within the UBI while also ending the tax-favored treatment of Social Security benefits.177 Those two changes are nowhere near offsetting: extending a $500-a-month UBI to 52.7 million OASI recipients would cost about $316 billion a year, while ending the tax-favored treatment of Social Security benefits would save only around $38 billion a year.178 The combination of changes does, however, neutralize the objection to treating OAS benefits differently from other income streams that might arise if senior citizens were put to an either-or choice.

The treatment of children also has proven to be controversial among UBI proponents. To be sure, the notion that children should be beneficiaries of redistribution should be shared by almost all welfarists and resource egalitarians, and most libertarian justifications for redistribution apply to children as well.179 Reflecting these notions, children have historically been viewed as among the most deserving recipients of government aid, and most current social programs favor families with children. The minimal aid to childless adults provided by the EITC180 —arguably the most UBI-like program in the United States—exemplifies this judgment. Moreover, as one of us has noted elsewhere, a UBI that excluded children and was funded through the elimination of existing cash and near-cash transfers would potentially make the problem of child poverty even graver than it already is.181

What then, are the arguments against including children? Murray has argued against extending a UBI to children on the grounds that doing so would amount to a fertility incentive for low-income adults.182 The empirical literature on the responsiveness of childbearing decisions to fertility incentives yields mixed results,183 but even accepting the empirical premise that extending a UBI to children would encourage low-income parents to have more kids, it is not clear that we should consider that to be undesirable. For example, resource egalitarians might argue that adults with substantial material resources on account of inheritance or luck already have the luxury of choosing how many children to have without worrying overly about resource constraints; why should the same not be extended to adults with less income or wealth?184 Welfarists, for their part, have been engaged in a robust decades-long debate about optimal population size—a debate that we relegate to the margins but that will shape the views of some as to whether fertility incentives embedded in a UBI are desirable or not.185

Aside from fertility considerations, others have argued against extending a UBI to children on administrative grounds. For example, Stern writes: “[J]ust think of all the issues that come up when you try to get your mind around giving [a basic income] to a ten-year-old.”186 We find this argument puzzling. Just as the child tax credit is paid to the child’s parent rather than to the child, we anticipate that the UBI would go to the child’s guardian. Granted, in some cases, there will be questions about which adult should qualify as a particular child’s guardian, though such questions already arise in the child tax credit and EITC contexts.187 While in no sense have the child tax credit and the EITC fully resolved the question of whose child is whose, neither has the challenge of matching children to parents paralyzed either program.

Because we find the arguments for including children to be much more persuasive than the arguments against, our view is that children should be included within a UBI. At that point, the tax system’s labyrinthine set of supports for families with children (the child tax credit,188 the child and dependent care credit,189 the exclusion for dependent care assistance programs,190 and the implicit subsidy for single parents embedded in the head-of-household filing status191 ) all could be laid to rest. This is one more way in which a UBI could be used to achieve substantial simplification of benefits.192

A final—and perhaps the most controversial—eligibility question concerns the treatment of noncitizens. The welfarist argument for redistribution would seem to apply regardless of whether the transferee is a citizen: there is little reason to believe that the marginal utility of income is higher for US citizens than for lawful permanent residents (Green Card holders) or other immigrants at equivalent income levels. The resource egalitarian argument for equality of opportunity also brooks no obvious distinction on the basis of nationality. And many of the libertarian arguments for a UBI would seem to apply with similar force to noncitizens. Immigrants, too, are separate persons for purposes of Mack’s argument; the Lockean proviso is not nationality specific; and the public-good benefits of poverty alleviation (for example, alleviation of the aesthetic and moral distress that poverty brings to the rest of society) would not seem to depend on whether the poor are citizens or not.193

Even so, there are pragmatic (whether or not persuasive) arguments for limiting a UBI’s scope. First, making the UBI available immediately to new arrivals—regardless of their legal status—might encourage illegal immigration. Second, extending the UBI to immigrants might make it more politically difficult to lift restrictions on immigration, because now the fiscal cost of allowing new immigrants to the country would be higher. Third, making the UBI available to immigrants might have effects on who chooses to come here. Perhaps for these reasons, some cash assistance programs (for example, SSI) and near-cash aid programs (for example, SNAP) already exclude most new adult arrivals.194

A reasonable compromise, we think, is to extend a UBI to lawful permanent residents but to exclude other noncitizens. Once the country has concluded that a person may remain here permanently, we see little reason to exclude that person from our community of concern. And because a lawful permanent resident must—by definition—come by that status lawfully, it seems unlikely that the extension of a UBI to lawful permanent residents would encourage individuals to cross the border illegally or overstay a visa. The exclusion of undocumented immigrants—including “Dreamers”—from a UBI may strike many readers as harsh, though we note that such individuals already cannot qualify for SNAP, Medicaid, DI, SSI, and many other cash and near-cash benefits.195 We think these concerns are best addressed through reforms to the immigration laws, though we acknowledge that this outcome is easier said than legislatively done.196

C. Uniformity: Equal Benefits for All

The claim that every citizen and lawful permanent resident should receive a UBI does not necessarily imply that all should receive the same amount. Here, we consider whether benefits should be uniform or whether they should vary on the basis of household size, assets, and geography.

1. A UBI should be implemented on a per-person rather than a per-household basis.

Some UBI proposals contemplate that each individual will receive the same amount; others envision an adjustment for household size (that is, a two-person household would receive less than double the amount that a single person receives, and a three-person household would receive less than 1.5 times the amount that a two-person household receives).197 While there is a plausible argument in favor of adjusting the UBI on the basis of household size, we ultimately conclude that the case for a per-person UBI is more compelling.

The nub of the argument for a household size adjustment is that household size affects ability to pay. Having a roommate—or a spouse—generally reduces per-person housing costs. The bill for groceries to feed two people is generally less than double the bill for one. And so on. The square-root scale often used in studies of income inequality and poverty implicitly assumes that the cost of living for two people is 1.41 times the cost of living for a single person (that is, doubling up saves approximately 30 percent), and that the cost of living for four people is twice the cost of living for a single person (that is, “quadrupling up” saves 50 percent).198

The arguments against a household size adjustment are several. Consider a UBI in which individuals living alone receive a $12,000 UBI while married or cohabitating individuals receive $8,000 each ($16,000 per couple).199 This structure creates an $8,000 penalty for childless adults who wed.200 Welfarists might emphasize that household size adjustments distort decisions to marry and cohabitate. Resource egalitarians might argue that the neutrality principle suggests that society should not reward or penalize individuals for household formation choices. Libertarians, too, might value the fact that a UBI is invariant to household size: a per-person rather than per-household UBI does less to intrude upon core decisions in an individual’s life.201 Readers with a range of philosophical commitments also may appreciate the independence that an individual UBI grants to household members; for example, a per-person UBI delivered as a biweekly check to each eligible adult may make it easier for a spouse to leave an abusive relationship.202

Another argument against a downward adjustment for individuals living in larger households—one that sounds in a welfarist register—builds on an important 1996 paper by Professor Louis Kaplow.203 Kaplow’s insight is that larger households likely enjoy economies of scale, and so may be better off than smaller households with the same per-person income, but that this does not necessarily mean that the marginal utility of income is lower for larger households because larger households have a greater ability to share resources.204 To illustrate: A two-person household likely needs only one microwave oven, one refrigerator, one washing machine, one Roku box, and so on. Thus, A and B, who each earn $50,000 and live together, may be better off than C, who earns $50,000 and lives alone, because A and B can share their microwave, refrigerator, washing machine, Roku box, etc., and spend more on other items. And we generally think that better-off individuals have a lower marginal utility of income. Yet if A and B buy a new microwave, refrigerator, washing machine, or Roku box, both of them can use it, whereas if C buys a household item, only C benefits. Thus, it is not clear whether social welfare is increased by transferring $30 (roughly the cost of a Roku box) from A and B to C. C arguably “needs” the money more, but A and B arguably can make better use of it.

Administrative and compliance concerns also weigh against household-size adjustments. Living arrangements among low-income individuals are often transitory, with roommates, romantic partners, and relatives moving in and out frequently yet irregularly. Constantly adjusting grant size to reflect these changes would impose a significant burden on individuals and on whatever agency implements the program.205 And beyond the administrative and compliance costs, a UBI dependent on fluctuating household size would hamper the ability of low-income individuals to plan for the future. Indeed, the fact that a UBI does not waver in size is one key advantage of a UBI over existing welfare programs that deliver benefits less reliably.

Finally, insofar as policymakers decide to make the tax-and-transfer system dependent upon household size, we think it is much easier to accomplish that objective through tax adjustments than through transfer adjustments. The IRS can (and does) collect information on household size once a year, and the tax system adjusts for household size in a number of ways (for example, filing status). Implementing these adjustments through changes to biweekly or monthly UBI payments strikes us as an unnecessarily complicated and costly endeavor.

2. A UBI should not depend on assets.

A UBI should not, in our view, depend on an asset test. This recommendation may surprise some. If the goal of a UBI is to redistribute from haves to have-nots, then an asset test initially appears to make sense.206 If A and B each have incomes of $3,000 a year, but A has $10,000 sitting in the bank while B has no savings, then A seems better able to fend for himself than B, and arguably should merit less assistance.

Several considerations counsel against conditioning a UBI on assets, however. One is administrative: while we already have a well-developed third-party reporting apparatus for measuring individual and household income, our administrative technologies for appraising individual and household wealth are much less sophisticated. A second is liquidity related: illiquid assets may do relatively little to improve an individual’s ability to pay in the short term. Both the administrative and liquidity concerns can be allayed through an asset test that depends solely on easy-to-value liquid assets (for example, cash, checking and savings accounts, and stocks and bonds traded on active markets), but that, in turn, would incentivize individuals to transmute their wealth into less liquid forms (for example, life insurance policies and gold bars). Anecdotally, some welfare recipients report buying more household durables such as furniture (which is not counted in asset tests for most existing welfare programs) for precisely this purpose.207 Some low-income individuals who wish to save might also eschew banks and instead simply stockpile cash if they knew that bank account balances would count against them for UBI purposes. This incentive would only magnify the burdens that lower-income individuals already face in obtaining traditional financial services.

An asset test also would distort the choice between present-period consumption and saving for the future.208 Such a test would amount to a new tax on capital income, likely at a regressive marginal rate (because higher-net-worth individuals who are well above the asset test threshold would face no marginal tax as a result).209 This strikes us as entirely backwards. Society arguably has an interest in encouraging savings among low-income households so that they can weather shocks such as a car breakdown or an unexpected medical bill.210 We see absolutely no reason, by contrast, why society should effectively discourage low-income households from accumulating savings through what amounts to a regressive wealth tax.

A final consideration is that asset levels often fluctuate dramatically for the poorest individuals and households. The amount in an individual’s checking account might increase temporarily on account of a prior landlord returning a security deposit, a tax refund, a one-time bonus, student loans, or a range of other possible events. Low-income individuals and households may spend these funds quickly to repay debts or purchase durables.211 Looking at temporary infusions of cash could lead to a reduction in an individual’s UBI unrelated to any actual change in ability to pay. Again, this “churn” in eligibility may interfere with the ability of individuals to plan their lives.212

Our view on an asset test for a UBI should not be interpreted as a verdict on capital income and wealth taxes more generally. While we think that there are strong arguments in favor of a progressive consumption tax that effectively exempts capital income and imposes no tax on wealth, our case against an asset test does not depend upon that premise.213 The most straightforward way to tax capital income or wealth would be through the tax system—not through an administratively messy mechanism that has the effect of taxing wealth at regressive rates.

3. A UBI should not vary by location.

A third dimension along which the size of a UBI might vary is geography: Should a UBI be adjusted for geographic differences in the cost of living? Our view is that the size of a UBI should be uniform regardless of where in the United States one lives, and we think that this view can be defended on welfarist, resource egalitarian, and libertarian grounds. It also comports with the overall approach throughout most of the Internal Revenue Code, which in general is geographically neutral.214 Federal policy is, however, somewhat inconsistent on this front: certain federal benefits—such as the Affordable Care Act premium tax credit215 and Section 8 housing choice vouchers216 —vary by location, as does the federal government’s pay scale.217

The welfarist argument in favor of geographic adjustments is that location is an indicator of need. For example, according to research by scholars at MIT, a single adult in San Francisco needs more than $42,000218 to cover basic expenses such as food, medical care, housing, transportation, and taxes, whereas a single adult in Illinois’s Cook County (which encompasses Chicago) can scrape by on just over $28,000.219 The higher cost of living in San Francisco might on first glance seem to justify a larger UBI for individuals living there.

The welfarist response is at least twofold. First, where one lives is at least to some extent a matter of choice: the individual who chooses to live in San Francisco may value mild winters and views of the Golden Gate Bridge more than the individual who chooses to live in Chicago. Insofar as location is a matter of choice, then the individual who lives in San Francisco is not necessarily needier than the Chicagoan but rather has different consumption priorities. To be sure, choice of location is also constrained by other factors. The average American lives only eighteen miles from her mother—suggesting that location is importantly a function of birthplace and family ties220 —and interstate migration has been on the decline for several decades.221 A UBI, though, would plausibly lower barriers to migration by enabling more individuals to afford moving costs as well as childcare and eldercare costs associated with living away from extended families.

Second, even if location is not freely chosen, the fact that the San Franciscan faces a higher cost of living than the Chicagoan does not necessarily mean that the San Franciscan’s marginal utility of income is higher as well. A San Franciscan earning $30,000 a year might be materially worse off than a Chicagoan earning the same amount, and for that reason the San Franciscan might prize an additional dollar somewhat more, but it is also the case that a dollar goes further in Chicago than in San Francisco. Thus, it is unclear whether the marginal utility of income is higher for the San Franciscan or the Chicagoan earning the same amount, and insofar as the welfarist case for redistribution is predicated upon differential marginal utilities of income, the argument for redistributing more to the San Franciscan is ambiguous at best. Add in the administrative costs of locational adjustments and the argument for a geographically differentiated UBI becomes rather weak.

For the resource egalitarian, meanwhile, the case for geographic adjustments arguably depends upon whether location should be considered a matter of choice or of chance. If the former, then the neutrality principle would suggest that society should not grant a larger benefit to individuals with a taste for mild winters and views of the Golden Gate Bridge—as expensive as those tastes might be to satisfy.222 If the latter, then perhaps we should think of being born in the San Francisco Bay Area as a chance event that raises the amount of income that one needs in order to function as an equal citizen. On this view, individuals unlucky enough to be born into high cost-of-living areas might have a claim to a larger UBI.

Yet even if location is a matter of chance rather than choice, the resource egalitarian argument for redistributing more to individuals in high cost-of-living areas is uncertain. The cost of living tends to be higher in areas with greater employment opportunities and higher wages.223 Should we think of the San Francisco Bay Area native as unlucky to have been born in a high cost-of-living area, or lucky to have been born in an area with an unemployment rate that is 1.1 percentage points below the national rate and average weekly wages that exceed the national average by more than 76 percent?224 Should redistribution be targeted at individuals who by chance land in areas with high wages and a high cost of living, or low wages and a low cost of living? The question illustrates the reality that even if cost-of-living adjustments are attractive in theory, they are challenging to implement in practice.225

From a libertarian perspective, the case for cost-of-living adjustments is especially weak. Cost-of-living adjustments effectively reward localities that—through stringent zoning regulations—drive up housing costs.226 The idea of the federal government subsidizing exclusionary local regulations should be anathema to most libertarians. Cost-of-living adjustments also favor individuals with expensive geographic preferences relative to individuals who would prefer to spend less on housing and more on other goods and services. Libertarians who believe that individuals should determine on their own what trade-offs to make are likely to chafe at the notion that the government should give a larger amount to those who choose to live on the island of Manhattan rather than in Manhattan, Kansas.

D. Assignability: Beneficiaries Should Have a Limited Ability to Borrow Against Future Payments

One largely overlooked but critically important implementation issue concerns the assignability of future UBI benefits. Should individuals be allowed to post their future UBI benefits as collateral for a loan or to trade those benefits away in a reverse-annuity transaction? To illustrate: Imagine that an individual faces a sudden and unexpected expense (for example, a bill for medical care or car repair) and lacks the cash on hand to make the payment. Should she be able to, say, borrow $5,800 today and assign her next twelve UBI payments to the lender (roughly the equivalent of borrowing at a 6.3 percent interest rate)?227 If so, financial institutions presumably would offer individuals the opportunity to trade their rights to UBI benefits in the future for a lump sum today. The result would be that individuals could convert periodic payments into something similar to the stakeholder grants suggested by Professors Ackerman and Alstott.228

From a welfarist perspective, the question of assignability is especially challenging. On the one hand, welfarists might be concerned about individuals taking out large loans early in life and then squandering their lump sums on unsuccessful investments or imprudent personal purchases. On the other hand, assignability—and the loans that it would facilitate—could expand credit access for millions of Americans who might use the lump-sum amounts to, say, start a business or pay college tuition. And as compared to alternatives such as payday loans or large credit card balances, borrowing against future UBI payments might allow low-income individuals to secure significantly lower interest rates.229 Predatory lending is a concern either way, but the market for UBI assignment loans may be easier for the government to regulate because the government effectively controls the collateral.

For resource egalitarians, the question of assignability seems somewhat more straightforward. The choice-chance principle arguably suggests that if an individual freely chooses to trade UBI payments in the future for a lump sum today, society is under no obligation to bail that individual out if she later regrets the decision. The neutrality principle, moreover, suggests that the government should not intervene to favor those who choose to save over those who choose to borrow.

For libertarians committed to the value of individual freedom, the idea of allowing UBI recipients to decide whether to assign their future benefits might seem quite attractive. On the other hand, minimal-state libertarians who believe in a UBI as a way to ensure that no indigent individuals are “faultlessly” poor might wonder whether we hold individuals at fault for decisions they made years or decades in the past. Meanwhile, Friedman’s public-goods argument for a UBI leads us to ask whether nonindigent members of society will experience moral or aesthetic distress when they see their fellow citizens living in abject poverty if they know that those living in poverty had the option of a stable UBI and yet squandered it.

Assignability also interacts with longevity in interesting and potentially significant ways. Presumably, financial institutions will make lump-sum offers based on their projections of the recipient’s life span. The financial institution will stand to lose on individuals who live shorter-than-expected lives and gain on individuals who live longer-than-expected lives. Within the reverse-annuity risk pool, we will see a sort of redistribution from the long-lived to the short-lived (similar to a whole life insurance risk pool, and the opposite of what occurs in a traditional-annuity risk pool). Some egalitarians may see this as desirable because it partially compensates the short-lived for their bad luck. Welfarists may be concerned because the marginal utility of an additional dollar is presumably higher the longer one will live, and in this respect reverse annuities amount to a transfer in the wrong direction.

We do not have a fully satisfactory answer to the assignability question, but we offer a tentative suggestion: a rule that allows adults over the age of eighteen to assign benefits for a limited period (for example, twelve months) in exchange for a loan. The one-year allowance would go some way toward expanding credit access for low-income individuals without generating the risk that some older Americans might be relegated to deep poverty on account of imprudent decisions that they made in their youth. UBI pilot programs can shed further light on this question by randomly assigning recipients to treatment and control groups that either can or cannot assign their benefits and then comparing outcomes. One year is an admittedly arbitrary figure, and further study might reveal evidence that weighs in favor of expanding or contracting that period.

While our suggestions are tentative, our firmer conclusion is that assignability—though rarely discussed—will be central to UBI design.230 No matter the size, a freely assignable UBI will not eradicate poverty entirely because some individuals will assign their UBIs away. On the other hand, a UBI with a prohibition on assignment reproduces some of the same paternalism that UBI supporters often point to as a flaw of our current welfare state. Put differently: If the goal is to ensure a subsistence-level income for all Americans, then a freely assignable UBI will not accomplish that end.231 If the goal is to increase financial opportunities for all Americans and then to let them chart their own course through life, then free assignability would seem to be the better approach. The question of assignability brings to the foreground philosophical differences among UBI proponents of various stripes.

E. Payment Mechanism: Direct Deposits on a Biweekly Basis via the Social Security Administration

Once we have decided that the government ought to provide a basic income, the question arises: How? That is, how exactly will payments be made to recipients, how frequently, and by whom?

Direct deposit to a bank account or debit card seems like the most straightforward method of payment, and just over 99 percent of Social Security beneficiaries receive their payments via direct deposit already.232 The system is imperfect: the SSA’s Office of the Inspector General reports that direct deposit payments for nearly 7,200 beneficiaries—totaling almost $11 million—were misdirected from 2014 to 2016.233 But for a program making roughly $1 trillion a year in payments to more than 60 million beneficiaries, the payment problems seem like mere hiccups.234

Still, further innovations on top of the existing Social Security payment system may be worthy of consideration. One possibility is to encourage—or even require—parents to set up savings accounts for their children in order for the children to receive a basic income, with payments deposited directly to those accounts. Accounts could be structured such that parents can withdraw funds to pay expenses for their children until those children reach age eighteen, at which point the children (now adults) would have full control over their accounts.235 This could help to reduce the ranks of the “unbanked” (that is, the portion of the population that lacks access to accounts at financial institutions).236 A similar program in Mongolia, which distributes mineral and mining revenues into accounts set up for children under eighteen,237 has contributed to greater financial inclusion in that country than in the United States on some dimensions. For example, 89 percent of Mongolian adults in the lower two quintiles of the income distribution had bank accounts in 2014, compared to 87 percent of American adults in the bottom two quintiles.238 The 2 percentage point difference between Mongolia and the United States may seem small, but the statistic is quite striking once one considers the wide gap in economic and financial development between the United States (per capita GDP of roughly $63,000) and Mongolia (per capita GDP of less than $4,200).239

Others have suggested using blockchain technology to power a basic income program.240 A nonprofit organization called the People’s Currency Foundation already has sprung up with the goal of implementing a blockchain-based UBI.241 We confess to finding little in this idea to recommend itself, as a blockchain-based UBI would seem to exclude the technologically unsophisticated (and, moreover, the existing direct-deposit system for Social Security seems to work just fine).

A somewhat more serious (in our view) question is whether payments should go to individuals whose tax liability far exceeds their UBI amount. For example, should Jeff Bezos receive a direct deposit into his bank account even while he withholds thousands of dollars from his paycheck and makes estimated tax payments of many millions? The answer matters partly for optics (will support for a UBI be stronger if everyone feels like a beneficiary, or will Americans view payments to Bezos et al. as a waste?) and partly for purposes of administrability. We consider the optical question at greater length in Part IV; from a perspective of administrability, we anticipate that paying the same amount to every American will be less cumbersome than trying to adjust individual payments on the basis of income.

The question of payment frequency has also divided supporters of unconditional, unrestricted cash transfers. Ackerman and Alstott have proposed a system whereby individuals would receive $80,000—paid out in four annual increments—upon entering college or turning twenty-one, whichever comes first.242 Others argue for monthly UBI payments—a suggestion in line with most current UBI proposals.243 Payments could be even more frequent than that; just as many banks pay interest daily,244 the government could make daily deposits to UBI recipients’ accounts.

In theory, payment frequency might be of little relevance if recipients can freely contract with financial institutions. For example, one can annuitize a once-in-a-lifetime lump-sum amount in order to transform it into a stream of annual or monthly payments; conversely, one can borrow against a future stream of annual or monthly payments in order to obtain a lump-sum amount today.245 There are, however, at least three aspects of the payment frequency issue that merit policymakers’ attention.

The first is administrability. Transaction costs increase with payment frequency. Monitoring costs do as well. Workers who are paid via direct deposit once every two weeks bear a time cost when they check their bank accounts to ensure that the payment came through. Those costs would rise (or the vigilance of monitoring would decline) if payments were made every day.

A second aspect of the frequency question involves consumption smoothing. The stresses of poverty—combined with limited access to credit and savings vehicles—potentially make it difficult for households to spread expenditures over a long payment cycle, with the consequence that spending may be concentrated toward the beginning of the cycle and households will go without adequate nutrition or other necessities near the end. Substantiating this concern, economists have observed a “Food Stamp nutrition cycle,” with caloric intake declining by 10 to 15 percent over the course of each month.246 There is also some evidence that stores respond to this pattern by raising prices on the day that SNAP beneficiaries receive payments.247 These findings have led to an emerging view in food-policy circles that more frequent payments (for example, on a biweekly or semimonthly basis) would improve outcomes for beneficiaries by allowing them to smooth their consumption more easily.248 The counterargument to the consumption-smoothing case for more frequent payments is that a longer window between disbursals operates as a “forced savings” mechanism: beneficiaries effectively “save up” for significant durable-goods expenses (for example, cars and washing machines).249 Low-income households—especially those with limited access to banks or other financial institutions—may otherwise find it difficult to save for larger expenditures.250

A third aspect of the frequency question is the interaction between payment frequency and longevity. As noted above in our discussion of assignability, one-time payments tend to favor shorter-lived individuals while more frequent payments tend to favor longer-lived individuals.251 That is, if payments are made each period that one is alive, then longer-lived individuals will receive more over their lifetimes. If one believes that premature death is an element of inequality that the government ought to offset through cash payments, then perhaps that is an argument for one-time payments.

The prevalence of biweekly or semimonthly payment schedules in the private-sector employment context seems to indicate that individuals prefer relatively regular payment rather than “forced savings” brought about by payment infrequency. If there were widespread demand among employees for less frequent payment, then presumably employers would economize on transaction costs by reducing payment frequency. Our own tentative view is that the government should likely follow the judgment of the majority of employers who have concluded that disbursals every other week or twice a month strike a reasonable balance between administrability and consumption smoothing. We acknowledge, though, that technological advances that lead to further reductions in transaction costs might reconfigure that balance. As for the forced savings counterargument, our hope is that a UBI via direct deposit will expand financial access, allowing low-income individuals to utilize savings mechanisms (for example, certificates of deposit) that are far superior to payment delay.

A final question in the design of a payment mechanism is the question of who should administer the program. While others have suggested that the IRS may be the agency best suited to disburse UBI payments,252 our view is that the SSA is a more natural fit—for three reasons. First, the SSA already makes monthly payments to more than 67 million beneficiaries, or over one-fifth of the total US population.253 For it, carrying out a UBI would mean scaling up an existing endeavor. The IRS, by contrast, only interacts with most taxpayers annually. Second, the American public appears to have a much more favorable view of the SSA than of the IRS. In the most recent Pew Research Center survey that asked respondents about their views of both agencies, the share of respondents saying that they had a favorable view of the SSA exceeded the share with an unfavorable view by 18 percentage points. The comparable figure for the IRS was −10 percentage points (that is, the share with an unfavorable view exceeded the share with a favorable impression).254 Third, at least in recent years, members of Congress have sought to vilify the IRS for a range of alleged infractions and starved the agency of cash.255 While ultimately a basic income will need to be funded by taxes that the IRS collects, it would seem wise in the near term to shield a UBI from the toxicity of tax politics as much as possible. Tasking the SSA rather than the IRS with carrying out a UBI is a step in that direction.

F. Funding Mechanism: Consolidate Cash and Near-Cash Transfers While Adding a Broadbased Surtax

The previous Section asked how a UBI should be paid; a harder question is how a UBI should be paid for. Vox’s Dylan Matthews has argued that “[w]hether or not basic income is a good idea depends entirely on how you pay for it,”256 and while we think that the success of a UBI depends on more than that, we certainly agree that the “how you pay for it” question is key. As of mid-2018, the total population of the United States was approximately 327 million,257 and if we exclude the approximately 11 million unauthorized immigrants and 2 million temporary lawful permanent residents,258 that leaves 314 million citizens and lawful permanent residents living in the country. Multiplying $6,000 per year by 314 million gives us a total annual cost for a UBI of $1.884 trillion. Clearly, a UBI would be expensive.

One way to offset that cost would be to discontinue some or all existing cash and near-cash transfer programs. Setting aside programs administered by the SSA (to which we will return momentarily), the primary federally funded cash and near-cash transfer programs include SNAP ($70 billion in 2018), TANF ($17 billion), the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) ($5 billion), unemployment insurance ($32 billion), and Section 8 rental assistance ($34 billion).259 The EITC and child tax credit will cost a combined $196 billion in 2018.260 So far, that sums to $354 billion, leaving a funding gap of $1.530 trillion remaining to be filled.

Potential elimination of the OASI, DI, and SSI programs run by the SSA raises especially difficult questions. At the end of 2017, the average monthly OASI benefit for retired workers was $1,404;261 for disabled workers under DI, it was $1,197;262 and for SSI recipients, it was $542.263 The estimated costs of these programs in 2018 was $848 billion for OASI, $147 billion for DI, and $56 billion for SSI.264 Eliminating OASI, DI, and SSI would generate $1.056 trillion, bringing us two-thirds of the way toward closing the remaining funding gap. At the same time, replacing these programs with a $500 per month UBI would leave millions of elderly and disabled Americans worse off—especially if it comes with the elimination of other programs (for example, SNAP and Section 8) from which many low-income elderly and disabled individuals also benefit. Requiring seniors to choose between a UBI and OASI benefits—as proposed by Murray—would reduce the cost of a UBI by roughly $240 billion,265 but would be vulnerable to the political and philosophical objections noted above.266

For now, we set aside further discussion of the interaction between a UBI and Social Security’s OASI program. Social Security (and OASI in particular) is often considered to be the “third rail of American politics”;267 if a UBI is pitched as a competitor to OASI, we are pessimistic about basic income’s political prospects. With respect to DI and SSI, we can anticipate welfarist, egalitarian, and libertarian reactions. Welfarists might argue that even if disability determinations are imperfect, DI and SSI eligibility are still rough proxies for ability to pay. Transfers to the disabled (or, more precisely, transfers to individuals whom the SSA has determined to be disabled) still will—on balance—move money from individuals with lower marginal utility of income to individuals with higher marginal utility of income. Some egalitarians (such as Professor Anderson) might argue that even if everyone ought to receive a basic income, disabled individuals ought to receive something more, because they require more resources in order to exercise equivalent freedoms.268 Libertarians, we expect, will be the most hostile toward maintaining DI and SSI because they (1) are likely the most skeptical about the informational value of government disability determinations and (2) are most concerned about the increase in taxes that will be necessary to close the UBI funding gap if DI and SSI are preserved.

A split-the-difference approach might be to reduce but not eliminate DI and SSI benefits. For purposes of our running tally, we will assume that benefits are reduced by one-third—an arbitrary fraction that we choose because we think that one of the three theories that form our philosophical foundation for a UBI would seem to suggest that those programs be eliminated entirely. (We understand that a real-world UBI might strike the balance differently.) That reduces the UBI funding gap by a further $68 billion. Adding in the previously discussed change to the tax treatment of Social Security benefits brings us $38 billion more.269 Of course, that still leaves us with a long way to go.

One option is to scale back federal funding for health care. The dollars at stake are significant: $383 billion for Medicaid, $18 billion for the Children’s Health Insurance Program (CHIP),270 $47 billion for Affordable Care Act premium tax credits, and more than $700 billion for Medicare.271 And the basic resource-egalitarian and libertarian arguments in favor of unrestricted cash transfers—that individuals should have the opportunity to choose for themselves how that money is spent—would seem to apply to health care just as much as SNAP. On the other hand, there is perhaps a plausible argument that the government enjoys informational and scale advantages vis-à-vis the free market in the provision of health coverage, which may be why most other advanced economies have chosen to channel health insurance through the public sector.272 Moreover, in some states, the average premium for an individual health insurance plan would eat up the entire $500 per month UBI.273 For many Medicaid recipients and other beneficiaries of federally subsidized health insurance, eliminating those subsidies and pouring the savings into a UBI would be of little help at all. We will therefore proceed on the assumption that these programs will remain in place. If they were eliminated in order to make fiscal room for a UBI, the funding challenge would be significantly less daunting.

All this leaves us with a $1.424 trillion funding gap. At least some of that money is likely to come from an increase in income taxes.274 To be sure, some could come from other sorts of taxes, such as a carbon tax. In fact, two prominent Republicans have recently called for a carbon tax coupled with a rebate that is essentially a UBI.275 Yet, a carbon tax set at the social cost of carbon would offset only a small portion of a UBI’s costs. US energy-related carbon emissions amounted to 5.27 billion metric tons in 2018; a carbon tax of approximately $36 per metric ton would—holding emissions constant—raise $190 billion per year.276 This is almost certainly an overestimate because the whole point of a carbon tax, aside from raising revenue, is to reduce carbon output. In any event, it would still leave us with a funding gap of $1.234 trillion.

Total income reported on individual income tax returns for tax year 2017 (filing year 2018), the most recent year for which comprehensive data are available, was approximately $11.170 trillion.277 We add US corporate pretax profits of $2.044 trillion, as measured in the first quarter of 2018.278 That yields an individual and corporate income tax base of $13.214 trillion. A crude way to think about the tax increase that would be necessary in order to offset the cost of a UBI is to imagine a flat surtax on all income (individual as well as corporate). Dividing the $1.234 trillion funding gap that remains after the imposition of a carbon tax by the $13.214 trillion base amount yields a rate of approximately 9.3 percent. Each additional $100 billion in cost savings would translate to a 0.75 percentage point reduction in the requisite tax. We should add that the 9.3 percent figure is almost certainly an underestimate, as it does not yet take into account the behavioral effects of higher taxes.279

Eliminating a variety of tax expenditures—such as the exclusion of employer-sponsored health insurance, the home mortgage interest deduction, the 20 percent deduction for pass-through income, and stepped-up basis at death280 —could generate additional revenues that would offset a UBI’s cost. The Joint Committee on Taxation’s expenditure estimates associated with those provisions for fiscal year 2019 sum to $290.9 billion ($172.8 billion for health insurance, $33.9 billion for mortgage interest, $50.2 billion for pass-throughs, and $34.0 billion for stepped-up basis).281 Eliminating those expenditures and using the revenue to offset the cost of a UBI would, very roughly speaking,282 reduce the size of the surtax needed to fund a $500 per month basic income from 9.3 percent to 7.1 percent. We should emphasize, however, that eliminating tax expenditures does not allow lawmakers to avoid the trade-off between a larger UBI and higher tax rates. Eliminating the abovementioned tax expenditures would still cause millions of taxpayers to pay higher effective tax rates.

Nor can that essential trade-off between a larger UBI and higher taxes be avoided by adding income-based phaseouts to a UBI. As noted above,283 an income-based phaseout is simply a surtax in sheep’s clothing. Consider a hypothetical UBI of $10,000 that phased out at a rate of 25 cents per dollar of income above $20,000. For every additional $100 of income, a taxpayer sees her UBI drop by $25. This is economically equivalent to a tax of 25 percent.284 The only difference is that a phaseout applies only to that subsection of taxpayers in the phaseout range.285

Ultimately, whether a UBI is funded through an explicit surtax on income or through a surtax that is partly redescribed as a “phaseout,” we think that policymakers should draw from more than a century of experience in the design of income taxes in the United States. A few key lessons emerge. First, there is no need to construct an entirely new definition of income for the taxes used to fund a UBI. The additional surtax can be layered onto the existing tax structure, given that Congress, the courts, and the IRS over the years have developed a rich set of standards and rules that aim to measure ability to pay while accommodating other policy concerns.286 Although this may seem obvious to those well-versed in tax policy, nontax welfare programs routinely use alternative definitions of income when imposing phaseouts.287 Given that several other UBI proposals incorporate taxes cast as phaseouts based on “income” without defining income, we think it important to clarify this point. Second, there is no need for a new agency to collect the funds that go toward a UBI: the task falls firmly within the IRS’s wheelhouse.

Third, cliff effects should be avoided. For instance, the current structure of the EITC causes taxpayers to lose their entire credit when their investment income crosses the $3,600 threshold.288 This structure—which in some cases imposes an implicit marginal tax rate of over 655,000 percent on the 3,601st dollar of investment income—is almost certainly not the best way to design a tax system.289 Unfortunately, some current UBI proposals contain such effects. Consider Chris Hughes’s proposal to give $500 per month ($6,000 per year) to every adult earning less than $50,000 a year “who is working in some way”—which, according to Hughes, should include college students as well as individuals who care for young children or care for aging or disabled adults.290 Hughes’s scheme is economically equivalent to a phaseout-free basic income for working individuals of $6,000 per year, combined with an additional tax of $6,000 once income reaches $50,000. The result is a strange (and likely quite distortionary) cliff effect at the 50,001st dollar of income: a marginal rate of 600,000 percent on that dollar.

Fourth, policymakers should be cognizant of the marriage incentives potentially embedded in phaseouts and surtaxes and should avoid arrangements that severely penalize marriage. Hughes’s phaseout, for example, would also impose a harsh marriage penalty: for example, two single individuals each earning $26,000 would lose a combined $12,000 per year by marrying.291 More sophisticated UBI phaseouts often embed marriage penalties as well. For example, Jessica Wiederspan and collaborators at the University of Michigan School of Social Work have modeled proposals that would set a UBI at 75 percent, 100 percent, or 133 percent of the federal poverty level, with a phaseout rate of 33 percent or 50 percent.292 All of these setups would involve steep marriage penalties because the poverty level is a function of household size. For example, with a UBI at 100 percent of the poverty level and a phaseout rate of 33 percent, two adults earning $20,000 per year would face a marriage penalty of $9,090 if they wed.293

Fifth and finally, policymakers should seek to avoid sky-high marginal rates. A surtax in the 7 to 10 percentage point range, spread across all taxpayers, would put the top marginal rate on individuals only slightly above where it stood before 2018 and would leave the rate on corporations well below pre-2018 levels.294 One consequence of a low, yet universal surtax of, say, 10 percent is that as a practical matter, one’s UBI is not fully offset by additional taxes until one’s income is $60,000. Put another way, our UBI would not phase out until $60,000 of income.

One advantage of rolling together various cash and near-cash transfer programs and imposing a single surtax is that it allows policymakers to avoid some of the perverse incentives generated by the status quo. For example, the combination of the EITC and SNAP phaseouts and the 10 percent federal income tax bracket results in marginal tax rates above 50 percent for some low-income households.295 Those same phaseouts plus Medicaid eligibility criteria can generate substantial marriage penalties for some couples as well.296 With fewer moving pieces, policymakers can structure a UBI and any accompanying phaseout or surtax to minimize these potential pitfalls.

Our UBI will, inevitably, produce winners and losers. The overwhelming majority of low-income individuals and families would be winners. A UBI of $500 per person per month—$2,000 per month for a four-member household—would more than compensate for the loss of SNAP benefits (up to $192 per month for a single person and up to $640 per month for a four-person household)297 and/or the EITC (the equivalent of up to $44.08 per month for a single person and up to $485.67 per month for a two-adult, two-child household).298 Throwing the refundable child credit (the equivalent of $233.33 per month for a two-child household) into the mix does not change this calculus.299 Even those who also qualify for WIC (average monthly benefit of roughly $41 per person) on top of SNAP and the EITC will almost always prefer the UBI.300 For those somewhat higher up the income ladder, the benefit from a UBI will more than offset the cost of the additional surtax until income reaches at least $60,000 for a single individual and at least $240,000 for a four-member household.

To be sure, there will be a relatively small number of low-income individuals and families who will receive less under our proposal than under the status quo—in particular, some Section 8 voucher recipients living in high-rent areas and some families in states such as Alaska and New Hampshire where TANF benefits are especially generous.301 Our view is that a more even distribution of benefits among low-income individuals and families is preferable to the status quo, in which programs such as Section 8 provide substantial benefits to a very small portion of low-income families but leave others languishing on long waiting lists.302 We also believe that a $500-per-month UBI that, as explained above, is not recaptured by increased taxes until one’s income is $60,000 is preferable to a larger UBI that is phased out at a lower income (such as Murray’s proposal). While the latter may better target certain very poor individuals, such frameworks necessitate regressive tax structures that import many of the existing work disincentives into the UBI context. Likewise, the fact that our proposal replaces most of the existing welfare state instead of adding to it renders it economically feasible and will cause fewer economic distortions. Almost any intervention that unsettles the existing public-assistance landscape will leave some people worse off, and choosing among assistance programs requires making hard trade-offs. But the upside of our proposal—a strong safety net with fewer holes and distortions—seems to us to be well worth it.

IV. The Outward Façade of a UBI

So far we have deferred discussion of whether a UBI should be truly “universal” or whether benefits should phase out after income exceeds a certain threshold. We have deferred that discussion primarily because the question is semantic: As illustrated in Part I.B, the exact same scheme can be described as (a) a UBI with no phaseout that is funded by a surtax on income, (b) a UBI that phases out, or (c) a negative income tax. The different framings may affect the political popularity of the proposal; they do not (or at least, need not) affect program design.

As a practical matter, we think that the easiest way to implement a UBI would be for the SSA to make uniform biweekly or semimonthly payments via direct deposit to each eligible individual’s account and for the offsetting surtax to be integrated into the existing individual income tax system. There would be no need for the SSA to track the income of 314 million citizens and lawful permanent residents on a biweekly basis to ensure that they remain below the phaseout threshold; if you are a US citizen or a Green Card holder, you will be entitled to receive $250 twice a month regardless of how much other income you make.

As an optical matter, the advantage of this arrangement is that it effectively frames a UBI as a universal program, rather than as a form of “welfare” that goes only to the poor. Basic income supporters hope that this “universal” framing will make a UBI as popular as OASI (which reaches most seniors) and Medicare (which reaches every US citizen over sixty-five and every lawful permanent resident over that age who has resided in the United States for the past five years).303 The disadvantage is that some may react negatively to the notion of the über-rich receiving government payments. Financial journalist Felix Salmon reflects this view in a recent critique of basic income proposals, writing that:

UBI is a pretty inefficient way of giving poor people money. Think about it this way: Just 40 percent of a UBI’s expenditure would go to the bottom 40 percent of the population, and a mere 10 percent would go to the 10 percent who need it most. What would happen to the rest of the money?

Study after study has shown that when you give money to the homeless and the very poor, they don’t spend it on frivolities like booze and tobacco: In fact, rates of drinking and smoking invariably go down rather than up. On the other hand, if you gave me an extra $1,500 per month, no strings attached, I’m sure a significant chunk of that would end up in my wine fridge. That might be popular with my local wine merchants, but as a means of redistributing society’s wealth in the interests of fairness and equality, it does leave something to be desired.304

In one sense, Salmon—despite being one of the smartest and most perceptive commentators on economic policy today—has been fooled by the framing of a UBI into thinking that high-income individuals like himself would be receiving “extra” money. Salmon reportedly earned more than $400,000 in 2017;305 if the income tax rate is raised by 7 percent to 10 percent in order to offset the cost of a UBI, Salmon will have tens of thousands of dollars less each year to spend on his wine collection, even though he will receive a UBI via direct deposit every two weeks.

In theory, a UBI could be implemented such that Salmon and other high-income individuals do not receive any payment. Either we could track each individual’s income throughout the year and withhold payment from high-income individuals, or we could switch to an annual payment cycle and implement a UBI as a negative income tax. The former approach would raise administrative costs substantially.306 The latter tack would make it harder for individuals to smooth consumption throughout the year. The “negative income tax” framing suffers from the additional disadvantage of including two words that do not elicit warm reactions: “negative” and “tax.”

If resistance remains fierce to the notion of the rich “receiving” a basic income, we could add an extra line to the individual income tax return on which high-income individuals “pay back” their UBI. This would be similar to the phaseout of personal exemptions prior to the December 2017 tax law307 and the income-based limitation on the child tax credit under the current Code.308 By relying on phaseouts rather than raising rates, lawmakers complicate the tax-filing process to achieve economically equivalent but less transparent results. Perhaps there are benefits of this approach from a political perspective; as a matter of policy, we see nothing to recommend it.

Finally, a UBI could be framed as a new “tax credit” rather than a new benefit without changing any of its essential features. There is some evidence to suggest that individuals react more positively to cash transfers that are described as “tax credits” rather than as spending programs.309 The difference is merely semantic: some tax credits are paid out more frequently than annually and administered in tandem with agencies other than the IRS.310 Further survey experiments can shed light on whether reframing would foment greater support; for present purposes, the key point is that whether a UBI is framed as a new tax credit or a new public benefit is a piece of the façade rather than an element of deep structure.

Conclusion

We have argued in favor of a universal basic income of $500 per month ($6,000 per year) that would be paid to every US citizen and lawful permanent resident via direct deposit in biweekly installments, administered by the SSA, and funded by the consolidation of some (but not necessarily all) existing cash and near-cash transfer programs as well as a modest increase in taxes. Up until now, we have considered a UBI as a sharp break from the status quo. Migration toward a UBI could occur more gradually, however. While our focus is on a UBI as an alternative to the status quo, our analysis also suggests ways in which existing transfer programs could improve incrementally, even if the goal of a UBI remains distant.

First, as noted above, many existing transfers come with severe restrictions on use. SNAP benefits cannot be used to buy toilet paper or Tylenol.311 Section 8 vouchers generally cannot be used to rent apartments with shared kitchens.312 Welfarist, resource egalitarian, and libertarian approaches all support changes to these rules that would allow recipients to exert more control over the use of their benefits. One move in a desirable direction would be to consolidate SNAP, TANF, and Section 8 into a single program that provides recipients with debit-like cards and allows them to spend their benefits without micromanaging their choices.

Second, many existing transfers are conditional upon work. For example, a parent cannot claim the child tax credit unless she receives at least $2,500 per year in earned income, and the full $1,400 refundable credit is available only if the parent earns at least $11,833.313 Denying support to the individuals and families that need it most makes little sense from a welfarist, egalitarian, or libertarian perspective. Changes to the child tax credit and EITC that extend benefits to those at the very bottom would not bring us all the way to a UBI, but they would put us somewhat closer to that ideal.314

Third, as noted above, the tangle of phaseouts for existing cash and near-cash transfer programs can generate high marginal tax rates at the bottom of the income distribution that discourage low-income adults from working.315 We have argued that a shift to a UBI could address this problem by replacing all of those phaseouts with a single surtax at a modest rate. Yet even without going all the way to a UBI, policymakers could “pop” these bubble rates by consolidating cash and near-cash transfers into one program with a less-punishing phaseout structure.

Such modesty may prove unnecessary, however, if voters and politicians from multiple points on the ideological spectrum can coalesce around a national basic income program. If and when that happens, the design questions canvassed in this Article will take on central significance. We do not claim to have found all the answers yet. We do believe, though, that goals of ending extreme poverty and empowering low-income individuals can be advanced through greater scholarly attention to a UBI’s nuts and bolts.

• 1See Karla Lant, Universal Basic Income Pilot Programs (Futurism), online at https://futurism.com/images/universal-basic-income-ubi-pilot-programs-around-the-world (visited Jan 25, 2020) (Perma archive unavailable). See also Leonid Bershidsky, In Finland, Money Can Buy You Happiness (Bloomberg, Feb 9, 2019), archived at https://perma.cc/QUW2-8AZ5 (noting the completion of Finland’s UBI pilot); Gretchen Frazee, Ontario Is Canceling Its Basic Income Experiment (PBS NewsHour, Aug 6, 2018), archived at https://perma.cc/SS3K-7ZL2 (noting the premature end of Ontario’s UBI pilot after a conservative political party took over the province’s government).
• 2See Sam Harnett, A California City Is Planning on Giving Money to Some Residents, No Strings Attached (Marketplace, Feb 27, 2018), archived at https://perma.cc/3BTN-TNNL.
• 3See Fran Spielman, Plan to Pay 1,000 Residents $1,000 a Month—No Strings Attached—Pitched by Panel (Chicago Sun-Times, Feb 7, 2019), archived at https://perma.cc/U43F-UF7B.
• 4See Hawaii House Concurrent Resolution No 89-17, Hawaii House of Representatives, 29th Legis (May 5, 2017), archived at https://perma.cc/K4PT-FK8W.
• 5In addition to the one thousand families receiving $1,000 a month, a control group of two thousand families will receive $50 a month. See Basic Income Project Proposal: Overview for Comments and Feedback *13–14 (Y Combinator Research, Sept 2017), archived at https://perma.cc/XT9Z-56H9; Nitashka Tiku, Y Combinator Learns Basic Income Is Not So Basic After All (Wired, Aug 27, 2018), archived at https://perma.cc/9N4S-B4RX.
• 6The First Study of Basic Income in the United States (Y Combinator Research, 2018), archived at https://perma.cc/75YB-EB8J.
• 7See Charles Murray, In Our Hands: A Plan to Replace the Welfare State 6–10 (AEI rev ed 2016).
• 8See Chris Hughes, Fair Shot: Rethinking Inequality and How We Earn 91–95 (St. Martin’s 2018).
• 9See Andy Stern with Lee Kravitz, Raising the Floor: How a Universal Basic Income Can Renew Our Economy and Rebuild the American Dream 201–02 (Public Affairs 2016).
• 10In addition to those of Hughes, Murray, and Stern, a number of other UBI books have garnered significant publicity. See generally, for example, Annie Lowrey, Give People Money: How a Universal Basic Income Would End Poverty, Revolutionize Work, and Remake the World (Crown 2018); Rutger Bregman, Utopia for Realists: How We Can Build the Ideal World (Little, Brown 2017) (Elizabeth Manton, trans); Philippe Van Parijs and Yannick Vanderborght, Basic Income: A Radical Proposal for a Free Society and a Sane Economy (Harvard 2017).
• 11See Obama’s Full Speech About the ‘Politics of Fear and Resentment’ (CNN, July 17, 2018), archived at https://perma.cc/C7TJ-4Z58 (“[W]e’re going to have to consider new ways of thinking about these problems, like a universal income. . . . [W]e’re going to have to worry about economics if we want to get democracy back on track.”).
• 12See Hillary Rodham Clinton, What Happened 238–39 (Simon & Schuster 2017). See also Dylan Matthews, Hillary Clinton Almost Ran for President on a Universal Basic Income (Vox, Sept 12, 2017), archived at https://perma.cc/KUX3-KTM5.
• 13See Joe Garofoli, Andrew Yang Doing Better Than Many Veteran Politicians in First White House Run (SF Chron, July 17, 2019), archived at https://perma.cc/D9RP-DVJB. Yang withdrew from the race in February 2020 after qualifying for all of the 2019 Democratic debates but failing to win delegates in the Iowa caucuses or New
  Hampshire primary. See Matt Stevens, Andrew Yang Drops Out: ‘It Is Clear Tonight from the Numbers That We Are Not Going to Win’ (NY Times, Feb 11, 2020), archived at https://perma.cc/W5MP-DLQV.
• 14See generally, for example, Michael R. Asimow and William A. Klein, The Negative Income Tax: Accounting Problems and a Proposed Solution, 8 Harv J Legis 1 (1970) (discussing accounting, timing, and income-averaging issues related to designing a negative income tax); William D. Popkin, Administration of a Negative Income Tax, 78 Yale L J 388 (1969) (analyzing administrative concerns with existing welfare programs and their implications for a negative income tax); James Tobin, Joseph A. Pechman, and Peter M. Mieszkowski, Is a Negative Income Tax Practical?, 77 Yale L J 1 (1967) (offering specific recommendations about various technical issues related to a negative income tax such as defining the family unit, defining income, and crafting a payment schedule); Sheldon S. Cohen, Administrative Aspects of a Negative Income Tax, 117 U Pa L Rev 678 (1969) (criticizing Professors Tobin, Pechman, and Mieszkowski’s proposal on administrative grounds); Comment, A Model Negative Income Tax Statute, 78 Yale L J 269 (1968) (proposing a model negative income tax statute while addressing policy considerations and whether complementary welfare strategies should be pursued).
• 15Professor Milton Friedman coined the term “negative income tax” in Milton Friedman, Capitalism and Freedom 192 (Chicago 1962). In August 1969, President Richard Nixon delivered a televised address to the nation in which he unveiled a proposal for a “Family Assistance Plan” that was loosely modeled on Friedman’s negative income tax idea. On the history of the negative income tax and the Family Assistance Plan, see Daniel P. Moynihan, The Politics of a Guaranteed Income: The Nixon Administration and the Family Assistance Plan 50–59, 136–46 (Random House 1973).
• 16See US Department of Labor, Bureau of Labor Statistics, Historical Consumer Price Index for All Urban Consumers (CPI-U) *4, archived at https://perma.cc/B7BW-MR7Z (ratio of July 2019 CPI to April 1970 CPI is 256.571 / 38.5, or 6.664, times $1,600 value of proposed minimum income for a family of four is approximately $10,662).
• 17See Miranda Perry Fleischer and Daniel Hemel, Atlas Nods: The Libertarian Case for a Basic Income, 2017 Wis L Rev 1189, 1203–44.
• 18One might ask: Why six? Why not, for example, split the “eligibility” block into two (citizenship and age) so there are seven? Fair question. We have found—after workshops and numerous drafts—that the six-block structure creates a manageable framework for studying and discussing UBI implementation. We do not claim that there is anything magical to the number six though, and we acknowledge that a list of this sort entails an inevitable dose of arbitrariness. That said, the content of these six blocks is neither arbitrary nor unique to a UBI; it reflects design decisions inherent to any program of redistribution.
• 19See John R. Brooks, The Definitions of Income, 71 Tax L Rev 253, 253 n 2 (2018) (noting the alternative definitions of “income” within various sections of the Internal Revenue Code).
• 20A UBI would almost certainly not be paid in physical currency. Instead, debit cards, direct deposit, or virtual currency are the most likely delivery mechanisms (although physical checks also remain a possibility). See Dom Galeon, Universal Basic Income Could Become a Reality, Thanks to This Technology (Futurism, Mar 10, 2017), archived at https://perma.cc/9B6J-4P2H (discussing UBI experiments that rely on blockchain-based payments). We discuss payment mechanisms at greater length in Part III.E.
• 21See Carole Pateman and Matthew C. Murray, Introduction, in Matthew C. Murray and Carole Pateman, eds, Basic Income Worldwide: Horizons of Reform 1, 2 (Palgrave Macmillan 2012); Van Parijs and Vanderborght, Basic Income at 8 (cited in note 10).
• 22We address whether a UBI should be limited to citizens, adults, or other non-need-based categories in Part III.B.
• 23Professors Philippe Van Parijs and Yannick Vanderborght have written that, as a definitional matter, a truly “universal” basic income must not be “subjected to an income or means test.” Van Parijs and Vanderborght, Basic Income at 8 (cited in note 10). As we discuss below, this is true only if it refers to the outward appearance of a UBI and not the inner workings. Any UBI that is financed by an income tax can be redescribed as a cash transfer subject to an income test because high-income individuals will contribute more through income taxes than they receive in the form of a UBI. Likewise, any UBI financed by a carbon tax can be redescribed as a cash transfer subject to a personal carbon emissions test because individuals whose carbon emissions exceed the mean will pay more through carbon taxes than their UBI.
• 24See Social Security Administration, Understanding Supplemental Security Income SSI Eligibility Requirements—2019 Edition, archived at https://perma.cc/77P5-K5S8.
• 25See US Department of Agriculture, Food and Nutrition Service, Able-Bodied Adults Without Dependents (ABAWDs), archived at https://perma.cc/M6GN-E23N (last updated July 17, 2018) (explaining that unemployed individuals without dependents may continue to receive SNAP benefits by complying with a state workfare program or by completing a SNAP Employment and Training Program).
• 26See 42 USC § 608(a).
• 27See Drug Testing for Welfare Recipients and Public Assistance (National Conference of State Legislatures, Mar 24, 2017), archived at https://perma.cc/MF8Y-LDLJ (noting that, as of March 2017, fifteen states had passed legislation requiring drug screening for public assistance applicants and twenty more states had proposed similar programs).
• 28See, for example, New York State Office of Temporary and Disability Assistance, Electronic Benefits Transfer (EBT) Card, archived at https://perma.cc/T34Z-HA63. See also Restrictions on Use of Public Assistance Electronic Benefit Transfer (EBT) Cards (National Conference of State Legislatures, May 8, 2015), archived at https://perma.cc/HZD3-F75V (noting that as of 2015, at least thirty-seven states use EBT cards to issue TANF benefits).
• 29See 42 USC § 608(a)(12)(A)(i)–(iii). The law does not use the term “strip club” but instead refers to “any retail establishment which provides adult-oriented entertainment in which performers disrobe or perform in an unclothed state for entertainment.” 42 USC § 608(a)(12)(A)(iii).
• 30See Kan Stat Ann § 39-709(b)(14).
• 31See Thomas Paine, Agrarian Justice 27 (J. Adlard 2d ed 1797).
• 32See Bruce Ackerman and Anne Alstott, The Stakeholder Society 3 (Yale 1999) (proposing a stake of $80,000 for each American at maturity and attributing the idea to Paine).
• 33While these examples all involve a flat tax rate, the equivalence extends to tax systems with graduated rate structures.
• 34See, for example, Marc Fleurbaey and François Maniquet, Optimal Income Taxation Theory and Principles of Fairness, 56 J Econ Lit 1029, 1031 (2018) (considering welfarist, egalitarian, and libertarian perspectives on optimal taxation); Michael Pressman, “The Ability to Pay” in Tax Law: Clarifying the Concept’s Egalitarian and Utilitarian Justifications and the Interactions Between the Two, 21 NYU J Legis & Pub Pol 141, 147–48 (2018) (stating that utilitarianism [welfarism] and egalitarianism “are two of the main principles thought to underlie our determination of how to distribute the tax burden,” while also considering libertarian perspectives); Dorothy Van Soest, Strange Bedfellows: A Call for Reordering National Priorities from Three Social Justice Perspectives, 39 Soc Work 710, 713–15 (1994) (considering utilitarian [welfarist], egalitarian, and libertarian perspectives on fiscal policy).
• 35See Edward R. Teather-Posadas, Universally Basic: An Ethical Case for a Universal Basic Income *12–14 (Colorado State University Department of Economics Working Paper, Mar 26, 2017), archived at https://perma.cc/6KBG-G9J8.
• 36See Philippe Van Parijs, Competing Justifications of Basic Income, in Philippe Van Parijs, ed, Arguing for Basic Income: Ethical Foundations for a Radical Reform 3, 23–24 (Verso 1992). On “community” and “fraternity” as justifications for a government guarantee of material well-being, see Thomas C. Grey, Property and Need: The Welfare State and Theories of Distributive Justice, 28 Stan L Rev 877, 894–97 (1976).
• 37See Stern with Kravitz, Raising the Floor at 166 (cited in note 9) (stating that Stern’s main reason for supporting UBI is “its potential to deliver economic justice and security at a time when globalization and technological progress make it harder for Americans to find jobs that pay a living wage”).
• 38See Catherine Clifford, Elon Musk Says Robots Will Push Us to a Universal Basic Income—Here’s How It Would Work (CNBC, Nov 18, 2016), archived at https://perma.cc/G5SF-BXAV (quoting Musk’s belief that the United States will “end up with a universal basic income, or something like that, due to automation. . . . I’m not sure what else one would do”).
• 39See What Is the Freedom Dividend? (Yang2020), archived at https://perma.cc/2AEC-NA65 (stating that Yang supports a UBI because of technological change).
• 40Haley Byrd, Democrats Tiptoe Around Universal Basic Income (Independent J Rev, Oct 2, 2017), archived at https://perma.cc/PY9H-X2RD.
• 41Kathleen Pender, Why Universal Basic Income Is Gaining Support, Critics (SF Chron, July 15, 2017), archived at https://perma.cc/9WLA-7HPC.
• 42As Professors Hilary Hoynes and Jesse Rothstein note, debates about a UBI’s potential work-disincentive effects potentially do depend upon predictions regarding automation-induced unemployment. If one believes that automation will lead to widespread job losses, then the concern that a UBI may discourage entry into—or encourage exit from—the workforce becomes less salient. See Hilary Hoynes and Jesse Rothstein, Universal Basic Income in the United States and Advanced Countries, 11 Ann Rev Econ 929, 932–33 (2019).
• 43See Daniel Hemel, Book Review, Bringing the Basic Income Back to Earth (The New Rambler, Sept 19, 2016), archived at https://perma.cc/7TJJ-TRM6, reviewing Stern with Kravitz, Raising the Floor (cited in note 9) and Murray, In Our Hands (cited in note 7).
• 44See Lowrey, Give People Money at 7, 32 (cited in note 10).
• 45On the dominance of welfarist approaches, see Sarah B. Lawsky, On the Edge: Declining Marginal Utility and Tax Policy, 95 Minn L Rev 904, 910–11 (2011).
• 46See, for example, Will Kymlicka, Contemporary Political Philosophy: An Introduction 10 (Oxford 2d ed 2002) (“[U]tilitarianism . . . claims that the morally right act or policy is that which produces the greatest happiness for the members of society.”); Liam Murphy and Thomas Nagel, The Myth of Ownership: Taxes and Justice 51 (Oxford 2002) (defining the goal of utilitarianism as “maximiz[ing] the total happiness of the members of a society”); John Rawls, A Theory of Justice 20 (Belknap rev ed 2003) (identifying the “main idea” of utilitarianism as the idea that a just society achieves “the greatest net balance of satisfaction”).
• 47See, for example, Joseph Bankman and Thomas Griffith, Social Welfare and the Rate Structure: A New Look at Progressive Taxation, 75 Cal L Rev 1905, 1947 (1987);
  Miranda Perry Fleischer, Charitable Giving and Utilitarianism: Problems and Priorities, 89 Ind L J 1485, 1506–07 (2014); Lawsky, 95 Minn L Rev at 915–17 (cited in note 45).
• 48See David F. Bradford, Untangling the Income Tax 153 (Harvard 1986); Bankman and Griffith, 75 Cal L Rev at 1947–48 (cited in note 47); Nancy C. Staudt, The Hidden Costs of the Progressivity Debate, 50 Vand L Rev 919, 939–42 (1997).
• 49See Natasha Frost, Jeff Bezos Just Lost $38 Billion. He’s Still the Richest Person in the World (Quartz, July 6, 2019), archived at https://perma.cc/6NH9-EUSF.
• 50On the relationship between income and happiness in cross-national studies, see Thomas D. Griffith, Progressive Taxation and Happiness, 45 BC L Rev 1363, 1372–74, 1378–79 (2004) (citing a study that found “[a]n extra dollar of income in a poor nation . . . produced thirty-seven times as much utility as an extra dollar in a middle nation and seventy times as much utility as an extra dollar in a rich nation”).
• 51See id at 1379–81.
• 52We should note that there is a subtle—but substantive—distinction between the claim that “the marginal utility of income is diminishing” and “the redistribution of income from the rich to the poor increases overall utility.” Let’s assume, at least arguendo, that total utility is lower in Scenario 1, where A earns $90x and B earns $10x, than in Scenario 2, where A earns $80x and B earns $20x. It does not necessarily follow that if we start from Scenario 1, impose a $10x tax on A, and transfer $10x to B, we will thereby raise total utility. First, A’s utility if she starts with $90x and then loses $10x to a tax may be lower than if she started with $80x in the first place. This could be a function of the well-documented phenomenon of loss aversion, or it might result from A being particularly averse to taxation. See, for example, Amos Tversky and Daniel Kahneman, Loss Aversion in Riskless Choice: A Reference-Dependent Model, 106 Q J Econ 1039, 1040 (1991) (loss aversion); Abigail B. Sussman and Christopher Y. Olivola, Axe the Tax: Taxes Are Disliked More than Equivalent Costs, 48 J Marketing Rsrch S91, S91–S92 (2011) (aversion to taxation). Second, B’s utility if he starts with $10x and then receives an additional $10x transfer may be lower than if he earned $20x in the first place. This result might obtain if, as suggested by some sociological work, earning income generates greater utility (for example, through self-esteem effects) than does receipt of an equivalent transfer. On self-esteem and earned income, see, for example, Katherine S. Newman, No Shame in My Game: The Working Poor in the Inner City 104 (Knopf 1999). While welfarism still probably can support income redistribution, the path from the diminishing-marginal-utility premise to the redistribution prescription requires an additional step.
• 53See, for example, J.A. Mirrlees, An Exploration in the Theory of Optimum Income Taxation, 38 Rev Econ Stud 175, 206–07 (1971). See also Bankman and Griffith, 75 Cal L Rev at 1946–47 (cited in note 47) (discussing Professor Mirrlees’s model and its consideration by other tax scholars).
• 54See, for example, Anne L. Alstott, Equal Opportunity and Inheritance Taxation, 121 Harv L Rev 469, 474–75 (2007) (noting the dominance of utilitarianism in tax analysis, advocating resource egalitarianism as an alternative, and analyzing inheritance taxes through a resource egalitarian lens); Lawrence Zelenak, Taxing Endowment, 55 Duke L J 1145, 1172 (2006) (identifying Professors Ronald Dworkin, Daniel Markovits, Liam Murphy, and Thomas Nagel as standard-bearers of the egalitarian position); Miranda Perry Fleischer, Equality of Opportunity and the Charitable Tax Subsidies, 91 BU L Rev 601, 627 (2011) (identifying and describing “resource egalitarians” and offering a resource egalitarian analysis of the charitable tax subsidies).
• 55Alstott, 121 Harv L Rev at 471 (cited in note 54).
• 56See id at 476–77; Fleischer, 91 BU L Rev at 627 (cited in note 54).
• 57Alstott, 121 Harv L Rev at 477 (cited in note 54).
• 58Id.
• 59Id.
• 60See, for example, Elizabeth S. Anderson, What Is the Point of Equality?, 109 Ethics 287, 289 (1999) (coining the term and criticizing the emphasis on luck); Richard J. Arneson, Luck Egalitarianism Interpreted and Defended, 32 Phil Topics 1, 1, 13–15 (2004) (defending the theory); Samuel Scheffler, What Is Egalitarianism?, 31 Phil & Pub Affairs 5, 5, 31 (2003) (critiquing luck egalitarians for not rooting luck egalitarianism in a “more general conception of equality as a moral value”).
• 61Scheffler, 31 Phil & Pub Affairs at 5 (cited in note 60).
• 62This assumes perfect insurance markets, which would enable individuals to insure against future bad luck. Imagine that a series of storms causes crop failure. Whether to insure against that possibility is a choice. If D does not insure and goes bankrupt but E does and survives the bad season, insurance has transformed the result to one of choice from one of weather-dependent luck. For more on the role of insurance, see Ronald Dworkin, Sovereign Virtue: The Theory of Practice and Equality 76–77 (Harvard 2000).
• 63Ackerman and Alstott, The Stakeholder Society at 25–26 (cited in note 32).
• 64See, for example, Anderson, 109 Ethics at 310–11 (cited in note 60).
• 65See id at 310 (discussing the smoker example).
• 66Id.
• 67Id at 289. The contributions of Ackerman and Alstott, Scheffler, and Anderson are but a portion of the rich “equality of what” debate in the philosophical literature. Although we address the aspects of that debate most crucial to UBI programs, a full exploration of it is beyond the scope of this Article. For a more expansive treatment, see Fleischer, 91 BU L Rev at 623–35 (cited in note 54) and sources cited therein.
• 68Anderson, 109 Ethics at 288–89 (cited in note 60).
• 69Id at 320.
• 70See id at 317–20.
• 71See Fleischer and Hemel, 2017 Wis L Rev at 1191–92 & nn 2–7 (cited in note 17).
• 72This misconception likely stems from Robert Nozick’s assertion that only “a minimal state . . . is justified” and that “[t]axation . . . is on a par with forced labor.” Robert Nozick, Anarchy, State, and Utopia ix, 169 (Basic Books 1974).
• 73See generally Fleischer and Hemel, 2017 Wis L Rev 1189 (cited in note 17).
• 74To be clear, we are not arguing that all strands of libertarianism support redistribution. We emphasize that libertarianism is a spectrum of views, some of which support redistribution and some of which do not. See id at 1204–05.
• 75For overviews of libertarian thought, see Harry Brighouse, Justice 86–94 (Polity 2004); Kymlicka, Contemporary Political Philosophy at 102–65 (cited in note 46); Eric Mack and Gerald F. Gaus, Classical Liberalism and Libertarianism: The Liberty Tradition, in Gerald F. Gaus and Chandran Kukathas, eds, Handbook of Political Theory 115, 124–29 (Sage 2004); Peter Vallentyne and Bas van der Vossen, Libertarianism (Stanford Encyclopedia of Philosophy, July 1, 2014), archived at https://perma.cc/YA2T-86M6. See also generally John Tomasi and Matt Zwolinski, A Brief History of Libertarianism (forthcoming) (manuscript on file with author).
• 76Eric Mack, Non-Absolute Rights and Libertarian Taxation, 23 Soc Phil & Pol 109, 119 (2006).
• 77See id at 140.
• 78See Matt Zwolinski, Property Rights, Coercion, and the Welfare State: The Libertarian Case for a Basic Income for All, 19 Indep Rev 515, 523 (2015); Fleischer and Hemel, 2017 Wis L Rev at 1211–17 (cited in note 17).
• 79John Locke, Second Treatise of Government 21 (Hackett 1980) (C.B. Macpherson, ed) (originally published 1689).
• 80See Fleischer and Hemel, 2017 Wis L Rev at 1228–31 (cited in note 17).
• 81Gerald F. Gaus, Social Philosophy 170 (M.E. Sharpe 1999).
• 82Id. Lomasky, who conceptualizes individuals as simultaneously project pursuers and social creatures, offers a similar argument. He argues that individuals who lack basic necessities therefore lack the prerequisites for project pursuit and have no “rational stake in the moral community established by [a] system of [private property] rights.” Loren E. Lomasky, Persons, Rights, and the Moral Community 127 (Oxford 1987). Lomasky concludes that a basic safety net is justified to give them such a stake:

  If a person is otherwise unable to secure that which is necessary for his ability to live as a project pursuer, then he has a rightful claim to provision by others who have a surplus beyond what they require to live as project pursuers. In that strictly limited but crucial respect, basic rights extend beyond liberty rights to welfare rights.

  Id at 126. For a critique of this argument, see generally Tibor R. Machan, Against Lomaskyan Welfare Rights, 14 Reason Papers 70 (1989).

• 83See Fleischer and Hemel, 2017 Wis L Rev at 1224–28 (cited in note 17).
• 84See Mark V. Pauly, Income Redistribution as a Local Public Good, 2 J Pub Econ 35, 37–38 (1973).
• 85Id at 37.
• 86Assuming record stores still exist. See Number of Independent Record and CD Stores in the United States from 2003 to 2013 (Statista, Apr 8, 2016), online at https://www.statista.com/statistics/192691/independent-cd-and-record-stores-in-the-us/ (visited Jan 25, 2020) (Perma archive unavailable) (noting that the number of independent record and CD stores in the United States declined by more than 50 percent from 2003 to 2013).
• 87Many in-kind fringe benefits are not taxed. Thus, for example, an employee facing a 35 percent tax rate might prefer $100 of untaxed employer-provided meals over $100 in wages because the wages will be reduced to $65 after taxes. The employee would likely be willing to forgo $100 of food in exchange for $100 of cash (which she could, if she wanted to, use to buy $100 of food), but she may prefer $100 of employer-provided meals over $65 of cash.
• 88See, for example, Kevin D. Williamson, The White Ghetto (National Review, Dec 16, 2013), archived at https://perma.cc/S5P3-EEAH (describing trading soda purchased with SNAP benefits for cash and other items).
• 89We believe, however, that the common concern that parents who are negligent (perhaps due to drug addiction or mental illness, for example) will waste their children’s UBI by spending it on themselves instead of their children is overblown. First, empirical studies of cash-transfer programs consistently find that these transfers do not result in more spending on alcohol and other drugs (and may indeed have the opposite effect). See David K. Evans and Anna Popova, Cash Transfers and Temptation Goods, 65 Econ Dev & Cultural Change 189, 204–07 (2017). Second, current aid programs often exclude parents suffering from addiction and mental illness because of employment requirements, complicated application procedures, drug tests, and bans on participation by drug felons. For the children of these parents, “something” from a UBI will be better than the “nothing” they may get under the current system. Third, a UBI would not replace the existing legal system that addresses abuse and neglect (even if imperfectly). See Fleischer and Hemel, 2017 Wis L Rev at 1247–48 (cited in note 17).
• 90See Hilary Hoynes, Doug Miller, and David Simon, Income, the Earned Income Tax Credit, and Infant Health, 7 Am Econ J: Econ Pol 172, 195–96 (2015) (finding that the EITC increases mean birth weight and reduces the incidence of low birth weight); Gordon B. Dahl and Lance Lochner, The Impact of Family Income on Child Achievement: Evidence from the Earned Income Tax Credit: Reply, 107 Am Econ Rev 629, 630 (2017) (looking at EITC recipients and finding that increased incomes lead to statistically significant increases in reading and math test scores); Randall K.Q. Akee, et al, Parents’ Incomes and Children’s Outcomes: A Quasi-Experiment Using Transfer Payments from Casino Profits, 2 Am Econ J: Applied Econ 86, 101–08 (2010) (finding that teens raised in households that received UBI-like cash transfers of approximately $4,000 per year were 22 percent less likely to have been arrested at ages sixteen to seventeen than teens in a comparison group whose families did not receive transfers).
• 91See, for example, Johannes Haushofer and Jeremy Shapiro, Policy Brief: Impacts of Unconditional Cash Transfers *17–18 (Oct 24, 2013), archived at https://perma.cc/RT8X-2WSQ. See also Esther Schuering, Social Cash Transfers in Zambia: A Work in Progress, in Degol Hailu and Fábio Veras Soares, eds, Cash Transfers: Lessons from Africa and Latin America, 15 Poverty in Focus 20, 20 (2008).
• 92See Evans and Popova, 65 Econ Dev & Cultural Change at 190 (cited in note 89).
• 93See, for example, Haushofer and Shapiro, Policy Brief at 16–17 (cited in note 91); Schuering, 15 Poverty in Focus at 20 (cited in note 91).
• 94See, for example, Harnett, A California City Is Planning on Giving Money to Some Residents (cited in note 2); Basic Income Project Proposal at *3 (cited in note 5).
• 95Elizabeth Anderson, A Basic Income for All: Optional Freedoms (Boston Rev, Oct 1, 2000), archived at https://perma.cc/2YZJ-KX4X (citation omitted).
• 96See Fleischer and Hemel, 2017 Wis L Rev at 1234 (cited in note 17).
• 97See Timothy Besley and Stephen Coate, Public Provision of Private Goods and the Redistribution of Income, 81 Am Econ Rev 979, 983–84 (1991) (discussing the quality argument without addressing effects from stigma); Hoynes and Rothstein, 11 Ann Rev Econ at 948–49 (cited in note 42) (finding that stigma associated with eligibility-restricted programs “reduces program take-up and the potential reach and benefits of the programs” but “may also help target the programs, maximizing impact by reaching those who stand to benefit the most”).
• 98See Fleischer and Hemel, 2017 Wis L Rev at 1236–37 (cited in note 17).
• 99For a summary of studies suggesting that in-kind benefit programs may screen out needier individuals and families in many circumstances, see Janet Currie, The Take-Up of Social Benefits, in Alan J. Auerbach, David Card, and John M. Quigley, eds, Public Policy and the Income Distribution 80, 112–15 (Russell Sage 2006).
• 100See Anjali Talcherkar, Timeline: History of Addiction Treatment (Recovery.org, Dec 7, 2018), archived at https://perma.cc/XH8Q-Y88Y.
• 101Contract with America Advancement Act of 1996, Pub L No 104-121, § 105, 110 Stat 847, 852–53. The ban remains in force to this day. See 42 USC §§ 422(e), 423(d)(2)(C).
• 102See Jennifer L. Erkulwater, Disability Rights and the American Social Safety Net 102–05, 129–31 (Cornell 2006).
• 103Social Security Administration, Annual Statistical Report on the Social Security Disability Insurance Program, 2016 *25 tbl 6 (Oct 2017), archived at https://perma.cc/278E-SGQ3; Social Security Administration, SSI Annual Statistical Report, 2016 *76 tbl 38 (Nov 2017), archived at https://perma.cc/Z9ZY-33ZZ; Social Security Administration, SSI Annual Statistical Report, 2017 *76 tbl 38 (Sept 2018), archived at https://perma.cc/SPN9-883L.
• 104See Hugo Benitez-Silva, Moshe Buchinsky, and John Rust, How Large Are the Classification Errors in the Social Security Disability Award Process? *5–6, 48 (National Bureau of Economic Research Working Paper No 10219, Jan 2004), archived at https://perma.cc/2ASR-XU3E (summarizing past research and reestimating false negative rate based on new methodology).
• 105For a tentative argument along these lines, see Robert A. Moffitt, The Negative Income Tax and the Evolution of U.S. Welfare Policy, 17 J Econ Perspectives 119, 124–31 (2003) (noting that the effects of a UBI on labor supply may be different for different groups, leaving the net effect uncertain).
• 106See Akee, et al, 2 Am Econ J: Applied Econ at 91 (cited in note 90) (describing the casino dividend program).
• 107See id at 108–09 & tbl 9.
• 108See Damon Jones and Ioana Marinescu, The Labor Market Impacts of Universal and Permanent Cash Transfers: Evidence from the Alaska Permanent Fund *2 (National Bureau of Economic Research Working Paper No 24312, Feb 2018), archived at https://perma.cc/CYK4-7MSY.
• 109See id at *30 tbl 2.
• 110See Social Security Administration, SSI Federal Payment Amounts for 2019, archived at https://perma.cc/8BSM-A2WW; Social Security Administration, Substantial Gainful Activity, archived at https://perma.cc/9SK7-EEHB.
• 111André Gorz, Beyond the Wage-Based Society, in Karl Widerquist, et al, eds, Basic Income: An Anthology of Contemporary Research 297, 298 (Wiley & Sons 2013).
• 112For a literature review, see Mylene Lagarde, Andy Haines, and Natasha Palmer, Conditional Cash Transfers for Improving Uptake of Health Interventions in Low- and Middle-Income Countries: A Systematic Review, 298 JAMA 1900, 1904–08 (2007).
• 113See Sarah Baird, Craig McIntosh, and Berk Özler, Cash or Condition? Evidence from a Cash Transfer Experiment, 126 Q J Econ 1709, 1732–36 (2011).
• 114See, for example, Laura Robertson, et al, Effects of Unconditional and Conditional Cash Transfers on Child Health and Development in Zimbabwe: A Cluster-Randomised Trial, 381 Lancet 1283, 1290–91 (2013) (finding that conditional cash transfers had larger average effect on school attendance but unconditional cash transfers had larger average effect on vaccination rates).
• 115As of about a decade ago, the pregnancy rate in Malawi was 154 per 1,000 females ages fifteen to nineteen and just 57 per 1,000 females in this age range in the United States. See Gilda Sedgh, et al, Adolescent Pregnancy, Birth, and Abortion Rates Across Countries: Levels and Recent Trends, 56 J Adolescent Health 223, 226 tbl 1 (2015) (using 2009 data for Malawi and 2010 data for the United States). The female secondary school enrollment rate is 98 percent for the United States and 37 percent for Malawi. See UNESCO Institute for Statistics, School Enrollment, Secondary, Female (% Gross) (World Bank), archived at https://perma.cc/7UY9-LGC7 (using 2016 data for the United States and 2017 data for Malawi).
• 116See Harnett, A California City Is Planning on Giving Money to Some Residents (cited in note 2).
• 117See Basic Income Project Proposal at *13–15 (cited in note 5).
• 118Cynthia Miller, et al, Testing a Conditional Cash Transfer Program in the U.S.: The Effects of the Family Rewards Program in New York City, 4 IZA J Labor Pol *22 (2015).
• 119See id at *5–7; Cynthia Miller, et al, Effects of a Modified Conditional Cash Transfer Program in Two American Cities: Findings from Family Rewards 2.0 *6–11 (MDRC, Sept 2016), archived at https://perma.cc/8WRV-FJYX.
• 120Indeed, the Family Rewards program, which offered a range of educational, health, and work incentives to families with children, found that CCTs raised household income and reduced poverty but had more ambiguous effects on children’s education, health care utilization, and parents’ work and training. See Miller, et al, Findings from Family Rewards 2.0 at *ES-8 (cited in note 119). These findings arguably suggest that it is the cash transfer—rather than the conditionality—that matters most.
• 121Anne L. Alstott, Work vs. Freedom: A Liberal Challenge to Employment Subsidies, 108 Yale L J 967, 1006 (1999).
• 122Anderson, A Basic Income for All (cited in note 95).
• 123See, for example, Bruce A. Ackerman, Social Justice in the Liberal State 115–20, 129–33 (Yale 1980) (concluding that equality requires giving relatively more resources to disabled individuals); Dworkin, Sovereign Virtue at 77–81 (cited in note 62) (same); Eric Rakowski, Equal Justice 120–21, 142 (Oxford 1991) (same); Richard J. Arneson, Equality and Equal Opportunity for Welfare, 56 Phil Stud 77, 78 (1989) (describing Dworkin’s argument that disabilities should be accounted for in considering equality of resources); G.A. Cohen, On the Currency of Egalitarian Justice, 99 Ethics 906, 917–21 (1989) (describing the egalitarian view of justice for those born with disabilities and offering a criticism of it); Amartya Sen, Equality of What?, in The Tanner Lectures on Human Values 195, 218–19 (May 22, 1979) (arguing for equality of “basic capabilities”), archived at https://perma.cc/WMF2-A6BA.
• 124Anderson, A Basic Income for All (cited in note 95).
• 125See 42 USC §§ 422(e), 423(d)(2)(C).
• 126On the disagreement among the SSA, the IRS, and the courts, see Letter from Karen L. Hawkins, Chair, Section of Taxation, American Bar Association, to David Kautter, Acting Commissioner, Internal Revenue Service, Comments on Information Collection Under Revenue Procedure 99-21 *13–17 (Feb 1, 2018), archived at https://perma.cc/3FLP-U3JV.
• 127See notes 262–64 and accompanying text (discussing the cost of the DI and SSI programs).
• 128See Fleischer and Hemel, 2017 Wis L Rev at 1234 (cited in note 17).
• 129See id at 1209–10, 1239–40.
• 130This argument may or may not apply to welfarism. A welfarist would compare the welfare gains and losses from being overinclusive (that is, providing a basic income to all) and from being underinclusive (leaving some individuals behind).
• 131See Hughes, Fair Shot at 92 (cited in note 8).
• 132See Harnett, A California City Is Planning on Giving Money to Some Residents (cited in note 2).
• 133See Spielman, Plan to Pay 1,000 Residents $1,000 a Month (cited in note 3).
• 134See Stern with Kravitz, Raising the Floor at 201 (cited in note 9).
• 135See Andrew Yang, The War on Normal People: The Truth About America’s Disappearing Jobs and Why Universal Basic Income Is Our Future 165–66 (Hachette 2018).
• 136See Scott Santens, What If You Got $1,000 a Month, Just for Being Alive? I Decided to Find Out (Vox, Nov 14, 2016), archived at https://perma.cc/2L3G-K5D5.
• 137Professors Van Parijs and Vanderborght propose a UBI set at 25 percent of a country’s per capita gross domestic product, which according to their calculations, translated to almost $1,200 per month in the United States using 2017 data. Van Parijs and
  Vanderborght, Basic Income at 11 (cited in note 10).
• 138See Liana E. Fox and José Pacas, Deconstructing Poverty Rates Among the 65 and Older Population: Why Has Poverty Increased Since 2015? *15 (US Census Bureau, Social, Economic & Housing Statistics Division Working Paper No 2018-13, Apr 6, 2018), archived at https://perma.cc/6XW2-GCUC (defining “deep poverty” as below 50 percent of the federal poverty level). See also Kayla Fontenot, Jessica Semega, and Melissa Kollar, Income and Poverty in the United States: 2017—Current Population Reports *47 (US Census Bureau, Sept 2018), archived at https://perma.cc/YJ84-AKVM (stating that the federal poverty level for a one-person unit is $12,752 per year for individuals under age sixty-five and $11,756 for individuals age sixty-five and over).
• 139Fontenot, Semega, and Kollar, Income and Poverty in the United States at *17 (cited in note 138). Ongoing work by University of Chicago economist Bruce Meyer and collaborators casts doubt on some of these Census Bureau poverty statistics. See generally Bruce D. Meyer, et al, The Use and Misuse of Income Data and Extreme Poverty in the United States (National Bureau of Economic Research Working Paper No 25907, May 2019), archived at https://perma.cc/RA2K-NT5N. Very little of our argument depends upon the precise number of people living in Census-defined deep poverty.
• 140Almost but not entirely. As we discuss in Part III.D, assignability and garnishment would mean that some individuals eligible for a UBI do not in fact receive a periodic check.
• 141See US Census Bureau, Number in Poverty and Poverty Rate: 1959 to 2016 (2017), archived at https://perma.cc/PC4T-K259.
• 142See US Census Bureau, Measuring America: How the U.S. Census Bureau Measures Poverty (2017), archived at https://perma.cc/J9DA-MKTB.
• 143The Census Bureau sets a slightly lower poverty threshold for individuals over age sixty-five. This apparently stems from a belief that older individuals have lower consumption needs. For a critique of this view, see Mary Borrowman, Understanding Elderly Poverty in the U.S.: Alternative Measures of Elderly Deprivation *12 (Schwartz Center for Economic Policy Analysis and Department of Economics, The New School for Social Research Working Paper No 2012-3, Apr 2012), archived at https://perma.cc/NL6S-EEKY.
• 144See US Census Bureau, How the Census Bureau Measures Poverty (Aug 16, 2018), archived at https://perma.cc/RB7W-GNKJ.
• 145Since a UBI is a cash benefit, we assume that it would be included in the Census Bureau’s “Census money income” definition. See US Census Bureau, Income: About (Feb 29, 2016), archived at https://perma.cc/JT8M-MKBB.
• 146New York City Housing Authority, Section 8 Housing Choice Voucher Program: Voucher Payment Standards (VPS) (Jan 1, 2018), archived at https://perma.cc/3V49-2CKE.
• 147Memorandum from Lizbeth Silbermann, Director of Program Development Division, Food and Nutrition Service, US Department of Agriculture to All Regional Directors, Supplemental Nutrition Assistance Program, SNAP — Fiscal Year 2018 Cost-of-Living Adjustments *3 (July 28, 2017), archived at https://perma.cc/F4NN-AU8X.
• 148The amount should presumably adjust with changes in price levels and other economic conditions, though we set aside for present purposes the question of what index to use. See generally Daniel Hemel, Indexing, Unchained, 83 L & Contemp Probs (forthcoming 2020) (arguing that all of the available inflation indices—including unchained and chained versions of the Consumer Price Index—provide normatively unattractive benchmarks for year-to-year adjustments in tax-and-transfer policy parameters).
• 149See Murray, In Our Hands at 7 (cited in note 7); Stern with Kravitz, Raising the Floor at 201 (cited in note 9); Yang, The War on Normal People at 165–66 (cited in note 135).
• 150See Akee, et al, 2 Am Econ J: Applied Econ at 90, 101–08 (cited in note 90).
• 151See Wankyo Chung, Hyungserk Ha, and Beomsoo Kim, Money Transfer and Birth Weight: Evidence from the Alaska Permanent Fund Dividend, 54 Econ Inq 576, 581–83 (2016) (finding an 8–14 percent reduction in the incidence of low birth weight—defined as less than or equal to 2.5 kg—infants as a result of the dividend).
• 152See US Census Bureau, QuickFacts: United States, archived at https://perma.cc/3BQX-H8ZF (approximately 327 million US residents); D’Vera Cohn, 5 Key Facts About U.S. Lawful Immigrants (Pew Research Center, Aug 3, 2017), archived at https://perma.cc/NSX8-X8NB (approximately 13 million unauthorized immigrants and temporary lawful residents); GDP (Current US$) (World Bank), archived at https://perma.cc/TFL3-JN2D.
• 153More than 2 million Americans are incarcerated, see Wendy Sawyer and Peter Wagner, Mass Incarceration: The Whole Pie 2019 (Prison Policy Initiative, Mar 19, 2019), archived at https://perma.cc/P3P3-XEEQ, and might (or might not) be excluded from a UBI on that basis. Whether to extend a UBI to people in prison is a challenging normative question that we do not tackle here. We thank André Washington for raising this point.
• 154See General Government Spending: Total, % of GDP, 2015 (OECD), archived at https://perma.cc/EV7D-T96L.
• 155See id. Note that our decision to include children under age eighteen has only a moderate impact on these magnitudes, as persons under eighteen years of age constitute less than a quarter of the US population. See QuickFacts (cited in note 152) (22.4 percent of population under eighteen).
• 156That is, $500 per month approximately four weeks per month and forty hours per week equals $3.125.
• 157Board of Governors of the Federal Reserve System, Report on the Economic Well-Being of U.S. Households in 2017 *21 (May 2018), archived at https://perma.cc/YTA8-2W6N.
• 158See, for example, Scott Santens, It’s Time for Technology to Serve All Humankind with Unconditional Basic Income (Medium, Apr 13, 2018), archived at https://perma.cc/2UFM-MASQ:

  If technological unemployment is the Gordian Knot of the 21st century, UBI is the sword that slices through it. By simply severing the connection between income and work through the unconditional provision of an income for life that’s always sufficient for basic needs, the fear of technological unemployment is eliminated.

• 159US Department of Labor, Bureau of Labor Statistics, The Employment Situation—February 2018 *tbl B-8 (Mar 9, 2018), archived at https://perma.cc/8TUK-ML8W.
• 160See text accompanying notes 37–44.
• 161See Robert J. Lampman, Nixon’s Family Assistance Plan *20 (University of Wisconsin-Madison Institute for Research on Poverty Discussion Paper No 57-69, Nov 1969), archived at https://perma.cc/7ZA6-3Z8S.
• 162See id at *19–20.
• 163See, for example, Murray, In Our Hands at 7 (cited in note 7); Stern with Kravitz, Raising the Floor at 201 (cited in note 9); Yang, The War on Normal People at 166 (cited in note 135).
• 164See Hughes, Fair Shot at 111–14 (cited in note 8).
• 165See French Elections, National Service, and Queens at War 30:45–32:40 (Vox: The Weeds, Apr 26, 2017), available at https://www.stitcher.com/podcast/voxs-the-weeds/e/
  french-elections-national-service-and-queens-at-war-49967422 (visited Jan 25, 2020) (Perma archive unavailable).
• 166For the number of Americans receiving OASI benefits, see Social Security Administration, Social Security Beneficiary Statistics: Number of Beneficiaries Receiving Benefits on December 31, 1970–2018, archived at https://perma.cc/PM8Z-ZVDM.
• 167See IRC § 32(c)(1)(A)(i), (ii)(II).
• 168See US Census Bureau, Current Population Survey, 2016 and 2017 Annual Social and Economic Supplements: Table 3: People in Poverty by Selected Characteristics: 2015 and 2016, archived at https://perma.cc/UU29-Z346. The figure was 18.0 percent for children under the age of eighteen and 11.6 percent for adults ages eighteen to sixty-four. Id.
• 169See Social Security Administration, Fast Facts & Figures About Social Security, 2018 *16 (Sept 2018), archived at https://perma.cc/QNT9-R7X6.
• 170This never-beneficiary population consists primarily of infrequent workers and late-arriving immigrants. See Kevin Whitman, Gayle L. Reznik, and Dave Shoffner, Who Never Receives Social Security Benefits?, 71 Soc Sec Bull 17, 20 (May 2011), archived at https://perma.cc/CR5Z-RF9S.
• 171See Murray, In Our Hands at 109 (cited in note 7) (describing the “simplest solution” for transitioning to a UBI as offering a choice between the current system and a UBI).
• 172See SNAP Helps Millions of Low-Income Seniors *1 (Center on Budget and Policy Priorities, Apr 26, 2017), archived at https://perma.cc/N9HZ-BYMR.
• 173See Social Security Administration, SSI Annual Statistical Report, 2018 *21 tbl 3 (Sept 2019), archived at https://perma.cc/8WH6-8L2S.
• 174See Demographic Facts: Residents Living in Public Housing *1 (National Center for Health in Public Housing, 2016), archived at https://perma.cc/K8YX-AW6Z.
• 175For example, the hypothetical household that the US Department of Agriculture uses to illustrate the SNAP special rules for the elderly and disabled—a two-person household receiving $500 per person in Social Security benefits and $200 of pension income, with $300 in excess medical expenses and $600 in shelter costs—would receive a SNAP benefit of $205 per month in fiscal year 2020. See US Department of Agriculture, Food and Nutrition Service, SNAP Special Rules for the Elderly or Disabled, archived at https://perma.cc/Z5PX-5WPL. A $500 per month UBI with an either-or choice between OASI benefits and the UBI would not affect the household’s cash income, so the loss of $205 in SNAP benefits would be effectively uncompensated.
• 176See, for example, A. Gandara, Letter, Americans Who Paid into Social Security Deserve to Get Their Contributions Back (Las Vegas Rev-J, June 22, 2018), archived at https://perma.cc/JK8R-73K6 (“If you paid in, no matter your income, you should be able to collect.”).
• 177See IRC § 86 (providing that 15–50 percent of social security benefits, depending on income, are excluded from taxable income).
• 178See US Department of the Treasury, Office of Tax Analysis, Tax Expenditures *34 tbl 3 (Oct 16, 2017), archived at https://perma.cc/24AG-TPH4 (fiscal year 2019 estimate). The Congressional Budget Office projects a slightly smaller yield ($34 billion in fiscal year 2020). See Congressional Budget Office, Options for Reducing the Deficit: 2019 to 2028 *242 (Dec 2018), archived at https://perma.cc/8UXW-K9JP. The discrepancy is sufficiently small that it will not substantially affect any of our big-picture estimates in Part III.F.
• 179Some libertarian theorists, however, argue that children are not yet autonomous individuals whose consent is needed in a just society. For a fuller discussion of libertarianism and the treatment of children under a UBI, see Fleischer and Hemel, 2017 Wis L Rev at 1253–56 (cited in note 17).
• 180See Elaine Maag, Who Benefits from Expanding the EITC or CTC? Understanding the Intersection of the EITC and CTC at the Household Level *4 (Urban Institute, July 30, 2018), archived at https://perma.cc/3DS5-LLX2.
• 181Consider how a single parent with two children fares under proposals for a basic income of $1,000 a month per adult as compared to the current system:

  [M]any families [living in poverty] receive more than $12,000 in federal benefits already. SNAP, which covers approximately 45 million Americans, provides a maximum benefit of $6,132 per year for a family of three and an average benefit of around $4,500. The maximum earned income tax credit (EITC) for a parent with two children is $5,572 per year, and the maximum child tax credit is $1,000 per child per year. [Note: $2,000 per year following the December 2017 tax law] . . . For a single-parent family of three eligible for the average SNAP benefit, close to the maximum EITC, and [receiving] some combination of other federal “welfare” programs, . . . to eliminate these benefits and replace them with a $12,000-a-year UBI would actually lower the floor.

  Hemel, Book Review, Bringing the Basic Income Back to Earth (cited in note 43). In contrast, the same family would receive a total of $18,000 a year from a basic income of $500 a month per person including children.

• 182See Murray, In Our Hands at 44–47 (cited in note 7).
• 183Compare, for example, Kevin Milligan, Subsidizing the Stork: New Evidence on Tax Incentives and Fertility, 87 Rev Econ & Stat 539, 541–43 (2005) (finding a strong effect on fertility when the Canadian province of Quebec introduced a policy that paid up to CAD $8,000 to families for having children), with Jeff Grogger and Stephen G. Bronars, The Effect of Welfare Payments on the Marriage and Fertility Behavior of Unwed Mothers: Results from a Twins Experiment, 109 J Pol Econ 529, 540–42 (2001) (finding smaller effects from fertility incentives embedded in US welfare policies).
• 184Several caveats are in order. First, we assume that a UBI in the range of $500 to $1,000 per month that extends to children would offset some of the financial burden of having more kids, but it probably would not make childbirth profitable (in the sense of leaving the parent in a more comfortable financial position than if she did not have an additional child). Second, we are keenly aware of the fact that even high-income adults do not necessarily get to choose the number of children they have. A variety of factors too numerous to list exhaustively here—including but not limited to infertility, adoption restrictions, multiple births, and unplanned pregnancies among individuals who object to or lack access to abortion—can cause individuals to have fewer or more children than they would desire.
• 185The “classical utilitarian” perspective posits that the social planner’s objective should be to maximize the sum of utilities across the population. That might suggest that we should encourage adults to have more children as long as the lives of those children would be sufficiently comfortable that their utility is positive. The “average utilitarian” perspective posits that the social planner should strive to maximize the average utility of members of the population. The implications of the average utilitarian perspective are ambiguous, though it too could militate in favor of a pro–population growth policy if each individual adds more to the available pool of resources than she consumes. See Douglas A. Wolf, et al, Fiscal Externalities of Becoming a Parent, 37 Population & Dev Rev 241, 241–43, 249–51 (2011). A third perspective, “critical-level utilitarianism,” supports population expansion up to the point that the utility of an additional member exceeds some threshold level α. Depending on a number of factors, critical-level utilitarianism may support policies that incentivize or disincentivize fertility. For an overview and analysis, see generally Thomas I. Renström and Luca Spataro, The Optimum Growth Rate for Population Under Critical-Level Utilitarianism, 24 J Population Econ 1181 (2011).
• 186Stern with Kravitz, Raising the Floor at 204 (cited in note 9).
• 187See, for example, Leslie Book, U.S. Refundable Credits: The Taxing Realities of Being Poor, 4 J Tax Admin 71, 86–87 (2018); Elaine Maag, H. Elizabeth Peters, and Sara Edelstein, Increasing Family Complexity and Volatility: The Difficulty in Determining Child Tax Benefits *6–8 (Tax Policy Center, Mar 3, 2016), archived at https://perma.cc/XDM8-E3N3.
• 188See IRC § 24.
• 189See IRC § 21.
• 190See IRC § 129.
• 191See Jacob Goldin and Zachary Liscow, Beyond Head of Household: Rethinking the Taxation of Single Parents, 71 Tax L Rev 367, 373–74 (2018).
• 192Replacing the various child-targeted tax provisions with a UBI also reduces, but does not entirely eliminate, the challenge of matching children with the most appropriate adult to receive the benefit on their behalf. The existing EITC generates incentives to shift qualifying children from one household to another so that the adult who would receive the largest credit from claiming the child does so. Since a UBI does not vary with earned income in the same manner, a UBI reduces this incentive. And while intrafamily conflicts would still arise over which adult should be able to claim a child, an important advantage of rolling various existing cash and near-cash transfer programs together into a single UBI is that families would not have to contend with different child attribution rules under different programs. So even while a UBI does not eliminate child attribution questions, it does serve to streamline those determinations.
• 193Perhaps some individuals experience greater aesthetic or moral distress when they observe a fellow citizen in poverty than when they observe a non-US national in equivalent poverty. We do not share this intuition, though we think it illustrates one of the challenges for the classical liberal public-goods argument for a UBI: the justification for a UBI comes to depend upon potentially fickle public opinions rather than on a robust philosophical basis. See Fleischer and Hemel, 2017 Wis L Rev at 1226–28 (cited in note 17).
• 194Lawful permanent residents who arrived in the United States after August 1996 generally must work for ten years or must serve in the military before they become eligible for SNAP and SSI (though children under the age of eighteen are eligible for SNAP, as are various categories of refugees). See US Department of Agriculture, Food and Nutrition Service, SNAP Policy on Non-Citizen Eligibility (Sept 4, 2013), archived at https://perma.cc/N2K3-DXZW; Social Security Administration, Spotlight on SSI Benefits for Aliens—2019 Edition, archived at https://perma.cc/4PMQ-W3FK.
• 195See Maria Santana, 5 Immigration Myths Debunked (CNNMoney, Nov 20, 2014), archived at https://perma.cc/UWR8-7L2K. Undocumented immigrant children are eligible to attend public schools; undocumented immigrants do receive emergency medical assistance; and undocumented immigrants may receive a limited set of additional in-kind benefits through, for example, the National School Lunch Program; Special Supplemental Nutrition Program for Women, Infants, and Children (WIC); and Head Start. See Tara Watson, Do Undocumented Immigrants Overuse Government Benefits? (EconoFact, Mar 28, 2017), archived at https://perma.cc/SFA7-4VBX.
• 196A separate question concerns the 9 million or so US citizens living abroad. See US Department of State, Bureau of Consular Affairs, Consular Affairs by the Numbers (Dec 2018), archived at https://perma.cc/TFH7-3K2V. The decision to exclude those citizens abroad from a UBI might be justified on the grounds that many are eligible for substantial benefits in the countries in which they live, and so layering a UBI on top of that would amount to a sort of “double payment.”
• 197Compare Murray, In Our Hands at 8–9 (cited in note 7) (not adjusting for household size), with Jessica Wiederspan, Elizabeth Rhodes, and H. Luke Shaefer, Expanding the Discourse on Antipoverty Policy: Reconsidering a Negative Income Tax, 19 J Poverty 218, 229 tbl 3 (2015) (adjusting for household size). Existing welfare programs generally adjust for household size, as did the early negative income tax proposals from the 1960s and 1970s. See, for example, Asimow and Klein, 8 Harv J Legis at 28–29 (cited in note 14); Popkin, 78 Yale L J at 403–11 (cited in note 14); Tobin, Pechman, and Mieszkowski, 77 Yale L J at 4–5 (cited in note 14); Comment, 78 Yale L J at 276–77 (cited in note 14).
• 198See What are Equivalence Scales? *2 (OECD Project on Income Distribution and Poverty), archived at https://perma.cc/7W3F-GMD4.
• 199This is similar to one scenario suggested by Jessica Wiederspan and collaborators in which each household would receive a UBI equal to the federal poverty level. See Wiederspan, Rhodes, and Shaefer, 19 J Poverty at 222, 229–30 (cited in note 197). The federal poverty level as of 2017 was $12,752 for a single person household under age sixty-five and $16,414 for a two-person household under age sixty-five. See Fontenot, Semega, and Kollar, Current Population Reports at *47 (cited in note 138).
• 200The arithmetic is straightforward: $12,000 × 2 − $16,000 = $8,000.
• 201In separate work, one of us has argued that the principle of “marriage neutrality” carries limited normative force. See generally Daniel Hemel, Beyond the Marriage Tax Trilemma, 54 Wake Forest L Rev *101 (forthcoming 2020), archived at https://perma.cc/MKJ9-5434. Insofar as “singles bonuses” or “marriage penalties” might be justified, however, they can be implemented through the tax system rather than through adjustments to the UBI.
• 202Some may also believe symbolic reasons counsel in favor of casting the UBI as an individual benefit. As described in Part II, some of the strongest justifications for a UBI are rooted in concern for individuals as such. Consider libertarianism, under which a UBI shows respect for the separateness of persons, legitimates private property rights by ensuring the Lockean proviso is met with respect to all, and provides a basis for all individuals to consent to a system of private property rights. Styling the UBI as an individual benefit acknowledges the autonomy of all individuals.
• 203Louis Kaplow, Optimal Distribution and the Family, 98 Scand J Econ 75 (1996).
• 204See id at 89–90.
• 205Indeed, the difficulty of applying household-size rules is one oft-cited problem with the federal EITC. See Robert Greenstein, John Wancheck, and Chuck Marr, Reducing Overpayments in the Earned Income Tax Credit *3 (Center on Budget and Policy Priorities, Jan 31, 2019), archived at https://perma.cc/6ND2-E4WP (citing the Treasury Department’s estimate that “70 percent of EITC improper payments stem from issues related to the EITC’s [complicated] residency and relationship requirements”).
• 206Many scholars make a similar argument in favor of taxing wealth in addition to income. See, for example, Lily L. Batchelder, What Should Society Expect from Heirs? The Case for a Comprehensive Inheritance Tax, 63 Tax L Rev 1, 58–59 (2009); Liam Murphy, Why Does Inequality Matter? Reflections on the Political Morality of Piketty’s Capital in the Twenty-First Century, 68 Tax L Rev 613, 628 (2015); David Shakow and Reed Shuldiner, A Comprehensive Wealth Tax, 53 Tax L Rev 499, 500 (2000). But see Miranda Perry Fleischer, Not So Fast: The Hidden Difficulties of Taxing Wealth, in Jack Knight and Melissa Schwartzberg, eds, Wealth: Nomos LVIII 261, 275–85 (NYU 2017) (discussing concerns such as valuation and constitutionality with administering a wealth tax).
• 207See, for example, Matthew Desmond, Evicted: Poverty and Profit in the American City 217–18 (Crown 2016) (describing how one SSI recipient views using layaway to purchase items like furniture and televisions as a logical alternative to savings since having more than $2,000 in her bank account results in a reduction to her benefits).
• 208To be sure, the existing federal income tax does so already insofar as it includes capital income (interest, dividends, capital gains, etc.). But a number of federal tax provisions offset this distortion by allowing households to save substantial sums in tax-free or essentially tax-free vehicles (for example, Roth and traditional IRAs, 401(k) plans, life-insurance policies, Section 529 plans, and owner-occupied housing).
• 209Decreasing one’s benefits as wealth rises is economically equivalent to imposing a tax on that wealth. Consider a UBI in which benefits decrease by 10 cents for each dollar of assets owned over $2,000. As long as Oliver’s assets remain under $2,000, he receives a UBI of, for example, $1,000 a month. But if Oliver manages to save some money and his assets increase to $2,100, his monthly benefit decreases by $10. For every $100 of wealth, he has $10 less and the government has $10 more. This is the very definition of a wealth tax. Cliff-like limits are even worse, imposing the equivalent of wealth taxes that exceed 100 percent. Consider a family that loses several hundred dollars of SNAP benefits each month because their countable assets climb above $2,250, the household “resource limit” to be eligible for SNAP in FY 2018. Memorandum from Silbermann, SNAP — Fiscal Year 2018 at *1 (cited in note 147). A few dollars of increased wealth can lead to a loss of hundreds of dollars’ worth of benefits.
• 210See Do Limits on Family Assets Affect Participation in, Costs of TANF? (The Pew Charitable Trusts, July 7, 2016), archived at https://perma.cc/X4M9-6W45. See also Rebecca Vallas and Joe Valenti, Asset Limits Are a Barrier to Economic Security and Mobility *3–4 (Center for American Progress, Sept 10, 2014), archived at https://perma.cc/VHP9-X2E7.
• 211See Do Limits on Family Assets Affect Participation in, Costs of TANF? at *3 (cited in note 210).
• 212Id.
• 213See generally, for example, Joseph Bankman and David A. Weisbach, The Superiority of an Ideal Consumption Tax Over an Ideal Income Tax, 58 Stan L Rev 1413 (2006).
• 214The rate tables, personal exemption amount, and standard deduction have never varied based on location, and the EITC and child tax credit do not depend on geography either. Some provisions—such as the deduction for state and local taxes (SALT), see IRC § 164—benefit taxpayers in certain jurisdictions more than others, but the SALT provision does not formally discriminate on the basis of location. Two notable exceptions are the Low Income Housing Tax Credit, see IRC § 42(d)(5)(B), and the newly enacted “opportunity zone” provision, see IRC § 1400Z-2, both of which grant favorable treatment to investments in certain census tracts.
• 215See IRC § 36B (tying the credit amount to the premiums for health insurance on the state-established exchanges in the taxpayer’s state).
• 216See US Department of Housing and Urban Development, Housing Choice Vouchers Fact Sheet, archived at https://perma.cc/4LM5-DVPC (stating that housing choice voucher amounts are based on the cost of housing in the “local housing market”).
• 217See US Office of Personnel Management, Pay & Leave: Salaries & Wages: 2018 General Schedule (GS) Locality Pay Tables, archived at https://perma.cc/MUR3-X494 (listing different pay tables for different geographic areas).
• 218See Amy K. Glasmeier, Living Wage Calculation for San Francisco County, California (Living Wage Calculator), archived at https://perma.cc/8SNX-BYF7.
• 219See Amy K. Glasmeier, Living Wage Calculation for Cook County, Illinois (Living Wage Calculator), archived at https://perma.cc/8ZM5-D2Z8.
• 220See Quoctrung Bui and Claire Cain Miller, The Typical American Lives Only 18 Miles from Mom (NY Times, Dec 23, 2015), archived at https://perma.cc/A4R7-RRY9.
• 221See Raven Molloy, Christopher L. Smith, and Abigail Wozniak, Job Changing and the Decline in Long-Distance Migration in the United States, 54 Demography 631, 633 fig 1 (2017).
• 222For more on expensive tastes, see Alstott, 121 Harv L Rev at 478–85 (cited in note 54); Fleischer, 91 BU L Rev at 630–31 (cited in note 54).
• 223On the relationship between wages and cost of living across urban areas, see generally Wendell Cox, The Center for Opportunity Urbanism (COU) Standard of Living Index, 3rd Annual Edition (Dec 2018), archived at https://perma.cc/R6GA-A9K8.
• 224See US Department of Labor, Bureau of Labor Statistics, San Francisco Area Economic Summary *1 (Oct 2, 2019), archived at https://perma.cc/QAK3-93KA.
• 225The Center for Opportunity Urbanism seeks to measure the standard of living across urban areas by adjusting the real average wage for the cost of living. San Jose, California—with high costs but also high wages—comes in first place in this ranking (indicating that workers there can afford more material goods and services than workers elsewhere). See Cox, COU Standard of Living Index at *11 tbl 1 (cited in note 223). San Francisco places in the top third. The implication is that larger UBIs for individuals in high-cost-of-living areas might redistribute in the wrong direction (that is, from individuals in places where job prospects are bleak to individuals in places where high-wage employment opportunities are more plentiful).
• 226See Edward L. Glaeser, Joseph Gyourko, and Raven E. Saks, Why Have Housing Prices Gone Up?, 95 Am Econ Rev 329, 332–33 (2005).
• 227We chose the $5,800 figure arbitrarily. Presumably, lenders would offer an amount less than the nominal sum of future payments to reflect (a) the time value of money and (b) the risk that the borrower will die before the loan is repaid.
• 228See Ackerman and Alstott, The Stakeholder Society at 3–5 (cited in note 32).
• 229To be sure, individuals could use their UBI benefits to make loan payments regardless of whether UBI rights are assignable. The advantage of assignability is that the additional security would likely lead lenders to offer much lower interest rates than they would for unsecured loans.
• 230A related question is whether to allow garnishment of a UBI for back taxes and child support. The arguments in favor of garnishment are straightforward: Why should an individual who already owes money to the federal government receive even more? And why should the federal government abide by a parent’s delinquency in supporting a child? At the same time, garnishment would mean that a UBI would not in fact lift all individuals out of deep poverty because some would see most or all of their UBI diverted.
• 231For this reason, Professors Van Parijs and Vanderborght argue that beneficiaries should not be allowed to assign their UBIs. See Van Parijs and Vanderborght, Basic Income at 10 (cited in note 10).
• 232See Social Security Administration, Social Security Administration Beneficiaries: Social Security Direct Deposit and Check Statistics (Aug 2019), archived at https://perma.cc/65CM-3MR6.
• 233See Social Security Administration, Office of the Inspector General, Unauthorized My Social Security Direct Deposit Changes in Calendar Years 2014 Through 2016 (Aug 2017), archived at https://perma.cc/US6F-NDL3.
• 234See Social Security Administration, Annual Statistical Supplement to the Social Security Bulletin, 2017 *2–3 (Mar 2018), archived at https://perma.cc/E83H-24F8.
• 235Although beyond the scope of this Article, we recognize that this proposal would require additional rules to address the allocation of access to such accounts when parents do not live together.
• 236On the causes and consequences of being “unbanked,” see Michael S. Barr, Banking the Poor, 21 Yale J Reg 121, 130–41 (2004).
• 237See Leora Klapper, Can Universal Basic Income Boost Financial Inclusion and Transparency? (Brookings Institute: Future Development, June 15, 2017), archived at https://perma.cc/68YZ-WA2W.
• 238See Asli Demirguc-Kunt, et al, The Global Findex Database 2014: Measuring Financial Inclusion Around the World *84 (World Bank Group Policy Research Working Paper No 7255, Apr 2015), archived at https://perma.cc/5Q3Q-KDEH.
• 239GDP (Current US$) (cited in note 152) (2018 statistics).
• 240See, for example, Alicia Naumoff, Why Universal Basic Income Should Be Paid in Bitcoin (Cointelegraph, Jan 19, 2017), archived at https://perma.cc/3KBD-88M7.
• 241See About Mannabase (Mannabase), archived at https://perma.cc/GE3M-XN3P.
• 242Ackerman and Alstott, The Stakeholder Society at 51–52 (cited in note 32).
• 243Most traditional cash and near-cash transfer programs (such as SNAP, WIC, TANF, and Social Security) are paid out monthly. See generally, for example, US Department of Agriculture, Food and Nutrition Service, Supplemental Nutrition Assistance Program (SNAP): Monthly Issuance Schedule for All States and Territories, archived at https://perma.cc/Z83C-LN6P. Cash transfers implemented through the federal tax system tend to be distributed annually. Efforts to facilitate more frequent payments to EITC recipients have so far floundered. For a discussion, see generally Steve Holt, Periodic Payment of the Earned Income Tax Credit Revisited (Brookings Metropolitan Policy Program, Dec 2015), archived at https://perma.cc/CHN4-5C3H.
• 244See, for example, Learn About Savings Accounts (Ally Bank, Aug 1, 2016), archived at https://perma.cc/D92D-EPU9.
• 245See Fleischer and Hemel, 2017 Wis L Rev at 1241–44 (cited in note 17).
• 246Jesse M. Shapiro, Is There a Daily Discount Rate? Evidence from the Food Stamp Nutrition Cycle, 89 J Pub Econ 303, 307–08 (2005). See also Karen S. Hamrick and
  Margaret Andrews, SNAP Participants’ Eating Patterns over the Benefit Month: A Time Use Perspective, 11 PLOS ONE 1, 14 fig 2 (2016) (finding that the probability of not eating on a given day rises as more time elapses from the receipt of benefits for SNAP recipients).
• 247See Justine Hastings and Ebonya Washington, The First of the Month Effect: Consumer Behavior and Store Responses, 2 Am Econ J: Econ Pol 142, 156–59 (2010).
• 248See, for example, United States Department of Agriculture, Economic Research Service, Informing Food and Nutrition Assistance Policy: 10 Years of Research at ERS 30 (Dec 2007); Tommy Tobin, Semi-Monthly Benefit Transfers Are a Simple Way to Improve Food Stamps (Forbes, Apr 23, 2018), archived at https://perma.cc/4WEN-6PVJ.
• 249For a discussion, see Damon Jones, Information, Preferences, and Public Benefit Participation: Experimental Evidence from the Advance EITC and 401(k) Savings, 2 Am Econ J: Applied Econ 147, 149–52 (2010).
• 250According to Bureau of Labor Statistics data, more than 56 percent of private-sector businesses pay their employees biweekly or twice a month, and only about 11 percent pay their employees less frequently than biweekly/semimonthly. See Matt Burgess, How Frequently Do Private Businesses Pay Workers, 3 Beyond the Numbers: Pay & Benefits (May 2014), archived at https://perma.cc/3QWZ-MJSF. If there were widespread demand among employees for less frequent payment, then presumably employers would economize on transaction costs by reducing payment frequency. However, insofar as payment frequency concerns arise from stresses and constraints unique to lower-income households, aggregate payroll data is arguably of limited utility.
• 251See text following note 229.
• 252See, for example, Michael Munger, One and One-Half Cheers for a Basic Income Guarantee: We Could Do Worse, and Already Have, 19 Indep Rev 503, 506 (2015) (arguing that using the IRS to implement a UBI will result in little new bureaucracy or rulemaking due to existing infrastructure).
• 253See Social Security Administration, Monthly Statistical Snapshot, July 2018 *1 tbl 1, archived at https://perma.cc/BJG5-7MLZ.
• 254See Beyond Distrust: How Americans View Their Government *58–59 (Pew Research Center, Nov 23, 2015), archived at https://perma.cc/J3FV-NTHX.
• 255See Leandra Lederman, IRS Reform: Politics as Usual?, 7 Colum J Tax L 36, 48–55 (2016).
• 256Dylan Matthews, A Basic Income Really Could End Poverty Forever (Vox, July 17, 2017), archived at https://perma.cc/7UW2-EDLB.
• 257See QuickFacts (cited in note 152); Cohn, 5 Key Facts About U.S. Lawful Immigrants (cited in note 152); GDP (Current US$) (cited in note 152).
• 258See Cohn, 5 Key Facts About U.S. Lawful Immigrants (cited in note 152).
• 259See Congressional Budget Office, Spending Projections, by Budget Account: Updated April 2018 Baseline, archived at https://perma.cc/3D3H-2CU7.
• 260See Congressional Budget Office, Revenue Projections, by Budget Category: Updated April 2018 Baseline, archived at https://perma.cc/N5DT-4V5H.
• 261Fast Facts About Social Security at *16 (cited in note 169).
• 262Id.
• 263Id at *25.
• 264Congressional Budget Office, Spending Projections, by Budget Account (cited in note 259).
• 265See Murray, In Our Hands at 109 (cited in note 7). As of the end of 2017, approximately 40 million OASI beneficiaries received benefits of $500 or more per month. See Social Security Administration, Retired Worker Beneficiaries in Current Payment Status at the End of December 2017, Distributed by Benefit Level, Sex, and Age Group (Jan 8, 2018), archived at https://perma.cc/YP3X-3BHM. Presumably, these individuals would choose to continue receiving OASI benefits and forgo a UBI. The $240 billion figure is 40 million multiplied by a $6,000-per-year UBI.
• 266See text accompanying notes 166–78.
• 267Rick Shenkman, When Did Social Security Become the Third Rail of American Politics? (History News Network, Mar 6, 2005), archived at https://perma.cc/NW9P-3G35.
• 268See Anderson, A Basic Income for All (cited in note 95) (“Disabled people typically require more resources to achieve equivalent freedoms—to move around, to get access to information, and so forth—than those who are not disabled.”). See also note 123.
• 269See text accompanying notes 177–78.
• 270See Congressional Budget Office, Spending Projections, by Budget Account (cited in note 259).
• 271See id; Juliette Cubanski, Tricia Neuman, and Meredith Freed, Issue Brief, The Facts on Medicare Spending and Financing *1 (Henry J. Kaiser Family Foundation, Aug 2019), archived at https://perma.cc/9JCU-MZBA.
• 272See David M. Cutler, Health Care and the Public Sector, in Alan J. Auerbach and Martin Feldstein, eds, 4 Handbook of Public Economics 2143, 2167–69 (Elsevier 2002) (discussing different justifications for public and private provision of health insurance and noting that countries with public health insurance tend to spend less on it).
• 273See, for example, Catherine Ho, California Health Insurance Premiums to Rise an Average of Nearly 9% in 2019 (SF Chron, July 19, 2018), archived at https://perma.cc/5Q2T-7MGZ (noting that the average premium for an individual plan without subsidies in California is $500 per month).
• 274Andy Stern has suggested using a value-added tax (VAT) of 5 percent to 10 percent to offset some of a UBI’s cost. See Stern with Kravitz, Raising the Floor at 212–13 (cited in note 9). But a VAT does not avoid the trade-off between a larger UBI and higher income taxes because a VAT is the rough economic equivalent of an additional tax on labor income. See Gilbert E. Metcalf, Value-Added Taxation: A Tax Whose Time Has Come?, 9 J Econ Perspectives 121, 131–32 (1995). Proposals for a UBI funded by a wealth tax also do not avoid this basic trade-off because a wealth tax is the economic equivalent of a tax on the risk-free component of capital income. See Deborah H. Schenk, Saving the Income Tax with a Wealth Tax, 53 Tax L Rev 423, 435–41 (2000).
• 275See John Schwartz, ‘A Conservative Climate Solution’: Republican Group Calls for Carbon Tax (NY Times, Feb 7, 2017), archived at https://perma.cc/R45S-XMHG (discussing a proposal by James A. Baker III and George P. Shultz—both former Republican secretaries of state—to tax fossil fuels and then give consumers a “carbon dividend”). This proposal demonstrates that one potential advantage of a carbon tax–funded UBI is that it may be able to garner bipartisan support. More recently, several chapters of the College Republicans have endorsed such a proposal. See Lisa Friedman, College Republicans Propose an Unusual Idea from the Right: A Carbon Tax (NY Times, Mar 6, 2018), archived at https://perma.cc/EW2U-BEHZ (reporting that Republican student groups at twenty-three universities joined together to form the coalition “Students for Carbon Dividends” to raise support for a carbon tax).
• 276For the emissions figure, see US Energy Information Administration, What Are U.S. Energy-Related Carbon Dioxide Emissions by Source and Sector? (May 15, 2019) archived at https://perma.cc/Q5E8-HUN5 (2018 figures). The $36 per ton figure reflects the $31 per ton figure in William D. Nordhaus, Revisiting the Social Cost of Carbon, 114 Proceedings Natl Acad Sci 1518, 1520 (2017). Professor Nordhaus’s figure is stated in 2010 dollars, so we update to reflect inflation between 2010 and 2018. See CPI Inflation Calculator at *4 (cited in note 16).
• 277See Internal Revenue Service, Individual Income Tax Returns, 2017 *20 fig A (2019), archived at https://perma.cc/VH4H-US2G.
• 278See National Income: Corporate Profits Before Tax (Without IVA and CCAdj) (Federal Reserve Bank of St. Louis, Sept 26, 2019), online at https://fred.stlouisfed.org/series/A053RC1Q027SBEA (visited Jan 25, 2020) (Perma archive unavailable).
• 279See Norton Francis, et al, Tax Policy Center’s Briefing Book: A Citizen’s Guide to the Fascinating (Though Often Complex) Elements of the Federal Tax System *45 (Urban Institute, 2018), archived at https://perma.cc/X6H8-TVDE.
• 280See IRC §§ 106, 163(h), 199A, 1014.
• 281Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 2017–2021 *37 (mortgage interest), *38 (pass-through income and stepped-up basis), *42 (employer-sponsored health insurance) (May 25, 2018), archived at https://perma.cc/NNU3-2UAD.
• 282We caution that tax expenditure estimates are not the same as revenue estimates, though they tend to be quite close. See Yair Listokin, Tax Expenditure Estimates Approximate Revenue Estimates, 145 Tax Notes 701, 704–05 (2014).
• 283See Part I.B.
• 284In fact, existing welfare programs contain this type of implicit tax. When a taxpayer loses SNAP benefits of $24 because her income increases by $100, that is equivalent to a 24 percent tax. See Congressional Budget Office, Effective Marginal Tax Rates for Low- and Moderate-Income Workers in 2016 *3 (Nov 2015), archived at https://perma.cc/S3X3-9YMH (discussing the phaseout of benefits and its impact on marginal tax rates).
• 285In this hypothetical, once a taxpayer’s income exceeds $40,000, her UBI is fully phased out ($40,000 × 25 percent = $10,000) and she faces no additional implicit tax due to the phaseout.
• 286Indeed, as Professor John Brooks illustrates, the Code already offers several definitions of income from which to choose, including gross income, adjusted gross income, taxable income, alternative minimum taxable income, and various definitions of “modified adjusted gross income.” See Brooks, 71 Tax L Rev at 253 n 2 (cited in note 19).
• 287California’s TANF program, CalWORKs, requires recipients to include in income free rent or utilities, bills paid by others, and “contribution[s] from persons, organizations, or assistance agencies”; many of these transfers would likely count as excludable gifts for income tax purposes. California Department of Social Services, Health and Human Services Agency, Manual of Policies and Procedures: Eligibility and Assistance Standards *376 (definition of income for benefit-eligibility determination includes contributions from others), *404–05 (inclusion of housing and utilities paid by others in income), archived at https://perma.cc/W3G4-DBUM. See also IRC § 102(a) (“Gross income does not include the value of property acquired by gift.”). Further, Section 8 requires the inclusion of child support and alimony payments, 24 CFR § 5.609(b)(7), even though the former have long been excluded from gross income and the latter are now excluded as well. See IRC § 71(a) (providing that alimony is included in gross income), repealed by the Tax Cuts and Jobs Act § 11051, Pub L No 115-97, 131 Stat 2054, 2089 (2017). This divergence is also true on the deduction side; for example, many programs allow deductions for “excess” shelter costs. See, for example, US Department of Agriculture, Food and Nutrition Service, SNAP Eligibility: Frequently Asked Questions, archived at https://perma.cc/CC89-UGJD.
• 288See Internal Revenue Service, 2019 EITC Income Limits, Maximum Credit Amounts and Tax Law Updates (July 10, 2019), archived at https://perma.cc/7E7R-S94L. The threshold is adjusted annually for inflation.
• 289One single dollar of extra income—specifically, the 3,601st dollar—causes one to lose one’s entire EITC, which can be as high as $6,557. On cliff effects more generally and the EITC-investment income cliff in particular, see Manoj Viswanathan, The Hidden Costs of Cliff Effects in the Internal Revenue Code, 164 U Pa L Rev 931, 936–39 (2016).
• 290Hughes, Fair Shot at 92–93 (cited in note 8).
• 291See Felix Salmon, Book Review, Chris Hughes Made Millions at Facebook. Now He Has a Plan to End Poverty (NY Times, Feb 25, 2018), archived at https://perma.cc/YMD2-3J34, reviewing Hughes, Fair Shot (cited in note 8) (noting the “nasty marriage penalty” implicit in Hughes’s proposal).
• 292See Wiederspan, et al, 19 J Poverty at 229 tbl 3 (cited in note 197).
• 293This calculation is based on a federal poverty level of $12,752 for a single individual and $16,414 for a two-person household. 2 × ($12,752 – 33 percent × $20,000) − ($16,414 – 33 percent × $40,000) = $9,090.
• 294The December 2017 tax law reduced the top marginal tax rate on individuals by 3.8 percentage points and the rate on corporations by 14 percentage points. The 3.8 percentage point reduction for individuals was a function of a 2.6 percentage point cut in the statutory rate plus the elimination of the Pease phaseout, which had added up to 1.2 percentage points. See Michael Kitces, Individual Tax Planning Under the Tax Cuts and Jobs Act of 2017 (Kitces, Dec 18, 2017), archived at https://perma.cc/MLA2-RHDD.
• 295See Congressional Budget Office, Effective Marginal Tax Rates at *3 (cited in note 284).
• 296See W. Bradford Wilcox, Joseph P. Price, and Angela Rachidi, Marriage, Penalized: Does Social-Welfare Policy Affect Family Formation? *15 (AEI Institute for Family Studies, July 26, 2016), archived at https://perma.cc/296M-NU6L.
• 297See Memorandum from Silbermann, SNAP — Fiscal Year 2018 at *3 (cited in note 147).
• 298See 2019 EITC Income Limits (cited in note 288).
• 299See IRC § 24(a), (d), (h)(5).
• 300See US Department of Agriculture, Food and Nutrition Service, WIC Program: Average Monthly Benefit per Person *2 (Mar 8, 2019), archived at https://perma.cc/XGW3-GAVT. The comparison should also include the effect of the new income surtax, the effect of which will be modest when income is modest.
• 301See Ashley Burnside and Ife Floyd, TANF Benefits Remain Low Despite Recent Increases in Some States *13 appx tbl 1 (Center on Budget and Policy Priorities, Jan 22, 2019), archived at https://perma.cc/JD9B-WMM7.
• 302See, for example, Max Nesterak, Section Wait: Federal Housing Vouchers Hard to Get, Hard to Use (Minn Public Radio, July 18, 2018), archived at https://perma.cc/3EQA-E33X.
• 303For Medicare eligibility requirements, see 42 CFR § 406.20.
• 304Felix Salmon, Universal Basic Income Is Not the Solution to Poverty (Slate, July 10, 2018), archived at https://perma.cc/4BT2-7PNL.
• 305See Silvia Killingsworth, Felix Salmon, “Fusion Money,” and Floating Upward (The Awl, Jan 31, 2018), archived at https://perma.cc/4NNW-QJ6H.
• 306Experience with the advance EITC and the high administrative costs of current means-tested programs demonstrate these challenges. See note 243 (discussing difficulties with distributing EITC payments throughout the year).
• 307See IRC § 151(d)(3) (phaseout of personal exemptions).
• 308See IRC § 24(b) (income-based limitation on the child tax credit); IRC § 24(h)(1)–(3) (income-based limits applicable for 2018–2025).
• 309See Conor Clarke and Edward Fox, Note, Perceptions of Taxing and Spending: A Survey Experiment, 124 Yale L J 1252, 1277–78 (2015).
• 310See, for example, IRC § 36B(b) (monthly payment of Affordable Care Act premium assistance credits); IRC § 36B(e)(3) (significant regulatory authority over premium tax credits allocated to Secretary of Health and Human Services).
• 311See US Department of Agriculture, Food and Nutrition Service, Supplemental Nutrition Assistance Program (SNAP): What Can SNAP Buy? (Sept 4, 2013), archived at https://perma.cc/YB7M-24XR.
• 312See 24 CFR § 982.401(d)(2)(i).
• 313See IRC § 24(d), (h)(5)–(6). The refundable portion of the tax credit is limited to 15 percent of earned income in excess of $2,500. $(11,833 − 2,500) × 15 percent = $1,400.
• 314See Benjamin M. Leff, EITC for All: A Universal Basic Income Compromise Proposal *20–30 (unpublished manuscript, 2019) (on file with authors); Dylan Matthews, A California Child Allowance (Medium: Economic Security Project, Aug 22, 2017), archived at https://perma.cc/NVT9-U4BV.
• 315See text accompanying note 295.