If We Can Afford Our Current Welfare System, We Can Afford Basic Income

Max Ghenis
Jan 24, 2016 · 7 min read

Basic income (BI) is getting a lot of press these days. From Switzerland’s upcoming referendum to give each citizen 2,500 francs per month, to GiveDirectly’s research upending the philanthropy world, to Finland’s plan for a 2017 BI pilot, more are warming to the idea that just giving people money may be the solution to poverty.

Many dismiss basic income as unaffordable. The reality is that basic income is only a form of redistribution, independent of the amount to be redistributed. It is therefore equally affordable to any existing social safety net. Specifically, budgets for existing programs can fund BI via these revenue-neutral steps:

  1. Replace current antipoverty programs dollar-for-dollar with cash transfers.
  2. Reform income eligibility requirements to ensure people are always better off upon earning an extra dollar.
    (1) and (2) can be consolidated into a “negative income tax” (NIT)
  3. Implement negative income tax as an equivalent basic income.

That is, NIT should be attractive to those looking to simplify the suite of antipoverty programs and improve work incentives. As I’ll argue below, BI and NIT are effectively equivalent, so BI would be just as attractive. This article lays out the conceptual case for getting to basic income without major changes to federal budgets or tax burdens.

Replacing the antipoverty bureaucracy

The US federal government currently spends over a trillion dollars per year helping the poor through over 100 programs (some shown below). Some are well-known, such as Supplemental Nutrition Assistance Program (SNAP, f.k.a. food stamps), housing assistance, Temporary Assistance for Needy Families (TANF, f.k.a. welfare), and the Earned Income Tax Credit (EITC). The majority are much smaller, each providing particular goods or services to those in need, and many more programs provide help at the state and local levels.

Source: US House Committee on Ways and Means

This assortment of programs creates three key issues:

  1. Overhead. Even large federal programs typically have considerable administrative costs. For example, a 2012 study from the Center for Budget and Policy Priorities found that SNAP and housing vouchers each cost 9–10% in overhead expenses. The same study found that the EITC (a cash transfer) cost under 1%.
  2. Burden on recipients. Administrative dollar figures fail to capture the time-consuming and demeaning experience for recipients, who must learn about, apply for, and comply with the tomes of forms and provisions of each program.
  3. Paternalism. When programs dictate the specific uses of funds, they ignore the varying preferences of the poor. For example, some may want to spend more on housing, while others may prefer to spend more on food. By purchasing the goods and services for them in fixed quantities, we remove their option to participate in markets according to their tastes. This results in what economists call deadweight loss, a form of inefficiency from lack of market equilibrium.

Some of these programs will likely be more effective than their cash value — especially those serving the physically disabled and mentally ill — and should remain intact. But most could be more easily administered and humanely received as cash grants. Assuming utilization rates remain constant, this is clearly revenue-neutral. Unfortunately, this not always the case due to stigma and difficulty of application and compliance (e.g., SNAP has ~75% utilization); if an easier cash experience attracts more recipients, benefits could be adjusted to remain revenue-neutral in aggregate.

Eliminating welfare cliffs

Replacing current benefits dollar-for-dollar with cash leaves a big problem on the table: welfare cliffs (or welfare traps). A welfare cliff occurs when a recipient, upon earning a marginal dollar, loses a means-tested benefit (one that goes away if you earn over a certain amount) of greater value than the dollar. This creates work disincentives which can be sizable. For example, as depicted in the below graph, a single mom in Pennsylvania collects more income+benefits earning $29k than she does earning $69k.

Source: Pennsylvania Department of Public Welfare

Non-cash benefits are particularly prone to welfare cliffs, since they can be real goods and services not easily divisible (e.g. childcare). Cash, on the other hand, can be distributed to eliminate cliffs, plus smooth out the curve. A smooth curve means that anyone can easily predict their total income+benefits as a function of their income, spending less time optimizing for benefits. To be revenue-neutral, some people will be worse off with a smooth curve (e.g. those earning $29k getting maximum benefits), and others will be better off (e.g. those who lose $6k of benefits after earning just over $29k). But every dollar earned will lead to improved livelihood.

Replacing non-cash benefits with cash, and smoothing out the curve, has been formalized into a policy called negative income tax (NIT). This policy provides a government payment to those below a certain income level (called the phase-out level) equal to a fraction of the difference between their income and the phase-out level (more on that below). Nobel prize-winning economist Milton Friedman —known as one of the leaders of the Chicago school of economics — popularized the policy when he advocated it in his 1962 book, Capitalism and Freedom.

While the proposal never passed, a partial solution called the (previously mentioned) Earned Income Tax Credit (EITC) was implemented instead in 1975. The EITC is now widely cited as one of the most effective antipoverty programs in the US, having lifted 6.2 million people out of poverty in 2013.

The negative income tax goes further by covering all below the phase-out level (EITC targets working families with children), ensuring full utilization (15–25% of households eligible for EITC don’t claim it), and increasing impact by absorbing budgets from other programs.

From negative income tax to basic income

By replacing the onerous bureaucracy of antipoverty programs and eliminating welfare cliffs, negative income tax would go a long way to improving the lives of the poor. However, some differences make basic income more compelling:

  1. Frequency of distribution. Like other tax rebates, negative income tax payments are distributed annually. Basic income can be distributed monthly or more often, which aligns better to its purpose of satisfying basic ongoing needs.
  2. Simplicity. Negative income tax still requires some math to determine the payment based on total earnings. Recipients of basic income know exactly how much they get every month.
  3. Social cohesion. A payment received by all equalizes and unifies a society, creating concrete consequences to government revenue and spending policies. The Nordic countries, for example, have favored universal government programs funded by broad-based taxes.

Fortunately, for any given negative income tax structure, an equivalent basic income structure exists. Consider a NIT with a phase-out level of $30k, a 50% marginal withdrawal rate, and a 50% income tax (unrealistic, but this simplifies calculations). This means that someone earning $0 will receive $15k (($30k - $0) * 50%), someone earning $30k will neither receive payments nor pay income taxes, and someone earning $60k will pay $15k in taxes (($60k - $30k) * 50%). We can visualize this as a chart mapping gross (earned) income to net (after taxes and payments) income (similar to the Pennsylvania welfare cliff chart):

Source: Model by Max Ghenis

As the title gives away, this same mapping from gross income to net income is achieved with a basic income of $15k and a flat income tax of 50%. The US progressive tax code complicates things a bit, but it’s provable that any negative income tax scheme — regardless of the income tax rates — can be implemented equivalently as a basic income (via income taxes, in some cases changing rates). More money passes through government, but for affected taxpayers, it’s money out one pocket and back in the other.

As the world moves to an age of increasing globalization and automation, more will be drawn to the benefits of basic income. Many questions remain about the idea, for example, its effectiveness, optimal payout, and whom it should include. While studies have addressed some questions (e.g. cash transfers don’t reduce work incentives), the various experiments planned for the next couple years will yield even better data.

But if basic income is achieved via the above strategy — replace existing non-cash programs with cash, smooth out the payout curve to avoid welfare cliffs, and substitute negative income tax with basic income — affordability does not belong on the list of concerns. If the BI produced by this approach is too low, that is an indication that the existing welfare state is too stingy; if it doesn’t adequately address the needs of society, then neither do today’s programs. Criticisms of the safety net’s size (in either direction) are orthogonal from the notion that basic income is a more efficient means of distribution.

Poverty is not a vague incurable condition, but one that can be directly eradicated via basic income. Finland’s proposal to move 100% to BI is admirable, but other countries can start with small steps like expanding EITC and reforming means tests. With basic income attainable, now is the time to seriously push toward its implementation.

Basic income

Articles discussing the concept of unconditional basic income (UBI)

Max Ghenis

Written by

Data scientist, economic policy nerd, basic income advocate, YIMBY

Basic income

Articles discussing the concept of unconditional basic income (UBI)