A Negative Income Tax, One Step at a Time

The negative income tax (NIT) sometimes seems like the carbon tax of social policy. Both are irresistibly appealing to economists and have long pedigrees. Both are supported by blindingly persuasive logic. Yet neither policy seems capable of mustering much political support in 21st-century Washington politics. I see two things as essential in repackaging the NIT for today’s America.

The first essential is to recognize the reality of path dependency — the need to start from where we are, not from a clean slate, and take things one step at a time.  Gerald Gaus calls that approach “exploring the adjacent possible.”

The second essential is to present the NIT in a value framework that has broad appeal across the political spectrum. As things stand, the NIT has about an even balance of progressive and conservative skeptics, yet properly implemented, it offers much that is in harmony with the values of both sides.

Here, then, are some ideas for nudging the NIT along from a merely an elegant concept toward something more concrete and workable.

One step at a time

When I moved to Vermont years ago, I found that the locals had a wealth of tongue-in-cheek jokes, sometimes at their own expense. In one, a New Yorker stops his car to ask a farmer how to get to the tiny town of North Pomfret. After thinking for a minute or so, the Vermonter answers, “You can’t get there from here.”

Path dependency is the social scientist’s way of saying, “You can’t get there from here.” The steps you took to get where you are now constrain what you can do in the future.

In either its rustic or academic version, path dependency is highly relevant to the design of an NIT. When Milton Friedman first started thinking about the concept in the late 1930s, the modern welfare system did not exist. Roosevelt’s New Deal had introduced some ad hoc jobs programs like the Civilian Conservation Corps and the Works Progress Administration, but there was little else. If an NIT had been introduced then, it could have been drawn on a blank slate.

Things are different today. Each of our many poverty programs has its own constituents, its own staff who depend on it for their jobs, and its own rent-seeking hangers-on. As a textbook exercise, throwing out the whole lot and replacing them with an NIT has a definite appeal, but it doesn’t look likely to happen.

Could an NIT simply be added on top of our current welfare system?  Some say yes. Consider, for example, a proposal called Guaranteed Income for the 21^st Century (GI21), about which I have written previously. Its authors designed GI21 as a Universal Basic Income with a phase-out, which made it, in effect, an NIT. However, they were specific that GI21 was to supplement existing poverty programs, not replace them. They were less specific as to just how the phase-out rates of GI21 were to interact with those of SNAP (also known as food stamps), the Earned Income Tax Credit (EITC), and so on. By my calculations, however, no matter which of several variants were chosen, it was clear that GI21 beneficiaries would keep little, if anything, of each additional dollar they earned, especially in the critical near-poor range of $10,000 to $50,000 per year. GI21 would have boosted the incomes of many families, but it would have been anything but work-friendly.

But, if starting from a blank slate is impossible due to path dependency, and if an add-on NIT would be destructive of work incentives, what is to be done? I see at least three ways forward that would not require the immediate repeal of current programs, but only changes in their phase-in and phase-out parameters.

One way would be simply to modify existing programs. Consider the EITC, the largest cash transfer program we have now. As of 2023, the EITC gives a credit of 34 cents per dollar earned to a family with two parents and one child, up to an income of $11,750. Starting at an income of $28,120, the credit phases out at a rate of 16 percent. Anybody in between gets the full credit amount, $3,995. It is, in effect, an NIT married to a wage subsidy for low-income workers.

A few steps would bring the EITC closer to an NIT. One would be to raise the minimum credit to a positive amount, say, $5,000 per family. Minnesota’s Working Family Credit, which is part of its version of the EITC at the state level, offers a possible model.  Another idea would be to broaden eligibility by making the EITC available to married workers on terms comparable to those who are unmarried, thus fixing the program’s existing marriage penalty. Paying benefits monthly, instead of once a year at tax time, would help, too. At the same time, other benefits that are also delivered through the tax system could be expanded or modified.

A second approach would be to offer an option to convert selected in-kind benefits to cash. For example, SNAP beneficiaries could opt to receive cash rather than the current electronic benefits transfer cards, which can be used only for food. Princeton economist Diane Whitmore Schanzenbach estimates that between 20 and 30 percent of food stamp recipients would spend less on food than their food stamp benefit amount if they received cash instead of stamps. She interprets that as meaning that they would be better off with cash. On average, she thinks that those stamp recipients value their total benefits at 80 percent of their face value. These estimates suggest that a cash option would likely have a good take-up even if the amount of cash offered were less than 1-to-1 with the face value of the in-kind benefits. Part of the budget savings from the lower maximum benefit could be used to improve work incentives by reducing the phase-out rate to something below its current 24 percent.

A third, more radical, approach would be what I call total income phase-out (TIP). Regardless of how many cash and in-kind welfare programs stayed on the books, TIP would introduce a harmonized phase-out system that would apply to all of them. Applied consistently, it would be mathematically indistinguishable from an NIT.

The central aim of TIP is to ease the biggest source of work disincentives for the poor, namely, the current practice that makes the phase-out rates of various programs additive. Today, for example, if you are in the phase-out zone for both EITC (16 percent) and SNAP (24 percent), 40 cents of every extra dollar you earn disappears in benefit reductions. The numbers worsen as a family qualifies for more programs, such as Medicaid, housing, or school lunches. As my earlier research has shown, these work disincentives hit the near-poor hardest. People who have worked their way up from nothing to the threshold of self-sufficiency do not infrequently encounter “disincentive deserts” where more than 80 cents of each extra dollar earned is taken away in benefit reductions. And that is before you factor in commuting costs, child care, or work-related clothing.

TIP would replace the additive phase-outs with a uniform rate of, say, 35 percent applied to total net income. Total net income would include the sum of earned income, cash benefits, and the cash value of in-kind benefits, minus payroll taxes, income taxes (if any), and work-related expenses. It might still be worth consolidating or cashing out some of today’s alphabet-soup of in-kind public assistance programs, but that would not have to be done all at once. Even if the TIP approach were, at first, used to harmonize the phase-outs of just a single pair of programs, such as EITC and SNAP, it would be a step forward. Ultimately, if applied to all public assistance programs, the overall profile of household earnings and total net incomes under TIP could be made mathematically indistinguishable from a classic NIT.

Value framing

Let’s turn now to the second big issue that must be addressed if an NIT is to become a politically realistic option in today’s America. That is to present the concept in a way that appeals to the values of voters and policymakers across the political spectrum. That is not as impossible a task as it might seem, given the current degree of polarization. Although progressive and conservative critics of the NIT often use different language, a careful examination shows that many of their underlying concerns are similar. Consider, first, the progressive point of view, which tends to emphasize the values of inclusivity, accessibility, opportunity, and empowerment.

Compared with our current welfare system, an NIT is hands-down more inclusive. Despite all the billions spent, the current system fails to help everyone who needs and deserves it. Even if you draw a line at undocumented immigrants (an admitted point of contention), a classic NIT would exclude fewer deserving recipients.

Accessibility is a closely related virtue of the NIT. The current system too often denies access to people who ought to be eligible because of excessive red tape, lack of English literacy, lack of internet access, processing backlogs, waiting periods, and a host of other reasons. A fully implemented NIT would make feasible automatic enrollment — the ultimate in accessibility. Short of that, each step toward an NIT would be a step toward simplicity and universality.

Next, progressives lament that although America is supposed to be a land of opportunity, the current welfare system can make that promise a sham. Work disincentives, discussed earlier, are a case in point. What good is it to offer someone the opportunity to work longer hours, seek a promotion, or apply for a higher-paying job if the lion’s share of any extra earnings would disappear into the black hole of benefit reductions? In contrast, an NIT promises to make opportunities for self-help a reality.

Finally, an NIT opens a path to empowerment and autonomy. Its inclusiveness and simplicity clear away the paternalism of a system that, in the name of the supposed best interests of beneficiaries, too often restricts their ability to lead their lives as they see fit. Those restrictions make welfare disempowering and stigmatizing. Food stamps, housing vouchers, even Medicaid suggest distrust of those they intend to help. Cash is liberating.

And what about conservatives? Although the NIT was famously championed by Milton Friedman, one of their intellectual icons, it has no shortage of conservative skeptics. Yet, on closer examination, an NIT is in harmony with key conservative values of work, family, freedom, and cost-effectiveness.

Perhaps the most stubborn myth among conservatives is that the poor are lazy, and that giving them cash would make them lazier still. But that myth crumbles under closer examination. For one thing, the “lazy poor” are, in reality, few in number. Data show that more than half of those who benefit from programs like SNAP and Medicaid already work at least part-time. Of those who do not work, the great majority cite caretaking duties, school, or poor health as the reason. If conservatives really want people to achieve self-sufficiency, they should use fewer sticks and more carrots. A properly designed NIT would encourage self-sufficiency by letting workers with low incomes keep a larger share of each extra dollar they earn.

Next, the current welfare system not only penalizes work, but also marriage and the family, which many conservatives (not without justification) view as bedrocks of economic security. The best way to lift the marriage penalties in the current welfare system would be to cash out welfare benefits and consolidate programs wherever possible – exactly what an NIT would do.

Freedom is the third key value that should attract conservatives to an NIT. What, do you say anything with the word “tax” is a threat to freedom? That is a narrow definition. There is a broader “republican” view of freedom, championed by Friedrich Hayek and other writers that conservatives revere. Republican freedom emphasizes the ability of individuals to live their lives according to their own decisions and plans, rather than being subject to the will of others. Compared with today’s welfare system, the NIT greatly enhances individual freedom in the republican sense. Cash gives greater latitude to live your own life as you choose than do food stamps or housing vouchers. And, since everyone is eligible for the NIT, the right to enrollment does not depend on the whim of some social worker.

Finally, morality aside, conservatives see a great deal of waste in our current welfare system. An NIT would enhance the cost-effectiveness of social programs by reducing administrative costs, by making sure that fewer of the poor fall between the cracks, and by encouraging work and self-sufficiency.

Running through these lists, we see how progressive and conservative values converge when applied to an NIT. Sure, the language may differ. But aren’t the autonomy and empowerment that concern progressives just a different way of expressing what conservatives call republican freedom? Aren’t opportunity and work incentives two sides of the same coin? Isn’t the inherent administrative simplicity of an NIT both a way of enhancing access and at the same time making social insurance more cost effective? The same features of today’s welfare system that upset both progressives and conservatives ought to make both of them love an NIT.

Getting there from here

I started by calling a negative income tax the carbon tax of social policy – attractive on paper but never enacted. In fact, though, an NIT and a carbon tax differ in an important way that makes the NIT even harder to implement. The difference is this: A carbon tax would improve environmental performance whether it were the sole emissions reduction policy or an add-on to existing regulations and subsidies. In contrast, a conventional NIT added to the already crushing work disincentives of our current welfare system would be a disaster.

It is utopian to think of wiping the slate clean and starting from scratch with a stand-alone NIT. Instead, NIT advocates (and advocates of the NIT’s close cousin, a UBI) need to figure out a step-by-step approach. But there is a catch: Not any random series of steps will work. What we need is a sequence of reforms, each of which meet three criteria:

1.  Each step should be small enough to enact with a single piece of legislation, regulatory action, or executive order. A refundable Child Tax Credit is an example of one such step, and one that is not totally beyond the realm of legislative feasibility.

2.  Each step should improve both the lives of the poor and the performance of the economy even while other programs remain in place. For example, given the evidence that people value cash more than equally costly in-kind benefits, a cash-out option for SNAP could improve the lives of the poor while cutting the burden on federal taxpayers.

3.  Each step should be a win-win as viewed from values of both progressives and conservatives. The progressive values of inclusivity, accessibility, opportunity, and empowerment are, after all, complementary to the conservative values of work, family, freedom, and cost-effectiveness. With a little imagination, each step toward reform can be sold as a plus to both sides.

This essay has been too short to offer a complete road map to repackaging the NIT for today’s America. I have tried, though, to give examples that make such a project look at least a little more feasible.

Adapted with permission from "A Negative Income Tax for Today's America," published in Hypertext, a newsletter of Niskanen Center. Image generated by Gemini.