The negative income tax (NIT) is a mechanism for integrating income support and income taxation into a single administrative structure such that persons below a defined income threshold receive payments from the state, the amount of which decreases as earned income rises, while persons above the threshold pay positive tax. The proposal was formalized by Milton Friedman in 1962, though its intellectual antecedents include earlier work by Juliet Rhys-Williams and, implicitly, the refundable tax credit mechanisms that had appeared in various national tax codes. Unlike Universal Basic Income, which pays all recipients the same amount regardless of income, the NIT targets its transfers toward lower earners while maintaining a work incentive through a partial phase-out rate: for every additional dollar earned, the transfer is reduced by less than a dollar, ensuring that working always improves net income. This structural design makes the NIT a fundamentally different policy instrument than a flat UBI, even when both are described as guaranteeing a minimum income.
The mechanics are straightforward. The government sets a guaranteed minimum income level — say, $15,000 per year for a single adult — and a phase-out rate — say, 50 percent. An individual with zero earnings receives the full $15,000. An individual earning $10,000 has their transfer reduced by $5,000 (50 percent of $10,000), leaving them with $10,000 in earnings plus $10,000 in transfer for $20,000 total. An individual earning $30,000 receives no transfer (since 50 percent of $30,000 equals the $15,000 guarantee), and earns positive income above the break-even point. The mathematics ensure that net income always rises with earned income, eliminating the poverty trap but not fully eliminating the work-disincentive effect, since the marginal return to earning is reduced by the phase-out rate across the entire support range.
The NIT's appeal spans political divides in ways that reveal both its strengths and its tensions. For conservatives and libertarians, the NIT replaces a fragmented, paternalistic welfare apparatus — food stamps, housing vouchers, Medicaid, TANF, SNAP, and dozens of categorical programs each with their own eligibility rules and administrative overhead — with a single cash mechanism that respects individual choice and eliminates the supervisory relationship between welfare bureaucracy and recipient. Friedman's version was explicitly designed as a rollback of the New Deal administrative state. For center-left economists, the NIT's work-incentive structure makes it more compatible with labor market participation than a flat basic income, addressing the labor supply concerns that critics of UBI raise. For antipoverty advocates, the NIT targets its benefits toward those who need them most, using means-testing logic while eliminating the cliff effects and benefit traps that make existing means-tested programs so punishing to recipients who find work.
The most sophisticated version of the NIT argument engages the poverty trap problem with precision. Under current U.S. benefit structures, a low-income person who increases their earned income may simultaneously lose Medicaid eligibility, housing voucher value, food stamp benefits, and child care subsidies, facing effective marginal tax rates that can exceed 70 to 80 percent on additional earnings. This creates a structural disincentive to formal employment that is often attributed to individual laziness but is in fact a product of benefit program design. The NIT's smooth phase-out eliminates these discontinuous cliff effects, replacing them with a predictable linear reduction that preserves work incentive throughout. In this respect, the NIT is better described as welfare state rationalization than as welfare state replacement.
The empirical record from the 1970s U.S. income maintenance experiments — conducted in New Jersey, rural Iowa and North Carolina, Gary (Indiana), and Seattle-Denver — provides the most extensive real-world evidence base. These experiments randomly assigned families to NIT-like transfer programs at varying guarantee levels and phase-out rates and tracked labor supply, family stability, and other outcomes over three to five years. The headline finding was a modest reduction in hours worked, averaging 5 to 7 percent for primary earners and 17 to 22 percent for secondary earners (predominantly wives). Critics took these findings as evidence that guaranteed income reduces work; later analysts argued that the reductions reflected rational choices to invest more time in job search (which raises long-run earnings) and to exit exploitative employment, and that the experiment's short duration prevented observation of the longer-run labor supply recovery that income security enables. The experiment also found, controversially, an increase in marital dissolution — a finding later attributed to statistical artifacts and measurement problems but which damaged the political case for the NIT at the time.
The Earned Income Tax Credit (EITC), enacted in the United States in 1975 and subsequently expanded, is a partial NIT variant: it provides refundable credits that rise with earned income up to a phase-in maximum, then remain constant, then phase out. The EITC's work requirement in the phase-in range makes it distinctly different from a pure NIT — it is conditional on work rather than universal — but it demonstrates that a refundable, earnings-sensitive transfer mechanism can be administered effectively through the tax system at scale. Research consistently finds the EITC to be one of the most effective U.S. antipoverty programs, with strong positive employment effects, particularly for single mothers, during the phase-in range. Its limitations — the phase-out range creating high effective marginal rates for moderate earners, coverage gaps for childless adults, complexity of the credit rules — point toward the NIT's fuller rationalization as a possible improvement.
The NIT, viewed through Law 4, is a structural mechanism for expanding the effective freedom of low-income workers by eliminating the poverty trap — the structural constraint that converts effort into net loss rather than net gain. Its design logic is: the material floor expands agency; the smooth phase-out preserves the agency-expanding effect of productive work; the single administrative mechanism reduces the surveillance and conditionality that diminish agency in categorical welfare programs. Whether this logic delivers its promised outcomes depends on the guarantee level (too low to matter), the phase-out rate (too steep recreates the poverty trap; too shallow is fiscally expensive), the institutions surrounding it (a good NIT embedded in a bad housing market helps less), and what programs it replaces.