The minimum wage and the Earned Income Tax Credit (EITC) are often discussed in the same breath. Both are pitched as ways to help low-income workers. Both are routinely defended and attacked as if they were interchangeable. They are not. They are structurally different machines, with different inputs, different costs, and different sets of people they reach.

Understanding the difference is the first step in deciding how much of each — if any — the United States should use.

Same goal, different machine

Both policies share a simple motivation: the belief that a full-time worker shouldn't live in poverty. Where they diverge is in how they try to produce that outcome.

  • The minimum wage sets a legal floor on the price of an hour of labor. The cost of the policy is paid by employers, and through higher prices, by consumers. It helps every worker whose pre-mandate wage was below the floor — and has no effect on workers who would have been paid above it anyway.
  • The EITC is a refundable federal tax credit. Workers receive it when they file taxes, and it phases in as earnings rise, reaches a plateau, and phases out. The cost is paid by taxpayers through the federal government. It helps workers whose household income is below a defined threshold, scaled by the number of children.

That is a large structural gap. One policy intervenes at the moment a job is offered. The other intervenes at the moment taxes are filed. Those different intervention points produce different side effects.

Mechanics, step by step

Consider a single worker — Maria, a 19-year-old starting her first job in a mid-cost city, producing roughly $12 of value per hour in a retail position. What happens to Maria under each policy?

Under a $15 minimum wage: the employer looks at Maria's expected productivity, compares it to the mandated $15 cost, and concludes that the role doesn't make sense. Maria doesn't get the job. Or Maria gets fewer scheduled hours. Or Maria gets the job, but the employer raises prices, invests in kiosks that reduce her tasks, or hires a 26-year-old with retail experience instead. Maria's pay-per-hour, if she works at all, is $15. Her probability of being hired is lower.

Under the EITC: the employer hires Maria at $12 per hour, because $12 worth of labor is still a profitable transaction. Maria's annual earnings are modest — roughly $18,000 at part-time hours — but the EITC adds a meaningful refundable credit on top of that. She takes home more than her wage alone would provide, and she keeps the job that builds her skills toward a higher-paying future role.

The minimum wage raises the pay of the workers who keep their jobs. The EITC raises the pay of the workers who take the jobs.

That sentence is the clearest structural contrast. A wage floor improves outcomes conditional on being hired. A tax credit improves outcomes conditional on being employed. The two policies answer slightly different questions.

Who actually gets helped

Both policies reach low-income workers, but they do so in strikingly different patterns.

The EITC is targeted by household income and family size. A single mother with two children gets a much larger credit than a 19-year-old second earner in a non-poor household — even if they earn the same hourly wage. This makes the EITC progressive at the household level.

The minimum wage is targeted by hourly wage, not by household income. CBO analyses and studies by the Economic Policy Institute have repeatedly shown that a large share of minimum-wage earners live in non-poor households — they are teenagers, spouses, or part-time workers adding a second income. As an anti-poverty instrument, the minimum wage is not well-targeted. Some of its benefits flow to middle-class households.

An uncomfortable nuance This does not make the minimum wage useless. It means the law is doing something other than pure poverty alleviation — namely, raising wages for a broad class of lower-paid workers, many of whom are not poor by household measures. Whether that's a good use of the instrument depends on what you're trying to accomplish.

Complements, not substitutes?

The clean Friedman argument treats the EITC as a strict replacement for the minimum wage. But a lot of serious modern research has argued they work best together — that the minimum wage can actually protect the EITC from being partially captured by employers.

Here's the concern. When the EITC boosts a worker's total compensation, some of that boost can, in theory, be absorbed by the employer through slightly lower wages. (The logic: if a worker is willing to take a job at a total after-tax pay of $X, and the EITC provides part of $X, the employer can pay a lower hourly wage without the worker quitting.) In this view, the EITC effectively subsidizes low-wage employers as much as low-wage workers.

A binding minimum wage prevents that capture by putting a floor under the hourly wage. Research from the San Francisco Fed and others has argued that a moderate minimum wage paired with an expanded EITC produces better outcomes than either alone — the minimum wage blocks the capture, and the EITC provides the targeted boost.

This is a meaningful concession from the free-market position. It means the honest answer to "wage or credit?" is often: some of both, calibrated to local conditions.

How to think about the mix

A serious policy framework would separate two questions.

Question one: how do we raise living standards for the working poor? Here, the EITC is the better tool. It targets household poverty, rewards work, and doesn't eliminate jobs. Every serious anti-poverty economist — left, right, and center — acknowledges this. Expanding the EITC, especially for childless adults (who currently receive very small credits), would do more for measured poverty reduction than any likely minimum wage change.

Question two: what minimum wage, if any, prevents employer capture and maintains wage discipline at the very bottom? Here, the answer depends on local conditions. A wage floor set at roughly half the local median wage has historically produced small or no detectable employment effects. A wage floor set at two-thirds or more of local median wages has produced measurable harm, especially for teenagers and entry-level workers.

Those two framings produce a pragmatic synthesis: a robust, expanded EITC paired with a minimum wage that is calibrated, not uniform. The federal $7.25 floor is almost certainly too low to do useful work. A national $20 floor would almost certainly cause significant damage in low-wage regions. The policy question that deserves more attention is: what mix produces the best outcome for the specific workers we're trying to help?

Friedman's deepest insight — the one that has aged the best — is that policy design matters. Two laws that look like they solve the same problem can work in radically different ways on the ground. Reading the minimum wage and the EITC as alternatives, rather than as opposites, is where a serious debate should start.

Further reading