In Defense of the Earned Income Tax Credit

May 26, 2020

Max B. Sawicky

This is a response to a recent critique by Matt Bruenig of the People’s Policy Project that has provoked a renewed scrutiny of EITC by the left. The critique, The Myths of the Earned Income Tax Credit, is hereinafter referred to as EITC Myths. In addition, there is an accompanying piece by Bruenig, It’s Time for Democrats to Abandon the Earned Income Tax Credit, in the socialist magazine Jacobin. 

Like other benefits that depend on work, such as the exclusion of employer-paid health insurance from taxable income, the EITC has strengths and weaknesses. In an era of mass unemployment, benefits conditional on employment are clearly inadequate.

Is there a case for work-conditioned benefits (WCBs) in the absence of income guarantees, or as a substitute for them? Here are a few:

  • WCBs incentivize work. Absent any sort of WCB, labor force participation will be inadequate. This rationale is unappealing. Macroeconomic policy can remedy employment gaps, and unemployment is not due to a lack of desire to work.
  • WCBs reward work. Under some circumstances, where jobs are freely available, rewarding work and declining to reward those able but unwilling to work could be defended.
  • WCBs can reduce income inequality. If you think the government should act affirmatively to reduce inequality, WCBs could be designed for such a purpose. As noted above, this is among the chief virtues of the EITC.
  • WCBs might be effective in subsidizing a shorter work week, a desirable end in and of itself. The EITC’s current structure does not fulfill this objective, but benefits that were available for those working less than full time might do so.
  • WCBs might not be preferable to income guarantees, but they are superior to an utter lack of such guarantees. WCBs may be politically feasible when income guarantees are not. Under current circumstances, this is a leading rationale for the EITC.

As suggested above, all of these justifications collapse in an era of mass unemployment. Without employment, there are no benefits, no objectives are met, and incentives to work are unlikely to be a cause of unemployment. Periods of mass unemployment bring to the fore the fundamental shortcoming of WCBs as social insurance or social welfare policy writ large, but they still raise the question of available, feasible alternatives.

Unsurprisingly, the older criticisms of the EITC have not anticipated the current crisis. The most common was to stigmatize the benefit as a subsidy to employers, a form of “corporate welfare.” More recently, a meme has circulated that damns any sort of means-tested benefit in favor of universal benefits such as the so-called Universal Basic Income (UBI).

EITC Myths raises two additional criticisms: that the EITC fails to reduce poverty as much as is claimed, and that it fails to increase labor force participation. Here, Bruenig  is really fighting the rationale surrounding Clinton’s welfare reform, which capped decades of attacks on the Aid to Families with Dependent Children (AFDC) program. Under this critique, AFDC (and by implication, other means-tested programs) promote indolence, personal irresponsibility, giving birth to illegitimate children, and abstention from the labor force. This is indeed a toxic narrative, and it continues to threaten surviving anti-poverty programs, particularly the Supplemental Nutritional Assistance Program (SNAP) and Medicaid. As noted at the outset, however, the EITC can be defended without reference to invidious aspersions on the moral character of the poor.

One of EITC Myths principal criticisms of the EITC is that it fails to reduce poverty as much as it claims. Actually, its primary objective is not to reduce poverty, but there should be no doubt that poverty reduction is one of its benefits. Bruenig’s criterion for rejection of the EITC — that its poverty reduction effect is less than is claimed elsewhere — is questionable. That something “isn’t as good as you thought,” whoever you are, begs the question of how good it has to be. Bruenig’s criticism calls to mind the old sociologists’ joke:

Q: “How’s your wife?”

A: “Compared to what?”

EITC Myths offers no comparisons of the effectiveness of existing or potential alternative methods of reducing poverty. And nowhere does it acknowledge that most EITC benefits go to those below median income, which means that most of its benefits reduce inequality. Nor does it admit that take-up rates — the extent to which eligible persons participate in the program — are well in excess of those for other anti-poverty programs. If we want families to receive income support, we have to credit a program providing benefits for which more of them are willing to apply. Lower-income workers like to apply for the EITC because less “welfare” stigma surrounds it, and with the aid of a tax preparer, application is less onerous than obtaining other means-tested benefits.

Bruenig provides comparisons of the administrative costs of other programs and notes correctly that the EITC’s costs should reflect the cost of tax preparers that beneficiaries employ. But his comparison to other programs is defective. It is not clear how he calculates the cost of participation for other programs, beyond government administration. Applying for SNAP or Medicaid benefits can be onerous in its own right, but in those cases it is the applicant who bears the cost in the form of time and aggravation. Nor is it clear that he has included the administrative costs of Medicaid incurred by health care providers, ordinarily one of the major criticisms of the tangled US health care system.

On the question of increased labor force participation, it is true that this has been touted, but under the same perverse logic of the 1996 welfare reform, namely that poverty results from individuals’ personal inadequacies.

 The implication is that individuals need some kind of push to seek employment. It is also true that the extent of this effect is doubted in new research by Henrik Klevin cited in EITC Myths.

Do we really need programs like the EITC to encourage people to work? One response to the contrary goes by a paraphrase of the “Field of Dreams” maxim: “If you pay them, they will work.” The EITC successfully gets money to a lot of people who badly need it, labor force participation aside.

EITC Myths calls for “replacing the EITC and the Child Tax Credit with a universal monthly child benefit administered by the Social Security Administration”, but his attack on the EITC is not tested against any specific alternative. To say a program is bad because it could be better is not a good analysis. The counterfactual is always in question. It is bad policy analysis 101 when a real-world program is unfavorably compared to an idealized alternative. 

A child allowance incurs the same problems of efficiency and targeting as a UBI, with the saving grace — compared to the UBI — that households with children are more plausible aid recipients than households in general. But the EITC is itself targeted to families with children. The same comparative factors come into play. A means-tested children’s allowance is preferable to a universal one because it can provide more aid to families in greater need.

The inadequacies and fallacies surrounding the UBI have been written about elsewhere, including in Jacobin and other places, but we ought to say a word about means-testing. (Under means-testing, benefits decrease the higher one’s income.) It goes to a broader problem in lefty policy discourse that entails negative connotations attached to concepts like corporate welfare and subsidies.

The fundamental reality of the US welfare state is that it makes heavy use of private sector vendors. Health care providers, farmers, and supermarkets get a cut of benefits under Medicaid and SNAP. They are not government employees. We might prefer different arrangements, but that is where the game begins. It is meaningless to disparage a particular program as “bad” in isolation because it entails private sector participation. Most anti-poverty programs do. The badness is only meaningful in light of politically and technically feasible alternatives.

The real onus against the EITC is that the US lacks a coherent system of income guarantees. We don’t disdain public transit because it fails to provide a guaranteed income, so criticism of the EITC on these grounds does not follow. It could be argued that the EITC discourages provision of more universal income guarantees by offering a substitute that is unsatisfactory, from the standpoint of income guarantees. Indeed, when AFDC was buried under the Clinton Administration’s push to “end welfare as we know it,” the expansion of the EITC in the late 1980s and early 1990s was touted as part of the reform.

There should be no argument that as an income guarantee, the EITC and other work-conditioned benefits leave much to be desired. Life without employment for those who are healthy and of working age is precarious. There is a patchwork system of unemployment insurance, available temporarily and only to those with an earnings history. There is a mean-spirited, fragmentary system of cash benefits provided by state governments through the Temporary Assistance for Needy Families (TANF) program. There is health care through the Medicaid program and food assistance through  SNAP.

In the final analysis, however, people without employment need money, on a weekly or at least monthly basis, and there is scant relief of that sort available. For political reasons, the US welfare state is degraded to the point where WCBs have become a substitute for universal income guarantees. While income guarantees in the absence of WCBs make sense, the converse is less obvious. WCBs without income guarantees imply a gross rejection of any anti-poverty policy. That is where we are. It does not follow that criticism of the EITC in isolation is a constructive response.

We already have a panoply of income support programs, especially in the form of social insurance, but also through Medicaid, SNAP, and the EITC. The logical question is, where are the gaps in coverage? The logical answer is to start with the yawning rift in the safety net where AFDC used to be, primarily very low-income families with children.

Imagine a population of 10 persons and a budget of $1,000. If we decide that the bottom quintile of the population (two people) needs income support the most, a benefit of $500 each is possible. Alternatively, spreading the thousand dollars over the entire population reduces the amount available to our least fortunate pair to $100 each. And incidentally, under the equal per capita UBI, to balance the budget, the top 50 percent will have to give up their benefit entirely. The “universal” angle to the UBI, which is supposed to mollify political opposition, is illusory.

It is true that after a decades-long campaign, AFDC was destroyed. So without doubt, a Negative Income Tax (of which AFDC is one example) is politically difficult to do. But so are alternatives that would cost much more. In the meantime, one-sided attacks about flawed but existing anti-poverty programs only serves the opposition.

Support Cepr


If you value CEPR's work, support us by making a financial contribution.

Si valora el trabajo de CEPR, apóyenos haciendo una contribución financiera.

Donate Apóyanos

Keep up with our latest news