Does the Earned Income Tax Credit Deliver?

Originally published in The American Prospect on May 26, 2020.
—-

As tens of millions of American workers file for unemployment amid the global economic and public-health crisis, we are reminded of how much of the nation’s welfare system is tied to jobs. This is not only true with employer-sponsored health insurance and other benefits, but with a lesser-discussed feature: the Earned Income Tax Credit (EITC).

The EITC is the largest federal subsidy for low-income workers. It’s a refundable credit that workers generally get in an annual lump sum when they file taxes. For childless low-income workers, the benefit is pretty small; in 2019, it maxed out at $529. But for low-income parents with children, it can rise as high as $6,557, depending on how many kids they have and how many hours they worked. About 22 million workers received EITC benefits in 2018, and it’s credited with lifting 5.6 million Americans out of poverty.

The EITC is also distinctive for being unusually popular among both parties. Democrats are fans because it gets more money to the poor, while Republicans like it because it rewards work. During the recent Democratic presidential primary, all the major candidates except Andrew Yang supported expanding the EITC, and in Congress, almost every Senate Democrat has signed on to a bill that would bolster it. In 2017, Gordon Berlin, then-president of the think tank MDRC, told The New York Times that he sees expanding the EITC as the best policy to pursue if one wanted to cut poverty.

But while the political class continues to rally around the EITC, their enthusiasm overlooks an ongoing debate among left-of-center policy wonks that has picked up steam over the last year. Some new research finds significant shortcomings with the EITC as a poverty-reduction tool; others have rebutted that critique. With a potential Democratic majority poised to return to this framework for aiding the working poor, the debate could clarify whether there are better options to reach that goal.

Last summer in the Prospect, New School economist Teresa Ghilarducci and her graduate student Aida Farmand laid out one critique: They argued that while the EITC does increase labor participation among the poor, it also effectively acts as a subsidy to low-wage employers and bears responsibility for driving down American wages overall.

A few months later, Princeton economist Henrik Kleven made a very different argument, in a working paper published by the National Bureau of Economic Research. Kleven argued that, contrary to popular consensus, the EITC is not responsible for increasing labor force participation at all. Specifically, he argues that the large increase in employment among low-income women in the mid-1990s, which is generally attributed to the 1993 EITC expansion, was actually driven by contemporaneous welfare reforms and the decade’s booming economy.

Most EITC experts have dismissed the concern that the tax credit may exert downward pressure on wages. If true, it doesn’t outweigh its benefits, and just underscores the need for a robust minimum wage, the theory goes. “If you put a strong minimum wage and a good EITC you get the best of both worlds,” said Bob Greenstein, the founder and president of the Center on Budget and Policy Priorities.

But while the political class continues to rally around the EITC, their enthusiasm overlooks an ongoing debate among left-of-center policy wonks that has picked up steam over the last year. Some new research finds significant shortcomings with the EITC as a poverty-reduction tool; others have rebutted that critique. With a potential Democratic majority poised to return to this framework for aiding the working poor, the debate could clarify whether there are better options to reach that goal.

Last summer in the Prospect, New School economist Teresa Ghilarducci and her graduate student Aida Farmand laid out one critique: They argued that while the EITC does increase labor participation among the poor, it also effectively acts as a subsidy to low-wage employers and bears responsibility for driving down American wages overall.

A few months later, Princeton economist Henrik Kleven made a very different argument, in a working paper published by the National Bureau of Economic Research. Kleven argued that, contrary to popular consensus, the EITC is not responsible for increasing labor force participation at all. Specifically, he argues that the large increase in employment among low-income women in the mid-1990s, which is generally attributed to the 1993 EITC expansion, was actually driven by contemporaneous welfare reforms and the decade’s booming economy.

Most EITC experts have dismissed the concern that the tax credit may exert downward pressure on wages. If true, it doesn’t outweigh its benefits, and just underscores the need for a robust minimum wage, the theory goes. “If you put a strong minimum wage and a good EITC you get the best of both worlds,” said Bob Greenstein, the founder and president of the Center on Budget and Policy Priorities.

The debate got a new jolt last week when Matt Bruenig, the founder of the People’s Policy Project think tank, put out a new, highly critical paper on the EITC. Bruenig cites Kleven’s work to show that the EITC does not increase labor supply, and he also argues that the tax credit’s administrative costs have been understated (mainly because they don’t take into account the private administrative costs of tax preparers that the majority of EITC beneficiaries use), while its poverty-reduction impacts have been overstated (by 47 percent, according to his calculations). Bruenig cites a 2019 Census study that also found the EITC overstated its impact on reducing poverty. He argued in Jacobin that Democrats “should abandon their EITC fetish.”

Steinbaum praised Bruenig’s paper for “putting the received wisdom of the EITC in a more questionable light.” Claudia Sahm, the director of macroeconomic policy at the Washington Center for Equitable Growth, called Bruenig’s arguments “thought-provoking and dangerous.”

Not everyone was taken with Bruenig’s arguments. Some criticized him for treating Kleven’s paper as dispositive, and others argued that his estimate of the EITC’s administrative costs—11 percent—was dubious. A different Rothstein estimate, which also accounted for private tax prep, clocked EITC administrative costs at about 5 percent. John Wancheck, an EITC expert at the Center on Budget and Policy Priorities, said the data underlying the Rothstein estimate is likely more reliable and reflective of national trends.

Rothstein, Greenstein, and Edin all told the Prospect they believed Bruenig had overstated the importance of his mismeasurement critiques.

Bruenig argues in his paper that it’s problematic that the Current Population Survey—which is sponsored by the Census and the Bureau of Labor Statistics—counts EITC benefits received in the subsequent year as having been received in the current year. In doing so, he argues, the survey makes the EITC appear more perfectly targeted to those in poverty than it really is.

Greenstein told me he doesn’t find that objection “very important” and thinks it’s given too much weight in Bruenig’s paper. Edin agrees that “it’s a technicality that I don’t think means very much.” She said in order to believe it was important, one would have to believe there was a serious difference between those who are living just above and just below the poverty line. Rothstein echoed them in saying, “I have a hard time believing it matters.” The researchers also pointed to evidence not included in Bruenig’s paper, like that the EITC has had significant benefits for children, and that many beneficiaries like that they can use their tax credit as an annual forced-savings vehicle, rather than a monthly wage supplement.

In an interview, Bruenig defended the mismeasurement critiques, saying they directly challenge the idea that the EITC is a “well-targeted program.” He also said that how policymakers measure the EITC’s effectiveness contrasts with how they measure the effectiveness of other welfare programs like food stamps, which evaluate uptake in the current year. “The [evaluations] have to be comparable,” he said. “And if you want to defend the EITC on the basis that a lump-sum payment can be beneficial to the poor, well that’s different than saying it’s a well-targeted program.”

KLEVEN AND BRUENIG both cite one piece of evidence as a reason they believe the EITC has been overhyped when it comes to incentivizing work. They argue that there’s no real proof that low-wage workers who were already employed then increased their hours to access more of the benefit, or phased out their hours once they had reached the maximum benefit. In economics jargon, this is known as an “intensive margin effect.” Even Rothstein agrees there’s been little good evidence for this, though he notes it’s a bit harder to study.

“How can it be that the EITC influences whether people work or not but does not influence how much they work?” asked Bruenig. “Defenders just wave away this question and say it’s ‘informational frictions.’”

Edin, for her part, says that when she did her qualitative research for her 2015 book, It’s Not Like I’m Poor, she found that while very few low-income people could name the EITC, virtually everyone she met knew they got a tax refund that was associated with their kids and that they had to work to get it. “People generally understood the more you work the more you get,” she said, later adding that no survey she knows of has asked that question.

With millions now losing jobs or seeing working hours reduced, problems with the EITC are cropping up. First, households generally receive the EITC as a refund in February and March, meaning it’s not something people can turn to if they face an unexpected crisis later in the year. And if they’re unable to find work in a depressed labor market, they won’t receive any assistance at all.

“It’s targeted and very effective at raising people’s incomes for low-income taxpayers … but it is not designed to be an effective recession stabilizer for families,” said Hilary Hoynes, a professor of public policy and economics at UC Berkeley.

In the aggregate, researchers find that EITC usage doesn’t vary all that much across economic cycles. But Hoynes says this obscures what actually goes on. During recessions, low-income workers may lose all their earnings, and therefore all their EITC benefits. Meanwhile, higher-income households that were not previously eligible for the EITC may suddenly “drop in” to eligibility, as they face a reduction in their own earnings.

AWARE THAT MANY workers in poverty or on the brink of it could soon lose their EITC, House Democrats included a provision in their latest proposed stimulus package, the HEROES Act, that would allow workers to substitute their 2019 earnings in the 2020 tax year. In other words, a middle-class household that was newly eligible for the EITC could still claim it with their 2020 earnings, but a low-income worker who lost their job could claim the EITC using their 2019 income status.

Other temporary modifications in the HEROES Act include expanding the EITC for workers without children (by nearly tripling the maximum benefit to $1,437) and expanding age eligibility.

It’s not clear how hard Democrats plan to fight for these measures—they’ve already stated that the HEROES Act is just a starting point for negotiations with the Senate. But these are not the only poverty-reduction programs on the Democratic side that use the EITC model. The Working Families Tax Relief Act—a bill led by Sens. Sherrod Brown (D-OH) and Michael Bennet (D-CO)—would substantially increase the size of the EITC, and has nearly every Senate Dem signed on as a co-sponsor. Rep. Ro Khanna (D-CA) and Brown introduced an even bigger EITC expansion bill in 2017, the GAIN Act, and Sen. Kamala Harris (D-CA) has her own version, too, the LIFT Act.

“The EITC is one of the largest anti-poverty programs that exists in America and it also needs to be strengthened, which is why I introduced the Cost of Living Refund, which would lift millions more American families and individuals into the middle class,” said Rep. Khanna, adding that the EITC can’t stand alone and we need a higher minimum wage, stronger union protections, and other safety net programs. “This crisis underscores the dire need for more anti-poverty programs, and I’m reviewing the latest research on how we can make the EITC more effective and what other approaches we can pursue to combat the scourge of poverty.” The offices of Sens. Bennet, Brown, and Harris did not respond.

The debate over the EITC won’t be ending anytime soon, though all involved agree that improving the EITC alone, or scrapping it, would not be enough.

Kleven told Vox’s Dylan Matthews last year that even if the EITC doesn’t incentivize labor participation, that doesn’t mean it’s a bad policy, as it would also suggest the EITC is not driving down wages as Ghilarducci and Farmand argued. It would make the EITC “a pure money dump on the working poor that doesn’t come with any labor supply distortions,” he said.

But Bruenig says that if the EITC doesn’t actually drive labor force participation, then it’s even more immoral to phase in the tax benefit to the working poor, instead of giving it to all low-income households. He argues in his paper for replacing the EITC and the Child Tax Credit with a universal monthly child allowance, though he told me he’d also support an EITC that had no connection to the number of children one has. “It’s not ideal, but I would be OK with that,” he said.

Greenstein said his decades working in tax policy have given him no reason to think that scrapping the EITC will then drive more money to poor people with no earnings. Getting rid of it is not the answer, he argues, but he agrees there needs to be more money channeled to low-income families. Rather than a child allowance that extends to the top of the income scale, the Center on Budget and Policy Priorities backs bolstering the Child Tax Credit for families earning up to $200,000 or so. “We don’t have to choose,” Greenstein said. “We should do both.”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s