The Insufficiency of Tax Credits

“Welfare” in the United States is a stigmatized term: only 23.6% of Americans in 2018 believed that the government is spending “too little” on welfare, and 38.9% believed the government is spending “too much.”1 But this cannot be interpreted as the American public hating the poor and wanting them to fend for themselves. When the subject of the question is reworded as “assistance to the poor,” there is a massive shift. In 2018, 73.2% of Americans said they believed the government spends “too little” on assistance to the poor, with only 7.2% saying the government spends too much. This indicates that the term “welfare” has a specific interpretation attached to it—different from government assistance to the poor—that makes people vastly less supportive of it. This stigma is generally attributed to preconceptions about welfare recipients that have many of their roots in racial prejudice from rhetoric by politicians, such as Ronald Reagan and Bill Clinton, who argued that welfare would lead to people not wanting to work.2 This has significantly influenced the design of present-day welfare programs. One such example is that Clinton and Reagan implemented numerous cuts to welfare while championing and expanding the Earned Income Tax Credit (ETIC), a fully refundable tax credit targeted towards low-income workers.3 So, what made the EITC different from other redistributive policies that Clinton and Reagan opposed?

Components of Tax Credits

A tax credit allows one to reduce their tax liability—how much tax they owe—by the value of the credit. For example, if someone receives a tax credit for $500 and owes $2,000 in taxes, the credit means they’ll only need to pay $1,500. This is akin to the government hanging out $500.

Tax credits have a fundamental issue in their design that policymakers have to decide whether to address. If one’s tax liability is less than the value of the tax credit, they will receive less money than someone with a higher tax liability; the benefit phases in with income. Referring back to the previous example, if one’s tax liability is $300 and they are eligible for the $500 tax credit, they can only receive $300 worth of the benefit because that is all they owe in taxes. This makes tax credits regressive towards people with very low to no income.

These types of tax credits are non-refundable. A refundable tax credit allows the recipient to “refund” the remaining value of the credit, even if they do not pay income tax. The person with a tax liability of $300, for instance, can refund the missing $200 to receive the full $500 credit. A refundable tax credit, however, can still have a phase-in, where the value of the credit gets bigger with income as opposed to income tax liability. This is how the EITC works. This means one can still receive some of the benefits if the tax credit is refundable, but the benefit amount will be less than someone who makes more. As a cost-saving measure, tax credits typically also include a phase-out at a certain level of income, where the benefit slowly reduces.

Tax credits can be implemented to apply to certain purchases, such as the recently implemented Clean Vehicle Tax Credit, through which the purchase of an eligible electrical vehicle grants a tax credit of up to $7,500.

The typical rationales for preferring tax credits and their specific designs over direct cash transfers for the purpose of alleviating poverty are insufficient and not rooted in empirical evidence, as will be shown through an examination of the Earned Income Tax Credit (EITC) and the Child Tax Credit (CTC).

Why Tax Credits?

The argument for tax credits being better at reducing poverty despite their regressivity rests on several premises. Firstly, proponents argue there are the “deserving poor,” who are hardworking, low-income earners, and the “undeserving poor,” who are unemployed and lazy. Thus, proponents say welfare should be conditional in order to separate the undeserving from the deserving. Additionally, regressive tax credit supports suggest that welfare should aim to promote work, not discourage it, and giving people free money with no strings attached means they will work less or stop working.4 

Tax credits technically have a built-in work requirement; if one does not work, they will not have income. Supporters suggest that the regressivity and phase-in can be beneficial in that they tie the value of the credit to income, which incentivizes people with low incomes to work more and find better-paying jobs. Then, once income has reached a higher level, the value of the tax credit is phased out. President Bill Clinton, who signed a massive expansion of the EITC, articulated this premise:

“By expanding this Earned Income Tax Credit to working families and especially to the working poor, this congress has made history by enabling us to say for the first time now if you work hard and have children in your home and you spend 40 hours a week at work, you can be a successful worker and parent, and you will be lifted out of poverty.”5 

However, the design of tax credits can be changed by policymakers to make them less regressive. The EITC is fully refundable, reducing the effective penalty on those with lower incomes, and the benefit phases in as earnings increase. The phasing in of the benefit encourages work while achieving a mitigation of the harmful “tax” part of the tax credit. To evaluate whether the properties of tax credits are beneficial towards reducing poverty and encouraging work, we can examine the expansion of the Child Tax Credit that increased the benefit, made it fully refundable, and added eligibility for non-workers. 

Impact of the CTC expansion

The CTC is a conditionally refundable tax credit geared towards helping families with the cost of childcare. The credit is paid out as an annual lump sum, with the amount varying based on marital status, number of children, and family income.6 The specifics, as explained by the Tax Foundation, are as follows.

“The value of the child tax credit (CTC) is 15 percent of a household’s adjusted gross income (AGI) above the first $2,500 of earnings until the credit reaches its maximum at $2,000 per child. Up to $1,400 of the credit is refundable as the Additional Child Tax Credit [ACTC], meaning the taxpayer can receive a refund of up to $1,400 even if they don’t owe any tax. The credit is reduced by 5 percent after adjusted gross income reaches $200,000 for single parents and $400,000 for married couples. Additionally, there is a $500 non-refundable credit for non-child tax credit dependents.”7 

Below is a visualization of the CTC illustrating the typical trapezoid shape of a tax credit created by the phase-in and phase-out.

The 2021 American Rescue Plan included a major temporary expansion of the CTC that expanded the maximum credit amount to $3,600 and removed the phase-in by making the benefit fully refundable, so it was fully available to low-income families and non-workers. According to the Center on Budget and Policy Priorities, this expansion led to 4.1 million children being lifted from poverty, with the child poverty rate dropping to 5.2% — almost halving from 9.7% in 2020. The Center calculates that 80% of this massive reduction in child poverty is attributed to removing the phase-in.8 It’s evident that the phase-in was a critical barrier to the effectiveness of the CTC in addressing child poverty for a very simple and obvious reason: excluding the poor doesn’t help reduce poverty. 

Some, such as Scott Winship, the Director of Poverty Studies and a Senior Fellow at the American Enterprise Institute, argue that the poverty-reducing effects of the expansion are temporary and that if the expansion was made permanent, there would be a significant decline in employment as parents stopped working leading to numerous negative societal impacts such as increased divorce that would culminate in an increase in child poverty.9 

The interesting thing about Winship’s argument and the concern he articulates is that he relies not solely on evidence but on appealing to concerns and uncertainty against massive tangible improvements in people’s lives. He cites a study by University of Chicago Economists Kevin Corinth and Bruce Meyer that “concluded that if the child allowance was made permanent, 1.5 million parents would stop working. As a result, rather than child poverty falling by one-third, the decline would be 22 percent.” A 22 percent decrease in child poverty would still be a massive accomplishment, and Winship agrees, calling it “an impressive figure.” In response, Winship brings up several hypothetical unintended long-term consequences, such as increasing divorce rates and declining marriage, for which he cites minimal evidence to support.10  

The Corinth and Meyer paper, when reduced down to its central claim, as described by Matt Bruenig of the People’s Policy Center. is that one out of ten single mothers with two kids who would receive $6,000 annually from the CTC rather than $2,685 will choose to drop out of the workforce and live in poverty.11 This conclusion should raise immediate skepticism and is inconsistent with comparative research on expanded child benefits in other countries. 

A 2021 study of the employment impacts of Canada’s expansions of their childcare benefits, the Universal Childcare Benefit in 2015 and the current Canadian Child Benefit from 2016 (neither of which have phase-ins) found “no evidence of a labor supply response to either of the program reforms on either the extensive or intensive margins.”12 

Even if the conclusions of Corthin and Meyer were true, there is still a net drop in child poverty and a rather significant one at that. The question then becomes one of tradeoffs and values: millions of children being lifted out of poverty or the hypothetical ones who would supposedly drop into poverty. Only one of these results has been proven to be certain. 

The debate around tax credits is reflective of the broader debate around the role of welfare in the first place. Tax credit proponents believe that welfare should be constructed to incentivize people towards or be conditional on the desired behavior. The goal is not to help the least advantaged directly but to guide them towards actions such as work and marriage. The problem with this view is that the least advantaged are those who cannot work at all: children, the elderly, and those with disabilities. At the very least, the goal of welfare should be to help these non-workers and under the current regime, they’re often excluded by design.