America’s most important policy for pulling economically struggling workers out of poverty is the Earned Income Tax Credit – which is now much bigger than cash welfare assistance disbursed through Temporary Assistance to Needy Families. Launched in 1975 to reimburse the Social Security payroll taxes paid by low-income workers with children, the Earned Income Tax Credit started small but has since grown to major proportions. By 2010, almost 27 million families received $59.5 billion from the federal Earned Income Tax Credit. In comparison, the United States spent only about $19.8 billion that year on block grants to help the states to cover welfare benefits for about two million families.
A Program with Built-In Incentives and Disincentives
Considered strictly from an economic calculus perspective, the federal Earned Income Tax Credit appears to create a complex pattern of incentives and disincentives with the potential to affect the work and family choices of recipients.
- Larger credits go to families with more children (up to a limit).
- The tax credit is based on family rather than individual income, which could have the effect of making some people who are individually eligible lose the credit when they marry.
- The size of the credit grows as wage earnings rise up to a certain point, and then phases out as wage earnings rise into what the government considers a middle range.
In fact, available research suggests that the Earned Income Tax Credit does encourage low-income parents to enter the labor market. Yet research documents few effects on the amount of work by recipients or on their choices about marriage or childbearing. To better understand what is going on, a colleague and I interviewed 115 low-income people. We asked about their views and reactions to the Earned Income Tax Credit, in light of their family and job situations.
Understanding “The Refund”
The first thing we discovered is that it is difficult to ask questions about the Earned Income Tax Credit. Low-income people who filed tax returns to claim this refund – often working with tax preparers – typically knew little about the details of the program and its incentive structures.
- To be sure, nearly nine out of ten had heard of the Earned Income Tax Credit and understood that they received a large tax refund because of it. Yet fewer than one in ten of those we interviewed could specify the amount of their personal tax refund.
- More than two-thirds of our interviewees knew they received the Earned Income Tax Credit because they worked and didn’t make high wages, and 38% realized that they benefited because they had children. But few understood the specific program incentives or disincentives related to work hours and earnings, marriage, and childbearing.
Our interviewees often talked about the Earned Income Tax Credit as a benefit that lacked the social stigma that could be associated with a visible cash-welfare handout from government. The benefit is claimed in a socially discreet way, and recipients described it as a “bonus,” like “winning the lottery,” or as a “reward” they “earned” for working and having children.
How Recipients Handle Program Rules
Low-income tax-filers who claim the Earned Income Tax Credit were, for the most part, uncertain about how their refunds could go up or down if they worked more or less, married or divorced, or had more children. Even those who did grasp such nuances were unlikely to respond given constraints in the labor and marriage markets. Recipients were not passive, however. Like all U.S. tax-filers do, they tried to maximize refunds by altering filing statuses:
- Employment rules. Low-income people find it hard to control their own work hours in precarious and unstable jobs. Rather than changing the number of hours they worked in order to maximize their refunds, many of them used a technique called “claiming zero” – claiming zero dependents or exemptions from their weekly paychecks. Like many Americans who adopt this strategy, they could save some money from each check and look forward to a large lump-sum refund at tax time.
- Marital rules. For many single working parents, suitable marriage partners seemed unavailable, so they did not worry about the possible impact of marriage on their tax credit. Furthermore, many told us that if they were to marry, they would “file separate” – that is, continue to file taxes as head of household, which counts their income but not their partner’s income. Regardless of marriage, they viewed their money as separate and their own.
- Rules about children. In contrast to the fuzzy understanding of incentives for work earnings and marriage, filers clearly saw that having more children could boost their tax credit. But this hardly mattered, because it costs more to raise a child than the tax refund provides. Child-claiming was influenced much more than child-bearing. In order to let as many family members as possible maximize their refunds, filers spread claims for children beyond mothers and resident fathers – to include non-resident fathers, resident boyfriends, and grandparents.
In sum, we learned that the opaqueness of the Earned Income Tax Credit as a major social program embedded in the tax code both frees it from stigma and masks its nuances. Low-income claimants of this benefit may not calibrate work and family choices in the ways analysts imagine, but they do use their own filing techniques to maximize refunds. Some observers may feel that improper manipulations of the tax system are afoot. But our results suggest, instead, that Americans who calibrate their claims for this tax credit so vital to personal and family wellbeing are simply responding to the unpredictability of work and family life faced by low-income households. They file in ways that will let the Earned Income Tax Credit deliver the greatest help to the largest number of low-income workers and their children.
Read more in Laura M. Tach and Sarah Halpern-Meekin, “Tax Code Knowledge and Behavioral Responses among EITC Recipients: Policy Insights from Qualitative Data.” Journal of Policy Analysis and Management 33, no. 2 (2014): 413-439.