State Policy and Practice Related to Earned Income Tax Credits May Affect Receipt among Hispanic Families with Children
The federal Earned Income Tax Credit (EITC) is designed to help low- to moderate-income families support themselves and their children while also encouraging workforce participation. In 2018, the EITC lifted about 5.6 million people out of poverty,[1] including 3 million children, and has been hailed as the largest and most effective anti-poverty program for U.S. families with children.[2],[3],[4] The increased income supplied by the EITC—which varies by earned income and family size, averaging $3,191 in 2017 for a family with children[5]—is associated with improved child educational[6],[7] and health outcomes.[8],[9],[10]
To benefit from the credit, however, potential beneficiaries must first file taxes and then claim the credit on their tax return. Estimates suggest that more than 1 in 5 eligible taxpayers, and 1 in 4 eligible workers, do not receive the credit,[11],[12],[13] with noted disparities by race and ethnicity. Overall, Hispanic families have historically been the least likely to know about or to ever have received the EITC.[14],[15],[16]
Reasons for lower rates of federal EITC participation among the Hispanic population are not fully understood, particularly since employment rates—and thus eligibility for any refundable tax credit—are high among low-income Hispanic families. The tax system, like other social service and public systems, may impose a number of administrative burdens on Hispanic families. These include high learning and psychological costs due to lack of information and related materials in multiple languages that can raise awareness of the tax credit, low levels of outreach to workers who are not connected to the public systems through which information about the EITC is often disbursed, and complex rules that may create confusion about eligibility among families with nontraditional living arrangements or those who are eligible but whose earnings are below the threshold required for filing taxes.[17] In addition, potential Hispanic EITC recipients may be uncertain about residency and citizenship status eligibility requirements, and may have additional concerns related to their immigration status.[18]
In the 1980s, states also began to offer earned income tax credits for their residents. Tax credits available via states are accessible through the same tax filing system as the federal EITC, but may differ with respect to the amount of the credit, whether it is refundable, and its eligibility requirements. States also adopt different strategies to inform residents about their federal and state tax credits, and state policies and outreach practices have the potential to address many of the administrative burdens that Hispanic families may face. This research brief explores the extent to which variation in state-level EITC policies and outreach practices may be associated with EITC participation among eligible Hispanic families with children.
Key Findings
For the 2013 to 2015 tax years, we generated new estimates that indicate lower reported rates of EITC receipt among eligible Hispanic families with children (46%), compared to eligible non-Hispanic White (55%) and Black families (58%). This pattern is consistent with previous work indicating that Hispanic workers are less likely to receive the EITC.[19],[20] Lower overall rates of receipt are on par with results from other studies that use national survey data[21],[22],[23] and may reflect families’ difficulty remembering whether they had received the credit, a lack of knowledge of credit receipt (due, for example, to the use of external assistance with tax preparation), self-report bias, and the inclusion of both non-tax-filers and tax filers among respondents.
Among states that offer their own state EITC, the refund value of credits varies from 3.5 percent to 43.0 percent of the federal refund; the structure of the credit may target certain segments of the total eligible federal EITC population based on earnings and family size. A higher level of generosity in refundable state EITC credits (at least 15% of the federal EITC amount) is associated with higher federal EITC receipt among Hispanic families, compared to states with no state EITCs. More structured state EITCs that are limited to segments of the total eligible federal EITC population are associated with slightly reduced EITC receipt among Hispanic families, whereas less generous refundable credits (less than 15% of the federal EITC amount) and nonrefundable credits show no statistical association with EITC receipt for Hispanic families.
Of the state earned income tax credit policies and practices examined for this study:
- Statutes requiring public programs to inform recipients and employers to inform employees, the availability of information about the EITC in Spanish, and the density of free tax preparation sites are associated with higher receipt among Hispanic families.
- States that offer state EITCs have higher proportions of Hispanic and Black populations and of residents who live in urban areas. These states are also more likely than states without a state EITC to provide EITC information in Spanish, engage in EITC social media campaigns, and have a greater density of tax preparation sites.
- Given the overlap, within states, of the policies and practices examined, it is not possible to empirically determine whether any one specific state EITC policy or practice is especially effective at increasing EITC receipt among Hispanic families. Nonetheless, states with high Hispanic populations appear to offer a bundle of policies and practices that collectively minimize administrative burden in ways that are predicted to facilitate EITC receipt.
Of the family and individual characteristics examined:
- Hispanic families with a parent who is self-employed or has limited English proficiency—or who live in households that include other relatives, naturalized U.S. citizens, or non-U.S. citizens—are less likely to receive the EITC.
- Families who participated in Medicaid, SNAP, WIC, or SSI are more likely to receive the EITC.
- Employer-sponsored insurance and periodic unemployment are not associated with EITC receipt.
Of the demographic and economic characteristics of states examined:
- Although Hispanic families are less likely to receive the EITC, those who live in states with a larger Hispanic share of the population are more likely to report receipt than Hispanic families living in states with lower proportions of Hispanic residents.
- Hispanic families who live in states that allowed unauthorized immigrants to obtain drivers’ licenses were more likely to receive the EITC than those who lived in states without such legislation. This is in line with prior research suggesting that state legislation that provides appropriate supports for immigrants may increase employment opportunities and access to services, and may influence participation in public programs among Hispanic families, including those in which all family members are U.S. citizens.[24]
- Overall, states with high Hispanic populations appear to offer a bundle of policies and practices designed to support EITC participation, and Hispanic families residing in those states appear to have higher likelihoods of receipt than Hispanic families residing in other states.
Policy Context
The federal EITC began in 1975 and has been expanded several times since. It is the largest government-funded cash transfer program for low- and moderate-income workers and is primarily targeted at working families with children.[25] Because the federal credit is refundable, it does not simply offset taxes owed; if the credit is more than what a household owes in taxes, the tax filer receives the difference back from the government. However, recipients receive the credit based on the adjusted gross income claimed on their tax filing; thus, to receive the credit, individuals must file a tax return, even if they do not owe any tax and are not required to file federal taxes.[26]
To qualify for the credit, a taxpayer’s earned income must be greater than zero but less than the defined federal income limits.[27] The amount of the credit rises as a taxpayer’s income increases until it reaches a maximum credit level known as the “plateau,” at which point the credit begins to phase out as income rises. Both the income limit and the credit amount vary by the tax filer’s number of qualified dependents. For example, a single parent with two dependent children who earns $17,000 a year would receive a credit of $5,828.
In the 1980s, states began to enact their own EITCs to supplement the federal EITC, building off its success in reducing financial hardship for low-income working families. Today, 28 states and the District of Columbia offer their own EITCs. Most states followed the federal EITC structure but may vary in terms of the amount of credit, its refundability, eligibility requirements, and outreach efforts.
About the Study
The present study leverages observed variation in state-level EITC policies and practices to address gaps in research about how state EITC policy design may influence disparities in receipt by racial and ethnic group. To determine this, we conducted a scan of state EITC policies and practices that are of particular relevance to low-income Hispanic families, including the refundability and generosity of state-level EITCs, statutes requiring that certain populations be informed of their potential eligibility for the EITC, and the availability of EITC information and services in both English and Spanish.
We combined information from the scan with data on EITC receipt from the Survey of Income and Program Participation (SIPP) to estimate the association of state-level policies and practices on EITC receipt, above and beyond a range of household- and state-level characteristics. We focused our analysis on policies and practices implemented during tax years 2013 to 2015, which corresponds to the reference years for reports of EITC receipt in the most recent SIPP data available (2015-2017).
Given some (albeit limited) evidence suggesting that the availability of state EITCs may increase federal EITC program participation among lower-income filers,[28] we anticipate that the availability of a state EITC and its generosity will increase both awareness of the EITC program more generally and the financial incentives for eligible taxpayers to claim the EITC. Prior research has also found that state and local outreach—such as notices reminding individuals of their eligibility—can increase both tax filing[29] and EITC receipt,[30],[31] supporting our hypothesis that statutory and administrative approaches to outreach will be associated with increased rates of EITC participation. The availability of free tax preparation services has also been found to increase both tax filing and EITC receipt,[32],[33] with some research suggesting that Hispanic families are more likely than their peers to seek external assistance with tax filing.[34],[35] Together, this work points to the potential of these types of state policies and practices to improve EITC participation and, in turn, the well-being of families and children who might otherwise miss out on this important credit.
Data and Methods
Data and sample
To estimate EITC eligibility and examine EITC receipt and its correlates, we drew on data from the Survey of Income and Program Participation (SIPP) Panel 2014. SIPP 2014 is a nationally representative, longitudinal survey that collected detailed information on demographics, incomes, and program participation from the same set of households during each year (called a wave) from 2014 to 2017. We used reports of EITC receipt from waves 2 to 4 and family characteristics from waves 1 to 3 to capture the incomes and demographics on which the wave 2 to 4 EITC questions were based (that is, tax years 2013 to 15). We treated nuclear families (i.e., parent[s] and their dependent child[ren]) as the tax filing unit. We then restricted our sample to EITC-eligible Hispanic families with at least one child under age 18 and with non-missing information on EITC receipt.[36]
Our final sample included 2,117 family-year observations of Hispanic families. Hispanic families are defined in our study as those in which any parent self-identifies as someone with a Spanish, Hispanic, or Latino origin and lives with at least one biological, step, or adoptive child. We derived a sample of EITC eligible families by approximating EITC eligibility by treating families with a positive total simulated amount of the federal and state EITCs as EITC-eligible families. We employed TAXSIM (Version 32), a widely used microsimulation program of federal and state taxes, and entered families’ SIPP demographic and income information in the program to identify eligible families. For filing status, we assumed married filing jointly for married couples with children and head of household for single and cohabiting parents.[37] For qualifying dependents, we included children from birth to age 18, full-time students under age 24, and children of any age with a disability, and excluded children who are married or are parents themselves.[38] All eligible families in our sample reported positive earnings.
Measures of EITC receipt and family characteristics
EITC receipt. EITC receipt is our main outcome of interest. In this measure, we consider families in which either parent reported receipt of the federal EITC. Families that did not file federal tax returns, or who filed but did not receive the EITC, are treated as non-receipt. While filers who did not claim the EITC and non-filers may have different reasons for not claiming the EITC, both groups are eligible for the credit and are the target of state-level policies and practices.
Family demographics. Family demographics include age of the primary taxpayer, family structure, age and number of children, household composition, immigration status, and English proficiency, as these are theoretically associated with economic needs, knowledge about the EITC, the filing status and benefit level, and (as a result) the likelihood of EITC receipt. For the age of the primary taxpayer, we use the parent who is the householder identified by SIPP (usually someone who pays most of the rent or mortgage) or the first parent interviewed in families that do not include the householder. We consider three types of family structure: married-couple, single-parent, and cohabiting-parent families. Cohabiting-parent families are led by two parents who are not married to each other, or by a single parent and their unmarried partner. For number of children and age of the youngest child, children are defined as EITC-qualifying dependent children. For household composition, we consider whether parents and children live with relatives outside of their nuclear family. With respect to immigration status, we examine presence within households of any non-U.S. citizens, any foreign-born naturalized citizens, or only U.S.-born citizens in the household. Finally, we include an indicator of English language proficiency for families in which any parent reported speaking English less than well. (See Table A.1 for a description of characteristic of eligible Hispanic families compared to all eligible families.)
Family economic well-being. The measure of annual earned income counts all parental earned income, including that of a parent’s unmarried cohabiting partner, who is often considered a social parent in the literature.[39] We categorize the highest levels of parental education into less than high school, high school graduate or equivalent (e.g., GED), some college, and bachelor’s degree or more. Self-employment is a proxy for employment that is less than formal, as the EITC benefit is based on formal incomes (as reported on a W2 or Schedule C). The indicator for any self-employment captures any parent who reported having been self-employed or having owned a business any time during the year. We include an indicator of no employer-sponsored health insurance to flag families in which any parent worked at any time during the year for an employer that did not provide health insurance coverage for the job. This correlate may reflect either informal employment or employment that is low-wage or low-benefit. Our periodic unemployment measure identifies families in which any parent was unemployed, looking for work, or on layoff for 14 weeks or more (roughly more than one quarter of a year) in the tax year. A public assistance indicator flags any family member participating in Medicaid; the Supplemental Nutrition Assistance Program (SNAP); the Women, Infants, and Children (WIC) Nutrition Program; or the Supplemental Security Income (SSI) program during the tax year.
Measures of state EITC policies and practices and state characteristics
We gathered information across three key dimensions of EITC policies and practices. In addition to information about the availability and structural characteristics of state-level EITCs, we also collected data on state statutes requiring that certain populations (e.g., public benefits recipients) be informed of their potential eligibility for the EITC, as well as the availability of EITC information and free tax preparation services in both English and Spanish.
Availability and structural characteristics of state-level EITCs. We used the National Bureau of Economic Research (NBER) table of state-level EITC provisions[40] to gather information on the availability of a state EITC, whether or not it was refundable, its generosity (as a percentage of the federal EITC amount), and whether its eligibility requirements differ from federal eligibility requirements for each tax year of interest (2013-2015). Guided by both empirical and conceptual considerations,[41] we combined these characteristics into a categorical variable that differentiated between the following mutually exclusive groups: (1) states that had no state EITC, (2) states that had a nonrefundable EITC, (3) states that had a refundable EITC that was less than 15 percent of the federal EITC, (4) states that had a refundable EITC that was greater than or equal to 15 percent of the federal EITC, and (5) states that deviate from the federal EITC eligibility requirements and structure. The threshold of 15 percent was chosen because it is the median level of generosity among states that offer their own EITC.
Legislation. We also gathered information from the Lexis Nexis US Public Laws/ALS database on statutes related to the EITC that were enacted during the tax years we examined. We identified two main types of statutes during the program period of interest, including whether (1) public benefits programs were required to inform recipients about their eligibility for the EITC, and (2) whether employers were required to inform employees about their potential eligibility.
Availability of information and services. We collected information on the availability of EITC information in Spanish directly from the website for each state’s administering agency. Additionally, we conducted a Google search of government websites that included each state’s name, “Earned Income Tax Credit,” and “Spanish” (or one of several Spanish language translations of EITC) to ensure that no sites were missed. Because this information could not be accessed for the years of our project period, we used current information as a proxy, with states that provided information about the EITC in Spanish coded as “yes” for all project years.
We used application programming interfaces (APIs) to collect data on each state’s total number of EITC-related Twitter and Facebook posts for each tax year examined. We also searched the social media accounts of each state’s administering agency for variations of “Earned Income Tax Credit.”
We obtained the number of free tax preparation sites (specifically, Volunteer Income Tax Assistance [VITA] programs) per state from the Internal Revenue Service (IRS). The IRS’s VITA program provides free tax preparation support to people who have low to moderate incomes, to persons with disabilities, and to limited English-speaking taxpayers. To better measure access to these services within each state, we calculated the number of VITA sites in each state and year per 10 square miles.
State characteristics of family in the tax year. State characteristics include the statewide median adjusted gross income (AGI) and employment population ratio, a binary indicator of having a metropolitan population over 250,000, the number of EITC-eligible filers, income thresholds for filing state income tax returns that are higher or lower than the federal thresholds (or no state income taxes), percentage of the population that is Hispanic and Black, and an indicator of whether unauthorized immigrants could obtain a driver license (as a proxy for a state’s general level of accommodation toward immigrants). We retrieved all these data from the Tax Policy Center’s EITC Interactive Database, except for the following: the employment population ratio, which is drawn from the U.S. Bureau of Labor Statistics; the metropolitan variable, which is drawn directly from SIPP; tax filing income thresholds, collected from each state’s Department of Revenue; and statutes allowing unauthorized immigrants to acquire drivers’ licenses, which were obtained from the National Conference of State Legislatures’ Immigration Laws and Current State Immigration Legislation database.
Analytic approach
We estimated multivariate analysis to examine how state EITC-related policies and practices are associated with EITC receipt among Hispanic families with children. Specifically, we estimated logistic regression models predicting EITC receipt with year-fixed effects and standard errors clustered by household. In this series of models, we included each state policy and practice separately to address potential multicollinearity among these policies and practices. In addition, we adjusted for detailed demographic, economic, immigration, and linguistic characteristics of Hispanic families, as well as important state demographic and contextual factors that are associated with both our state policies of focus and EITC receipt. EITC receipt and family characteristics included in this analysis vary by family, and state characteristics including policies differ by the family’s state of residence.
While our models include state contextual factors (e.g., taxpayer population, racial/ethnic diversity, urbanicity, employment population ratio, state income tax filing threshold) that might be associated with both policy and EITC receipt, we acknowledge the limitations of this method. We cannot adjust for unobserved characteristics, either time-varying or time-invariant. We also cannot examine the influences of policies within states due to the lack of changes in those states over time. Therefore, our results should be interpreted as associations rather than causal relationships.
Results
Rates of EITC participation among Hispanic families are lower than among their non-Hispanic peers.
Our analysis of SIPP data shows that 46 percent of Hispanic families who were identified as EITC-eligible reported EITC receipt; 54 percent reported non-receipt, including 46 percent that filed but did not receive an EITC and another 8 percent that did not file (see Figure 1). The receipt rate is 7 percentage points lower among eligible Hispanic families than among eligible families of all races and ethnicities.
States vary with respect to whether they offer a state EITC, the key characteristics of their state credit, and their policy and outreach practices.
Based on our policy scan, in tax year 2013, 24 states and Washington, DC offered an EITC; one state ended its EITC in 2014 and two states added one in 2015 (see Table 1). States that adopted state EITCs mostly provided benefits through a refundable credit and used an income limit identical to that of the federal EITC. States exhibited considerably more variation with respect to the generosity of their EITCs, which ranged from 3.5 percent of the federal credit to 43 percent.
a Data were not available for tax years of interest, so current year data were used as a proxy for all years.
Only 14 states engaged in any sort of social media outreach related to the EITC during the three years examined. Among states that posted at least once during this time period, New York and California were the most active, posting a total of 49 and 29 EITC-related social media posts, respectively, in 2015. Most states, however, posted only once or twice in a given year, if at all, with the mean number of EITC-related posts increasing slightly, from 1.8 in 2013 to 2.0 in 2015.
All 50 states and Washington, DC had free tax preparation (VITA) sites. On average, states had 1.1 tax preparation sites per 100 square miles. DC has the highest density of free tax prep sites, with 27.9 per 100 square miles, while Alaska has the lowest density, with 0.01 free tax prep sites per 100 square miles.
States with more generous state EITCs are also more likely to engage in other strategies to support EITC participation and tend to have higher proportions of Hispanic residents.
States with more generous state EITCs (at least 15% of the federal EITC) tend to have higher proportions of Hispanic and Black residents, a higher proportion of residents using a language other than English as their primary language, and a higher proportion of residents who live in urban areas, compared to other states (see Table A.2). States with no state EITCs or less generous state EITCs (less than 15% of the federal EITC) are smaller by population size and have a larger share of households that receive SNAP, relative to other states. States without state EITCs also have a higher median adjusted gross income than other states.
States with a refundable state EITC (of any amount) were more likely to require their public assistance programs to inform participants of EITCs than those with nonrefundable EITCs. Among states with no EITC, or less generous state EITCs, only two required employers to inform employees about the EITC. States with state EITCs that were at least 15 percent of the federal EITC were more likely to promote EITCs on social media, publish EITC information in Spanish, and have a higher density of tax preparation sites than those with lower state EITCs or no state EITCs. However, there was still variation in states with less generous or no state EITCs. For example, one quarter of those with no state EITCs nevertheless made EITC materials available in Spanish, which provided the variability needed for us to examine the association of EITC information in Spanish and EITC receipt in states without state EITCs.
More generous refundable state EITCs are associated with higher receipt among Hispanic families, as are statutes that require public programs to inform recipients and employers to inform employees, the availability of information about the EITC in Spanish, and the density of free tax preparation sites.
Our multivariate analysis showed some associations between state EITC policies and the likelihood of EITC receipt among Hispanic families. Figure 2 summarizes the relative odds of receipt given a specific policy or practice. An odds ratio of 1 means the policy is not associated with the odds of receipt; an odds ratio of 1.5 means the odds of receipt for families in a state with the policy is 50 percent higher than (or 1.5 times) the odds for families who live in states without that policy.
We find that more generous state EITCs (at least 15% of the federal EITC amount) are associated with a higher likelihood of EITC receipt among eligible Hispanic families (see Figure 2 and Table A.3). By contrast, state EITCs that target more limited segments of the total federal EITC-eligible population are associated with slightly reduced EITC receipt. Less generous refundable credits (less than 15% of the federal EITC amount) and nonrefundable credits are not associated with EITC receipt. Further, we did not detect a statistically significant association with a simple measure of the availability of a state EITC, in part because only two states had newly implemented a state EITC over the time period under study.
In addition, several statutory and administrative approaches to outreach are positively associated with EITC receipt among Hispanic families, including the requirement that employers inform employees of the EITC, the requirement that programs inform their participants, the provision of EITC information in Spanish, and the provision of a higher density of tax preparation sites. By contrast, the number of social media posts is negatively associated with the likelihood of receipt among Hispanic families.
Notes: The points in the figure above represent odds ratios; the lines represent the 95% confidence interval. All models include EITC-eligible Hispanic families with dependent children (N=2,117). Additional covariates include a range of demographic, economic, immigration, and linguistic family-level characteristics; state demographic, economic, and contextual factors; and year fixed effects. For model specification, see Table A.3 in the appendix.
We ran two additional models with interaction terms to examine specific hypotheses: (1) that statutory efforts to inform public benefits program recipients about the EITC would be more strongly associated with EITC receipt for families who received Medicaid, SNAP, WIC, and SSI; and (2) that the availability of EITC information in Spanish would be more strongly associated with EITC receipt for families in which at least one parent spoke English less than well. Our results support these hypotheses.
Hispanic families in which a parent is self-employed or less fluent in English—or that includes foreign-born household members—are less likely to receive the EITC, while Hispanic families who participate in other public programs are more likely to receive the EITC.
With respect to family-level characteristics, we find that Hispanic families that include a parent who is self-employed are less likely to receive the EITC than those that do not include a self-employed parent. In addition, Hispanic families that include a parent with limited English proficiency, or that live in a household with other relatives, are less likely to receive the EITC than Hispanic families without either of these characteristics. Families that live in the same household with a non-U.S. citizen or a naturalized U.S. citizen are less likely to receive the EITC than families for whom all household members are U.S.-born citizens (see Figure 3). We find higher likelihoods of EITC receipt for Hispanic families participating in any of the Medicaid, SNAP, WIC, and SSI programs. On the other hand, we do not find a statistical association between lacking employer-sponsored insurance or periodic unemployment and EITC receipt.[42]
Notes: The points in the figure above represent odds ratios; the lines represent the 95% confidence interval. All models include EITCeligible Hispanic families with dependent children (N=2,117). Additional covariates include a range of demographic and economic family-level characteristics; state policy, demographic, economic, and contextual factors; and year fixed effects. For model specification, see Table A.3 in the appendix.
Hispanic families who live in states that have a larger proportion of Hispanic residents are more likely to receive the EITC than Hispanic families in other states.
For the most part, the state-level demographic and contextual factors included in our models are generally not associated with EITC receipt among Hispanic families; however, there are a few exceptions. First, although Hispanic families are less likely to receive the EITC, those that live in states with a larger Hispanic share of the population are more likely to report receipt than Hispanic families living in states with lower proportions of Hispanic residents. This suggests that the size of the Hispanic community within a state may increase the likelihood of receipt. Second, Hispanic families living in states that allow unauthorized immigrants to obtain drivers’ licenses are more likely to report receipt, suggesting that this particular policy may increase the likelihood of receipt—perhaps through the increased employment and information exchange that come with mobility, or as an indicator of a state environment that is more broadly supportive of immigrants.
Discussion
Millions of moderate- and low-income working families claim the EITC each year and benefit from the additional income it provides.[43] However, millions more eligible families stand to benefit as well, and it is critical to understand the extent to which the EITC—an important poverty reduction strategy[44]—is not equitably reaching all eligible families. Research suggests that Hispanic families, in particular, are less likely to be aware of or claim the EITC,[45],[46] with Hispanics making up 24 percent of the EITC-eligible population in 2009 but only 15 percent of those receiving the federal EITC.[47]
Our results generated new estimates showing lower rates of EITC receipt among eligible Hispanic families with children (46%), compared to White (55%) and Black (58%) families with children. We find that more generous state EITCs, laws requiring that certain populations be informed about EITC, the provision of EITC information in Spanish, and higher densities of free tax preparation sites are associated with a greater likelihood of EITC participation among Hispanic families. We also find that states with higher proportions of Hispanic residents tend to engage in multiple policies and practices designed to support EITC participation, and that Hispanics in these states are more likely to receive the EITC.
Generated estimates of EITC receipt are lower than prevailing estimates of 75 percent.[48],[49] The lower rate of EITC participation among Hispanic families is on par with findings from other studies,[50],[51],[52] as are the lower overall estimates of EITC participation.[53],[54],[55],[56] Several factors may be at play in the underreporting of EITC receipt, especially in the SIPP—namely, SIPP only asks the question about EITC receipt of respondents with incomes below a certain level. Additionally, the question specifically asks about EITC receipt on the previous year’s tax return, which may not be easily recalled. Moreover, prior research suggests that some EITC recipients may be unaware that they received the credit because they received external assistance with tax preparation or they conflated it with the child tax credit.[57],[58] Thus, underreporting of EITC receipt may instead reflect a lack of knowledge of the credit.
Non-filers are an important population to consider when assessing EITC participation. Many have incomes below the threshold required to file taxes, which means they are more likely to face additional burdens to claiming the credit and less likely to be aware of the credit (or aware they can receive the credit even if they have no tax liability).[59] Our descriptive analyses indicate that a greater percentage of Hispanic families (8%) fall into this category, when compared to the sample as a whole (6%).
Our findings offer several insights about reaching Hispanic families, particularly among those families with a parent who is self-employed or has limited English proficiency, or among households that include other relatives, naturalized citizens, or non-U.S. citizens. Self-employed workers, for example, are less likely to benefit from formal systems through which they might receive information about the EITC.[60] Parents whose first language is Spanish and who are less fluent in English are also less likely to be provided support to better understand the complex eligibility rules and guidelines for completing the application process.[61]
The presence of extended family members in the same household may diminish the need for the income support that the EITC provides.[62] And families who live in a household that includes non-U.S. citizens may be more likely to be unaware of the EITC, unsure about their eligibility, experience difficulty navigating the tax filing process, or have concerns about privacy and use of personal information;[63] research suggests that these challenges can extend to households that include naturalized citizens as well.[64] Each of these factors point to critical avenues for policy and outreach that can support increased EITC participation among Hispanic families.
Finally, our analysis uncovers several promising strategies that states have pursued; collectively, these strategies reduce burden and can support EITC receipt. Strategies include mandated administrative outreach efforts that are relatively simple and cost-effective to implement, such as the provision of EITC information in Spanish and the requirement that employers inform employees and public benefits programs inform recipients about EITC. Low program awareness and informational complexity contribute to low EITC participation[65],[66],[67] and are in line with field research that finds that notices and reminders can increase both tax filing[68] and EITC receipt.[69],[70] Second, increased availability of and access to free tax preparation services can increase both tax filing and EITC receipt. Two policies that stand out—even after accounting for other detailed family, state, and policy characteristics—are higher levels of refundable state EITC amounts and greater density of free tax preparation sites.
Conclusion
The federal EITC is currently the most promising anti-poverty strategy in the United States that supports the well-being of children and families. Additionally, many states implement companion policies and practices that increase EITC participation among their residents. State investment in higher rates of EITC receipt not only benefits individual families but also states themselves: This investment puts cash in the hands of families who spend in their local economies,[71] boosts labor force participation,[72] and reduces use of other state supports.[73] These benefits far outweigh the costs, which are low for many of these policies and practices (e.g., public campaigns, translation of information). Moreover, it is critical that states invest now in Hispanic children, who will comprise a large portion of the future U.S. workforce and taxpayers. Several states, which tend to have higher proportions of Hispanic populations, are already engaging in a bundle of strategies to increase awareness and reduce administrative burden; these efforts appear to be associated with increased receipt among Hispanic families with children. More states should follow suit.
Lisa. A. Gennetian is the Pritzker Associate Professor of Early Learning Policy Studies at Duke Sanford Center for Child and Family Policy. Santiago Deambrosi is an Associate in Research at Duke Sanford Center for Child and Family Policy.
Lisa A. Gennetian and Yiyu Chen contributed equally to this work.
Support for this research was provided by the Robert Wood Johnson Foundation and the William T. Grant Foundation. The views expressed here do not necessarily reflect the views of the Foundations.
Download
References
[1] Center on Budget and Policy Priorities. (2019). Policy Basics: The Earned Income Tax Credit. Washington, D.C.: Center on Budget and Policy Priorities. Retrieved from https://www.cbpp.org/research/federal-tax/policy-basics-the-earned-income-tax-credit
[2] Hoynes, H. W., & Patel, A. J. (2018). Effective policy for reducing poverty and inequality? The Earned Income Tax Credit and the distribution of income. Journal of Human Resources, 53(4), 859-890.
[3] Hoynes, H., & Rothstein, J. (2016). Tax Policy Toward Low-Income Families (No. w22080). Cambridge, MA: National Bureau of Economic Research. Retrieved from https://www.nber.org/system/files/working_papers/w22080/w22080.pdf
[4] National Academies of Sciences, Engineering, and Medicine. (2019). A Roadmap to Reducing Child Poverty. Washington, DC: The National Academies Press. https://doi.org/10.17226/25246.
[5] Center on Budget and Policy Priorities. (2019). Policy Basics: The Earned Income Tax Credit. Washington, D.C.: Center on Budget and Policy Priorities. Retrieved from https://www.cbpp.org/research/federal-tax/policy-basics-the-earned-income-tax-credit#:~:text=The%20EITC%20is%20%E2%80%
9Crefundable%2C%E2%80%9D,for%20a
%20family%20without%20children.
[6] Chetty, R., Friedman, J. N., & Rockoff, J. (2011). New evidence on the long-term impacts of tax credits. IRS Statistics of Income White Paper. Washington, DC: Internal Revenue Service. Retrieved from https://www.irs.gov/pub/irs-soi/11rpchettyfriedmanrockoff.pdf
[7] Dahl, G. B., & Lochner, L. (2012). The impact of family income on child achievement: Evidence from the earned income tax credit. American Economic Review, 102(5), 1927-56.
[8] Hoynes, H., Miller, D., & Simon, D. (2015). Income, the earned income tax credit, and infant health. American Economic Journal: Economic Policy, 7(1), 172-211.
[9] Braga, B., Blavin, F., & Gangopadhyaya, A. (2020). The long-term effects of childhood exposure to the earned income tax credit on health outcomes. Journal of Public Economics, 190, 104249.
[10] Markowitz, S., Komro, K. A., Livingston, M. D., Lenhart, O., & Wagenaar, A. C. (2017). Effects of state-level Earned Income Tax Credit laws in the US on maternal health behaviors and infant health outcomes. Social Science & Medicine, 194, 67-75.
[11] Internal Revenue Service (IRS). (2020). Statistics for Tax Returns with EITC. Washington, D.C.: Internal Revenue Service. Retrieved from https://www.eitc.irs.gov/eitc-central/statistics-for-tax-returns-with-eitc/statistics-for-tax-returns-with-eitc
[12] Plueger, D. (2009). The Earned Income Tax Credit Participation Rate for Tax Year 2005. Internal Revenue Service Bulletin. Washington, DC: Internal Revenue Service. Retrieved from https://www.irs.gov/pub/irs-soi/09resconeitcpart.pdf
[13] Jones, M. R. (2014). Changes in EITC eligibility and participation, 2005-2009 (Working Paper #2014-04). Washington, DC: U.S. Census Bureau: Center for Administrative Records Research and Applications. Retrieved from https://www.census.gov/content/dam/Census/library/
working-papers/2014/adrm/carra-wp-2014-04.pdf
[14] Phillips, K. R. (2001). Who Knows about the Earned Income Tax Credit? (Series B, No.B-27). Washington DC: Urban Institute. Retrieved from: http://webarchive.urban.org/publications/310035.html
[15] Maag, E. (2005). Paying the Price? Low-Income Parents and the Use of Paid Tax Preparers. (Series B, No. B-64). Washington, DC: Urban Institute. Retrieved from http://www.urban.org/sites/default/files/publication/51786/411145-Paying-the-Price-.PDF
[16] Scholz, J. K. (1994). The earned income tax credit: Participation, compliance, and antipoverty effectiveness. National tax journal, 63-87.
[17] Bhargava, S., & Manoli, D. (2015). Psychological frictions and the incomplete take-up of social benefits: Evidence from an IRS field experiment. American Economic Review, 105(11), 3489-3529.
[18] Jones, M. R. (2014). Changes in EITC eligibility and participation, 2005-2009 (Working Paper #2014-04). Washington, DC: U.S. Census Bureau: Center for Administrative Records Research and Applications. Retrieved from https://www.census.gov/content/dam/Census/library/working-
papers/2014/adrm/carra-wp-2014-04.pdf
[19] Phillips, K. R. (2001).
[20] Scholz, J. K. (1994).
[21] Phillips, K. R. (2001).
[22] Maag, E. (2005).
[23] Scholz, J. K. (1994).
[24] Perreira K.M., & Pedroza, J.M. (2019). Policies of exclusion: implications for the health of immigrants and their children. Annual Review of Public Health, (40), 147-166.
[25] Crandall-Hollick, M. L. (2018). The Earned Income Tax Credit (EITC): A Brief Legislative History. Washington, DC: Congressional Research Service. Retrieved from https://fas.org/sgp/crs/misc/R44825.pdf
[26] Internal Revenue Service (IRS). (2020). How Do I Claim the EITC? Washington, D.C.: Internal Revenue Service. Retrieved from https://www.irs.gov/credits-deductions/individuals/
earned-income-tax-credit/claiming-earned-income-tax-credit-eitc
[27] Internal Revenue Service (IRS). (2020). Do I Qualify for the EITC? Washington, D.C.: Internal Revenue Service. Retrieved from https://www.irs.gov/credits-deductions/individuals/earned
-income-tax-credit/do-i-qualify-for-earned-income-tax-credit-eitc
[28] Neumark, D. & Williams, K. E. (2016). Do State Earned Income Tax Credits Increase Participation in the Federal EITC? (Working Paper #20163). Irvine, CA: Economic Self-Sufficiency Policy Research Institute. Retrieved from https://www.esspri.uci.edu/files/docs/2016/ESSPRI%20Working%20Paper%2
020163%20Neumark%20and%20Williams.pdf
[29] Guyton, J., Manoli, D. S., Schafer, B., & Sebastiani, M. (2016). Reminders & Recidivism: Evidence from Tax Filing & EITC Participation among Low-Income Nonfilers (No. w21904) Cambridge, MA: National Bureau of Economic Research. Retrieved from https://www.nber.org/papers/w21904
[30] Manoli, D., & Turner, N. (2016). Nudges and Learning: Evidence from Informational Interventions for Low-Income Taxpayers (Working Paper #20718). Cambridge, MA: National Bureau of Economic Research. Retrieved from: https://www.nber.org/papers/w20718
[31] Bhargava, S., & Manoli, D. (2015).
[32] Maag, E. (2005).
[33] Marcil, L. E., Hole, M. K., Wenren, L. M., Schuler, M. S., Zuckerman, B. S., & Vinci, R. J. (2018). Free tax services in pediatric clinics. Pediatrics, 141(6), e20173608.
[34] Maag, E. (2005).
[35] Mammen, S., Lawrence, F. C., Marie, P. S., Berry, A. A., & Knight, S. E. (2011). The earned income tax credit and rural families: Differences between non-participants and participants. Journal of Family and Economic Issues, 32(3), 461-472.
[36] Because reports of EITC receipt were drawn from the subsequent wave of SIPP data (relative to the income and demographic data used to determine eligibility), a family who was not interviewed at the next wave was considered missing.
[37] For cohabiting-parent families, we run simulations separately for the two parents. In each round, we include income information from only one parent and treat both joint children (with the cohabiting partner) and own children (with someone else) as dependents. A cohabiting-parent family is included in the EITC-eligible sample as one family when a positive total EITC amount is simulated for at least one of the parents.
[38] Note that we limit our final sample to families with children under 18, although some of those families can claim their older children as dependents.
[39] Kenney, C. (2004). Cohabiting couple, filing jointly? Resource pooling and US poverty policies. Family Relations, 53(2), 237-247.
[40] Shapiro, I. (2019). State EITC provisions 1977-2018. Cambridge, MA: National Bureau of Economic Research. Retrieved from https://users.nber.org/~taxsim/state-eitc.html
[41] Lenhart, O. (2019). The effects of state‐level earned income tax credits on suicides. Health Economics, 28(12), 1476-1482.
[42] Additional demographic factors were also associated with EITC receipt. Hispanic families with higher annual wages, more dependent children, with parents who were older were more likely to receive EITC. Single parents were more likely than married parents to receive EITC, and living in a household with other relatives was associated with reduced likelihood of receipt.
[43] Bitler, M., Hoynes, H. W., & Schanzenbach, D. W. (2020). The social safety net in the wake of COVID-19 (No. w27796). National Bureau of Economic Research.
[44] Hoynes, H. W., & Patel, A. J. (2018).
[45] Phillips, K. R. (2001).
[46] Maag, E. (2005).
[47] Short, K., Donahue, D., & Lynch, G. (2012). EITC estimates in the CPS ASEC simulations of after-tax income: Hispanic population (Working Paper #2012-19). Washington DC: U.S. Census Bureau: Social, Economic, and Housing Statistics Division. Retrieved from https://www.census.gov/library/working-papers/2012/demo/SEHSD-WP2012-19.html
[48] Scholz, J. K. (1994).
[49] Plueger, D. (2009).
[50] Phillips, K. R. (2001).
[51] Maag, E. (2005).
[52] Scholz, J. K. (1994).
[53] Bitler, M. P., Currie, J., & Scholz, J. K. (2003). WIC eligibility and participation. Journal of Human resources, 1139-1179.
[54] Mikelson, K. S. & Lerman, R.I. (2004). Relationship between the EITC and Food Stamp Program Participation Among Households with Children. Washington, DC: Urban Institute. Retrieved from https://www.urban.org/research/publication/relationship-between-eitc-and-food-stamp-program-participation-among-households-children
[55] Meyer, B. D., Mok, W. K., & Sullivan, J. X. (2009). The Under-Reporting of Transfers in Household Surveys: Its Nature and Consequences (No. w15181). Cambridge, MA: National Bureau of Economic Research. Retrieved from https://www.nber.org/papers/w15181
[56] These studies suggest that self-report survey data captures about 70-80% of administratively reported participation, but note that the characteristics of families reporting receipt of the specific programs of interest were similar to administratively reported characteristics of recipients nationally, suggesting that the undercount may be mostly random (BItler, Currie, & Scholz, 2003; Mikelson, 2004). If we assume similar amounts of underreporting in our own SIPP data, this would translate to an overall EITC participation rate of between 65-74%.
[57] Maag, E. (2005).
[58] Mammen, S., Lawrence, F. C., Marie, P. S., Berry, A. A., & Knight, S. E. (2011).
[59] Guyton, J., Manoli, D. S., Schafer, B., & Sebastiani, M. (2016). Reminders & recidivism: evidence from tax filing & EITC participation among low-income nonfilers (No. w21904). Cambridge, MA: National Bureau of Economic Research. Retrieved from https://www.nber.org/papers/w21904
[60] Nightingale, D. S. (2011). Informal and Nonstandard Employment in the United States: Implications for Low-Income Working Families. Washington, DC: Urban Institute. Retrieved from http://webarchive.urban.org/UploadedPDF/412372-informal-nonstandard-employment-in-us.pdf
[61] Fremstad, S. (2003). Immigrants, Persons with Limited Proficiency in English, and the TANF Program: What Do We Know? Washington, DC: Center on Budget and Policy Priorities. Retrieved from https://www.cbpp.org/archiveSite/3-18-03tanf.pdf
[62] Reyes, A. M. (2019). Mitigating Poverty through the Formation of Extended Family Households: Race and Ethnic Differences. Social Problems, Volume 67, Issue 4, Pages 782-799.
[63] Twersky, S. E. (2019). Restrictive state laws aimed at immigrants: Effects on enrollment in the food stamp program by US citizen children in immigrant families. Plos one, 14(5), e0215327.
[64] Perreira K.M., & Pedroza, J.M. (2019).
[65] Bhargava, S., & Manoli, D. (2015).
[66] Manoli, D., & Turner, N. (2016).
[67] Chetty, R., Friedman, J. N., & Saez, E. (2013). Using Differences in Knowledge across Neighborhoods to Uncover the Impacts of the EITC on Earnings. American Economic Review, 103(7), 2683-2721.
[68] Guyton, J., Manoli, D. S., Schafer, B., & Sebastiani, M. (2016). Reminders & Recidivism: Evidence from Tax Filing & EITC Participation among Low-Income Nonfilers (No. w21904). National Bureau of Economic Research. Retrieved from https://www.nber.org/papers/w21904
[69] Manoli, D., & Turner, N. (2016).
[70] Bhargava, S., & Manoli, D. (2015).
[71] Aval.os, A., & Alley, R. S. (2010). The economic impact of the Earned Income Tax Credit (EITC) in California. California Journal of Politics and Policy, 2(1).
[72] Eissa, N., & Hoynes, H. W. (2006). Behavioral responses to taxes: Lessons from the EITC and labor supply. Tax policy and the economy, 20, 73-110.
[73] Grogger, J. (2003). The effects of time limits, the EITC, and other policy changes on welfare use, work, and income among female-headed families. Review of Economics and statistics, 85(2), 394-408
