The earned income tax credit: a study of eligible participants vs. non-participants.

Using data from the National Longitudinal Survey of Youth, this study (N = 1,504) showed that about half the EITC eligible tax fliers in 2001 did not file EITC tax returns and that differences between EITC tax fliers and non-EITC tax fliers varied by birth place, Food Stamp program participation, marital status, race, residence, sex, socioeconomic history, and worker classification. Findings suggested that the EITC is well targeted in the sense that economically marginalized groups are likely to participate and that increased outreach efforts are also needed to ensure greater participation among tax fliers eligible for the EITC but who are less likely to claim it, especially self-employed persons and those residing in the Northeast.

Keywords: Earned income tax credit (EITC), poverty, social policy, tax policy, welfare reform

**********

The Earned Income Tax Credit (EITC) is one of the most significant cash-transfer programs for low-income families in the United States. It serves single and/or married individuals, with or without children, as long as they meet income eligibility. The EITC offsets in part the erosion of Federal responsibility to poor families, formalized by the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 which ended the entitlement nature of means-tested cash-based welfare provision (Bok & Simmons, 2002; Careley, 1996; Ozawa, 1995). EITC expenditures exceed those of Food Stamps and the Temporary Assistance for Needy Families (TANF) program, although they fall shy of the Supplemental Security Income (SSI) program. Although the EITC currently targets working poor individuals regardless of whether they have children, it lifts more children out of poverty than any other means-tested or social insurance program (U.S. Congress, 2004b, 2004c). It raises the income of many low-income workers above the poverty line, increases labor force participation among low-income single mothers, and enables many low-income families in general to meet immediate consumption as well as longer-term investment needs (Center on Budget and Policy Priorities, 1998).

Despite some misgivings in regard to compliance, the EITC has enjoyed bi-partisan political support since its inception in 1975 when it was viewed as a means of providing an offset to Social Security (payroll) taxes paid by low-income workers with children. It retains a pro-work, pro-family, and anti-welfare appeal that both Republican and Democratic presidential administrations have used to justify the 30-year Federal subsidy to low-income families. The EITC has become one of the main Federal programs to subsidize the working poor, to provide an economic incentive to retain their labor force attachment, and thereby to decrease their likelihood of remaining in poverty and of relying on TANF, Food Stamps, and other forms of public assistance.

This paper provides an overview of the EITC program and it reports results of a study that examined the likelihood of EITC participation among eligible taxpayers in 2001. The comparison between eligible participants and non-participants should provide insight into why some eligible households fail to claim the credit and help increase the effectiveness of EITC outreach efforts.

Literature Review

About the EITC

The EITC is one of nine major means-tested programs. The others are Food Stamps, Medicaid, Supplemental Security Income (SSI), TANF, the child support enforcement program, programs subsidizing child care, housing programs, and employment and training programs. The EITC is a unique means-tested program because it is a refundable tax credit. That is, if the amount of the credit exceeds the taxpayer's Federal income tax liability, the excess is payable to the taxpayer as a direct transfer payment. As a transfer payment, the EITC is like other Federal programs that provide poor and low-income families with public benefits. Since it requires earnings, the EITC is unique from other Federal benefit programs (U.S. Congress, 2004c). As a tax expenditure, the EITC is like the deduction of home mortgage interest (HMI) and the exclusion of employer pension (EPP) contributions from income, the two largest and best-known tax expenditures, which, however, are not means-tested (Holtzblatt, 2000). Howard (1997) provides a useful guide to the historical and political precedents of tax expenditures with social policy objectives, including the targeted jobs tax credit (TJTC), in addition to the EITC, HMI, and EPP. Weisbach and Nussim (2004) provide a more technical discussion of organizational and institutional design issues relevant to an integration of tax and spending programs.

Enacted during the Ford administration in 1975 as a way to offset the burden of Social Security tax on low-income working parents, the EITC generally equals a specified percentage of wages up to a maximum dollar amount. The maximum amount applies over a certain range of income and diminishes to zero over a certain income range. The EITC thereby has three ranges: phase-in, maximum credit, and phase-out. In the phase-in range, the EITC acts as a wage subsidy--as the family earns more, the transfer increases. In the maximum credit range, the transfer remains constant regardless of earnings. In the phase-out range, the EITC acts like a negative income tax--as the family earns more, the transfer is reduced. While the phase-in range provides a work incentive, the maximum and phase-out ranges have work disincentives for some families (Horowitz, 2002; see Moffitt (2003) for a history of the negative income tax in U.S. welfare policy). The income ranges and percentages have increased several times since 1975, expanding the credit, as have the numbers of participants. The 1975-2003 EITC parameters can be found in U.S. Congress (2004c); the corresponding number of recipient families and amount of credits can be found in U.S. Congress (2004b). It should be noted that unlike public assistance programs in some states, single-parent and two-parent families with similar income levels receive the same EITC benefit. Additionally, two-parent families with similar income levels receive the same EITC benefit regardless of whether one or both parents work (Greenstein & Shapiro, 1998).

It is also important to keep in mind that throughout the 1980s and 1990s the AFDC program was increasingly subjected to arguable criticism for sustaining, if not also creating, a class of "dependent" poor able-bodied persons, especially unmarried women with children. Ending welfare dependency had come to displace poverty reduction as a major social problem. Macroeconomic factors as causally relevant to the plight of poor persons in general and of poor mothers in particular were virtually drowned out or ignored in the policy debates about welfare. Poor mothers with young children were no longer considered "deserving" and the legitimacy of claims for cash assistance had eroded, aided in part by sustained attacks on the U.S. welfare state in general and on the AFDC program in particular since the early 1980s (Bowen, Desimone, & McKay, 1995; Caputo, 1997a, 1997b; Ellwood, 2000a; Handler, 1995; Hoynes, 1995).

Rules to tighten eligibility compliance and to improve enforcement of the EITC were incorporated into the Personal Responsibility and Work Opportunities Act of 1996, the Taxpayer Relief Act of 1997, and the Balanced Budget Act of 1997, each during the Clinton administration. It should also be kept in mind that the minimum wage, an alternative mechanism to raise the income of low-income workers, was increased to $5.15 in 1997 and that the EITC had been found to deliver a much larger proportion of a given dollar of benefits to low-income workers between 1989 and 1992 than increases in the minimum wage from $3.35 to $4.25 (Burkhauser, Couch, & Glenn, 1995). The Economic Growth and Tax Relief Reconciliation Act of 2001, the GW Bush administration's tax cut package, increased the beginning and ending amounts of the phase-out ranges for married taxpayers who filed a joint return and, among other provisions expanding the potential reach of EITC in general, excluded nontaxable employee compensation from the definition of earned income (Gale & Potter, 2002; Kiefer, D., et al., 2002; Smeeding, Ross, & O'Connor, 2000; U.S. Congress, 2004c). The three benefit formulas (for no children, one child, and two or more children) have remained in place since 1994 and the subsidy rate stabilized at 40 percent in 1997 (See U.S. Congress, 2004c). Although the levels have changed, Ozawa and Hong (2003) reported that a modified formula which included an additional category for three or more children with a subsidy rate of 46 percent would significantly reduce the poverty rates of children. This would be the case more so when combined with the child allowances that were part of the Economic Growth and Tax Relief Reconciliation Act of 2001. (1)

Prior studies of EITC

In addition to documenting the broad ideological support for the EITC, Greenstein and Shapiro (1998) and the Center on Budget and Policy Priorities (1998) have summarized much of the earlier research of its use and effects (Also see Blank, Card, & Robbins, 1999; Essa & Hoynes, 1999). More than half the increase in the proportion of single mothers in the labor force was found to be due to the effects of the EITC, especially those occurring after the expansion of maximum credit benefits in 1993 (Also, see Blank, 2000; Ellwood, 2000a & 2000b). The EITC was also found to offset between one-fourth and one-third of the decline in the national share of income received by the poorest fifth of households with children over a twenty-year period. It reduced poverty among children by one-fourth. U.S. Census data revealed that the EITC moved more children out of poverty than any other program or category of programs. This was especially the case among Hispanic children as well as among children in the South, where lower wages prevail and more low-income workers are likely to qualify.

More recent data showed that in 2002 the anti-poverty effectiveness of the EITC and Federal taxes removed roughly the same percentage of persons from poverty as Social Insurance (10.9% vs. 11.7% respectively) and higher percentages than means-tested non-cash programs (9.7%) and means-tested cash programs (3.5%). These percentages were a striking contrast to the 1979 figures of 15.3% for means-test non-cash programs, 10.9% for Social Insurance, 7.7% for means-tested cash program, and -1.6% for EITC and Federal taxes. The last, negative, percent was due to the regressive nature of the Social Security (payroll) tax which the EITC did not fully offset in the aggregate until 1993 (U.S. Congress, 2004a). Relying on data from the Panel Study of Income Dynamics (PSID) for the years 1975-1992, Horowitz (2002) reported that eligible EITC households were mobile and that most spells were short. Among those who were just starting a spell, 51 percent of spells lasted one year and 74 percent of spells were over in two years or less. Only 9 percent of spells lasted five years or longer. The average spell was 2.135 years.

Several of the earlier EITC studies focused on marriage penalties and bonuses, which are not the main foci of the present study. Suffice it to say for present purposes that research on taxes in general had consistently indicated that as the tax penalty on marriage increased, individuals were less likely to marry and more likely to divorce (Alm, Dickert-Conlin, & Whittington, 1999). The phase-in and phase-out levels of the EITC were thought to penalize married women who worked and to subsidize married women who did not work, thereby distorting marriage decisions and married women's labor force participation decisions (Dickert-Conlin & Houser, 2002). The EITC structure also implied that a single mother with no earnings who married a man with low earnings would qualify for the subsidy, thus providing a marriage subsidy; however, an EITC-eligible mother would become ineligible if she married and their combined income placed them beyond the phase-out range (Dickert-Conlin & Houser, 1999. Also, see Ellwood, 2000a&b; Holtzblatt & Rebelein, 2000).

Phillips (2001) noted that since the EITC is administered through the Federal tax system its use was contingent on how knowledgeable low-income taxpayers and likely taxpayers were about it. Relying on data from the 1999 National Survey of America's Families (NSAF), Phillips showed that knowledge about and use of the EITC varied by a number of sociodemographic characteristics. Nearly two-thirds of parents from the NSAF sample had heard of the EITC and nearly 30 percent had received it at some time. Low-income Hispanic parents were less knowledgeable about the program than low-income non-Hispanic parents of any race. In addition, among knowledgeable low-income parents, Hispanic parents were less likely to receive the credit. Married low-income parents were less likely than divorced/separated and never married parents to know about or receive the EITC. Divorced/separated parents were less likely than all parents to have ever used the program. Parents who had a history of welfare receipt had greater knowledge of EITC than were those with no such history. Whether or not a state implemented its own EITC program was found to have no effect on either knowledge or use of the Federal program.

Relying on 1990 SIPP data, Scholz (1994) found that higher income taxpayers were more likely to file returns among EITC eligible persons than lower-income taxpayers, although greater levels of income derived from self-employment decreased the likelihood of EITC participation. In addition, the likelihood of EITC participation among eligible tax filers was positively related to the size of the potential EITC payment. Taxpayer characteristics associated with non-participation included AFDC participation, Social Security participation, larger family size, being unmarried, being male, and being of Spanish origin. Those with college degrees were also less likely to participate than were those without high school diplomas. Those in private household occupations such as launderers, cooks, housekeepers, and child care workers, as well as equipment cleaners, helpers, and laborers were less likely to receive the EITC than were executives, managers, and those working in specialty professional occupations.

Finally, as the preceding survey of the literature indicates, most studies of the EITC relied on micro-level data and drew conclusions about individual responses to the program's incentive structure, especially in regard to marriage, work, and welfare. A notable exception, which goes beyond the scope of the present study, is Edwards (2004) who reported that the EITC is a better fiscal stimulus tool than broad-based tax refunds.

The aforementioned micro-level studies suggested that the EITC is a well-known and well-targeted program that works. Specific subgroups of the population, however, especially Hispanic parents, were reported to be less likely to know about and hence to use EITC. The present study sought to add to the knowledge base of EITC in several ways. It included single low-income workers without children, as well as those with children. The present study took into account past tax filing behavior, which prior studies have omitted. The assumption here was that persons with a history of not filing taxes would probably be less likely to use the EITC. The present study also took advantage of longitudinal panel data by making use of family background items and cumulative measures of income status and work history obtained at the time of interview rather than relying on respondents' recall of such measures as is the case in retrospective cross-sectional EITC studies. For example, it includes measures of work history because evidence suggested that the EITC induces labor market entry in families that do not initially have an adult worker (Neumark & Wascher, 2001). Hence a hypothesized positive relationship between EITC receipt and work history might be somewhat offset by those with weak to moderate work histories who might be EITC-eligible but perhaps not apply. Results were meant to further the outreach efforts of the Federal government and advocacy groups. The author thought that it was important to continue to monitor the use of the EITC to ensure that it remains a well-targeted and effective program and by extension, stands a reasonable probability of surviving at a time when reduction of the Federal deficit may necessitate tax increases to offset lost tax revenue from the EITC despite the efforts of the anti-tax, anti-welfare G.W. Bush administration.

Method

Data and Subjects

Data were obtained from the National Longitudinal Survey of Youth (NLSY), a nationally representative sample of 12,686 young men and women who were 14-22 years old when they were first surveyed in 1979. In 2002, the most recent year of available data, there were 7,724 respondents, representing 60.9 percent of the original sample. The primary purpose of the NLSY was collection of data on each respondent's labor force experiences, labor market attachment, and investments in education and training. The NLSY was deemed suitable for purposes of the present study because it asked questions about tax-filing behavior in general and about the EITC in particular (Center for Human Resource Research, 2004).

Reliance on the NLSY as the single source of data was not seen as problematic in light of the battery of significant legislative measures designed to curb the types of errors identified by IRS and other studies (Alstott, 1994; Greenstein & Shapiro, 1998; Kiefer, et al., 2002; McCubbin, 2000). In addition, Federal and advocacy efforts inform low-income persons about the program and about free tax preparation services, so public awareness is likely to have increased significantly since the time of earlier studies and reports (e.g., see Shipler, 2004). Furthermore, questions about the EITC were raised after respondents were asked about whether they filed taxes in the previous calendar year, a question that had been asked in prior survey years. Nonetheless, several limitations of the NLSY are of note. The 39 percent attrition rate invariably fell disproportionately on the lower income individuals and their families. The NLSY also represented only one cohort of the U.S. population first surveyed in 1979 and hence this study excludes more recent cohorts of young families, which regarding poverty, are of interest. Study findings and implications are presented and discussed with these limitations in mind.

Measures

EITC eligibility: To be eligible for EITC in 2001, federal guidelines required that a respondent had to live in a household with reported regular/military income less than or equal to $4,760 with no children, $7,140 with one child, and $10,020 with more than one child (U.S. Congress, 2004c, Table 13-12). The determination of EITC eligibility was made mechanically at the time of interview by the survey software on the basis of respondents' responses to "lead-in" questions about sources of income, household composition (specifically for the presence of a spouse), whether any biological children had ever been reported, and whether they filed tax forms for 2001 (McClaskie, 2005). Since this procedure in all likelihood resulted in an over-count of persons eligible for the EITC, the presence of children was determined from household composition items in the NLSY79. The study sample included only those respondents who had been determined EITC eligible by the mechanical method and about whom all other information was available.

Sociodemographic characteristics: These included age, education (highest grade completed), ethnicity (Hispanic vs. other), health status (whether health limited the amount or type of work one could do), marital status (married [reference category in multivariate analysis], never married, or separated/widowed/divorced), number of children in household, residence (urban vs. rural), race (White vs. other), SES history (the number of years respondents lived in families whose incomes fell below official poverty lines that accounted for family size), and sex. Background control measures included country of birth (US born vs. other) and mother's education, region of residence (Northeast, North Central, South, or West).

Class of worker: This measure comprised a series of dummy variables. It signified whether respondents were employed by government, private for profit companies, non-profit organizations (including tax-exempt and charitable), or family businesses (including self-employed).

Work history and related measures: Work history was captured by three measures: the average number of weeks worked per year between 1979 and 2000, the average number of week unemployed (vis-a-vis out of the labor force) per year between 1979 and 2000, and the number of weeks worked in calendar year 2001. In addition, the percentage unemployed in the area in which respondents lived was included because, as Ellwood (2000a) notes, reforms designed to support people who are working fail if people cannot find work. This measure was used as a control. Finally, the average annual percent of weeks unaccounted for in the NLSY in accounting for weeks worked was also included as a control variable.

Public Assistance Use: The public assistance measures included whether respondents reported they participated in the Food Stamps, Supplemental Security Income (SSI), or Temporary Assistance for Needy Families (TANF) programs. Each measure was coded such that 1=Yes, 0--No. Since the income eligibility thresholds are higher for Food Stamps and SSI than for TANF, participation in EITC was thought to be more likely for Food Stamp and SSI recipients.

Results

Nearly half (49%) the eligible taxpayers in the study sample filed for the EITC. As can be seen in Table 1, EITC filing status differed by age, class of worker, country of birth, marital status, mother's education, public assistance receipt, race, region of residence, SES history, sex, and work history. As Panel A shows, EITC tax fliers were younger than non-EITC tax filers (40.7 vs. 41.1 years old), less educated mothers (9.9 vs. 10.3 years of completed schooling), lived in poor families for a greater number of years (5.5 vs. 3.4), and on average worked fewer weeks per year between 1982 and 2000 (31.5 vs. 34.8). As Panel B shows, higher percentages of those born in the U.S. (94.9% vs. 92.5%), separated/divorced/widowed (49.3% vs. 32.4%), female (71.6% vs. 46.9), government employees (17.2% vs. 12.7%), Food Stamp recipients (9.8% vs. 2.2%), and TANF recipients (1.9% vs. 0.5%) were EITC tax filers than were non-EITC tax filers. Higher percentages of those married (43.7% vs. 28.3%), White (57.3% vs. 48.2%), Northeast residents (15.7% vs. 9.8%), and self-employed or in family businesses (15.0% vs. 7.4%) were non-EITC tax filers than were EITC tax filers.

As can be seen in Table 2, country of birth, marital status, number of children in household, number of years as a tax filer, number of years lived in a poor family, region of residence, sex, class or worker, use of public assistance, and work history were robust correlates of EITC tax filing status. When accounting for all study measures, those more likely to file for the EITC were born in the U.S. (1.8 times as likely), never married vs. married (1.6 times), separated/divorced/widowed vs. married (2.9 times), women (1.9 times), government employees vs. self-employed (1.9 times), private for profit sector employees vs. self-employed (1.8 times), non-profit organization employees (2.1 times), and Food Stamp recipients (3.1 times). Also, each increase in the number of children in a household increased the odds of filing for the EITC by 71 percent, while each additional year of being a tax filer increased the odds of filing for the EITC by 11 percent, of living in a poor family by 10 percent, and of working each additional week by 1.0 percent. Residence in the Northeast US decreased the odds of filing for the EITC by 38% in comparison to residence in the South.

Discussion

Findings of this study suggest that the EITC remains an increasingly well-targeted program. Food Stamp recipients, women, those with greater numbers of children, and separated/divorced/widowed persons are more likely than their low-income tax filing counterparts to participate in the EITC. These findings are consistent with much prior research (e.g., Blank, 2000; Blank, Card, & Robbins, 1999; Center on Budget and Policy Priorities, 1998; Ellwood, 2000a & 2000b; Essa & Hoynes, 1999; Greenstein & Shapiro, 1998). Corroboration of previous research demonstrating the well-targeted nature of the EITC is important for policy makers and EITC advocates to know. If nothing else, it signifies that government policy based on work-related incentives can work as intended to the benefit of low-income individuals and their families. Furthermore, as Alstott (1999) shows, the EITC has to compete with other federal wage subsidy schemes that go directly to employers rather than to individuals (e.g., the Welfare-to-Work Tax Credit launched during the Clinton administration) and it has several competitive advantages over such tax credit programs that can hold up under empirical scrutiny. For example, the EITC is more likely than wage subsidies that go to employers to bring non-employed persons into the labor market and to increase the amount of work that part-time workers do.

Findings of this study differ somewhat from Phillips (2001) who reported that divorced/separated parents were less likely than all parents to have ever used the program and that Hispanics were less likely than others to know about or participate in the EITC. The disparate findings are probably an artifact of different samples. The present study shows that low-income married people were less likely to receive EITC. Alternatively, greater outreach efforts by the Federal government and community groups in the late 1990s and early 2000s as noted below may have also contributed to different findings. Both the present study and Phillips, however, show that married low-income persons and parents respectively are less likely than never married and divorced/separated persons and parents to receive the EITC. On the whole, findings of the present study as well as those of Phillips and others suggest that the EITC rewards economically marginalized groups of working taxpayers with a greater level of income than would be the case otherwise. In doing so, the EITC continues to serve the purposes of social justice by expanding the opportunity set among those families whose economic circumstances are precarious at best (Barry, 2005).

Findings of the present study also suggest that increased outreach efforts are also needed to ensure greater participation among those eligible for the EITC. This is especially the case for married persons, those who file tax returns less frequently over the years, and corroborating Scholz (1994), those who are self-employed or work for a family business. In 2001 married persons may have been reluctant to file due to purported penalty effects reported in prior research if filing would make them worse off than not filing for the EITC (Dickert-Conlin & Houser, 1999; Ellwood, 2000a&b; Holtzblatt & Rebelein, 2000). To the extent the EITC is perceived as rewarding married parents less than others, especially never married, separated, or divorced parents, its political popularity with both Democratic and Republican politicians and other policy makers may erode despite the program's pro-work emphasis. The EITC might suffer the same fate as AFDC did in 1996 when, as previously noted, it was replaced with TANF (Bowen, Desimone, & McKay, 1995; Ellwood, 2000a; Fraser & Gordon, 1994; Handler, 1995; Haveman & Scholz, 1994). Also as noted however, prior to 2002 persons with "married filing separate" status were ineligible for the EITC. Separate filing status may remove the marriage penalty for many couples and their families and thereby enable them to benefit from the EITC program. In addition, married joint tax filers after 2001 should also benefit from the increased beginning and ending amounts of the phase-out ranges for married taxpayers who file a joint return as enacted in the Economic Growth and Tax Relief Reconciliation Act of 2001. The effects of such legislation on the EITC take-up rate for low-income married couples warrant future research. When providing services to low-income working families, social workers and other professionals should be aware of the EITC eligibility criteria and how they might affect married couples as tax filers.

Information about the EITC can be found at the Internal Revenue Service Internet site (http://www.irs.gov/individuals/ article/0,,id=96406,00.html). When no economically adverse consequences are likely to follow, couples should be encouraged to pursue the EITC when filing their taxes and to have an assessment done to determine the effects of filing taxes as a couple or as individuals. If they do not already do so, social and family service agencies with low-income working clients may consider incorporating financial counseling to their mix of services with the EITC in mind. This can be done by ensuring that professional providing psychotherapeutic counseling or other interpersonal services expand their clinical treatment regimen and attend more explicitly than might be the case otherwise to their clients' socioeconomic well-being. It can also be done by offering "concrete" financial counseling services whose practitioners are well-versed in public programs targeting low-income individuals and families. At the least, low-income working clients ($36,000 and below) can be directed to the Volunteer Income Tax Assistance program (VITA), whose local sites can be obtained by calling 1-800-8291040. Low-income working elderly clients (60 years and over) can be directed to the Tax Counseling for the Elderly program (TCE), of which the AARP's Tax-Aide program is a part. Information about TCE can be found at 1-800-829-1040, while information about AARP's Tax-Aide program can be found at 1-888-227-7669 or by visiting the AARP Internet site (http://www.aarp.org/ money/taxaide/).

As noted, among eligible taxpayers those who file tax returns less frequently, as well as those who are self-employed or work for a family business, are also less likely to file tax returns for the EITC. Apparently offers of free tax preparation services in low-income neighborhoods and on the Internet fail to capture these two groups of individual taxpayers who can benefit from filing for the EITC. Social service agencies that provide individual and family financial counseling services might pay closer attention to the tax-filing histories and self-employment status of their otherwise EITC-eligible clients and make sure they understand that they are both eligible for the EITC and that it is to their economic advantage file for it, especially for families with two low-income workers who could benefit from filing their tax forms separately.

In conclusion, this study adds to the body of knowledge about the EITC, a program that remains well-targeted. It identifies groups of individual taxpayers whom social workers and other professionals working with low-income working clients can benefit from participation in the EITC program. Several free tax preparation services are also identified so clients can locate nearby offices to help them determine their eligibility and to file tax returns accordingly. Future research should explore to what extent the EITC provides a boost for those who are already on their way escaping poverty, and to discern the proportions of families who fall below qualifying for the credit, who are helped but still destitute, and who make it beyond what might be considered an acceptable economic marker, such as 150-200 percent of the poverty line. As noted above, future research should also determine the effects of related legislation passed in 2002 and in 2001.

References

Alm, J., Dickert-Conlin, S., & Whittington, L.A. (1999). Policy watch: The marriage penalty. The Journal of Economic Perspectives, 13, 193-204.

Alstott, A.L. (1999). Work vs. freedom: A liberal challenge to employment subsidies. The Yale Law Review, 108, 967-1058.

Alstott, A.L. (1994). The earned income tax credit and some fundamental institutional dilemmas of tax-transfer integration. National Tax Journal, 47, 609-619.

Barry, B. (2005). Why social justice matters. Cambridge, UK: Polity Press.

Bittker, B.I. (1975). Federal income taxation and the family. Stanford Law Review, 27, 1389-1463.

Blank, R.M. (2000). Fighting poverty: Lessons from recent history. Journal of Economic Perspectives, 14(2), 3-19.

Blank, R.M., Card, D., & Robbins, P.K. (1999). Financial incentives for increasing work and income among low-income families. Revised. Paper presented at the Joint Center for Poverty Research Conference, November 5-6, 1998. Retrieved February 10, 2005 from http://www.jcpr.org/wpfiles/Card_Blank _Robins.pdf?CFID=4124124&CFTOKEN=57570865

Bok, M., & Simmons, L. (2002). Post-welfare reform, low-income families and the dissolution of the safety net. Journal of Family and Economic Issues, 23, 217-238.

Bowen, G.L., Desimone, L.M., & McKay, J.K. (1995). Poverty and the single mother family: A macroeconomic perspective. Marriage & Family Review, 20(1-2), 115-142.

Burkhauser, R.V., Couch, K.A., & Glenn, A.J. (1995). Public policies for the working poor: The earned income tax credit versus minimum wage legislation. Institute for Research on Poverty Discussion Paper no. 1074-95. Retrieved February 19, 2005 from http://www.irp.wisc.edu/publications/dps/pdfs /dp107495.pdf

Caputo, R.K. (1997a). Escaping poverty & becoming self-sufficient. Journal of Sociology & Social Welfare, 24(3), 5-23.

Caputo, R.K. (1997b). Family poverty & public dependency. Families in Society, 78, 13-25.

Careley, D. (1996). Dismantling the safety net: Fictions versus realities. Political Science Quarterly, 111,225-258.

Center for Human Resource Research (2004). NLSY79 user's guide: A guide to the 1979-2002 National Longitudinal Survey of Youth data. Columbus: The Ohio State University.

Center on Budget and Policy Priorities. (1998). Strengths of the safety net: How the EITC, Social Security, and other government programs affect poverty. Washington, DC: Author. Retrieved February 15, 2005 from http://www.cbpp.org/ snd98-rep.htm

Dickert-Conlin, S., & Houser, S. (1999). EITC, AFDC, and the female headship decision. Institute for Research on Poverty Discussion Paper no. 1192-99. Retrieved February 20, 2005 from http://www.irp.wisc.edu/publications/dps/pdfs/ dp119299.pdf

Dickert-Conlin, S., & Houser, S. (2002). EITC and marriage. National Tax Journal, 55, 25-40.

Edwards, R.D. (2004). Macroeconomic implications of the earned income tax credit. National Tax Journal, 57, 45-65.

Ellwood, D.T. (2000a). Anti-poverty policy for families in the next century: From welfare to work--and worries. The Journal of Economic Perspectives, 14, 187198.

Ellwood, D.T. (2000b). The impact of the earned income tax credit and social policy reforms on work, marriage, and living arrangements. National Tax Journal, 53, 1063-1105.

Essa, N., & Hoynes, H.W. (1999). The earned income tax credit and the labor supply of married couples. Institute for Research on Poverty Discussion Paper no. 1194-99. Retrieved February 19, 2005 from http://www.irp.wisc.edu/ publications/dps/pdfs/dp119499.pdf

Gale, W.G., & Potter, S.R. (2002). An economic evaluation of the economic growth and tax relief reconciliation act of 2001. National Tax Journal, 55, 133-186.

Greenstein, R., & Shapiro, I. (1998). New research findings on the effects of the earned income tax credit. Washington, DC: Center on Budget and Policy Priorities. Retrieved February 15, 2005 from http://www.cbpp.org/311eitc.htm

Handler, J. (1995). "Ending welfare as we know it": Another exercise in symbolic politics. Institute for Research on Poverty Discussion Paper no. 1053-95. Retrieved February 21, 2005 from http://www.irp.wisc.edu/publications/ dps/pdfs/dp105395.pdf

Hepner, M., & Reed, W.R. (2004). The effect of welfare on work and marriage: A view from the states. Cato Journal, 24, 349-370.

Holtzblatt, J. (2000). The hidden welfare state. [Review of the book The hidden welfare state]. National Tax Journal, 53,323-328.

Holtzblatt, J., & Rebelein, R. (2000). Measuring the effect of the EITC on marriage penalties and bonuses. National Tax Journal, 53, 1107-1133.

Horowitz, J.B. (2002). Income mobility and the earned income tax credit. Economic Inquiry, 40, 334-347.

Howard, C. (1997). The hidden welfare state. Princeton, NJ: Princeton University Press.

Hoynes, H.W. (1995). Does welfare play any role in female headship decisions? Institute for Research on Poverty Discussion Paper no. 1078-95. Retrieved February 21, 2005 from http://www.irp.wisc.edu/publications/dps/pdfs/ dp107895.pdf

Johnson, N., & Lazere, E. (1998). Rising number of states offer earned income tax credits. Washington, DC: Center on Budget and Policy Priorities. Retrieved February 20, 2005 from http://www.cbpp.org/9-14-98sfp.pdf

Kiefer, D., Carroll, R., Holtzblatt, J., Lerman, A., McCubbin, J., Richardson, D., & Tempalski, J. (2002). The economic growth and tax relief reconciliation act of 2001: Overview and assessment of effects on taxpayers. National Tax Journal, 55, 89-117.

McClaskie, S. (2005, February 18). Personal email communication with details provided by CHRR archivist Dr. Kim Nagi. Columbus, OH: Center for Human Resource Research.

McCubbin, J. (2000). EITC noncompliance: The determinants of the misreporting of children. National Tax Journal, 53, 1135-1164.

Moffitt, R.A. (2003). Milton Friedman, the negative income tax, and the evolution of U.S. welfare policy. Institute for Research on Poverty, Discussion Paper no. 1260-03. Retrieved February 21, 2005 from http://www.irp.wisc.edu/ publications/dps/pdfs/dp126003.pdf

Neumark, D., & Wascher, W. (2001). Using the EITC to help poor families: New evidence and a comparison with the minimum wage. National Tax Journal, 54, 281-317.

Ozawa, M.N. (1995). The earned income tax credit: Its effects and its significance. Social Service Review, 69, 563-582.

Ozawa, M.N., & Hong, B-E. (2003). The effects of EITC and children's allowances on the economic well-being of children. Social Work Research, 27, 163-178.

Phillips, K.R. (2001). The earned income tax credit: Knowledge is money. Political science quarterly, 116, 413-424.

Scholz, J.K. (1994). The earned income tax credit: Participation, compliance, and antipoverty effectiveness. National Tax Journal, 47, 63-87.

Shipler, D.K. (2004). The working poor: Invisible in America. New York: Alfred A. Knopf.

U.S. Congress (2004a). Green Book: Background material and data on programs within the jurisdiction of the Committee on Ways and Means. Appendix H--Data on poverty. House of Representatives, Committee on Ways and Means. Washington, DC: GPO. Retrieved February 10, 2005 from http://frwebgate.access .gpo.gov/cgi-bin/getdoc.cgi?dbname=108_green_book&docid=f:wm006_ 23.pdf

U.S. Congress (2004b). Green Book: Background material and data on programs within the jurisdiction of the Committee on Ways and Means. Appendix K--Spending for income-tested benefits, fiscal years 1968-2002. House of Representatives, Committee on Ways and Means. Washington, DC: GPO. Retrieved February 10, 2005 from http://frwebgate.access.gpo.gov/cgi-bin/getdoc .cgi?dbname=108_green_book&docid=f:wm006_26.wais

U.S. Congress (2004c). Green Book: Background material and data on programs within the jurisdiction of the Committee on Ways and Means. Section 13: Tax provisions related to retirement, health, poverty, employment, disability and other social issues. House of Representatives, Committee on Ways and Means. Washington, DC: GPO. Retrieved February 10, 2005 from http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=108_green _book&docid=f:wm006_13.pdf

Weisbach, D.A., & Nussim, J. (2004). The integration of tax and spending programs. The Yale Law Journal, 113, 955-1028.

Note

(1.) It should also be kept in mind that prior to 2002 low-income taxpayers with a "married filing separate" status were ineligible for the credit. As Holtzblatt and Rebelein (2000) have shown, this policy created a bonus for two very low-wage workers with children because their joint return entitled them to a higher credit than was the case had they filed separately, given the eligibility and phase-out levels of the credit at the time of their study. Contrarily however, a two-earner couple with children and $35,000 of combined income was ineligible for the EITC if married, but eligible for a sizable credit if they did not marry, lived together, and raised a family. Although these two latter couples had similar income and family responsibilities, they were not treated the same under the tax code, violating the principle of horizontal equity. Many of the compliance problems that the IRS faced and that were addressed in the aforementioned legislation during the mid-to-later 1990s were a function of the same issue, namely the relative treatment of single and married taxpayers. Achieving marriage neutrality, progressiveness, and equal taxation of couples with the same income was (and remains) a longstanding tax problem (Bittker, 1975). Finally, it should also be kept in mind that 14 states and the District of Columbia had also implemented their own EITC programs by 2000, adding between 5-25 percent more credit on top of the Federal one, and that these were not taken into account in the present study of the Federal program (Johnson & Lazere, 1998). For exemplary studies that did account for state EITC programs, see Neumark and Wascher (2001) and Scholz (1995).

Richard K. Caputo

Wurzweiler School of Social Work

Yeshiva University--Wilf Campus

Table 1
Panel A
T-test results by EITC Tax Filing Status

                                 EITC Tax Filing Status

                                 EITC       Non-EITC
                                 Tax        Tax
Measure                          Filer      Filer       T-value

Sociodemographic Measures
  Age                           40.69      41.06        -3.18 **
  Education                     12.27      12.46        -1.81
  Mother's education             9.90      10.29        -2.32 *
  Number of children
   in household                  1.65       0.93        11.22 ***
  Number of years
    filed tax return             4.83       4.82         0.16
  Number of years
    lived in poverty             5.52       3.44         9.96 ***
Work history measures
  Average number of
    weeks worked                31.50      34.80        -4.78 ***
  Average percent of
    unaccounted                 38.97      51.54        -1.42
    number of weeks
    worked, 1982-2000
  Number of weeks
    worked in 2001              44.95      43.57        1.81
  Percent of unaccounted
    number of weeks             19.54      34.59        -0.60
    worked in 2001
  Unemployment rate
    in area of residence         2.74       2.67        1.69

*** p [less than or equal to] .001, ** p [less than or equal to] .01,
* p [less than or equal to] .05

Panel B
Cross Tabulation Results
by EITC Eligibility Status
(within group percent)

                                 EITC Tax Filing Status

                                 EITC Tax   Non-EITC    Chi-square
Measure                          Filer      Tax Filer   value

Sociodemographic Measures
  Born in the U.S.               94.90      92.50       03.84 *
  Health limits work on can do   12.20      12.40       00.03
  Hispanic                       16.80      16.30       00.63
  Marital Status
    Married                      28.30      43.70       38.47 ***
    Never married                22.40      24.00       00.51
   Separated/divorced/widowed    49.30      32.40       44.67 ***
  Race--White                    48.20      57.30       12.30 ***
  Region of residence
    Northeast                    9.80       15.70       11.45 **
    North Central                23.80      23.40       00.02
    South                        49.60      44.40       04.02 *
    West                         16.80      16.50       00.03
  Sex--Female                    71.60      46.90       94.65 ***
  Urban residence                75.40      75.50       00.00
Class of worker status
  Government employee            17.20      12.70       06.05 *
  Private, for profit sector     68.40      66.80       00.44
    employee
  Non-profit organization        7.00       5.40        01.51
    employee
  Self-employed or
    family business              7.40       15.00       21.93 ***
Public Assistance
Participation
  Food Stamps                    9.80       2.20        39.33 ***
  SSI                            2.90       1.90        00.38
  TANF                           1.90       0.50        06.19 *

*** p [less than or equal to] .001, ** p [less than or equal to] .01,
* p [less than or equal to] .05

Table 2
Unstandardized Coefficients (B),
Standard Errors (SE), and Odds of
Filing for the EITC

Measures                                     B        SE      Odds Ratio

Sociodemographic Measures
  Age                                      -.037       .027   0.964
  Born in the U.S.                          .611       .277   1.843 *
  Education                                -.037       .035   0.964
  Health limits work on can do              .058       .186   1.060
  Hispanic                                 -.060       .209   0.941
  Marital Status
    Married (reference)                      --         --           --
    Never married                           .491       .181   1.633 **
    Separated/divorced/widowed             1.068       .146   2.909 ***
  Mother's education                       -.012       .023   0.988
  Number of children in household           .537       .055   1.711 ***
  Number of years filed tax return          .103       .048   1.108 *
  Number of years lived in poverty          .099       .019   1.104 ***
  Race-White                               -.171       .141   0.843
  Region of residence
    Northeast                              -.476       .197   .621 *
    North Central                          -.238       .170   .788
    South (reference)                      --      --         --
    West                                   -.001       .208   0.999
  Sex-Female                                .664       .133   1.942 ***
  Urban residence                           .038       .151   1.038
Class of worker status
  Government employee                       .662       .251   1.938 **
  Private, for profit sector employee       .576       .207   1.778 **
  Non-profit organization employee          .727       .308   2.068 *
  Self-employed or family business           --         --           --
    (reference)
Public Assistance Participation
  Food Stamps                              1.136       .329   3.113 ***
  SSI                                      -.229       .397   0.795
  TANF                                      .193       .661   1.212
Work history measures
  Average annual # of weeks worked,         .004       .006   1.004
    1982-2000
  Average % unaccounted # of weeks         -.035       .037   0.965
    worked, 1982-2000
  Number of weeks worked in 2001            .010       .004   1.010 *
  Percent unaccounted # of weeks worked     .003       .014   1.003
    in 2001
Unemployment rate in area of residence      .129       .088   1.138
-2 Log likelihood                                  1703.316

Hosmer and Lemeshow Goodness of Fit        Chi-square = 3.35, df = 8,
  Test                                     p = .91

*** P [less than or equal to] .001, ** P [less than or equal to] .01,
* P [less than or equal to] .05