State Online Resource Center, “State EITC Basics: FAQs about State EITC,” http://www.stateeitc.com/basics/index.asp, accessed November 24, 2008.
 As discussed elsewhere, custodial families would also benefit in cases in which the noncustodial parent’s tax refund is intercepted to repay child support arrears owed to the family.
 The NCP EITC policies considered here are targeted to noncustodial parents who pay their child support in full, so estimated eligibility and costs are much lower than under most recently proposed broad expansions to the childless EITC. See Carasso and colleagues (2008) for discussion and estimates of various proposals to expand the childless EITC. Coan (2008) compares an NCP EITC with a childless worker and all-worker EITC expansion.
 S. 1626 is ambiguous in this regard. It states that noncustodial parents must have “paid child support during the taxable year in an amount not less than the amount of child support due during the taxable year.” This could be interpreted to mean that the noncustodial parent must pay all current support as well as any child support arrears from prior years before being eligible for the NCP EITC.
 Information about the New York and District of Columbia noncustodial parent EITC policies is obtained from the 2006 New York and DC tax forms and instructions, as well as New York Tax Law Subsection 606 (d-1) and District of Columbia Official Code (DC ST 1981 § 47-1806.4) (both downloaded on October 25, 2006).
 The child-based EITC does not include age restrictions, except that parents who are themselves qualifying children for the EITC cannot claim the EITC. The same approach could be used for an NCP EITC.
 State and local EITCs are typically lower than the federal EITC. New York’s EITC is equal to 30 percent of the federal EITC and Washington, D.C.’s, EITC is equal to 40 percent of the federal EITC.
 The New York-based scenario does not include an option for the noncustodial parent to claim twice the childless EITC, because 2/3 of the federal EITC for one child is always greater than twice the federal childless EITC.
 Our estimates capture the higher beginning and end of the phaseout range for married taxpayers who file joint returns (the beginning and ending points are $3,000 higher in 2008 dollars, deflated to 2004 dollars). The resulting end of the phaseout for married couples is $20,265 under the S. 1626-based scenario, $33,004 under the New York–based scenario and for noncustodial parents with one child under the D.C.-based scenario, and $37,124 for noncustodial parents with two or more children under the D.C.-based scenario.
 A key element of S. 1626 is that it doubles an expanded childless EITC. If S. 1626 had proposed an NCP EITC that was simply twice the existing EITC, income eligibility would end at $11,490 instead of $17,599 and substantially fewer noncustodial parents would be found to meet the child support eligibility criteria.
 TRIM3 is developed and maintained by the Urban Institute under contract with the Office of the Assistant Secretary for Planning and Evaluation at the U.S. Department of Health and Human Services. Documentation of the TRIM3 model is available at http://trim3.urban.org/T3Technical.php. TRIM3 requires users to input assumptions and/or interpretations about economic behavior and the rules governing federal programs. Therefore, the conclusions presented here are attributable only to the authors of this report.
 Earnings include earnings of a spouse, where present.
 This issue is discussed in further detail in the section on NCP EITC design and implementation considerations.
 These results include all noncustodial parents. Of those who pay child support in full and have earnings under $30,000, 5 percent are eligible for the childless EITC and 30 percent are eligible for the child-based EITC.
 About 30 percent of noncustodial parents with earnings below $10,000 are ineligible for the EITC, primarily because they are outside the 25 to 64 age range for the childless EITC or have unearned income that raises them above the eligibility limit.
 The S. 1626-based NCP EITC estimates presented here show the effect of the S. 1626 NCP EITC relative to 2008 EITC rules (deflated to 2004 dollars and modeled on 2004 data). Therefore, for noncustodial parents eligible for the NCP EITC, the costs and benefits reflect the combined effects of extending the childless EITC and doubling it for noncustodial parents who pay their support in full. Estimated costs and benefits are lower if the S. 1626 NCP EITC is considered as an add-on to other EITC provisions of S. 1626 (rather than as a stand-alone policy, as shown here). For example, we estimate that the total cost of the EITC provisions in S. 1626 is $2,760 million (in 2004 dollars), of which $39 million represents the incremental effect of adding an NCP EITC (for 25- to 64-year-olds) to the other provisions of S. 1626 (compared with $70 million if the same NCP EITC is implemented as a stand-alone policy).
 Eligibility under the S. 1626-based scenario extends to $17,599 for single individuals and $20,265 for married couples.
 In particular, we do not require child support to be paid through a government agency (as would likely be required) and we do not capture noncustodial parents whose children live with someone other than the other parent (such as a grandparent).
 The IRS estimates that in 1999, at least 1.2 million taxpayers receiving the child-based EITC were ineligible because the child did not meet the criteria for a qualifying child (IRS 2002). The most common qualifying child error was the failure of the child to meet residency requirements.
 As stated elsewhere, our estimates do not apply this restriction.
 Over all income ranges, an estimated 74 percent of custodial families who receive support participate in the IV-D program (Hong et al. forthcoming).
 There are no published national data on the extent to which noncustodial parents who pay their current support orders in full have arrears. A study of nine large states (Sorensen et al. 2007) shows that most noncustodial parents with a child support order have arrears, but that arrears are less common among those who pay their current support in full. The study finds that 94 percent of child support obligors with reported incomes of $1 to $10,000 and 75 percent of those with reported incomes above $10,000 have child support arrears. Among those who paid at least some support during the year, 80 percent had arrears. Less than half of those paying their full order for the calendar year had arrears, but the median reported income of the full payers was $30,579—so many would be outside the income range of an NCP EITC.
 Wage withholding by employers is a common method of child support enforcement. Nationally, more than two- thirds of child support collections are made through wage withholding (OCSE 2009).
 Earnings include earnings of a spouse, where present.
 The only exception to this rule is under the D.C.-based scenario, where noncustodial parents with one resident child and two nonresident children would receive a higher benefit by claiming the NCP EITC. In this case, we count the noncustodial parent as eligible for the NCP EITC.
 The 2006 D.C. tax forms and instructions do not appear to prevent a noncustodial parent from receiving both the child-based and NCP EITC. However, D.C. law indicates that the NCP EITC is available only to taxpayers who are ineligible for the federal EITC (and would therefore be ineligible for the standard D.C. EITC).
 About 58 percent of child support obtained from federal tax intercepts in 2006 was forwarded to custodial families, with the remainder going to the government (OCSE 2009).
 An NCP EITC could contribute to marriage penalties because the noncustodial parent would no longer be eligible for the NCP EITC once married to the custodial parent. However, the additional income from an NCP EITC might increase marriageability of the noncustodial parent to another partner, because the noncustodial parent would continue to receive the credit once married as long as the additional income from the new spouse did not cause the noncustodial parent to become ineligible (Coan 2008).
 The Washington, D.C., NCP EITC form (Schedule N) goes even further and requires the name and Social Security number of the custodial parent.
 Before mid-1992, the IRS automatically assigned the EITC to those who appeared eligible but did not claim the credit (Scholz 1993–94). The IRS currently calculates the EITC for taxpayers upon request, but requires the taxpayer to submit Schedule EIC if claiming resident children.