Parenting across State Lines: Effect of the Uniform Interstate Family Support Act

Yunhee Chang, University of Illinois at Urbana-Champaign
Andrea H. Beller, University of Illinois at Urbana-Champaign

Since U.S. social policy proclaimed a substantial cutback in means-tested welfare benefits in 1996, state-initiated child support enforcement programs have gained increasing attention as an essential policy alternative for the poor. Replacing governmental spending with private support from the responsible parents seems appealing to the public, since single-parent families comprise about two-thirds of the nation’s welfare caseload (Department of Health and Human Services, 2000) and nearly half (48%) of them reported no financial assistance from the noncustodial parent (NCP) (Sorensen and Zibman, 2000). While each state developed various child support enforcement measures under Title IV-D of the Social Security Act, enforcement of cases with parties residing in different states -- namely interstate enforcement -- remains highly problematic. Interstate cases account for only 7 percent of the total child support collections, while they represent at least 25 percent of all child support cases in the nation (DHHS, 2002; U.S. General Accounting Office, 1992). Lower compliance for interstate cases can be explained by two factors: physical distance between the NCP and the child (“distance factor”), and the difficulties due to the involvement of multiple states (“state-line factor”). Physical distance can reflect emotional detachment between the NCP and the child. Economic theory also suggests that physical distance associated with interstate cases would discourage NCPs’ compliance due to the higher cost of providing support and monitoring the spending (Weiss and Willis, 1985 and 1993). Confusion and complexity in legal and administrative procedures created by conflicting legislation between states are said to have multiplied the problem. Proposed in 1992, the Uniform Interstate Family Support Act (UIFSA) replaced the Uniform Reciprocal Enforcement of Support Act (URESA) of 1950 and the Revised Uniform Reciprocal Enforcement of Support Act (RURESA) of 1968, which had defined interstate procedures of CSE for the preceding decades. Under UIFSA, “one controlling order” must be defined for each case and one state with “continuing exclusive jurisdiction” may modify it. Adoption remained optional for states, however, until the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996 required all states to implement UIFSA by 1998 in order to stay eligible for federal funding. If successful, UIFSA is expected to remove the “state-line factor” from the obstacles in interstate enforcement by making legal and administrative procedures across the states more efficient and simplified. This study addresses the problem of interstate child support enforcement and evaluates whether the 1996 federal welfare reform improved it by mandating the states to adopt the uniform legislation. We hypothesize (1) state borders between the parties affect enforcement outcomes negatively even after controlling for the physical distance. (2) UIFSA, the new uniform standard for interstate enforcement, improved the likelihood of compliance and sped up the enforcement processes among interstate cases. We benefit from a rich and unique dataset extracted from the state administrative system enabling an application of rigorous analytic techniques. The data from the administrative system of the Illinois Department of Public Aid (IDPA) and Department of Human Services (DHS) offers extensive information on demographic and program participation status for the sample of 55,582 IV-D cases which were on welfare at some time between June 1995 and July 1997 in Central Illinois. It contains detailed information on residential location of both parents, which allows us to incorporate geographic measures to better understand the characteristics of interstate cases. The data also contains histories of past child support activities for each case and provides grounds for dynamic analyses of enforcement duration. The sample includes 7,257 interstate cases (13%) and 7,907 cases in which the NCP’s address is unknown (14%). We use Geographic Information Systems (GIS) to compute the distance between the NCP and the child for each NCP-child pair using their postal zip-codes and the point-to-point distance calculation module. This is the first time GIS has been applied to the child support area to our knowledge and thus this paper makes a unique contribution to understanding the role of the distance factor. The enforcement outcomes used as dependent variables in this study include establishing paternity for the cases with children born out-of-wedlock, obtaining support orders, and collecting voluntary and involuntary child support payments. For the hypothesis (1), we estimate Probit and duration models Pr(Y=1) = F(a0 + a1*D + a2*D^2 + a3*INTER + a4*X) lambda(t) = lambda0(t)*exp(a0 + a1*D + a2*D^2 + a3*INTER + a4*X) where Y is an enforcement outcome, F is the standard normal distribution function, D is the distance measured in mileage, D^2 is distance squared, INTER is a dummy variable for an interstate case, X is a vector of case characteristics, lambda(t) is the hazard of achieving a success in enforcement at time t, lambda0 is the baseline hazard, and a0-a4 are unknown parameters. a3 measures the state-line factor and is expected to be negative. For the hypothesis (2), difference-in-differences estimators are obtained through Probit and duration models using in-state cases as a comparison group. Our baseline specifications for Probit model is Pr(Y=1) = F(b0 + b1*INTER + b2*UIFSA + b3*INTER*UIFSA | D,X) and the duration model is lambda(t) = lambda0(t)*exp(b0 + b1*INTER + b2*UIFSA + b3*INTER*UIFSA | D,X) UIFSA is a dummy for a case opened after UIFSA adoption in Illinois, and b0-b3 are unknown parameters. b3 is our differences-in-difference estimator of the effect of UIFSA, which is expected to be positive. In order to correct for the bias arising from defining interstate cases based upon availability of the NCP’s address in the administrative dataset, we follow van de Ven and van Praag’s (1981) application of Heckman’s method (1979) to a dichotomous dependent variable case. This paper assesses additional empirical models incorporating the responding state’s UIFSA adoption status. The across-the-states variations in the timing of adoption provide natural experiment settings and make it possible to measure the effect of UIFSA with difference-in-differences-in-differences (DDD) estimator. Duration models with time-varying policy treatment status are also estimated. Preliminary results partially support our hypotheses. This paper makes a significant contribution to the empirical literature on a little-documented policy area. REFERENCES Heckman, James J. “Sample Selection Bias as a Specification Error.” Econometrica, 1979, 47, pp.153-162. Sorensen, Elaine and Zibman, Chava. Child support offers some protection against poverty (Policy briefs No. B-10). Washington D.C.: Urban Institute, 2000. U.S. Department of Health and Human Services. Third annual report to Congress: Temporary Assistance for Needy Families program. August 2000. Available: http://www.acf.dhhs.gov/programs/opre/annual3.pdf ________. “Annual Statistical Report For Fiscal Years 1999 and 2000.” Retrieved in September 2002 at http://www.acf.dhhs.gov/programs/cse/pubs/2002/reports/datareport/table_13.html U.S. General Accounting Office, “Interstate Child Support Enforcement: Mothers Report Receiving Less Support From Out-of-State Fathers.” GAO/HRD-92039FS, July 1992. Van De Van, Wynand P.M.M. and Van Praag, Bernard M.S. “The Demand for Deductibles in Private Health Insurance: A Probit Model with Sample Selection.” Journal of Econometrics, 1981, 17(2), pp. 229-52. Weiss, Yoram and Willis, Robert. J. “Children as Collective Goods and Divorce Settlements.” Journal of Labor Economics, July 1985, 3(3), pp. 268-292. ________. “Transfers among Divorced Couples: Evidence and Interpretation.” Journal of Labor Economics, October 1993, 11(4), pp. 629-679.