The Role of Child Support

in Texas Welfare Dynamics

Executive Summary

 
 
 

Deanna T. Schexnayder

Jerome A. Olson
Daniel G. Schroeder
Jody L. McCoy
September 1998

 
Center for the Study of Human Resources
Lyndon B. Johnson School of Public Affairs
The University of Texas at Austin


This report was prepared with funds provided through Grant No. 96ASPE283A from the U.S. Department of Health and Human Services and an interagency agreement between the Texas Workforce Commission and the Center for the Study of Human Resources at The University of Texas at Austin. The views expressed here are those of the authors and do not represent the positions of the funding agencies or of The University.

Table of Contents

Executive Summary

Research Questions

Research Methods

Research Findings

Conclusions 
 

Executive Summary

The Family Support Act of 1988 (FSA) mandated a number of policy changes to increase the employability of caretakers receiving Aid to Families with Dependent Children (AFDC) and to improve the collection of child support from noncustodial parents.Some states, including Texas, enacted state legislation to strengthen their child support enforcement procedures prior to these federal mandates.This research study by the Center for the Study of Human Resources at The University of Texas at Austin (CSHR) measures the influence that increased child support enforcement strategies have had on welfare dynamics in Texas, which has one of the lowest AFDC grant levels in the U.S but relatively high rates of paternity establishment and child support collection.

Research Questions

  1. Which factors have the greatest influence on the award and collection of child support in Texas?
  2. To what extent does child support influence AFDC exits in Texas?
  3. To what extent does child support reduce AFDC recidivism in Texas? 

  4. What is the combined influence of child support and earnings of the custodial parent in removing families from poverty?

Research Methods

This analysis was conducted by statistically analyzing a sample of 65,616 Texas AFDC caretakers who were receiving AFDC as of September 1, 1992. Individual-level administrative data records for these AFDC caretakers, their children, and the noncustodial parents of the children were analyzed for a four-year period from September 1992 through August 1996.The longitudinal dataset used in the analysis was developed by linking AFDC demographic, spell, and benefits data; child support award, collection, distribution, and demographic data; Job Opportunity and Basic Skills (JOBS) program participation data; and Unemployment Insurance (UI) wage data for this time period. 
Two main statistical approaches were used in the analysis—descriptive statistics and statistical inference.Descriptive statistics were used to provide an understanding of the nature of Texas AFDC caretakers and noncustodial parents, and changes in various child support statistics over time.Statistical inference was used to measure the extent to which each of the AFDC- and child support-related attributes contributed to the probability of having a child support award, receiving child support collections, and exiting and returning to AFDC.
Research limitations of this study primarily centered on missing or inadequate data in the administrative files used for the analysis.In particular, more information was needed on demographic characteristics and out-of-state employment of noncustodial parents.The lack of historical records for some variables also limited their usefulness.

Research Findings

This study has revealed additional information about the characteristics of Texas AFDC families and noncustodial parents for whom the state is trying to collect child support, the factors influencing the award and collection of child support, and the influence of child support on AFDC exits, recidivism, and families’ ability to leave poverty.Results for each of these areas are summarized below.

Characteristics of AFDC Caretakers and Noncustodial Parents

While the demographic characteristics of the AFDC caretakers have been explored in prior Texas welfare dynamics studies, this study provides additional information about the nature of their relationship with the noncustodial parents of their children.As learned from prior studies, most Texas AFDC caretakers had one or two children, were either Hispanic or Black, and less than half had graduated from high school.Over half had received AFDC for more than 30 months at the time the sample was drawn and 14 percent were participating in the JOBS program.While slightly more than a quarter were employed, their earnings were very low.About one third of the sampled caretakers had given birth to all of their children out of wedlock, one third had some out of wedlock, and the remaining third had all of their children while married.
By the end of the four-year study period, only 35.9 percent of the sampled caretakers were still receiving AFDC.While the share who were employed had risen to 40 percent, earnings were still low enough that almost all families still qualified for Food Stamps.The share of caretakers with all children born out of wedlock had fallen, suggesting that some women had married prior to having additional children.
Although demographic information about the noncustodial parents was limited, earnings were calculated through the use ofTexas quarterly wage records. Of the noncustodial parents for whom SSNs were known (64 percent of all noncustodial parents in the OAG files), approximately 45 percent were employed in Texas at the beginning of the observation period.This rate fluctuated little during the four-year period. Median quarterly earnings for these noncustodial parents rose from $2,751 during the fourth quarter of 1992 to $3,525 four years later, an increase of 28 percent.While an additional number of noncustodial parents were probably employed outside of Texas, this rate could not be determined to the inadequacy of the earnings and residency data used for this study.Learning more about the demographic characteristics and the employment and earnings patterns of noncustodial parents is an area needing further study.

Award and Collection of Child Support

At the time the sample was taken, only six percent of the sampled AFDC caretakers were receiving any child support payments.For another 55 percent of the cases, the OAG was performing the steps necessary to establish a child support order or obtain collections.For nearly 40 percent of the cases, the OAG had not received sufficient information to open a case.By the end of four years, 90 percent of the cases had been acted upon by the OAG; 16 percent of the AFDC caretakers were receiving payments while child support cases had been opened and were being processed for another 74percent of AFDC cases.On average, each caretaker had 1.35 child support cases, due to the need to file separate proceedings for each noncustodial parent.Over the course of the study, paternity tests to determine the noncustodial parent were conducted for four percent of the sample, with paternity established for 88 percent of the administered tests.
The amount of child support obligated for all sample members ranged from approximately $16.5 million per quarter at the beginning of the study to $35.3 million per quarter four years later.Quarterly collections increased from $5.2 million to $21.6 million over the same time period.
The probability of an AFDC caretaker with an OAG case having an established child support award in place during anygiven quarter of the observation period was 47.7 percent.Variables increasing the probability of a child support award being established were:cumulative effort spent by the OAG to process the case; more than one noncustodial parent per AFDC case; older age of the youngest child in an AFDC case; Black or Hispanic race/ethnicity of noncustodial parent; and higher earnings of the noncustodial parent.Factors decreasing the probability of an awardbeing established include: more than one child on the AFDC grant; male AFDC caretaker; noncustodial parent younger than 25 or older than 45; the AFDC caretaker’s education being less than high school or unknown; and having children born out of wedlock
On average, the amount of the current child support obligation was $467 per quarter.Factors increasing the amount of the child support obligation include: cumulative effort by the OAG, parents of different race/ethnic backgrounds, had more than two children, and higher earnings of noncustodial parents.Obligations were significantly smaller if caretakers were racial/ethnic minorities, or had not completed high school, and if some or all children were born out of wedlock.While the analysis included estimates of factors affecting the amount of past due child support obligations, these results were the least predictable of those being studied.

In this study, child support collections were analyzed on an annual basis for collections obtained through IRS intercepts and on a quarterly basis for other types of collections.Quarterly collections occurred for 38.3 percent of the cases in which there was a child support obligation.The most important factor influencing these collections was the amount of the noncustodial parent’s earnings, with the probability of collection increasing by 0.81 percentage points for every $100 of quarterly earnings.The probability of collection increased with the cumulative effort by the OAG to process cases, age of the noncustodial parents, if the noncustodial parent was Hispanic, or if children were born out of wedlock. The probability of collection was reduced if the noncustodial parent was Black, the caretaker had multiple child support cases, had more children, or the youngest child was older.

Quarterly child support averaged $472 when collections were received.For every additional $100 earned by the noncustodial parent in a quarter, the collection increased by $3.52. Collection also increased with cumulative effort to process the case, age ofnoncustodial parents, or for families with more than one child.Factors contributing to smaller collections were female noncustodial parents, noncustodial parents of minority race/ethnic groups, the older age of the youngest child, a custodial parent with less than a high school education, or some or all children being born out of wedlock.

An IRS intercept was made in approximately one out of five of the years in which a noncustodial parent had an obligation, with an annual average of $936 collected per case. Earnings of noncustodial parents had the strongest influences on IRS intercepts.Other factors had weaker and generally opposite effects than for quarterly collections, suggesting that IRS intercepts may be a substitute for regular payments.

Several of the desired demographic attributes of noncustodial parents that could influence the award and collection of child support (such as education level, marital status, or out-of-state residence throughout the four-year study period) could not be included in the regressions described above due to data limitations.Acquiring such data from other sources and including these variables in the analysis remain areas for future research.

AFDC Exits and Recidivism

On average, 11.7 percent of the sampled caretakers exited from AFDC during each quarter of the study.Sixteen percent of total AFDC exits occurred as a result of child support collections.Of the total child support collected, 47 percent was retained by the state to recoup the costs of AFDC, with the remaining 53 percent distributed to families.
Child support received by the custodial parent contributed strongly to the probability that a caretaker would exit AFDC.A $100 increase in quarterly child support collections induced a 2.5 percentage point increase in the probability of exit, raising the quarterly probability of exit from 11.7 percent to 14.2 percent.The amount of child support received by an AFDC caretaker had approximately three times as much power as an equal amount of caretaker earnings in inducing exits from AFDC.
Other economic and demographic factors also influenced the probability of AFDC exit.As found in earlier research studies, caretakers who have completed high school are more likely to exit than those with lesser educational attainment while individuals with a long history of past AFDC dependence, minority caretakers, caretakers with many and younger children, and those currently participating in JOBS activities are less likely to exit. In addition, those with multiple child support cases or all children born out of wedlock also are less likely to leave the AFDC rolls.Higher unemployment rates or living in rural or large urban areas also reduced the chances of leaving AFDC.
Approximately 35 percent of all caretakers who exited from AFDC returned to the rolls within a year.For every $100 of child support received per quarter, the probability of recidivism in that quarter was reduced by 1.0 percentage points.As with exits, the receipt of child support had over a three times larger effect on recidivism than an equivalent dollar amount of the caretaker’s own earnings.Most of the other demographic and economic variables that reduced exits also increased the probability of recidivism.

By the end of the four-year study period, 8.5 percent of the original sample and 17.5 percent of families no longer on AFDC had left poverty through a combination of caretaker earnings and child support received.Of the families no longer on AFDC, 16.4 percent left poverty through custodial parent earnings alone, while the addition of child support enabled an additional 1.1 percent to leave poverty.No families left poverty due to child support alone. 

Conclusions 

Even with Texas’ strong commitment to child support enforcement, child support was being collected for only 16 percent of sampled caretakers by the end of the study period.Processing child support cases is a difficult and lengthy process.Major impediments to successfully completing this process appear to be the absence ofidentifying information and current addresses for noncustodial parents.New provisions within PRWORA should help with these efforts.
In sharp contrast to results from studies in high-benefit states, the findings from this study indicate that child support strongly influences Texas AFDC exits and recidivism.However, the total amount of child support received by these families is so low that it does little to raise them above the poverty level.The low amounts of child support occur primarily because of low noncustodial parent earnings.Attempts to increase child support collections should be coupled with strategies to increase the employability and earnings capacity of both AFDC caretakers and noncustodial parents.
This study has contributed to the existing literature by analyzing the influence of child support on welfare dynamics in a low-benefit state in the post-FSA era.Although some of the rules influencing the relationship between welfare receipt and child support have changed with the passage of PRWORA, research of this type is essential because substituting child support for public assistance is a public policy goal that has assumed even greater importance under the new law.
Further research is needed to account for the unanticipated gaps in the administrative data files available for this research so as to explore the influence of noncustodial parent attributes such as education level, current marital status, incarceration status, and out-of-state earnings on the award and collection of child support.Exploration of such factors, along with further analysis of the employment and earnings patterns of noncustodial parents, could greatly enhance our understanding of the degree to which these parents have the ability to pay regular child support.