University of Wisconsin–Madison

Child Support Policy Research Agreement, October 1, 2018–December 31, 2020

This research agreement between IRP and the Wisconsin Department of Children and Families (Daniel R. Meyer, Principal Investigator) supports data collection and research related to the child support system. The primary projects and summary descriptions appear below. Related publications and reports will be posted upon completion.

1. Court Record Data (CRD) data collection

Collections from the court record allow us to address a range of child support enforcement policy issues that cannot be addressed using KIDS data alone. The collected data will include information on legal custody and physical placement, visitation, and details concerning the specific provisions of each order (for example, child care and child physical placement provisions). Other information collected will include records of deviations from the use of the guidelines, and information on returns to court for purposes related to child placement, child support order revision, or enforcement of child support. We will collect any information about serial family child support obligations that parents have from prior orders in other cases.

Collection of cohort 37 of paternity and divorce cases (cases coming to court in 2017) will begin in 2020. We will collect the court history of approximately 700 adjudicated paternities, 450 voluntary paternity acknowledgment cases, and 850 divorce cases from court record files in 21 counties. Data for cohort 37 will be ready for analysis by December 2021, so analyses based on these data would be part of another research agreement.

2. Use of Guidelines

Building on prior research, we will assess the use of the guidelines in recent cases, examining cases coming to court in 2013 (the most recent data available). We will follow the methodology and assumptions used in previous reports to the extent they still apply. Considering information on income of both parents, the number of children, the placement arrangement, and prior obligations of the noncustodial parent (serial families), we will estimate what an order consistent with the guidelines would be. We will compare this to the actual order in the case, estimating the proportion of cases that are consistent with the guidelines versus above or below. We will also examine whether any reasons for deviation appear in the court record. Finally, we will present characteristics of cases that are more likely to be consistent with the guidelines.

3. Cost of Raising Children

In this task, we will conduct a literature review of the empirical research examining the costs of raising children. We will explain the strengths and limitations of the extant research and present the most recent cost estimates. We will compare these recent estimates of the costs of raising children to child support orders that would result from the use of the Wisconsin percentage-of-income standard.

4. Low-Income Guidelines

The Flexibility, Efficiency, and Modernization in Child Support Programs final rule of December 2016 contains new requirements for state guidelines. One of the requirements is that “guidelines must provide the child support order is based on the noncustodial parent’s earnings, income, and other evidence of ability to pay that . . . takes into consideration the basic subsistence needs of the noncustodial parent (and at the State’s discretion, the custodial parent and children) who has a limited ability to pay by incorporating a low-income adjustment, such as a self-support reserve or some other method determined by the State.” [45 C.F.R. § 302.56(c)(1)(ii)]. While compliance with this requirement does not immediately need to be addressed in state quadrennial guideline reviews, we expect that many states will already be revisiting the way that low-income noncustodial parents are treated in their guidelines. This task has two parts. First, we will gather information from other states, documenting whether changes to their guidelines for low-income cases were proposed, and whether, how, and why the guidelines were changed. Second, we will calculate the expected child support order in Wisconsin and in other selected states for several types of low-income cases. This allows us to assess whether expected order amounts in Wisconsin are similar to, higher than, or lower than those in other states.

5. Child Support Payments, Income Imputation, and the Low-Income Guidelines

This task uses data from the Wisconsin Court Record to explore the extent to which child support payments are related to several characteristics of child support orders, including (a) whether the order was set by default, (b) whether the order was based on imputed or actual income, and (c) whether the regular guidelines, the low-income adaptation of the guidelines, or neither was used to set the orders. The court record contains relatively complete information on whether income was imputed; information on default orders and the type of guideline used is less complete. We conduct both descriptive and multivariate analyses. We anticipate using propensity score matching techniques to increase the comparability of various types of cases. Most of the focus of this task will be on cases in which the low-income adaptation could have been used (e.g., noncustodial parents whose income is less than 150% of the federal poverty guidelines. This task complements task 4, which is focused on the low-income guidelines (but does not include payment information), and task 6, which is related to income imputation, but is focused on methods of imputation rather than the relationship between imputation and payments.

6. Income Imputation in Setting Orders

When the court does not have any evidence of the level of noncustodial parent income, income is typically imputed. In the past this was often based on an assumption of a full-time (or sometimes 30 hours per week) minimum wage job, and a child support order was assigned accordingly. The Flexibility, Efficiency, and Modernization in Child Support Programs final rule of December 2016 requires that if a state’s guidelines allow for income imputation, the order must take into consideration “the specific circumstances of the noncustodial parent . . . to the extent known, including such factors as the noncustodial parent’s assets, residence, employment and earnings history, job skills, educational attainment, literacy, age, health, criminal record and other employment barriers, and record of seeking work, as well as the local job market, the availability of employers willing to hire the noncustodial parent, prevailing earnings level in the local community, and other relevant background factors in the case.” [45 C.F.R. § 302.56(c)(iii)]. For this task we will write a brief report that provides alternative ways that income could be imputed that would be consistent with the new regulation. We will examine the types of information on the above list that the court would typically have available and show how imputed income could be constructed. We will also explore the feasibility of developing an on-line calculator to impute income.

7. Barriers to Child Support Payment

Understanding the reasons for non-payment of child support orders is a key concern as different explanations suggest different policy responses. In several recent KidStat meetings, concern has focused on three potential barriers: drug use, mental health issues, and incarceration. This task will assess the relationship between these barriers and child support payment. We will consider the extent to which these barriers are connected to employment and earnings, to orders (e.g. are those with evident barriers likely to have lower—or higher—orders?), and to child support payments. We will use baseline data from Supporting Parents Supporting Kids (SPSK) to identify noncustodial parents who stated they had barriers to employment that included issues of mental and behavioral health or incarceration. The report for this task will focus on the extent of these barriers, and the later child support orders and payments of those with and without these barriers.

8. Exploring Paths to Child Support Compliance

In planning for Supporting Parents Supporting Kids (SPSK) and the Child Support Noncustodial Parent Employment Demonstration (CSPED), the primary outcome of interest was compliance with child support orders. The model was that the enhanced-service group: (a) will receive different (better) child support services, which may change their attitude towards/cooperation with the child support system; (b) will receive employment services, leading to more earnings; (c) will receive parenting services, leading to an increased sense of responsibility for children; and (d) have a higher level of compliance with child support orders as a result of these three changes (the changed attitude toward child support, the increased earnings, and the increased sense of responsibility). The final CSPED impact report (to be released spring of 2019) will provide information on whether there were detectable impacts in each of these outcome areas, for all grantees as a whole, and for each grantee (including Wisconsin SPSK). In this task we will extend the basic impact analysis with a more finely-grained analysis of the steps toward increased compliance. Considering Wisconsin SPSK only, we will examine each step in the hypothesized path to compliance. That is, we will assess whether child support services changed attitudes, whether employment services changed earnings, whether parenting services changed the sense of responsibility, and, in particular, whether the attitudes, earnings, and responsibility together produced compliance. We will also assess whether any observed changes in compliance are the result of the SPSK model, or were more affected by other factors or services.

9. Changes in Perception of Child Support Enforcement and Performance

One of the primary outcomes of interest in the Supporting Parents Supporting Kids (SPSK) and the Child Support Noncustodial Parent Employment Demonstration (CSPED) program was attitudes towards the child support program. It has been suggested that when low income noncustodial parents hold negative attitudes about child support enforcement agencies both noncustodial parent outcomes and child support enforcement agency performance suffer. We will use survey measures of noncustodial parent attitudes, and change in attitudes, for SPSK participants, together with KIDS data on child support enforcement activities and child support outcomes (beyond those captured in the original evaluation), to analyze the factors that contribute to attitudes, and changes in attitudes, and to assess the impact of attitudes on child support outcomes.

10. Child Support in Military-Connected Wisconsin Families

In this task, we will examine the responsiveness of Wisconsin’s courts and child support agencies to the unique needs of military-connected families with child support orders. Unmarried military- connected (service member and veteran) parents are more likely than their civilian counterparts to have a formal child support order, yet they face distinct challenges in meeting their child support obligations, including relocations, extended deployments, changes in income corresponding with changes in assignment, combat-related stress, and status transitions (e.g., from reserve status to active duty, from service member to veteran) (Osborne & Dillon, 2013). The Wisconsin Department of Children & Families guides child support agencies “to give the highest priority and flexibility under Wisconsin law” to parents requesting an order change in response to being called to active duty. We will select a sample of counties representing higher and lower proportion of service members and veterans, and examine the experiences of military-connected parents seeking order modifications for reasons related to their military service, including but not limited to being called to active duty. We will conduct interviews with noncustodial and custodial parents to learn about their service-connected reasons for seeking order modifications, their experiences doing so, and their perceptions of the responsiveness of the courts and child support agencies to their changing circumstances, including response time and perceptions of the fairness of resulting order modifications. We will recruit participants through veterans’ organizations in selected counties.

11. Stability of Placement Arrangements and Satisfaction with Placement Arrangements

Mother sole placement no longer accounts for the majority of post-divorce placement arrangements. Shared placement continues to increase, and this increase has occurred across income levels. We will conduct a survey of divorced mothers and fathers from the Cohort 33 CRD who have shared placement at the time of their divorce judgment and who would still have at least one minor child at the time of the survey—an estimated 325 couples—and an additional 275 mothers with sole placement arrangements. The survey will allow us to assess the stability of arrangements (e.g. the actual number of nights spent with each parent several years after the arrangement was put in place), and also to assess parents’ satisfaction with the arrangements, including how these outcomes differ across income levels and other dimensions of shared- placement households. These analyses will provide insight into the implications of the recently documented decline in sole-placement, and any related decline in orders. In addition to addressing stability and satisfaction—the two topics that will be the focus of our initial analyses—the survey will generate data that would enable subsequent analyses of child well-being in diverse placement arrangements, which will add substantially to our understanding of shared placement as a growing family form. Our survey approach is designed to compare mothers’ and fathers’ perspectives in shared placement cases, while also providing an appropriate comparison group of sole placement cases.

12. Changes in Placement and Implications for Child Support Policy

In our previous work using the Wisconsin Court Record Data (CRD), we have shown that shared placement now accounts for half the outcomes in recent divorce cases in Wisconsin, with equal- shared placement now 35 percent of cases and unequal-shared 15 percent of cases. Yet in the available national data (from the Child Support Supplement of the Current Population Survey), only 29 percent of divorced custodial parents report joint physical custody (shared placement). In this task we will analyze the apparent discrepancy, assess the evidence on whether placement arrangements in Wisconsin are different from other states, and investigate the extent to which the difference between estimates using the CRD and national data reflects that national point-in-time estimates are based on a sample that includes those whose divorces were finalized many years ago. Because the likelihood of an order varies by placement arrangement, differences in placement arrangements may have implications for child support agency performance. The report will discuss the implications of the findings for comparisons of Wisconsin’s and other states’ proportion of cases with child support orders and payments.

13. Shared Placement and Post-Divorce Economic Well-Being

Past work at IRP has provided mixed evidence about the economic implications of shared placement. For example, we have documented fairly substantial declines in economic well-being among mothers with shared placement, relative to those with traditional sole placement. This stems from mothers experiencing relatively larger losses from their ex-partners’ earnings than they gain in child support, public assistance, and their own earnings gains. At the same time, because shared placement is much more common than sole placement among couples that have higher pre-divorce incomes, shared placement parents tend to fare better economically in absolute terms, even as they experience more sizable declines relative to pre-divorce circumstances. Our past work has been limited to four cohorts of data and two years of post-divorce outcomes. We will extend this work to more recent cohorts and a longer follow-up period. This will allow us both to assess the extent to which economic well-being stabilizes after the initial couple of years, and to look at differences across subgroups, including those with lower pre-divorce incomes.