Child Support Policy Research Agreement, September 2014–December 2016

This research agreement between IRP and the Wisconsin Department of Children and Families (Jennifer L. Noyes, 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) Modernization Project

The CRD is a unique data system that allows us to address a range of child support enforcement policy issues that cannot be addressed using KIDS data alone. In addition, the CRD’s long historical record allows analysis of important changes over time. However, given changes in child support enforcement policy, court record systems, and data available from other systems, it would be useful to re-evaluate the current design of the CRD effort, and consider revising the data collection strategy. We will not collect another cohort of data during this research agreement period, and will instead focus on assessing changes in the data collected in recent cohorts, and the implications for developing a new CRD collection strategy for the future. The new strategy will be designed to maximize the quality of future data collection and take advantage of new technology while preserving comparability and our ability to consider changes over time by combining data from the earlier CRD collections and future collections.

2. Guidelines Review

Building on prior research, we will undertake a traditional assessment of use of the guidelines, following the methodology and assumptions used in previous reports to the extent they still apply. We will focus in particular on the use of the guidelines in high income cases. Using an earlier report as a starting point, we will consider changes in the guidelines and in data availability, and the implications for the ability to determine use of the guidelines in different types of cases.  Issues to be considered include the frequency of necessary information missing from the court records (e.g., payer’s income, the percentage of time with each parent in shared placement) and the frequency of complicated situations where the appropriate guidelines amount is difficult to determine (e.g., orders based on imputed income, orders where multiple adjustment factors apply, multi-part orders).

3. County Performance and the Role of Incarceration

This project will have two new tasks considering the relationship between child support and incarceration. The first new task will focus on the extent to which child support performance outcomes for individual counties in Wisconsin vary according to the extent to which payors are incarcerated, or have a recent history of incarceration. We will use data from KIDS as well as the Department of Corrections to assess the extent to which individuals who owe child support are incarcerated or have a recent history of incarceration. Differences in the extent to which caseloads are incarcerated will be used to estimate the extent to which differences in performance may be explained by differences in caseload characteristics, especially incarceration history. Results will provide important information about the implications of incarceration on county performance and, related, the receipt of incentives payments from the federal government.

The second new task will build on work completed under previous contracts that evaluated the Milwaukee Prison Project and culminated in the September 2012 report “Holding Child Support Orders of Incarcerated Payers in Abeyance: Final Evaluation Report.” That study found that holding open the child support orders of incarcerated parents may not only produce the mechanical effect of reducing their arrears at the time of release, it may also have an effect on the behavior of these parents in ways that are consequential to their children and the custodial parents as well as related systems. Specifically, although a custodial parent would not receive the amount due to her for support of a child as delineated in an order while the NCP is incarcerated, the likelihood she will receive payments subsequent to release may increase. Under this task, additional analysis will be completed to further test the suggestive findings.

4. Outcome of Change in Interest Charging Policy

Effective April 1, 2014, interest charged on past-due support changed from 1 percent per month (12 percent per year) to 0.5 percent per month (6 percent per year). This change was authorized in 2013 Act 20, the 2013–2015 Wisconsin State Budget, based on the belief that a reduction in the interest rate, as well as the reduction in the amount of interest that accumulates, would incentivize payment of arrears. Currently, interest makes up 41 percent of arrears. The Department of Children and Families has estimated that assigned child support collections would increase by 3 percent as a result of the decrease in the interest rate on arrears. In this task, we will use KIDS data to analyze patterns of arrears accumulation before and after the policy change. In addition to analyzing the mechanical effect of the change on the accumulation of arrears, this task will assess the extent to which a reduction in the interest rate, as well as the reduction in the amount of interest that accumulates, incentivizes payment of arrears. The results of this analysis will be important to understanding the extent to which adoption of the policy had its intended effect.

5. Serial Obligors’ Investments in Children

There is very limited information available on serial payers’ investments in their children. Most current knowledge about serial payers, child support payment, and father involvement has been garnered from surveys of custodial mothers and has, predominantly, focused on a single (focal) child. The few studies that have solicited such information directly from noncustodial fathers have also focused on single child. Thus, outside of administrative data on formal child support payment, we have little empirical information about fathers’ patterns of resource transfers (particularly informal and in-kind contributions) and involvement with the full range of their children born to each mother with whom they have had a child. Documenting these patterns has important implications for better understanding how family structure and birth order, particularly in the context of serial families (often known and complicated families, or multiple partner fertility families), may be associated with nonresident fathers’ investments in children. Recent qualitative evidence suggests that many nonresident fathers invest heavily in a recent child, particularly if still involved with the child’s mother, but substantially shift those investments to a new child if conceived with a new partner. This task will draw on baseline data from the Child Support Noncustodial Parent Employment Demonstration (CSPED) evaluation, which offers the unique opportunity to describe the full range of father involvement (time and activities with children), informal and in-kind financial transfers, and formal child support investments that fathers make in all of their (resident and non-resident) children. The sample is expected to include at least 1,000 noncustodial parents from Wisconsin (the focus of this analysis), and up to 7,000 from other states (used for comparison).

6. Does Joint Legal Custody Engage Never-Married Fathers in Financial Support for Children?

Children who live with only one parent are at high risk of poverty and may lose the emotional support of their nonresident parent. Several child support enforcement policies attempt to increase the involvement of nonresident parents. One of these policies, presumptive joint legal custody, enacted in 1999 in Wisconsin, gives noncustodial parents some decision-making power over child-rearing. Already common in divorce cases, it was quickly implemented in nonmarital cases: the proportion of cases with joint legal custody rose from 23 percent to around 70 percent after the policy change. This project investigates the relationship between joint legal custody and child support compliance. Using Wisconsin Court Record Data, we will estimate multivariate models comparing parents with and without joint legal custody, including economic and demographic controls. To improve the comparison of cases with and without joint legal custody, we will also use propensity score matching (PSM) techniques.

7. Evaluation of Children First Program

Noncustodial parents may be court ordered to participate in the Children First Program if the parent is legally able to work full-time but is currently unemployed or underemployed. Services may include case management, employment services, skills training, and parenting. There is substantial variation in program implementation across counties. This report will use available documentation and preliminary analysis of administrative data to select, in consultation with Wisconsin Bureau of Child Support staff, three to five programs for more intensive case study. For the case study programs we will use data from CARES (program activities) and KIDS (child support orders and payments) to document participation patterns and pre-post child support outcomes for program participants. With this background, we will conduct individual interviews with managers, caseworkers, and program participants to evaluate program implementation and the use of Child First resources. This report will complement the analysis of related programs in Brown and Kenosha Counties being conducted as part of the CSPED evaluation.

8. Child Support Regularity, Moves and School Changes

A recent report (“Child Support Receipt and the Quality and Stability of Housing,” completed as part of the 2012–14 research agreement) finds an association between the regularity of child support mothers’ receive, holding the total amount constant, and housing stability as well as moves to higher quality housing. Though we observe rates of mobility for the SNAP subsample that range between 14 and 21 percent (between 2002–2006), we expect these are underestimates since we cannot capture within zip codes moves and because lower income families are most likely to move short distances. Moreover, prior research unambiguously finds that moving more than once a year is associated with poor health and developmental outcomes for children, though it is not clear if it is the disruption of moving, neighborhood quality changes or school changes that drive observed effects. We plan to use geo-coded KIDS address records to measure the distance of moves, appending external data on school district to detect whether a resulting move triggered a change in a child’s assigned school. The impact of the regularity of child support receipt on housing mobility is understudied. Using KIDS data from 2006–2012, this study will capture all moves and look with refinement at the relationship between child support regularity and the types of moves (distance and, if data are available, school change).