Guidelines - Chapter1 Vol I

Chapter 1.INTRODUCTION

Note that the Tables and Exhibits are at the end of the chapter.

In October 1994, CSR, Incorporated, and its subcontractor, the American Bar Association's Center on Children and the Law (ABA), were awarded a contract by the Office of Child Support Enforcement (OCSE) to evaluate the implementation of child support guidelines. This chapter discusses the history of child support guidelines, describes the components of the research undertaken to perform the evaluation, and provides a plan for the report.

1.1A HISTORY OF CHILD SUPPORT GUIDELINES

Historically, determining a child support award was at the discretion of the decisionmaker. He or she typically considered such factors as the financial resources of the child, the financial resources of the custodial parent (CP), the child's standard of living had the parents' marriage not dissolved, the physical and emotional conditions of the child, the child's educational needs, and the financial resources and needs of the noncustodial parent (NCP). [ Uniform Marriage and Divorce Act, 1973, 309, 9A U.L.A. 400.]

The results of allowing the decisionmaker to determine the award amount was not always adequate to support the children or equitable among the parties. One study [ Yee, L.M. 1979. " What Really Happens in Child Support Cases: An Empirical Study of Establishment and Enforcement of Child Support Orders in the Denver District Court. " Denver Law Journal 57(1):21 – 68.] found that the average monthly child support award in Colorado was less than average obligor monthly car payments. In addition to small award amounts, the decisionmaker's discretion often resulted in awards that were widely different for persons of similar circumstances. For example, the same study discovered that child support awards made by a single district court were 6.0 to 51.7 percent of an obligor's income for one child and from 5.6 to 40.0 percent of an obligor's income for two children. [ id ., at 27, Tables 3, 4, 5, 6 at 52-4.] Motivated by such findings, in 1984, Congress required every State seeking Federal funding for public welfare programs to establish child support guidelines. [ The Child Support Enforcement Amendments of 1984, Public Law (P.L.) 98 – 378, 98 Stat. 1305 (1984).] Initially these guidelines were required to serve only in an advisory capacity. However, some States adopted mandatory guidelines. The Family Support Act of 1988 required that State child support guidelines operate as rebuttable presumptions of the proper support amount. In order to deviate from these guidelines, a decisionmaker must submit written findings or findings on the record on why the guideline amount would be unjust or inappropriate for his or herparticular case. [ Section 102 of the Family Support Act of 1988, P.L. 100 – 485, 102 Stat. 2343 (1988), codified at 42 U.S.C. 667(b)(2) (1991).] Implementing final regulations are published at 45 C.F.R. 302.56. These regulations require that any order deviating from the guideline amount must include a finding on what the presumptive guideline amount would have been and a justification for the deviation.Currently, no nationwide child support guideline exists. However, the following Federal requirements guide States in developing support guidelines:•Support guidelines must be uniform throughout the State.•Support guidelines must be numerical formulas that, at a minimum, consider all of an obligor's income and provide for the child's health care needs. [ See 45 CFR 302.56.] •Support guidelines must provide rebuttable, presumptive amounts in any judicial or administrative proceeding involving the establishment of a child support order. (Based on this requirement, it is clear that guidelines apply equally to (1) cases handled by a State child support agency [IV-D cases], (2) cases handled by a private or legal services attorney, and (3) cases in which the parties are appearing pro se [i.e., without an attorney]. Guidelines apply to administrative, judicial, and quasi-judicial determinations of support. In addition, guidelines apply to interstate and intrastate cases, negotiated agreements, [ See 56 Fed. Reg. 22,347 (1991).] and contested cases. They must be used in setting temporary support as well as "final" support orders. Decisionmakers establishing child support in the context of a child protection or foster care proceeding, as well as in conjunction with a temporary restraining order in a domestic violence case, should establish such orders pursuant to child support guidelines.)

•Support guidelines must be used both to establish an initial child support order as well as to determine any subsequent award modification. [ 42 U.S.C. 666(a)(10)(A), (B).]

•A State must apply the guidelines to all cases; it cannot exclude an entire category of cases (e.g., high-income or low-income cases). [ See 56 Fed. Reg. 22,335 (1191).]

•State criteria for deviating from a guideline must take into consideration the child(ren)'s best interests. [ 45 CFR 302.56(g).]

The Family Support Act of 1988 also requires that States reevaluate their guidelines at least once every 4 years in order to ensure that their applications result in the determination of appropriate child support awards (effective October 13, 1989). [ Section 103(b) of the Family Support Act, supra note 5 to be codified at 42 U.S.C. 667(a) (1991).] As part of the reevaluation, Federal regulations require States to consider economic data on the cost of raising children and to analyze case data, gathered through sampling or other methods, on the application of and deviations from the guidelines. [ See 45 CFR 302.56(h).] Findings from the resulting analysis must inform each State's guideline review to ensure that guidelines truly apply to the majority of cases in that State.

Finally, the Family Support Act of 1988 requires States to provide procedures for the review and adjustment of IV-D child support orders. Beginning October 13, 1993, a review must occur at least once every 3 years unless (1) in an Aid to Families with Dependent Children (AFDC), Title IV-E foster care, or medicaid case, neither parent requests a review and the State determines that such a review would not be in the child's best interest or (2) in a non-AFDC IV-D case, neither parent requests a review. [ 42 U.S.C. 666(a)(10)(B).] Any adjustment must be pursuant to child support guidelines.

1.2MAJOR CHILD SUPPORT GUIDELINE MODELS

Current Federal law does not require States to adopt a particular type of support guideline. Although State guidelines vary, they often are based on the following models: the percentage-of-income model, the income-shares model, and the Delaware Melson formula. Each model is discussed below.

1.2.1Percentage-of-Income Model

Fewer than 15 States establish child support awards based on a percentage of the obligor's income. Of these States, some use the obliger's gross income to determine awards; other States use the obliger's net income. Most States that base their guidelines on the percentage-of-income model use a fixed percentage that remains constant at varying income levels. For example, Wisconsin's guidelines set awards at 17 percent of the noncustodial parent's (NCP's) gross income for one child, 25 percent for two children, 29 percent for three children, 31 percent for four children, and 35 percent for five or more children. The guideline does not explicitly consider the custodial parent's (CP's) income. It is assumed that the CP is contributing an equivalent amount of support through direct expenditures and in-kind services.

1.2.2Income-Shares Model

The income-shares model, which is used by more than 30 States, was developed by Dr. Robert Williams and was based on the work of Thomas Espenshade. Espenshade analyzed the 1972-1973 Consumer Expenditure Survey to determine the costs of raising children in the United States. According to Espenshade (1984), although the dollar amount spent on behalf of a child increases with income level, the actual percentage of parental income spent on a child decreases as income increases.

Calculating an award using this model involves consulting a chart that lists support amounts based on Espenshade's economic studies for varying income levels. The appropriate support amount is then prorated between the parents based on each parent's proportion of the total parent income. The support amount allows for prorated shares of child care and medical expenses to be added to the basic support amount. [ See Advisory Panel on Child Support Guidelines and Williams, R. 1987. Development of Guidelines for Child Support Orders: Advisory Panel Recommendations and Final Report, II-68 to II-75. Washington, DC: U.S. Department of Health and Human Services, Office of Child Support Enforcement.]

1.2.3Delaware Melson Formula

The Delaware Melson formula was developed by Judge Elwood F. Melson, Jr., of Delaware. It has been used statewide in Delaware since 1979 [ See Dalton v. Clanton, 559 A.2d 1197 (Del. 1989).] and was revised in 1990. The Melson formula is based on the net income of both parents, from which a self-support reserve is subtracted. From that balance, an amount for the primary support needs of the children is subtracted. After providing for these basic needs, the Delaware Melson formula also allows a child to benefit from each parent's remaining income. This additional support amount is determined by multiplying the remaining income by a percentage determined by the number of children needing support. Finally, child care expenses are added as part of the primary support allowance. Versions of the Delaware Melson formula have been enacted in Delaware, Hawaii, Montana, and West Virginia.

1.3SCOPE OF THE PRESENT STUDY

The evaluation was designed to investigate (1) the application of guidelines in actual case records, (2) the subjective perceptions of individuals involved in the child support system with regard to the overall operation of guidelines, (3) the extent and quality of mandated State guideline reviews, and (4) the effect of guideline implementation on award amounts by analyzing the 1992 Current Population Survey (CPS) Matched March–April Data File. The fifth component of the study was an expert panel. This remainder of this chapter provides a brief overview of the research questions and methodology for the five majorcomponents of the study. (Appendix A provides a more comprehensive description of the methodology.)

1.3.1Overall Research Questions

This study addressed a variety of questions related to the formulation and application of guidelines, including the following:

•How are State child support guidelines applied?

•What are the extent, amount, direction, and causes of deviations from the guideline formula?

•How do States account for families' special circumstances, such as second-family members (e.g., former or other children, stepchildren, and subsequent spouses), work-related and other child care, health insurance or health care expenses, and visitation and custody expenses? In addition, how do States extend guideline application to children who are past the age of majority or emancipation? To what degree are guidelines applied in the case of children who are students in postsecondary vocational or academic schools? Do decisionmakers address these circumstances by adjusting the award amount or by deviating from the guideline formula? When guidelines are implemented, is a verified income statement used? Is income imputation mentioned in the guidelines?

•What special issues relate to interstate child support cases? Have any problems arisen in applying guidelines to such cases? What aspect(s) of establishing interstate support orders using guidelines present particular challenges (e.g., gathering financial information and seeking or preventing deviations)?

•What is the extent and quality of each mandated State guideline review? What findings resulted and what actions were taken as a result of these mandated reviews? Are the guideline reviews up to date, and do they meet the regulatory requirements (e.g., collect data on deviations and provide evidence of analyzing the cost of raising children when determining guidelines)?

•What is the impact of mandated guidelines on the level of the award amount and on compliance? What is the effect of case status (in-state versus interstate) on the award amount?

These questions were addressed in five areas of investigation undertaken in the evaluation. Each area is described below in the following section.

1.3.2Overview of the Study Design

The study design has the following five major components:

•Case Data Collection.—Child support case records were collected and coded for quantitative analysis of guideline outcomes for families in 21 counties in 11 States.

•Stakeholder Interviews.—Various participants in the child support system in each of the 21 study counties were interviewed to obtain their views on guideline implementation and operation. The information was collected in each study county.

•Analysis of State Guideline Review Studies.—State guideline review studies were requested from each State and U.S. territories (49 of 55 provided written responses). These studies were analyzed for their extent and quality. In addition, any findings and actions taken as a result of these mandated reviews were described and analyzed.

•Analysis of the CPS Supplement and the March CPS Matched File.—A secondary analysis of the 1992 CPS Matched March–April Data File was performed to examine the effects of States' presumptive guidelines on child support award amounts and compliance. This analysis was designed to provide a national context for the collected case information collected.

•Expert Panel.—A panel of experts was assembled on two occasions to provide advice on (1) the direction that would be taken for research and (2) interpreting the findings and preparing the final report in order to make meaningful recommendations to the child support community. A variety of individuals was invited to join the Expert Panel to ensure representation from a full spectrum of interests, including representatives of CP and NCP advocate groups, administrators, judges, attorneys, and researchers. (See below for a complete list of Expert Panel members and a discussion of their recommendations.)

The case data collection and the stakeholder interviews were designed to address the first four research questions. The analysis of state guideline review studies addressed the fifth research question. Each component of the evaluation is discussed in greater detail below.

1.3.3Case Data Collection

This component involved collecting and analyzing case records in which a child support order was established or modified. The case records were obtained various forums, such as administrative agencies and courts. The data obtained from these cases were analyzed to answer the first four research questions.

1.3.3.1State and County Selection

The study sites were obtained by selecting 2 counties in each of 11 States. It should be noted that achieving a random sample was not essential to this selection process. The criteria used to choose the initial study States were representation of all specific guideline models (i.e., percentage-of-income, income-share, andDelaware Melson models) and willingness to participate. Many States refused to participate because of the involved work burden.

After a State confirmed its participation in the study, CSR contacted a designated liaison who secured the cooperation of State and local IV-D agencies and courts. CSR provided assistance to this State liaison in ways such as developing a preliminary data collection protocol for that State. The liaison used this protocol to assist discussions with local staff. CSR periodically monitored the liaison's progress in obtaining their cooperation.

The following 11 states elected to participate in the study:

•Arkansas;

•Delaware;

•Florida;

•Massachusetts;

•Minnesota;

•Missouri;

•New Hampshire;

•New Jersey;

•Pennsylvania;

•Washington; and

•Wisconsin.

1.3.3.2Data Collection Methodology

The aim of this study was to collect 200 cases with complete documentation from each participating county (4,200 cases total). To ensure that a diversity of case types was represented, the child support award process in each county was carefully researched. The end goal was a sample of cases from each county that reflected the various forums in which orders could be established or modified and that also included a mix of IV-D and non-IV-D cases. The methodology is further discussed in greater detail below.

Cases were prospectively selected; that is, for a case to be included in the study, a child support award must have been established or modified during the period of the study. The collected data were of two types: (1) worksheet information containing data on the income and expenses of the parties, adjustments, and the calculations of award amounts and (2) supplemental information from the case records on deviations, the size and composition of families, the type of award, case type, and other descriptive data. Although strenuous efforts were made to obtain complete case records containing both worksheet and supplemental data, some case files contained incomplete information.

Table 1-1 following this page lists (1) the 21 counties from which data were collected and (2) the number of case records of each type (i.e., worksheet or supplemental data) included in the research database. One county in New Jersey provided no usable worksheet data. Most analyses reported in Chapter 2 of thisreport combined worksheet with supplemental data, while the analysis reported in Chapter 3 used the supplemental data. The study methodology was designed to collect data on the full range of support cases decided in the counties, including both AFDC cases and non-AFDC cases handled by the IV-D agency, as well as non-IV-D cases. The case status, summarized by county in Table 1-2, shows that cases across all counties were nearly equally divided among three categories: (1) AFDC cases, (1) IV-D non-AFDC cases, and (3) non-IV-D cases. In Pennsylvania all child support cases are handled by the IV-D agency, and New Jersey does not make an administrative distinction between IV-D and non-IV-D cases. Child support orders applicable to the study included the following:

•Temporary orders;

•Permanent orders;

•Interim orders; and

•Award modifications.

Table 1-3 shows the distribution of cases by type of award across all sites. The most frequent type of award was a permanent award (in 63.0 percent of the cases). Modifications were made in 20.9 percent of cases, and temporary awards were reported in 10.5 percent of cases.

These awards could be made in any type of case, such as separations, dissolutions, or paternity establishments. Table 1-4 summarizes the types of cases and shows that paternity establishments were most frequent (29.3 percent of cases), followed by dissolutions (26.2 percent). Modifications to existing awards did not report a separate case type, and 18.6 percent of the records had missing case types.

Cases were collected in all relevant forums where cases are heard in the county, including courts, quasi-judicial settings, and administrative bodies. Table 1-5 shows that 38.9 percent of the cases were heard in courts and 51.0 percent in quasi-judicial settings. In addition, 6.7 percent were heard before administrative bodies. There was less uniformity than expected within States—only two States (Arkansas and Massachusetts)heard all cases in judicial forums, while in most States, many forums were used. Different case types were heard in different forums in some counties. As shown in Table 1-6 following this page, only 14.4 percent of non-IV-D cases were heard in judicial forums, compared with 38.9 percent of all cases, while 62 percent were heard in quasi-judicial forums, and 22.4 percent were heard in administrative forums.

The study counties varied substantially in case characteristics. The following characteristics were notable:

•Overall, more than one-third of the cases were IV-D AFDC cases (see Table 1-2). Among the counties, the percentage of cases that were AFDC cases ranged from a low of 12.2 percent (in a Wisconsin county) to a high of 54.8 percent (in a Massachusetts county). Across all counties, 33.0 percentwere IV-D non-AFDC cases, 3.0 percent were IV-D cases for which AFDC status was not reported, and 29.0 percent were non-IV-D cases.

•Fathers were the obligors in 90 percent of the cases, as expected (see Table 1-7). Counties varied little among each other with respect to the parent who was most often the obligee. Mothers were obligees in 78 percent of the cases, fathers were obligees in 5 percent, and the State was the obligee in 6 percent (see Table 1-8).

•The number of children covered by the order was one child in 67 percent of the orders, two children in 25 percent, and 3 or more children in the remainder of orders. This differed somewhat from the national CPS data, of which 52 percent of cases involved one child and 31 percent involved two children. [ Table 1 in: Census Bureau. 1995. Child Support for Custodial Mothers and Fathers: 1991. Current Populations Reports, Series P60-187. Washington, DC: U.S. Government Printing Office.] The age of the youngest child was younger than age 6 in 57 percent of the cases and was age 6 to 10 in 19 percent of the cases.

1.3.3.3Data Limitations

A variety of factors compromised the extent to which the study data and findings can be applied to the general population of child support cases. It is important to emphasize that the data and findings presented in this report cannot be used to make statements about all child support cases in any particular study State or in the entire United States for the following reasons, which are primarily linked to the research methodology.

•The participating States were chosen as a convenience sample, not a representative sample of all States. Although an effort was made to include a diversity of guideline model types and child support formulas in the study, States ultimately made the decision to participate, and several States chose not to participate for a variety of reasons, some of which were discussed earlier in this chapter.

•The method and period of the data collection were implemented to minimize the burden on court and agency staff. The effect was that collection in some counties was not continuous.

•The case record collection was conducted prospectively, and county personnel were fully aware that a study was underway in their counties. This awareness possibly influenced caseworkers, hearing officers, masters, judges, and others in the child support field, prompting them to follow procedures more closely and document proceedings more completely than usual. Although the quality of the records received from States suggests the contrary, this study sample cannot be described as "representative."

•The data collection methodology emphasized a balanced sample between IV-D and non-IV-D cases. In practice, this could not be attained because IV-D cases were more numerous that non-IV-D cases.

•The cases were collected by county personnel and, in some instances, a temporary staff or consultants. It is possible that some cases were inadvertently overlooked by the data collectors and other cases were not sent because they were incomplete.

•The documents collected for coding and analysis dictated what information could be gleaned. For example, if a second-family matter was discussed with the decisionmaker, but was not recorded on the written order, the information was not available for use in this study. Some documents were not kept in the permanent record due to storage limitations, for example (usually true of income verification papers). Unless the record explicitly discussed how income was verified, the information was lost and thus not included in this study.

1.3.3.4Conducting Unstructured Telephone Interviews With State and Local Officials

Unstructured telephone interviews were designed to collect information from those involved in the formulation and application of guidelines and the parents who have had personal experience with child support guidelines. The questions and topics discussed in the interviews were designed to gather information on the respondent's experience with issues such as the following:

•Adherence to guidelines by the court, referee, or administrative hearing officer;

•Specificity of findings supporting deviations from guideline;

•Representation by an attorney during proceedings;

•Procedures for obtaining relevant information (e.g., income information) from parties, including parties in interstate child support cases;

•Application of guidelines in interstate cases;

•Adequacy and efficiency of State guideline worksheets;

•Effectiveness of training offered in the State on guideline application and worksheet completion;

•Process used to establish and modify State guideline formulas or schedules;

•Respondent perceptions of guideline "fairness"; and

•Recommended changes.

Identifying Interview Participants

The ABA contacted State liaisons to identify selected individuals in 2 counties of the 11 study states. The participants to be interviewed included CPs, NCPs, IV-D caseworkers, IV-D attorneys, private attorneys, and judges. In those States where they played a role in child support, mediators, masters and commissioners were interviewed.

Parent advocacy groups also were contacted to identify CPs and NCPs willing to speak with an ABA representative regarding personal child support experiences. In those States where CPs and NCPs could not be identified, [ CP organizations were quick and helpful in providing names and numbers of persons who might agree to share their personal experiences with the ABA representative. Unfortunately, many provided CP telephone numbers were not valid and the organizations did not always refer a CP listed in the studied county. Some CPs contacted provided names of other CPs who would provide information. NCP organizations were less cooperative, and many State chapters or affiliates did not return initial contact phone calls. Other organizations expressed discomfort in providing a government study with their members' names or telephone numbers. The NCP organizations that cooperated did so without any hesitation and the provided names and phone numbers were a valuable source of information. ] the ABA contacted the State child support agency and asked if some of their clients would be willing to talk with CSR staff. This avenue proved effective in contacting CPs as well as NCPs.

This final report analyzes and summarizes 215 unstructured telephone interviews that were conducted in an effort to gain a real-life, as opposed to perceived, look at how State child support guidelines are formulated and applied.

1.3.3.5Evaluating State Guideline Reviews

CSR and the ABA developed the following list of factors and issues against which each State's support guideline review was analyzed:

•Basis for study (e.g., anecdotal information or case records);

•Number of cases reviewed by the State;

•Type of cases (e.g., IV-D AFDC, IV-D non-AFDC, or non-IV-D);

•The entity conducting the review;

•Factors listed as justification for deviation;

•Frequency of particular factors;

•Direction of deviation (i.e., upward or downward);

•Average amount of deviation;

•Completeness of case files, if the study provides this information;

•Recommendations resulting from State review;

•Quality of the research design; and

•Any changes made to State guidelines as a result of the review.

The findings from this analysis are presented in their entirety in Volume II of this final report. Specific results, as they pertain to child support issues, are presented in the main body of this report.

1.3.4Analysis Plan for the CPS Supplement and the March CPS Matched File

Current Population Survey (CPS) data were analyzed to provide a national context for the review of child support guideline worksheets and deviations and the analysis of State guideline review reports. This secondary analysis examined the effect of State presumptive guidelines on award amounts. A primary research question guided this analysis: What effect, if any, did presumptive State guidelines have on child support award amounts?

The monthly CPS is an in-person or telephone survey administered by the U.S. Bureau of the Census. Interviewers ask each household member older than age 15 questions on labor force participation, and in addition to basic CPS questions, interviewers ask supplementary questions. Each March the CPS collects economic data on individuals and families for the previous year, and each April of every other year (even-numbered years), interviewers ask CPs supplementary questions on child support. The Census Bureau matched responses to the April 1992 CPS supplement to responses to the March 1992 survey, creating a single data set with demographic, economic, employment, and child support information.

A total of 2,277 cases were in the March-April Matched CPS file, representing 5.3 million persons in the United States with established child support awards in 1992. Cases excluded from the selected subsample included cases in which the award was established in 1992, if no dollar amount was given for a child support award for 1991, or if the supported child was born after 1991. Therefore, 2,218 cases were eligible for inclusion in the analysis.

The key demographic characteristics of the population of child support obligees were as follows:

•Sex.—In 1992 most obligees with child support awards were women. Eight percent of obligees were males, compared with 92 percent who were females. Female obligees received larger award amounts than males.

•Race/ethnicity.—In 1992 three-fourths of obligees were white and non-Hispanic, sixteen percent of the obligees were black non-Hispanic, and seven percent were Hispanic. White obligees received larger award amounts than obligees who were racial or ethnic minorities. Hispanics received the smallest average award amount.

•Age.—In 1992 almost one-half (48 percent) of the obligee population were over age 35 and 1 percent were teenagers. Seven percent were teenagers at the time of the award/modification, 19 percent were ages 20 to 24, 32 percent were ages 25 to 30, 22 percent were ages 31 to 35, and 21 percent were over age 35. Older obligees (age 31 and older), received larger award amounts than younger obligees.

•Number of children.—One-half of obligees (53 percent) had one child, one-third (34 percent) had two children, 11 percent had three children, and3 percent had four or more children. [ Table 1 in: Census Bureau. 1995. Child Support for Custodial Mothers and Fathers: 1991. Current Populations Reports, Series P60-187. Washington, DC: U.S. Government Printing Office.] Average award amounts for obligees with one child covered by the award were smaller than for obligees with more than one child covered by the award. The average award amounts for one child and three children were $2,770 and $5,144, respectively.

Data analysis of the 1992 CPS child support supplement reveals that larger child support award amounts are associated with the following obligee characteristics:

•Persons who were older than 19 in 1992 and at the time that the award was established or modified;

•White;

•Female;

•Currently separated or divorced;

•Had more than one child;

•Resided in the Northeast or Midwest or in a suburban area;

•Had an award modified; and

•Did not receive public assistance in 1991.

For the analysis, CPS data were linked with State-level information on implementation dates of advisory and presumptive guidelines. Sources for this information included OCSE, the Policy Studies Incorporated, and Dr. H. Elizabeth Peters at Cornell University, who is conducting a study of guideline effects using the National Longitudinal Survey for Youth. States' implementation dates widely varied. Some States had slowly phased-in implementation, first adopting presumptive guidelines initially for AFDC cases or in selected counties. Other States made presumptive guidelines that were applicable statewide for all cases by a specific date. Therefore, the dates when States enacted presumptive guidelines is open to interpretation.

The mean and median amounts of awards were larger after guidelines became presumptive, but the average increase was only $400, from an average of $3,131 for awards that were made 2 or more years before presumptive guidelines were adopted to an average of $3,503 for awards 2 or more years after presumptive guidelines were adopted.

The average award amount for blacks decreased after presumptive guidelines were adopted by States (from $3,235 before to $2,362 after). The average amount for whites increased by $600.

Few differences existed between the proportions of persons who had an award established or modified before or after presumptive guidelines were adopted. However, the postpresumptive guidelines group were older at the time that the award was established/modified, although they tended to be younger in 1992 compared with the prepresumptive guidelines group. A smaller proportion were married at the time and a larger proportion had never married than was the case for the prepresumptive guidelines group. More in the postpresumptive guidelines group received public assistance in 1991 than in the prepresumptive guidelines group.

1.3.5Limitations of the CPS Data

A number of important data limitations have shaped and restricted this analysis, including the following:

•Information on parties' income at the time the award was established or modified was not available for this analysis. In addition, the best predictor of a child support award is the obligor's income. Without this information, issues of adequacy could not be addressed.

•The CPs' residences at the time of award establishment or modification were unknown, which hampers analyzing the effect of guideline implementation on awards. It also was unknown if the party lived in a State that had implemented guidelines (either for the specific type of case or for child support matters in general) at the time of award.

1.3.6Summary of Expert Panel Recommendations

As discussed earlier, a panel of experts was assembled to provide advice on the general nature of the research and to make recommendations based on the findings. Their following recommendations have guided the conclusions in this report:

•The current guideline periodic reviews should continue. States should continue to research and design consistent and equitable guidelines and also strive to enforce more consistent application of guidelines in all child support matters.

•The Federal role in child support should be to (1) provide advice to States on guideline construction and application areas they might incorporate into their review processes and (2) suggest solutions to potential problems.

•No steps should be taken to adopt a national child support guideline.

The Expert Panel was comprised of the following individuals:

•Charles Adams, Ph.D., Professor, School of Public Policy and Management, Ohio State University;

•Burt Barnow, Ph.D., Economist, Institute for Policy Studies, Johns Hopkins University;

•Donald Bieniewicz, Consultant, Children's Rights Council;

•Joan Entmacher, Senior Policy Counsel for Family and Economic Security Programs, Women's Legal Defense Fund;

•The Honorable Aubrey Ford, District Judge, Macon County, Alabama;

•Lynne Gold-Bikin, Esquire, Partner, Gold-Bikin, Clifford, & Young;

•Lowell Groundland, Director, Administrative Office of the Courts, State of Delaware;

•The Honorable C. Philip Nichols, Jr., Associate Judge, Seventh Judicial Circuit of Maryland, Prince Georges County; and

•Debbie Stabenow, Michigan State Senator.

1.4SUMMARY AND PLAN OF THIS FINAL REPORT

The remainder of this report presents the findings of this evaluation. Chapter 2 presents the characteristics of cases that (1) follow the presumptive guideline amount of child support, (2) deviate from the presumptive amount for a specified reason, and (3) differ from the presumptive amount for an unknown reason. These cases are referred to respectively, as "guideline cases", "deviation cases", and "discrepant cases".

Chapter 3 presents findings from the analysis of case records, stakeholder interviews, State guideline review reports, and State deviation case studies with respect to seven factor areas. These factor areas represent the most pressing issues in the design and implementation of child support guidelines, including multiple families; income determination, verification, and imputation; health care needs; day care; postsecondary educational support; tax considerations; and custody and visitation.

Chapter 4 is a summary of conclusions and recommendations resulting from this evaluation.

Following these chapters are a series of appendixes: Appendix A contains a detailed look at the research methodology, Appendix B contains typical child support orders and worksheets for the 11 study States, Appendix C contains the child support guidelines of the 11 study States, Appendix D contains summary tables for the 7 factor areas discussed in Chapter 3, Appendix E contains data tables and charts, and Appendix F contains a report on CPS analysis.