The Secretary of Health and Human Services Report to Congress on the Implementation of the Performance-Based Incentive System

Interim Report

Department of Health and Human Services
Office of Child Support Enforcement

October 2003

INTRODUCTION

Attached is the interim "Study of the Implementation of the Performance-Based Incentive System" report. This report was prepared for the Department of Health and Human Services, Office of Child Support Enforcement (OCSE) under contract with the Lewin Group in response to a Congressional mandate for such information. The study reviews the implementation of the performance-based incentive funding system through which the Federal government awards payments to state child support enforcement (CSE) programs.

The Federal Office of Child Support Enforcement implemented the new incentive formula over the fiscal year 2000 to 2002 period. The statute provided a gradual phase-in, in part, so that state officials would have time to perfect their measurement of performance and identify factors that affect determination of incentives.

BACKGROUND

Since 1975, the Federal government has paid incentives to state child support enforcement programs to encourage improvement in collections through efficient establishment and enforcement techniques. The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) required the Secretary of Health and Human Services to develop, in consultation with states, a performance-based incentive funding system through which the Federal government would award payments to state child support enforcement programs.

The 1998 Child Support Performance and Incentive Act (CSPIA) created the new structure to reward states that operated effective CSE programs. CSPIA requires HHS to produce interim and final reports that detail the implementation of this new system and offer recommendations for its improvement. This interim report describes the development of the new incentive system, components of the system, and initial program results. The final report will explore state experiences implementing the new system and include any recommendations for changes in the system deemed useful to improve the operation of the child support enforcement program. Under this Act, incentive payments are linked to a state's performance in five areas:

1. Paternity establishment percentage,
2. Percent of child support cases with orders established,
3. Current child support collections as a percent of total amount due,
4. Percent of cases making a payment toward arrears, and
5. Cost-effectiveness (i.e., total collections divided by total administrative costs).

THE STUDY ON THE IMPLEMENTATION OF THE PERFORMANCE-BASED INCENTIVE SYSTEM

• Describes the original incentive system and the development and structure of the new system.
• Describes the incentive calculation and payment processes under the new system. It details data collection methods, the data reliability audit process, the incentive calculation steps, and the payment steps.
• Provides information on program results and incentive payments for the 1999 to 2001 reporting period. It describes state trends on each of the five performance measures, outlines state experiences with the data reliability audits, and explores national and state trends in incentive payments.
• Explains the next steps for the project. It describes the empirical research and the interviews with state officials that will be conducted for the final report.
• Provides detailed state-level information on performance and incentive payments.

STUDY FINDINGS

The 1998 Child Support Performance and Incentive Act (CSPIA) performance-based incentive system is working. Many states received more money under the new system than they would have under the old system. States are changing their practices which has enabled many of them to improve and maintain higher performance levels. Post audit results show that state performance has increased for all incentive measures, except for the cost-effectiveness measure. Early on, many states were unable to pass data reliability audits, now, states can put their efforts toward maintaining performance. In addition, the study found that:

• OCSE paid fiscal year 2000 incentives based, in part, on state performance. As called for in CSPIA, OCSE successfully began implementation of the performance-based system in fiscal year 2000. One-third of the fiscal year 2000 payments were based on the new system - the remaining two-thirds were determined using the old system. In 2001, two-thirds were determined using the new system.
• States face on-going challenges with data reliability, particularly for the paternity establishment measure. Data reliability improved slightly in fiscal year 2000 - the first year the five performance measures factored in the incentive calculation. Between fiscal year 1999 and fiscal year 2000, the number of states failing the paternity establishment audit declined from 17 to 13. The number of states failing the current collections and cases paying toward arrears measures dropped from 12 in fiscal year 1999 to 7 in fiscal year 2000. States continued to report accurate data for the cases with orders measure, with only two states failing the audit. Only one state failed the cost-effectiveness measure. However, this one year trend of improvement did not continue in FY 2001, partially due to an increase in the data reliability standard from 90 percent to 95 percent. Overall, 26 states failed an audit on at least one measure in 2001.
• Reasons for unreliable data varied. If a state fails any part of its audit, OCSE provides an in-depth description of the reasons the data were found to be unreliable in the audit report. Although the specific justification for OCSE's finding differs for each state, examination of the fiscal year 2000 audit reports indicates that the failures are attributable to a few general reasons, such as programming errors, incomplete or inadequate audit trails and clerical, data entry, and conversion errors.
• Incentives as a percent of the maximum varied widely in first year of implementation. If a state reported high performance on each of the five measures, the state would have earned 100 percent of its potential incentive. In fiscal year 2000, no state achieved the 100 percent standard. The proportion of each states' potential incentive payments received ranged from 23 percent to 87 percent. States that received less than 40 percent of their potential incentive payments typically failed audits on one or more measures.
• In the aggregate, states received higher payments under the new system than they would have under the old system. In fiscal year 2000, OCSE paid $391 million in incentives. By contrast, the old system would have generated only $375 million in incentive payments to all states. This is due largely to the fact that the old system was strongly tied to levels of TANF collections, and most states have experienced declines in their TANF caseload.
• Transition to the new system created financial winners and losers in fiscal year 2000. Thirty-five states received more incentive payments with the performance-based system partially phased in than they would have under the old system, and 23 of those states had more than a 10 percent improvement in the level of incentives earned. On the other hand, 16 states earned less incentives, and 12 of those states saw more than a 10 percent decline in incentives relative to the old rules.
• Reports from states with reliable data suggest performance improvement on most measures. A review of audited fiscal year 1999 to fiscal year 2001 data suggests that performance improved in most areas. Median state scores increased for the IV-D paternity establishment (12 percentage points), cases with orders (3 percentage points), current support (5 percentage points), and arrears measures (3 percentage points)¹. On the other hand, the statewide paternity and cost-effectiveness measures declined.
• Although the cap on total incentives makes forecasting incentive payments challenging, many states benefit from the cap. States receive incentive money that other states did not earn. This is because the other states did not have reliable data or did not receive an incentive for one or more measures because of poor performance.

¹ Median calculated using only states that passed the 1999 and 2001 audits.