Second Error Rate Pilot Report
Download this document in Word (1,076 KB) or PDF (1.28 MB)
Congress responded to the issue of improper payments in Federal programs by enacting the Improper Payments Information Act (IPIA) of 2002. This law requires Federal agencies to identify programs that are vulnerable to improper payments and to estimate annually the amount of underpayments and overpayments made by these programs.
The Executive Branch also has worked to address the improper payments issue. A goal of the President’s Management Agenda (PMA) is “eliminating improper payments.” To this end, agencies are to establish a baseline of the extent of improper payments and to set goals to reduce improper payments.
Because the PMA and IPIA provide for establishing a child care error rate as one on-going measure of program efficiency, the Child Care Bureau (CCB) launched the project: "Measuring Improper Payments in the Child Care Program" to identify and describe methods that could help States identify, measure, and prevent improper payments in the administration of the Child Care and Development Fund (CCDF). A major focus of the CCB project was to work in partnership with the States to determine the feasibility of computing a child care improper payments error rate.
During Fiscal Year 2004 of the project, the CCB worked with 11 partner States6 to gather information and recommend methods to identify, define, measure, and prevent improper payments. During FY 2005 of the project, the CCB contracted with Walter R. McDonald & Associates, Inc. (WRMA) to develop and pilot test in four States—Arkansas, Colorado, Illinois and Ohio—a common methodology that States could use to compute an error rate.
The purpose of this report is to summarize the findings from the second pilot to further refine and develop a methodology to compute an error rate in five States—Florida, Kansas, New Jersey, Oregon and West Virginia. Following a brief overview of the Child Care Program, this chapter provides the background on the pilot project activities to date and the objectives of the current pilot.
The CCDF is a block grant that allotted over $5 billion to States, Territories and Tribes to support child care subsidies for low-income working families during FY 2006.7 The CCDF block grant allows maximum flexibility for States to set policies regarding eligibility and fiscal management approaches, as well as define administrative structures that allow maximum choice for parents. As a result, States' eligibility criteria, rates, regulation of child care providers, and payment mechanisms vary widely among jurisdictions. This flexibility makes it difficult to develop common approaches for identifying and measuring improper payments.
In FY 2004, the CCB visited six States (Arkansas, Connecticut, Indiana, Ohio, Oklahoma, and Virginia) and worked extensively with five additional partner States (Georgia, Maryland, Oregon, South Carolina, and Wisconsin). Together, the CCB and State partners gathered information on the range of strategies that States use to prevent and identify errors and appropriate enforcement actions taken when improper payments occur. Results revealed that efforts varied extensively among States. Few States had implemented systematic methodologies to estimate the amount of improper payments.
In FY 2005, the CCB contracted with Walter R. McDonald & Associates, Inc. (WRMA) to develop and pilot test a common methodology that States could use to compute an error rate and work with States to validate existing protocols or develop new approaches to address improper payments and fraud. Specific requirements, including statistically valid sampling, guided the conceptual design of this pilot. The CCB, in consultation with a number of “partner States,” chose to focus the analysis on eligibility error in order to measure an element that is common to every State and to mitigate some of the variation among State definitions encountered during the first error rate pilot. The methodology—conducted in Arkansas, Colorado, Illinois, and Ohio—had four main components:
- The contractor assisted each pilot State to select a random sample of up to 150 cases (children), using a sampling frame of all children in the State authorized to receive a child care payment8 during October 2004. The contractor designed the sample size to produce a statistically valid estimate of erroneous payments.
- Pilot States customized a Record Review Worksheet template to reflect child care policies in the State. The States used this instrument to guide a record review of the sampled cases to identify administrative errors in eligibility determination. States collected data regarding the number of cases with errors and whether the errors led to an improperly authorized payment. Although all of the worksheets contained common elements, the definitions pertaining to those elements varied from State to State.
- The contractor conducted site visits and provided technical assistance to the pilot State representatives who conducted record reviews and collected data. The contractor computed the error rates using the data submitted by the pilot States.
- The contractor also conducted telephone interviews, using a consistent protocol to gather additional information about improper payment activities in five States. These States were Arizona, California, Kansas, Nebraska, and New Hampshire.
B. Objective of the Second Pilot
The purpose of the FY 2006-2007 pilot was to further refine and develop a methodology to compute an error rate in five States—Florida, Kansas, New Jersey, Oregon, and West Virginia. Similar to the first error rate pilot, the methodology of the current pilot focuses on client eligibility and employs a case record review process to identify cases with errors, cases with errors that result in improperly authorized payments9, and percentages of authorized payments in error. However, in the second pilot the methodology provided pilot States with their results so they could analyze the types and sources of error. Following an analysis of the findings from the case record review process, the contractor forwarded the results to the pilot States for their internal review. Pilot States responded to a short survey providing an explanation for the causes of the errors and a description of next steps or corrective actions to be taken as a result of participation in the error rate pilot.
___________
6 In FFY 2004, the CCB visited six States (Arkansas, Connecticut, Indiana, Ohio, Oklahoma, and Virginia) and worked extensively with five additional partner States (Georgia, Maryland, Oregon, South Carolina, and Wisconsin). Return to text.
7 Child Care Development Fund Fact Sheet (October 2006) available on the Child Care Bureau website: http://www.acf.hhs.gov/programs/ccb/ccdf/factsheet.htm. Return to text.
8 For both pilots, States calculated and recorded the authorized payment amount based on the eligibility determination process as indicated within the case record. In both pilots the term “payment” refers to the amount authorized for payment. Return to text.
9 For both pilots, States calculated and recorded the authorized payment amount based on the eligibility determination process as indicated within the case record. In both pilots the term “payment” refers to the amount authorized for payment. Return to text.
June, 2007