Alternative for Estimating Improper Payment Amounts
CFO Council Working Group on Improper Payments
In section 2(a), "Identification of Susceptible Programs and Activities," and section 2(b), "Estimation of Improper Payments," the Improper Payment Information Act of 2002 (IPIA) requires heads of Federal agencies to:
- annually review, in accordance with the guidance prescribed by the Director of OMB, all programs and activities that may be susceptible to significant improper payments (risk assessment)
- estimate the annual amount of improper payments for programs and activities that may be susceptible to significant improper payments
Although IPIA requires an annual review of all programs and activities and an estimate of improper payments for programs and activities that may be susceptible to significant improper payments, it does not specify the methodology for estimating such payments. As directed by the law, OMB issued guidance (M-03-13) in May 2003 that prescribes minimum requirements for developing a statistically valid improper payment estimate on an annul basis. Working with relevant agency and program staff, the CFO Council Working Group on Improper Payments has determined that for a small number of large and complex high-risk programs, the M-03-13 requirements may not be achievable, given program realities and resource constraints.
This document, therefore, identifies an alternative approach agencies should consider in cases where obtaining an annual improper payment rate consistent with the requirements of OMB M-03-13 is not practical or feasible. Agencies should obtain OMB approval of this conclusion before deciding to utilize the alternative explained below. Use of this alternative should not preclude agencies from performing their annual risk assessments as required by M-03-13.
Scenario 1. Agency has a previous (less than five years old) baseline improper payment rate, and has a funded plan in place to obtain another full program improper payment rate within five years from the baseline year. The baseline rate should meet the requirements for statistical rigor as set out in M-03-13.
Step 1: Aging the baseline rate. The agency should use statistical methods to update or "age" the baseline improper payment rate in the intervening years, until the next program rate is established. Specifically, the agency should use available data to extrapolate updates of the baseline rate. At a minimum, the analysis should conclude whether the baseline rate is trending upward, downward or level.
Step 2. Program component annual measurement. The agency should develop an annual error rate for a component of the program. The component can be defined based on population, program area, or known problem area. To the extent possible, the component chosen for analysis should be based on risk so that the agency is targeting an area of the program where a significant amount of improper payments are expected to occur. This program component should be statistically sampled annually to obtain an error rate consistent with the statistical rigor requirements of M-03-13. The goal for the component study is not to extrapolate an improper payment rate for the program as a whole. The goal, rather, is only to estimate an improper payment amount for the relevant program component being studied. Component-specific baseline and target rates, as well as corrective action plans, should be developed to measure and assess agency progress in reducing improper payments in the program component.
Please note, that both Steps 1 and 2 in Scenario 1 are recommended if this alternative is chosen by the agency and approved by OMB.
Scenario 2. No baseline comprehensive improper payment rate established and no statistically valid methodology yet developed to obtain one.
Step 1. Plan for comprehensive baseline measure. A methodology to obtain a comprehensive baseline improper rate must be developed with a timeline that would allow for the first measurement to occur within three years. Statistical rigor must meet, at a minimum, the requirements of M-03-13.
Step 2. Program component annual measurement. While the agency is working toward a comprehensive baseline rate, the agency should identify a component to annually measure, and begin and report on this measurement within one year. (See Step 2 in Scenario 1 above.)
Step 3. Once the baseline rate is established, and unless the rate can be re-measured annually, the agency should perform both Steps 1 and 2 of Scenario 1 above to ensure that adequate information on improper payments is obtained on an annual basis.
It is important to note that agencies are not restricted to using only these two approaches; different strategies may be necessary because of pre-existing legislative requirements and/or prohibitions, or because a different method would be more appropriate in providing results for a particular program. However, whether an agency decides to use one of these two scenarios, or proposes a different process, all deviations from M-03-13 must be approved in advance by OMB.