Department of Health and Human Services
DEPARTMENTAL APPEALS BOARD
AFDC QUALITY CONTROL REVIEW PANEL
SUBJECT: New Jersey Department of Human Services
Docket No. A-96-113
Decision No. QC101
DATE: August 5, 1996
DECISION
The New Jersey Department of Human Services (New Jersey) appealed a March 18, 1996 quality control (QC) determination by the Regional Administrator (RA) of the Administration for Children and Families (ACF) in State Quality Control Review No. 259 (Federal Quality Control No. 124-0408). This was a prospectively budgeted case in which the client became employed after the most recent best estimate. The RA determined that the assistance unit (AU) received an Aid to Families with Dependent Children (AFDC) net overpayment of $302 for the December 1994 review month.
New Jersey conceded that an overpayment occurred in this case but disputed the method by which the amount of the overpayment was calculated. New Jersey took the position that the client's review month income should be determined pursuant to New Jersey's permissible state practice (PSP) for estimating income. ACF took the position that the amount of the client's review month income must be determined on the basis of her actual review month income.
For the reasons discussed below, we conclude that the Quality Control Manual (QCM) provides that the amount of this overpayment should be calculated on the basis of actual review month income. Therefore, we uphold ACF's determination that the AU was overpaid by $302 in the review month.
I. Relevant Federal Authority
Title IV, Part A of the Social Security Act (Act) establishes the AFDC program to provide assistance to needy children and their caretakers. Under section 408(a) of the Act, the Secretary of the Department of Health and Human Services must establish a quality control system to determine the amount of any erroneous AFDC payments made by a state. Under this system, states review a sample of AFDC payments made during the review period in order to determine the level of erroneous payments. The Act then provides for federal QC re-review of a subsample of the cases reviewed by the state. See section 408(b) of the Act. Pursuant to this statutory mandate, the Secretary has issued regulations for the operation of the federal and state AFDC QC systems. 45 C.F.R. §§ 205.40 through 205.43. Those regulations provide that a state agency must operate its QC system in accordance with applicable regulations and the policies and procedures prescribed in the Quality Control Manuals (QCM) issued by the Department. 45 C.F.R. § 205.40(d)(1).
QC reviews are conducted against PSP. 45 C.F.R. § 205.42(b). PSP is defined as "written rules and policies relating to eligibility and payment that are in accordance with existing, approved State plan provisions or with proposed plan amendments submitted to, but not acted upon, by the Department." 45 C.F.R. § 205.40(b)(12).
States use prospective budgeting to determine eligibility for AFDC and have the option of using that method to determine the amount of AFDC assistance payments. 42 C.F.R. § 233.31. Under prospective budgeting, the state AFDC agency computes the amount of assistance for future months based on the state agency's best estimate of income and circumstances which will exist in those months. 45 C.F.R. § 233.31(b)(1); QC Manual (QCM) § 3420 at IV-6 and IV-7. This estimate is based on the agency's reasonable expectation and knowledge of current, past or future circumstances. Monthly assistance payments thereafter are based on this estimate unless there is a change in the AU's circumstances or the estimate is recalculated.
A change in circumstance means a change occurring after the date of authorization of the initial payment which may affect the AU's eligibility or payment amount. 45 C.F.R. § 205.42(d)(1).
As to errors resulting from a change in an individual's circumstances subsequent to the most recent estimate, the QCM provides that:
If a change in income/income related circumstances occurred since the agency's most recent estimate, which was not acted upon by the agency, use actual RM [review month] income and circumstances to determine the amount of the error.
QCM, § 3420 C.2.(c) at IV-19 (emphasis added); see also QCM § 3420 B.2.b at IV-11.
II. Background
The AU in this case consisted of the client and her dependent children. The case was prospectively budgeted and the most recent best estimate was established in July 1994. At that time, the client had no income and the estimate was correctly based on zero income. The client became employed on September 29, 1994. The local agency was unaware of her employment, did not revise its best estimate, and continued to pay the client a grant based on zero income.
During the review month, December 1994, the client earned $696.97 which was paid in five paychecks, three of which included overtime.
To establish the amount of the error, the state QC reviewer relied on New Jersey's estimating procedure which provides that "non-representative income (or lack of income) shall not be used in calculating the best estimate." New Jersey Assistance Standards Handbook 10:82-1.10(e)4. In determining December income, the state reviewer therefore excluded the client's earnings for the weeks of December 10 and December 17 because those two weeks included overtime which could not be anticipated in the future and were consequently non- representative. The state reviewer did include one week of overtime in the earnings calculation because the client typically worked overtime one week per month. The state reviewer added the amounts earned in those three weeks, divided by three, and multiplied by a conversion factor of 4.333. Pursuant to this process, the state reviewer attributed $490.88 in earned income to the client for the review month.
The federal QC reviewer reached a different result in calculating review month income. The federal reviewer determined earned income by adding all five paychecks for December 1994, dividing by five, and converting the weekly average to a monthly figure by multiplying by a 4.333 conversion factor. RA redetermination letter of March 18, 1996. Using this method, the federal reviewer attributed $604.84 in earned income to the client for the review month. New Jersey appealed to the RA.
Relying on the instructions in section 3420 of the QCM, the RA calculated the amount of the error solely on the basis of the client's actual review month income without averaging or converting the client's weekly earnings. The RA therefore determined that the client had $696.97 in earned income during the review month, which resulted in a net overpayment of $302.
In its appeal to the Panel, New Jersey argued that under the QCM "actual review month income should be used only in the absence of other specific procedures to follow." New Jersey Notice of Appeal at 2. Since New Jersey had a PSP for calculating a best estimate in the type of circumstances presented by this case, New Jersey concluded it should rely on its PSP to calculate the amount of the error. New Jersey argued that its PSP presented "the most accurate picture of the client's employment." Id.
New Jersey also represented that the RA's decision was contrary to ACF's past practices in reviewing New Jersey QC cases. New Jersey explained that it consistently calculated the amount of similar errors on the basis of its estimating procedures and that federal reviewers had never questioned this practice. New Jersey further represented that, in reviewing similar cases in the last six month review period, it had determined that use of its estimating procedures resulted in higher error determinations than use of actual review month income.
ACF did not deny New Jersey's representations concerning federal review of similar cases. ACF maintained that these cases were irrelevant to the decision in this case.
Analysis
For the reasons discussed below, we uphold the RA's error calculation.
The QCM sets forth how to calculate the amount of an error in prospectively budgeted cases. Section 3420 C.2. at IV-18 and IV-19 provides:
(a) If PSP does not provide a method for calculating an estimate for the case situation, use actual RM income and circumstances to determine the amount of the discrepancy.
* * *
(b) If the agency's most recent best estimate is incorrectly calculated according to PSP . . . and, no income/income-related change in circumstance has occurred since the date of the most recent estimate, recalculate the agency's estimate following the State's methodology for estimating.
* * *
(c) If a change in income/income-related circumstances occurred since the agency's most recent estimate, which was not acted upon by the agency, use the actual RM [review month] income and circumstances to determine the amount of the error.
Thus in two situations (the absence of relevant state PSP and when there has been an income/income related change of circumstances), the reviewer is directed to use actual review month income. Only where the original estimate was incorrect and there has been no income/income related change in circumstances does the QCM direct the reviewer to rely on the state's PSP for formulating an estimate.
Calculating the amount of review month income by using the actual review month income rather than by applying a state's PSP to review month income does not necessarily increase the amounts of errors. In fact, New Jersey represented that its most recently reviewed cases would have had lower error results under the QCM standard. The important consideration in the QC process is that all cases are subject to the same method. While ACF could have chosen to instruct states to calculate the error as New Jersey proposes, it did not. We must apply the method that ACF set forth in the QCM unless the method is inconsistent with the Act or regulations.
While we find that the QCM is sufficiently clear as to how these errors are to be calculated, New Jersey represented, and ACF has not denied, that federal reviewers of New Jersey cases allowed New Jersey to apply its PSP to actual review month income in many prior cases. New Jersey asserts that to use actual review month income in this case and to fail to address these other cases would artificially inflate its error rate. If the Panel upheld the ACF error determination, New Jersey wrote that--
[W]e will request the Department of Health and Human Services to take the following actions: 1) adjust New Jersey's payment error rate retroactive to FFY [federal fiscal year] 91 to account for the differences in the error rate attributable to conversion 2) correspondingly adjust the national payment error rate as a result of New Jersey's adjustment and 3) poll other states to determine that its policy is being uniformly applied across the nation.
New Jersey Notice of Appeal at 3.
It is not clear whether New Jersey is requesting the Panel to take this action. However, the Panel's role involves the review of individual difference cases and does not include redetermining state or national error rates. See section 408(b)(4) of the Act. Therefore, we view any such request by New Jersey to be beyond our authority.
However, assuming that New Jersey is correct that the federal reviewers have applied a different standard to prior similar cases, we recognize that this case changes the methodology for New Jersey's cases. Such a change may affect the statistical integrity of QC error determination process as it applies to New Jersey and possibly other states. If it does, ACF presumably will take steps to rectify this problem. If ACF does not and ACF imposes a disallowance on New Jersey, New Jersey may choose to raise this as an issue pursuant to a section 408(j) or (k) appeal.
Conclusion
For the foregoing reasons, we uphold ACF's determination that a net overpayment of $302 occurred in this case.
____________________________
Sara Anderson
____________________________
Peggy McFadden-Elmore
____________________________
Andrea M. Selzer
(..continued)