UNITED STATES DISTRICT COURT, DISTRICT OF COLUMBIA.
May 14, 1976
STATE OF MARYLAND et al., Plaintiffs, State of California, Intervenor,
F. David MATHEWS et al., Defendants
The opinion of the court was delivered by: GREEN
JUNE L. GREEN, District Judge.
This is an action brought by thirteen states, one county and one intervening state
against F. David Mathews, successor Secretary of Health, Education and Welfare (HEW) and the administrator, Social and Rehabilitation Service in HEW. Plaintiffs, all participants in the Aid to Families with Dependent Children (AFDC) program, seek to have the HEW regulation contained in 45 C.F.R. § 205.41 published at 40 Fed. Reg. 32954 on August 5, 1975, declared invalid.
Plaintiffs assert that they do not question HEW's right to set quality controls. What is contested is HEW's authority to establish a percentage of financial disallowance by regulation and the reasonableness of the specific percentages imposed. The challenged regulation provides for the withholding of federal financial participation under the AFDC program when erroneous payments to AFDC recipients exceed prescribed tolerance levels of 3% of payments to ineligibles and 5% of overpayments. The case is presently before the Court on the defendants' motion to dismiss or in the alternative for summary judgment, and the plaintiffs' cross-motion for summary judgment.
For a more complete understanding of the issues raised by the parties, a brief background of the program is necessary.
The AFDC program, one of several income maintenance programs under the original Social Security Act of 1935, provides federal matching funds in specified amounts to states to support payments to plan recipients and to cover a percentage of administrative expenses. 42 U.S.C. §§ 603(a), 1301(a)(8). To participate, a state must submit a plan which meets criteria established by federal statute and the plan must be approved by the Secretary of HEW. 42 U.S.C. § 602.
The regulations at issue are the department's efforts to effectuate quality control and to facilitate corrective action. Under the regulations the states are required to implement a system by which a prescribed number of sample AFDC cases are selected during a six-month period. An investigation is then made to determine whether each case was correctly paid. The figures regarding the number of overpayments, underpayments
and payments to ineligibles are projected to the universe of all AFDC grants within each state during the period. The difference between each state's percentage of payments to ineligibles and as overpayments which exceed the tolerance levels of 3% for ineligibles and 5% for overpayments set by the challenged regulation will be converted into a dollar figure and constitute the federal financial participation (FFP) withholding.
The purported basis offered by the defendants for the selection of the respective percentage sanctions is a study made by Westat where intensive in-depth investigations were made of a sample of recipients under AFDC.
The Westat report
states, however, "It was mutually agreed that Westat's evaluation was to be of the systems without regard to the potential use of QC as a sanctioning tool . . .".
Further, in discussing the percentage tolerances for overpayments, the report states,
"5b. The current 5% level for overpayments does not seem to be achievable under present circumstances. Westat recommends that this level be changed to 9% as a temporary national tolerance. This level is also rigorous under present conditions . . .".
These comments were made despite a March 20, 1973 memorandum from HEW's Director of Division of Program Evaluation to the Acting Commissioner of the Assistance Payments Administration (APA) in HEW, which states as follows:
"In view of the fiscal disallowance policy requiring zero tolerance, DeGeorge and Cohen were apprehensive that the Westat report might include statements to the effect that this goal was unreasonable. As a result Joel Cohen, Tom Scarlett, Harris and I had a number of meetings with Westat to ensure that they limited their observations to the terms of the contract and did not offer opinions on the 'proposed' use of QC for fiscal disallowances. They agreed to do this.
The importance of this approach is that Mr. Cohen did not want our legal case weakened by one of our own consultants stating that QC should not be used in this manner or zero tolerance is an impossibility."
For an understanding of what is involved in each application filed in each state under AFDC, the following is quoted from the affidavit of Bill B. Benton, Jr.
Mr. Benton is an expert in the field of social and economic policies in welfare matters with special emphasis on quality control and error corrections.
"The eligibility determination process works in the following way. A prospective recipient must complete and sign an application for public assistance. In so doing an applicant must respond to 100 or more inquiries, the answers to each of which may affect the applicant's eligibility or the correct amount of the assistance payment. In addition, an applicant is commonly required to satisfy various other requirements, such as registering for the WIN program (a federally-established work training program), registering at the State Employment Service, providing information concerning an absent parent and/or the amount of child support payments, if any, etc. Many of the factors called for in the application process involve interpretations or judgments by the eligibility worker. For example, the eligibility worker must decide what of the applicant's expenses are 'work related', what portion of the applicant's rent is properly allocated to the assistance unit, what portions of the applicant's income are 'regular' income, how child support or other benefit payments should be allocated among various children, etc.
"In addition, the eligibility worker must, on the basis of the information supplied, make numerous calculations in determining eligibility and the correct amount of the assistance payment. Basically, the level of assistance is a function of an applicant's need and resources compared against an existing state standard. The usual first step is the determination of the applicant's income. This must take into account the value of in-kind income as well as income in cash form. The latter must be broken down between earned income, support payments, other benefits and unearned income because each of these categories is treated differently in determining eligibility or the amount of payment.
"Once income is determined it must be divided into regular and irregular income. Those figures must then each be converted to a monthly figure. As a general rule, applicants do not have a steady fixed monthly income; they may be reimbursed on a daily basis, they may work at irregular intervals, etc. Accordingly, the conversion to a monthly rate is itself a complicated arithmetic process. Once this conversion is made, the eligibility worker must determine what portions of that income are non-countable under federal regulations. (Examples of non-countable income are scholarships, income from certain work training programs, income earned under the Older Americans Act, etc.) Often some portion of the monthly income figure is exempt and some portion not exempt. As to that portion of an applicant's income that is not exempt, the eligibility worker must then make computations as required under income disregard regulations.
"Perhaps the most difficult of all the calculations the eligibility worker is required to make involve income disregards. For example, certain types of income are subject to a flat $5 per month disregard. Even this basic disregard, however, is subject to exception; and amendments to Title IV establish a new formula to be used in lieu of the flat $5 disregard in areas of child support. This, however, will be true for one year only. A different amount of Social Security benefits is disregarded under federal regulations. Likewise in computing the amount of an applicant's resources to determine if her assets are in excess of the maximum allowable, some assets are disregarded entirely. Certain portions of the value of some liquid assets are disregarded under the regulations; other portions of the value of fixed assets are disregarded. However, the amount of value to be disregarded will vary, depending upon the use to which the asset is put. For example, a client's automobile is excluded altogether as a resource if it is used to obtain or retain employment; if, however, the automobile is not used in connection with employment, only a portion of its value is disregarded.
"The most complicated of all the disregard calculations arise under the earned income disregard regulations. Whether this disregard applies at all is not easy to know. It depends on whether an initial determination or a payment recomputation is involved, on the level of earned income of the applicant and others within her assistance unit in each of the preceding four months, and on whether there has been a recent change in employment status or earnings and why. Where this disregard is determined to be applicable the first step is to compute a monthly earned income figure for each person in the assistance unit. This is not easy in those cases where income includes payments under various federal work training or job development programs for some of these payments are counted and some are not. Once the monthly earned income is arrived at, expenses determined to be 'work related' are subtracted. Then $30 plus 1/3 of the balance of that figure is disregarded. These deductions are the monthly amount spent for such items as baby-sitting, transportation, Social Security deductions, union dues, retirement contributions, state and federal taxes, lunches, etc. These expenses, which can fluctuate, sometimes on a daily basis, must be tabulated on a monthly basis. The calculations must be made not simply for the head of the household but for each member of the assistance unit with earned income. This includes a child who earns money baby-sitting or a child who has a paper route.
"In addition to income computations, determinations and calculations of the level of need of the applicant must be made. Some states, such as Maryland, have simplified these computations by the adoption of what is known as a 'flat grant' or consolidated standard of need. Under Maryland's flat grant system the only variable is the size of the assistance unit. Some states' 'flat grants' include shelter as a variable, and other states have as many as a dozen variables under a 'flat grant'. However, flat grants are typically more expensive than individual need determinations because states tend to set the flat grant figures high enough to avoid reducing benefits to most of the recipients under the previous individual determination system.
"Federal regulations do not require flat grants and many states continue to operate under itemized need grants. These require taking into account a large number of factors which affect an assistance unit's level of need, such as rent, living arrangements, insurance, utilities, fuel, clothing, special diets, enrollment in school, etc. These computations can be quite complicated and usually vary from month to month. For example, calculating shelter allowance and utility costs would involve arriving at a monthly rental (many clients pay by the week), proration in instances where shelter is shared, averaging fluctuating utility expenses, determining whether a telephone is 'essential', applying a special allowance for a family that lives in public housing, and ascertaining whether rental payments, including internal rental arrangements among members of the family unit, were in fact paid in the amount stated.
"The determination and calculation process described above must be made in the case of each applicant for assistance."
After the initial screening out of ineligibles and a determination having been made that a family is entitled to assistance under AFDC, Federal regulations require a redetermination of eligibility and level of payment of each client at least every six months. Although the rules state at least every six months, the state is penalized under the challenged regulation if it does not determine a change of circumstance and make the change in payment by the end of the second month following the change of circumstance.
The Court will now turn to the issues raised by the parties. Defendants assert that the Court lacks jurisdiction over the subject matter pursuant to 28 U.S.C. § 1346(a)(2); and that the action is not ripe for judicial review, as no withholdings have occurred.
At issue herein is the validity of a regulation enacted in accordance with the Administrative Procedure Act by an agency subject to the Act. The Court, therefore, has jurisdiction pursuant to the Administrative Procedure Act, 5 U.S.C. §§ 701-706. Scanwell Laboratories, Inc. v. Shaffer, 137 U.S. App. D.C. 371, 424 F.2d 859, 874 (D.C.Cir. 1970); and see Pickus v. Board of Parole, 165 U.S. App. D.C. 284, 507 F.2d 1107 (D.C.Cir. 1974).
The ripeness argument is similarly without merit since pre-enforcement review of a regulation has long been recognized. Abbott Laboratories v. Gardner, 387 U.S. 136, 87 S. Ct. 1507, 18 L. Ed. 2d 681 (1967).
As to the merits of the action, defendants have argued that the Act does not authorize federal financial participation to states for erroneous payments since such payments are not made in accordance with the approved state plan. To assure the efficient administration of the Act, however, the Secretary may provide matching funds for erroneous payments in any amount he deems suitable.
The plaintiffs have responded that, whether erroneously made or not, all payments made under the operation of an approved state plan are to be provided federal financial participation. Thus, the promulgation of the regulation in question was inconsistent with the Act, arbitrary, capricious and an abuse of discretion. Furthermore, even if statutory authority exists for such action, the regulation lacks a rational basis, is not reasonable, and therefore is contrary to the law.
The basic provision setting forth the states' right to federal financial participation states in pertinent part:
"(a) From the sums appropriated therefor, the Secretary of the Treasury shall . . . pay to each state which has an approved plan for aid and services to needy families with children. . . .
"(1) an amount equal to the sum [of a specified proportion] of the total amounts expended during such quarter as aid to families with dependent children under the State plan. . . ." 42 U.S.C. § 603(a)(1).
From the statutory language, it can be concluded that payments which are not made properly, pursuant to the approved plan, are not to be matched by federal funds. This interpretation is further reflected in the Act under 42 U.S.C. § 603(b)(2)(B) which provides for the recoupment by the federal government of their proportionate share of an erroneously made payment which is recovered by the state, and by 42 U.S.C. § 604(a)(2) which provides for the withholding of a percentage of federal funds if the state has failed to substantially comply with certain duties in the administration of the plan and is deemed to have acted outside the authority of the plan.
To assure the efficient administration of the Act, which would include the proper allocation of federal funds to states pursuant to 42 U.S.C. § 603, the Secretary has been vested with extensive power to promulgate regulations. 42 U.S.C. § 1302;
Serritella v. Engelman, 339 F. Supp. 738, 752 (D.N.J.) aff'd. per curiam, 462 F.2d 601 (3rd Cir. 1972). It is under this authority that the regulation in question was issued.
The standards are clear by which the validity of the challenged regulation is to be tested. The regulation must fall if it is shown to be unreasonable and inconsistent with the Act, or not "reasonably related to the purposes of the enabling legislation." National Welfare Rights Organization, et al. v. Mathews, 174 U.S. App. D.C. 410, 533 F.2d 637, 645 (D.C.Cir. 1976), and cases cited therein. Furthermore, the regulation will be held invalid if any of its "provisions were framed in an 'arbitrary' or 'capricious' manner." Amoco Oil Co. v. Environmental Protection Agency, 163 U.S. App. D.C. 162, 501 F.2d 722, 739 (D.C.Cir. 1974); 5 U.S.C. § 553(c) and § 706(2)(A).
The general purpose of the AFDC program is to encourage "the care of dependent children in their own homes or in the homes of relatives by enabling each State to furnish financial assistance and rehabilitation and other services, as far as practicable under the conditions in such State. . . ." 42 U.S.C. § 601. Consistent with this purpose is the overall concept of quality control regulations which would assure that a minimum of erroneous payments will occur and thereby ensure that a satisfactory amount will be available for those who properly qualify as recipients under the state plan. At issue here, however, is the implementation of federal financial withholdings as a means to effectuate quality control.
As previously indicated, under 42 U.S.C. § 603, erroneous payments are payments made outside the state plan. It is recognized, however, that under the existing administrative structure, the elimination of all erroneous payments is totally unrealistic. Comments to 45 C.F.R. 205.40 and 205.41; 40 Fed.Reg. 32956, Tuesday, August 5, 1975. It is agreed, though, that error rates need to be and can be further reduced. The effect of the regulation, therefore, is to define an acceptable level of erroneous payments within which the state will be deemed to be complying with the Act in its administration of its state plan. Consequently, under the Secretary's rulemaking power to assure the efficient administration of the Act, it can be concluded that a regulation establishing a withholding of federal financial participation in a specified amount set by a tolerance level is consistent with the Act. It must also be concluded, however, that the tolerance level set must be reasonable and supported by a factual basis and not established in an arbitrary or capricious manner. To establish a level under any lesser standard would be to fashion a tool which is inconsistent with the Act in that it does not promote any further efficiency and it prevents the state from furnishing assistance as far as practicable given the conditions in the state.
The defendants have not contested, and the record supports, that the initial establishment of tolerance levels at 3% and 5% was done so arbitrarily with no previous empirical study.
As has been set forth in the record, these levels were set because the Secretary felt that they were attainable and were ones which the public would accept. After the initial adoption of the 3% and 5% levels, a study was prepared by Westat Co. Research, Inc.
Nonetheless, the defendants now seek to justify the tolerance levels on the basis of the Westat study.
As has been set forth in the initial portions of this opinion, the Westat study was not designed to evaluate the use of tolerance levels as a sanctioning tool. Indeed, defendants expressly desired that Westat not offer any opinions in this regard for fear that the levels would be found inconsistent with their legal position. Therefore, defendants' assertions that this study provides the foundation for the tolerance levels clearly is not supported by the record.
Setting aside the fact that the study was not meant to evaluate the use of tolerance levels as a sanctioning tool, defendants' citation to the study's evaluation of the levels in support of the agency's assertions that the 3% and 5% levels are reasonable is equally misplaced. Both levels were found to be too rigorous as a permanent standard. Additionally, the study indicated that the levels should be set at a number which some states are achieving currently, and which other states can be reasonably expected to reach. According to HEW statistics reflecting case error rates for January-June 1975, only 9 states were meeting the target rate for ineligibles and only 25 states were attaining or were within three percentage points of meeting the target level.
For the same period in 1974, only 5 states were in compliance and only 17 were meeting or were within three percentage points of the set level. As to the statistics for overpayments for January-June 1975, only 2 states were within the 5% tolerance level and 8 states were meeting or were within three percentage points of the level. For the same period in 1974, only 1 state met the level, and 3 were meeting or were within three percentage points of the level. In addition, for the above period from 1974 to 1975, the statistics reflect that 19 states increased in their error rate as to ineligibles, and 9 states increased in their error rate as to overpayments.
Thus, the levels appear to be set unreasonably low in view of the Westat criteria.
It should also be noted that the substantive shortcomings of the regulation are reflected in its procedural history. Fundamental questions addressed to the agency concerning the foundation of the regulation included "the use of Quality Control for purposes of disallowing Federal funds; states' ability to attain the established tolerance levels for erroneous payments; inclusion of erroneous payments occasioned by 'client error' [as opposed to agency error] in state Quality Control samples; and the alleged violation of the Federal-State partnership by the disallowance of Federal financial participation pursuant to the Quality Control program." 40 Fed.Reg. 32954 (August 5, 1975). In the preamble to the final regulation in question, the agency merely indicated that these key concerns were responded to in 1973, and that the agency's position was unchanged. Id.
However, upon reading the 1973 proposed regulation, published as modified 38 Fed. Reg. 8743 (April 6, 1973), one searches in vain to find any findings or discussion of relevant factors concerning the key questions raised. Thus, not only has the agency been shown to have failed to articulate factors and findings pointing to the substantive basis for the particular levels, but the procedural deficiencies of the regulation are highlighted by this omission. See National Welfare Rights Organization v. Mathews, 174 U.S. App. D.C. 410, 533 F.2d 637, 648-49 (D.C.Cir., 1976); 5 U.S.C. § 553(c).
Consequently, in view of the facts that the tolerance levels were arbitrarily established at 3% and 5% without the benefit of an empirical study, and that defendants' reliance on the Westat study is misplaced, both as to the role of the study as providing the basis for the levels and as to the reasonableness of the levels, the Court holds that the regulation as it concerns the 3% and 5% levels was framed in an arbitrary and capricious manner and as an abuse of discretion. It is thereby inconsistent with the Act by unreasonably making funds unavailable to fulfill the purpose of the Act and by preventing the states from furnishing assistance as far as practicable given the conditions of the state, and is therefore invalid.