Pennsylvania Department of Public Welfare, DAB No. 485 (1983)

GAB Decision 485
Docket No. 82-193

December 21, 1983

Pennsylvania Department of Public Welfare;
Settle, Norval; Teitz, Alexander Garrett, Donald


The Pennsylvania Department of Public Welfare (State) appealed a
determination by the Social Security Administration (SSA) disallowing
$4,602,594 in federal financial participation (FFP) relating to alleged
erroneous payments made to certain recipients of Aid to Families with
Dependent Children (AFDC). SSA based its determination on audit
findings that the recipients were not registered for the Work Incentive
Program (WIN) and were not exempt from the registration requirement
under section 402(a)(19)(A) of the Social Security Act (the Act) and the
implementing regulations. The central issues concern whether the
statute requires federal participation in erroneous payments, whether
the applicable SSA policy provided for disallowing payments for
erroneous determinations about WIN registration by extrapolating from a
sample, and whether the auditors' findings support SSA's conclusions
that the recipients were not exempt from the WIN registration
requirement. We also address several issues concerning the calculation
of the amount of overpayments involved in the disallowance. With
certain exceptions discussed later in this decision, we sustain the
disallowance with regard to the majority of the payments for the 120
individual cases identified by the auditors. We reverse the
disallowance, however, to the extent it was based on extrapolation from
the sample because we find that, although extrapolation from a valid
sample generally can be used, here its use was inconsistent with SSA's
own guidance.

Our decision is based on the written record. The State requested a
hearing pursuant to 5 U.S.C. 554 rather than one under the Board's
procedures at 45 CFR Part 16. The Board denied that request in a ruling
dated April 29, 1983 and reaffirmed the denial in a ruling dated October
6, 1983. The Board offered the State an in-person hearing pursuant to
the Board's procedures, but the State rejected the opportunity (see
Confirmation of Telephone Conversations, dated October 11, 1983, and
State's Written Submission in Lieu of Oral Argument, dated November 9,
1983).

The Board's Jurisdiction

The State contended that the Board does not have jurisdiction over
this appeal because SSA allegedly failed to meet the requirements (2) of
45 CFR 74.304 by not pursuing informal negotiation procedures with the
State before taking the disallowance. Further, the State argued that 45
CFR 201.13 required that the State have an opportunity to appeal the
audit findings to the Commissioner prior to the Agency's issuing a
notice of disallowance, and that SSA failed to offer the State this
opportunity. In a ruling dated June 10, 1983, the Board rejected the
State's arguments and concluded that it has jurisdiction over this
appeal.

Background

The WIN Program. Congress created the WIN program in 1967 by enacting
Title IV-C of the Social Security Act. The program's purpose was to
assist parents and other caretaker relatives receiving assistance under
the AFDC program to become employed. In 1971 Congress made WIN
registration mandatory for nonexempt individuals, as a condition of
eligibility for AFDC. Section 402(a)(19)(A). Section 402(a)(19)(A) of
the Act requires that a State plan must provide --

(A) that every individual, as a condition of eligibility for aid
under this part, shall register for manpower services, training, and
employment as provided by regulations of the Secretary of Labor, unless
such individual is --

(i) a child who is under age 16 or attending school full time;

(ii) a person who is ill, incapacitated, or of advanced age;

(iii) a person so remote from a work incentive project that his
effective participation is precluded;

(iv) a person whose presence in the home is required because of
illness or incapacity of another member of the household;

(v) a mother or other relative of a child under the age of six who is
caring for the child;

* * *

SSA implemented these requirements at 45 CFR 224.20 (1978). State
welfare agencies are responsible for referring AFDC recipients to the
Department of Labor for WIN registration. The caseworkers' decisions
about whether AFDC recipients must register (3) are verified by the
states' Income Maintenance Units (IMUs), and the IMUs determine the
current WIN status of recipients at the time of eligibility
redeterminations (every six months). WIN registration does not affect
the need and income factors which are used to calculate the amount of
AFDC assistance provided.

The Audit. The HHS Audit Agency performed an audit of the WIN
registration program in the State for the period March 1, 1978 to March
31, 1979. The auditors concluded that the only major problem in the
State's administration of the WIN program was recipient registration.

The auditors used statistical sampling techniques to review WIN
registration. Only 27 of the State's 67 counties participated in the
WIN program. The audit sampled the five most significant counties of
the 27, selecting 1100 cases from a list of the State's eligibility
files. The auditors reviewed the cases to determine if recipients who
were not registered for WIN were exempt. The auditors' final report
stated that the auditors found 120 or 10.9% of the recipients sampled
were not exempt. The auditors extrapolated these errors to the universe
of AFDC cases in the five counties.

The auditors made these findings after reviewing the case files and
records at the Separate Administrative Units (SAUs) and IMUs. They also
obtained information from the Department of Labor and discussed each
case they concluded was in error with the caseworker responsible for
that decision. The auditors concluded that the major reason that
nonexempt recipients were not registered was a lack of sufficient
emphasis on this aspect of the program. The Audit Report stated, at
page 6:

The prevailing attitude among caseworkers as well as supervisors was
that since many recipients already registered did not have jobs, it was
not that important to register additional recipients.

The auditors also noted inconsistencies in procedures used by
caseworkers and IMUs, and stated that, in some instances, caseworkers
did not appear to know enough about the registration requirements.

The auditors concluded that because WIN registration is a condition
of eligibility for AFDC benefits, those recipients who were improperly
exempted from registration were ineligible for AFDC benefits otherwise
payable. Thus, the auditors recommended a disallowance of FFP for
payments made to these ineligible recipients.

(4) The Issues

I. Whether Title IV-A Requires Federal Participation in Erroneous
Payments.

The State argued that Title IV-A provides for FFP in erroneous AFDC
payments. The State's position was as follows. Section 403(g)(1) of
the Act provides for reimbursement of expenditures made as part of the
State's AFDC program. The State alleged that, under this statutory
provision, the Department of Health and Human Services (Department)
consistently reimbursed states for all AFDC expenditures during the
first 30 years of the program and did not adopt the policy generally
prohibiting FFP in erroneous payments until the Department began to
adopt Quality Control (QC) sanctions in 1971. Thus, the State
contended, Departmental practice showed that the statute could be
interpreted to allow FFP in erroneous payments. Moreover, the State
argued that section 403(j) of the Act, which was enacted in 1977 and
established incentive payments to states for reducing erroneous
payments, showed that Congress intended federal sharing in erroneous
payments. Finally, the State argued that the only remedy available to
SSA in response to erroneous payments by the State is section 404(a)(
2), which provides that the Secretary may withhold payments under the
program on a prospective basis, where the Secretary finds that in the
administration of the State plan there is a failure to comply
substantially with provisions of section 402(a). Thus, the State argued
that SSA must participate in erroneous payments, except where a state
has failed to substantially comply with the statutory provisions.

This Board has previously concluded that nothing in the Social
Security Act requires the Department to participate in erroneous
payments. California Department of Social Services, Decision No. 319,
June 30, 1982; Maryland Department of Human Resources, Decision No.
358, November 29, 1982. The Board relied on the language of the
statute, which says that reimbursement for state expenditures is an
amount equal to the amount expended "under the State plan." Section
403(a)(1). Moreover, section 403(b)(2) provides that the Secretary
shall reduce or increase the amount estimated by what "should have been
paid to the State." These provisions support the proposition that FFP is
generally not available in payments which do not meet state plan
provisions. This position was also reached by the court in Maryland v.
Mathews, 415 F. Supp. 1206 (D.D.C. 1976). /1/ The State's allegation
here that the Department participated in erroneous payments during the
early years of the program, even if true, should not detract from the
Secretary's general authority to disallow nor prevent the Secretary from
changing policy.


(5) The State also argued that section 403(j), enacted in 1977,
implied that the Department must participate in erroneous payments made
by the states. Section 403(j) provides that if the dollar error rate of
aid furnished by a state is

at least 4%, the amount of FFP in the expenditures made by the state
"shall be determined without regard to the provisions of this
subsection;"

and if the dollar error rate of aid furnished is

less than 4%, the amount of FFP is still determined without regard to
the provisions of section 403(j), but the State receives a certain
percentage of the amount by which the State's expenditures are less than
they would have been with an error rate of 4%.

Thus, the statutory provision sets out a plan whereby those states
with error rates of at least 4% receive only that FFP which the
Secretary would otherwise pay under the statute and regulations. Those
states which maintain an error rate of less than 4% receive the amount
of FFP otherwise paid under the statute and regulations, but also
receive an incentive payment or bonus equal to a percentage of what they
saved by keeping their error rate under 4%. Nothing in the provision
specifies an amount of FFP the Secretary must pay where erroneous
payments have been made, or that the Secretary must actually participate
in any erroneous payments. In fact, section 403(j) specifically
indicates that federal payment for expenditures is to be determined
"without regard to the provisions of this subsection." Thus, incentive
payments are expressly outside the normal federal participation in AFDC
expenditures and must be viewed merely as a Congressional attempt to
encourage the states to reduce their error rates. /2/ Accordingly, we
see nothing in section 403(j) which requires the Secretary to provide
FFP in erroneous payments made by the State.


Finally, the State argued that section 404(a) is the only remedy
available to SSA where erroneous payments have been made. Section 404(
a) provides that the Secretary may withhold federal funds prospectively
where a state on a continuing basis fails to comply substantially with
state plan or statutory provisions. Section 404(a) does not provide for
withholding the amount of unallowable (6) or erroneous costs identified
but provides for general withholding of federal funds in whole or in
part. Thus, section 404(a) contemplates a remedy designed to deal with
a major failure to comply with federal requirements on an ongoing basis.

Nothing in section 404(a) or elsewhere in the Act provides that the
section is an exclusive remedy. Federal courts have uniformly
recognized the availability of the disallowance remedy referred to in
section 1116(d) of the Act. The Third Circuit considered both these
remedies in several decisions, and acknowledged that they are not
mutually exclusive. State of New Jersey v. Department of Health and
Human Services, 670 F.2d 1284 (3rd Cir. 1983); accord, State of
Illinois v. Schweiker, 707 F.2d 273 (7th Cir. 1983). Thus, as this
Board has discussed, the Secretary may disallow for past errors in the
amount of the specific error or withhold funds prospectively until a
state substantially complies, or may exercise both remedies.
Massachusetts Department of Public Welfare, Decision No. 438, May 31,
1983. Limiting the Secretary to a remedy under section 404(a) would
read the disallowance remedy out of the statute and make section 1116(
d) meaningless. See also section 403(b)(2). Commonwealth of
Massachusetts v. Departmental Grant Appeals Board, 698 F.2d 22 (1st Cir.
1983). This would preclude the Department from ever retrospectively
recouping improperly spent funds on a basis directly related to the
errors made, where the error is less than substantial non-compliance or
where a substantial error was not continued. This approach is not
consistent with the statutory scheme, nor does it make good fiscal
policy. It has been consistently rejected by the courts. Therefore, we
conclude that, where a state erred in the past (substantially or not),
the Department may still disallow for the specific unallowable or
erroneous expenditures.

Nevertheless, in support of its position, the State pointed to a case
decided by the U.S. Court of Appeals for the Third Circuit, Shands v.
Tull, 602 F.2d 1156 (3rd Cir. 1979). In that case the court said that
section 403(j) and section 404(a)(2) of the Act together implied an
intent on the part of Congress to hold states to a standard only of
substantial compliance rather than total compliance with the statute and
to make some allowance for the difficulties of administering an
extensive bureaucracy.

We do not believe that Shands v. Tull controls this situation. The
Third Circuit there considered whether the State of New Jersey must
comply totally with a federal regulation which required states to take
final administrative action within 90 days of an appeal from a denial of
benefits under the AFDC program. The case did not involve erroneous
payments and did not discuss the issue of remedies available to the
federal government where expenditures by a state are in question. The
fact that Congress provided a (7) statutory remedy for major on-going
failures on the part of states does not necessarily mean that a state
need only substantially comply with a mandatory statutory requirement to
receive FFP. /3/

Thus, we think it is clear that generally the Secretary has statutory
authority to disallow FFP for erroneous payments made by a state under
the AFDC program, and is not legally required to participate in such
erroneous payments. Here, SSA determined that for the period between
March 1978 and March 1979 the State made certain erroneous payments.
SSA elected to use its disallowance authority under section 1116(d) to
disallow payment for these errors. Nothing in the statute precludes SSA
from doing so.

We now consider what SSA policy applied during the time period in
question for disallowing erroneous payments under AFDC.

II. SSA Policy for Disallowing Payments for Erroneous Determinations
Concerning WIN Registration.

A. Background

In Decision No. 319 and its Reconsideration, the Board concluded that
SSA had a specific disallowance policy for AFDC eligibility
determinations during the period 1977 through 1979. That policy (8) was
that SSA would disallow errors in AFDC eligibility determinations only
when they were identified individually and would not disallow based on
extrapolation from statistical samples until such time as SSA had
established new tolerance levels in accordance with the standards set
out by the court in Maryland v. Mathews, supra. The Board reached this
conclusion about SSA policy after examining the complex background of
the Quality Control (QC) program which applied to AFDC eligibility
determinations, the revocation of the regulation providing for
disallowances under the QC program (45 CFR 205.41), and the language in
the Action Transmittals issued by SSA /4/ explaining disallowance policy
after the revocation. Although SSA argued that the scope and intent of
the policy conceivably could have been interpreted narrowly to cover
only erroneous cases identified in the QC process and not cases
identified in a federal audit, the Board concluded that a broader
interpretation including errors from all sources was the more reasonable
interpretation. The Board also concluded that, even if the policy
issuances were interpreted narrowly, it would have been unreasonable
under the circumstances of the case for SSA to base its disallowance on
extrapolation from a sample. The Board pointed particularly to the
similarity between the sample taken by the audit and a QC sample.
Decision No. 319, p. 7.


The State argued that Decision No. 319 and the Reconsideration should
control this appeal and that SSA should not be able to extrapolate from
the individually identified errors included in the disallowance.

SSA argued that the Board should not continue to rely on the
interpretation of Agency issuances developed in Decision No. 319 and the
Reconsideration, but did not provide any new arguments about why that
interpretation was incorrect. SSA further argued that Decision No. 319
was limited to the particular circumstances of that case, which involved
an audit of the AFDC-Foster Care program over a four-year period and an
extrapolation to the entire universe of the state's AFDC-Foster Care
claims for that period. SSA argued that the Board had relied on the
fact that the AFDC-FC audit was indistinguishable from a QC sample as a
basis for concluding that the use of statistical extrapolation without a
tolerance level was unfair. In this case, SSA argued, the audit was
significantly restricted in geographical area, duration and (9) scope.
The audit was limited to five counties over a one-year period, /5/ the
extrapolation was made only to the cases in the five counties rather
than to the statewide caseload, and the disallowance involved only
errors concerning the failure to register nonexempt AFDC recipients for
the WIN program.


B. Analysis

This Board will normally accept statistical sampling evidence and the
results of extrapolation, where valid, as reliable evidence of the
amount of unallowable costs. See, e.g., University of California -
General Purpose Equipment, Decision No. 118, September 30, 1980.
Moreover, the federal courts have acknowledged and upheld, in cases
arising under the Social Security Act, the use of audit techniques which
involve projection of a relatively small number of components to a large
population. Rosado v. Wyman, 322 F. Supp. 1173, 1180-1183 (E. D.N.Y.
1970), 397 U.S. 397, 419 (1971); Georgia v. Califano, 446 F. Supp.
404, 409 (N.D. Ga. 1977). However, we conclude here, as in Decision No.
319, that SSA had deliberately limited its more general authority to use
statistical sampling and extrapolation techniques as a basis for
disallowing erroneous AFDC payments when identified from any source
including federal audit.

Our reasons are as follows:

* The Department had developed a policy through regulations for
disallowing eligibility errors in the AFDC program as a whole. This
policy was to use samples taken according to specific methods and to
disallow based on extrapolation of the results of these samples, but
only for errors above prescribed tolerance levels. The Department
stated that the tolerance levels were based on a recognition that, under
the administrative structures then existing in most states, "a
requirement . . . that states eliminate all erroneous payments, with a
resultant disallowance of Federal financial participation in any
erroneous payments is unrealistic." 40 Fed. Reg. 21737, May 19, 1975.

* SSA has not denied that determinations about exemption from the WIN
registration requirement fall within the scope of the policy. As noted
above, Title IV-A of the Act provides, in section 402(a)(19)(A), that
registration for the WIN program is a condition of eligibility for AFDC
benefits, and the regulations define "case error" as an overpayment,
underpayment, or payment for ineligibility. 45 CFR 205.40(a)(2) (1975).

(10) * Following litigation which struck down the tolerance levels as
arbitrarily established, the Department revoked its regulations and in
an accompanying notice stated that, in the absence of tolerances, the
policy for disallowing erroneous payments for AFDC eligibility
determinations would be for "individually identified" errors. 42 Fed.
Reg. 14717, March 16, 1977. As we noted in our Reconsideration, p. 3,
in a historical context this clearly meant error identification not
based on extrapolation from a sample.

* Following revocation of the regulations, the Agency clarified the
scope of the disallowance policy for the period at issue here through
Action Transmittals. AT-77-30, March 16, 1977; AT-77-55, May 17, 1977.
The phrase "through Federal reviews" used in these documents can
reasonably be read to mean more than federal review of a subsample of a
QC sample, and indeed, in the later Action Transmittal, SSA-AT-78-16,
May 4, 1978, the Agency clarified that the fiscal adjustment policy in
AT-77-55 applied to "all cases in which errors in payment are identified
from any source. . . ." p. 2 (emphasis in original).

The latter statement clearly does not limit the policy to errors
identified through QC procedures.

* While the Agency's Action Transmittals may have been primarily
concerned with how the states should adjust their claims for errors
identified through the QC system, they also clearly deal with a change
in disallowance policy for eligibility errors in the AFDC program. The
possibility of disallowance is the force behind the requirement to
adjust. Both the revocation notice and AT-77-30 state that, because of
the revocation of the disallowance provisions in section 205.41, the
"Department is now required to disallow" federal financial participation
in all individual incorrect payments.

Thus, we think that it is reasonable to conclude from the Agency
issuances applicable to the particular dispute in question that there
was a policy to disallow WIN registration determination errors only when
individually identified and not when extrapolated from a sample.

SSA nevertheless argued that there are significant differences
between the type of sample in which these errors were identified and the
sample used in the appeal involved in Decision No. 319, and that the
policy relied upon in Decision No. 319 need not control this case as a
result.

(11) We disagree that the differences are significant. Even though
the audit in this case involved only a one-year period and examined only
one type of determination affecting AFDC eligibility, neither of these
limitations provide a substantive basis to distinguish the Board's
previous decision.

SSA had developed a policy for disallowing AFDC eligibility
determination errors based on samples, and this policy contained
safeguards for the states, including allowance of tolerance levels for
errors when extrapolation was used. Under these circumstances, we do
not think it is reasonable for SSA to apply instead a separate policy
for case errors identified outside the QC process where the policy for
case errors identified outside the QC process where the policy remains
unarticulated in any Agency publication. Although we believe that SSA
had statutory authority to institute a separate disallowance policy for
AFDC case errors identified through federal audit, we are here faced
with the absence of any written policy to that effect. Under such
circumstances, it is impossible for the Board or the states to clearly
draw a line between the scope and effect of any such purported policy
and the QC policy with its inherent protections of either tolerances or
individual case identification.

Further, we do not see why it would be any less unreasonable or
unfair to extrapolate without a tolerance when only one aspect of the
AFDC eligibility determination is examined. If anything, there may be
greater unfairness because the State does not have the opportunity to
offset weaknesses in one area of its program administration (e.g.
monitoring WIN registration) against strengths in its performance
elsewhere so as to achieve a program-wide error rate below tolerance
levels.

In conclusion, a logical reading and reasonable application of SSA's
published policies argues that SSA narrowed its disallowance authority
in the area of AFDC eligibility determinations. We conclude that
determinations about the exemption of certain AFDC recipients from the
requirement to register for the WIN program fall within the scope of
that policy and that for the time period involved here, SSA may not
extrapolate to the universe of WIN cases as it proposed. /6/


(12) Although we reverse the disallowance to the extent of the amount
of extrapolation, we have concluded that SSA may disallow for
individually identified errors. Here the State challenged the validity
of the auditors' findings for some of the identified errors. Therefore,
we now discuss the auditors' findings.

III. The Validity of the Auditors' Findings.

The State contended that the audit workpapers do not support the
auditors' findings of ineligibility under the standards set out by the
General Accounting Office (GAO). These standards require that auditors
obtain "sufficient, competent and relevant evidence . . . to afford a
reasonable basis for the auditors' opinions, judgments, conclusions, and
recommendations." Standards for Audits of Governmental Organizations,
Programs, Activities and Functions, GAO, 1972, p. 7. The standards also
provide that the workpapers should contain the supporting evidence. At
p. 37. The State's primary argument is that the auditors had no
evidence that the recipients were actually nonexempt from the WIN
registration requirement, but reached their conclusions because the case
files lacked documentation supporting the caseworkers' conclusions that
the recipients were exempt. The State also alleged that the auditors
demanded an "arbitrary and unrealistic level of documentation" not
required by federal regulations. Principal Brief, p. 13.

SSA pointed out that a standard which applies to all types of audits,
such as the GAO standard, must be interpreted in light of the program
and issues being examined. SSA argued that in this case the State had a
burden to document that recipients were, in fact, exempt and that the
State failed to do so when the case files contained only the
caseworkers' unexplained conclusions that a person was exempt for health
or medical reasons.

In our discussion below we conclude that the federal auditors met the
standards set out by GAO because the auditors' workpapers contained
evidence that the State had not met its burden of documenting that the
recipients were exempt. Although the workpapers may not have contained
sufficient evidence that the recipients were in fact nonexempt, that is
not necessary as a basis for a conclusion that the costs are
unallowable.

This Board has consistently concluded that the burden to document
claims rests with the grantee. New York Department of Social Services,
Decision No. 204, August 7, 1981; New Jersey Department of Human
Services, Decision No. 154, March 18, 1981. The GAO standards do not
alter this basic requirement. The auditors examined the case files to
determine whether the federal standards for exemption from registration
were met and documented. If they (13) did not find evidence that the
State made the determinations in accordance with federal regulations,
they could reasonably conclude that the recipient was not exempt and,
therefore, the claimed costs were unallowable.

Federal regulations require that AFDC eligibility decisions be
supported by "facts in the . . . recipient's case record." 45 CFR
206.10(a)(8). In addition to this general requirement, federal
regulations set out certain criteria for establishing the "facts"
supporting exemption from the WIN registration requirement. 45 CFR
224.20(b) provides for the following exemptions:

* * *

(2) Regularly attending school and age 16 but not yet 21 years of
age. . . . There must be verification according to State welfare
regulations that the person is enrolled or has been accepted for
enrollmment for the next school term as a full-time student . . .;

(3) Ill, when determined by the IMU on the basis of medical evidence
or on another sound basis that the illness or injury is serious enough
to temporarily prevent entry into employment or training;

(4) Incapacitated, when verified by the IMU that a physical or mental
impairment, determined by a physician or licensed or certified
psychologist, by itself or in conjunction with age, prevents the
individual from engaging in employment or training under WIN;

* * *

(7) A person in the home, and the IMU has verified that a physical or
mental impairment, as determined by a physician or licensed or certified
psychologist, of another member of the household requires the
individual's presence in the home on a substantially continuous basis,
and that no other appropriate member of the household is available;

(8) A mother or other caretaker relative of a child under age 6;

* * *

(14) Thus, federal regulations clearly require not only that certain
circumstances exist but that they be verified and supported by evidence.
The burden is on the State to show that it has met these requirements.
Although the State contended that a caseworker's conclusion that a
recipient is exempt is a "fact," we do not agree.

The State argued that if documentation is required for the cases
granted exemption on the basis of "medical" or "health" reasons, only a
sound basis is required and not a physician's certificate. It is true
that the regulations do not require that a physician certify that a
recipient is temporarily ill. /7/ However, the regulations do require
that the exemption for temporary illness be given on the basis of
"medical evidence or on another sound basis." Since caseworkers are not
physicians and are not qualified to make medical judgments, they must
have some evidence that the recipient's temporary illness is serious
enough to prevent entry into employment. A physician's certificate or
some proof of illness would be acceptable evidence. The possibility
exists that the State could have met its burden if there were a notation
in the case file that a caseworker's decision was based on having seen a
physician's certificate or equivalent evidence; however, the State has
not even alleged that such statements were contained in the case files.
Moreover, the regulations require that IMUs verify that the
circumstances leading to the exemption existed (i.e., were a fact).
Since IMUs must be able to verify such facts from the case files or from
other appropriate sources, the implication is that in order for the
State to meet the regulations, proof of the facts must be in the case
files so that the IMUs can verify them. Thus, a caseworker's conclusion
without more would not be acceptable evidence. If the State had
submitted some proof of serious illness, we might have been able to
conclude that the federal regulations had been met. However, the State
has not rebutted the conclusions that there was no evidence to support
an exemption.


Of the 120 cases identified by the auditors as erroneous, the State
challenged 29 on the basis that the auditors' workpapers did not contain
sufficient evidence to support their conclusions. In 20 of these cases
the State submitted only a copy of the auditors' workpapers.
Examination of these workpapers shows that the auditors concluded that
the recipients in question were not exempt because (15) evidence of
facts supporting exemptions was not in the case files. The State did
not submit any documentation to rebut the auditors' conclusions.

In nine of the 29 cases the State submitted some form of additional
evidence to support its assertion that the recipients were in fact
exempt. We briefly discuss these cases in Appendix A. In six of these
cases we conclude that the auditors' findings should be upheld. For the
case in Exhibit 11, the amount of overpayment should be recalculated.
In Exhibit 96, the State submitted a physician's certificate for which
there is no patient name. If the State can show that the certificate is
indeed for the patient exempted, the finding may be reversed. The case
in Exhibit 105 shows information on the caseworker's form which is
sufficient evidence of exemption. Thus, of the 29 cases challenged by
the State, we uphold 26, tentatively uphold another, reverse one case
and another in part. See Appendix A.

The State raised a number of issues about the recalculation of the
specific overpayment amounts involved. We address these below.

IV. Issues about Recalculation.

Inconsistent Methods of Computing Overpayments

The State asserted that the auditors had incorrectly computed the
amount of overpayment in five cases and SSA agreed. Respondent's Brief,
p. 12. When SSA recalculates the amount of disallowance, it should
correct the computation of these amounts.

Exclusion of One Case from the Sample

The State asserted that one case should be excluded from the sample
because the overpayment occurred in a non-audited county. SSA contended
that the case was on the eligibility rolls at the time the sample was
made and the later transfer of the case to another county is irrelevant.
Since we are now only concerned with individually identified errors and
not extrapolation from a sample, the validity of the sample for
statistical purposes is not in question. The case was identified as an
error, and the State has not refuted the accuracy of that finding.
Therefore, the case should be included in the disallowance.

Proration of One Overpayment

In one case a payment was made to three family members, two of whom
were eligible. All but $231.40 of the assistance paid was voluntarily
refunded by the family. The auditors found that the $231.40 was also an
overpayment. The State contended that not (16) all of the $231.40 could
be an overpayment if some family members were eligible. SSA responded
simply that it did not understand this argument. The State has not
provided us with any documentation demonstrating that the amount in
question was in fact payment for an eligible family member and not an
overpayment on some other basis. Therefore, we sustain the disallowance
for this amount.

Overpayments Computed According to Action Transmittal SSA-AT-78-16

SSA explained its policy about how to compute overpayments resulting
from a change in client circumstances in Action Transmittal
SSA-AT-78-16. The Action Transmittal requires states to make fiscal
adjustments for erroneous payments that they could have prevented.
Adjustments are effective beginning the second month after the month in
which the change took place. The State alleged that the auditors did
not apply this policy to 42 cases; instead, the auditors computed the
amount of overpayment from the time the change took place.

SSA responded that although SSA-AT-78-16 had not been issued at the
time the auditors calculated the overpayments, the policy was in effect
at the time the disallowance was taken. Therefore, SSA agreed that some
of these overpayments would need to be recomputed in accordance with
SSA-AT-78-16. SSA pointed out that the State had the information
necessary to make the recalculations and therefore, it should do the
computations, documenting them in order to allow verification for
accuracy. If the calculations have not already been performed, the
State should make them and submit them to SSA within 30 days of the
receipt of this decision so that they can be included in the
recalculation of the disallowance.

Recovery of Child Support Payments

The State contended that the amount of the disallowance did not take
into account repayments which may have occurred through the Title IV-D
(Child Support Enforcement) program. Under that program, an AFDC
recipient assigns her child support payments to the State and the State
then obtains court orders to recover unpaid support. Federal
regulations set out how recovered payments are to be allocated. 45 CFR
302.51 and 302.52. A certain portion of the child support is returned
to the family and this amount reduces the amount of AFDC assistance
received. 45 CFR 232.20. The State acknowledged that the auditors took
these repayments into account and reduced the amount of overpayment by
those amounts. Principal Brief, p. 28. Therefore, these repayments are
not in issue here.

The remaining portion of the recovered child support payments are
allocated by providing incentive payments under the Child Support
Enforcement Program (45 CFR 302.52) and by returning some portion (17)
to the AFDC program. Section 302.51 indicates that the Child Support
Enforcement program is supposed to calculate the federal share of any
funds returned to the AFDC program. The State argued that this was also
repayment of federal funds and should be taken into account. SSA
contended that this portion of child support recovery was not
appropriately offset against AFDC assistance payments.

There is no evidence in the record that this portion of the recovered
child support payments would offset AFDC assistance payments. This
disallowance concerns only overpayment of assistance and not any other
type of Title IV-A costs. The State has admitted that repayment related
to overpayments in assistance have already been taken into account. If
the AFDC program received a portion of the recovered child support
payments which it used, for example, as administrative costs, that
repayment would have been reflected in other portions of the Title IV-A
program. There is no need to consider it a duplicate payment with
regard to a disallowance of overpayment of assistance. Therefore, we
agree with SSA that other repayments to the AFDC program are not
appropriately considered repayment of assistance overpayments and need
not be taken into account here.

Two Cases for Which SSA Has Accepted Evidence that the Recipients Were
Exempt

The State submitted evidence for a small number of cases, not
included in the group of 29 discussed above, allegedly showing that the
recipients were in fact exempt from the requirement to register for WIN.
SSA agreed to accept the evidence for two of these cases. See Letter of
October 31, 1983. Therefore, the disallowance should be reduced by the
amount of overpayment allocated to these two cases.

Conclusion

For the reasons stated above, we conclude that the disallowance
should be upheld in the amount of the overpayments calculated for the
individually identified errors only.

With regard to the recalculation of the disallowance based on
overpayments for individually identified cases, the following
calculations should be made:

* The number of cases for which identified errors are sustained is
reduced from 120 to 117. This reflects a reversal of two cases for
which SSA accepted evidence submitted by the State directly to SSA, and
one case referred to as Exhibit 105. See Appendix A.

(18) * If the State can show that the physician's certificate
submitted in Exhibit 96 is for the recipient involved in that case, the
disallowance should be reduced to 116 cases. See Appendix A.

* The amount of the overpayments calculated by the auditors for the
case referred to in Exhibit 11 should be recalculated. See Appendix A.

* The disallowance should be recalculated taking into account
inconsistent methods of computation used by the auditors, and
recomputation using the methods set out in SSA-AT-78-16.

The State should submit all appropriate calculations and information
to SSA within 30 days of receipt of this decision, and SSA should then
notify the State of the revised amount of disallowance, taking into
account the conclusions reached in this decision.

(1) APPENDIX A

FINDINGS FOR NINE CASES FOR WHICH STATE REBUTTED THE SUFFICIENCY OF
AUDITORS' FINDINGS

State's Exhibit 11: This overpayment involved a mother and daughter.
The State submitted caseworker notes from one year after the period in
question. The caseworker notes showed that the caseworker verified in
late April of 1980 that the daughter had graduated from high school in
June 1979. We can conclude, therefore, that the daughter was enrolled
in school during the period for which the auditors determined that she
should have been registered for the WIN program. Although the
caseworker apparently did not verify this fact at the time the exemption
was given, we think that the caseworker notes are sufficient to show
that the daughter was exempt. Therefore, we believe that a
recalculation of the overpayment relating to the daughter should be
made.

However, the caseworker notes for May 1980 simply indicate that the
caseworker received a doctor's letter regarding the health and
employability of the mother. The notes do not indicate to what period
this letter pertained, nor what the letter indicated. Therefore, we
cannot accept the caseworker's notes about the doctor's letter as
conclusive evidence that the mother was exempt.

State's Exhibit 14: The State submitted certifications showing that
the recipient's son needed psychiatric care. Some of the certifications
showed that he had been admitted to a hospital, and the others certified
that he needed care in a mental institution. The recipient was exempted
on the basis that she was needed to care for her son at home "on a
substantially continuous basis"; however, these certifications, while
indicating that the son was mentally impaired, also show that he may
have been cared for in an institution. Therefore, the evidence is
inconclusive that the son was at home, and we cannot find that the
recipient was correctly exempted.

(2) State's Exhibit 32: The wife was exempted for reasons of "ill
health." The State submitted only the caseworker's form, but no evidence
showing the nature of the illness or why it prevented her from deing
employed. Thus, we cannot conclude that this recipient ws exempt.

State's Exhibit 74: The recipient was exempted on the basis of
temporary illness in September 1978. A physician certified that she was
recovering from an injury and that he anticipated three to four months
of treatment. In March 1979, seven months later, the caseworker
indicated that the recipient was still under a doctor's care, but did
not note any evidence that the injury was still serious enough to merit
a continued temporary exemption. Thus, we do not find evidence to
warrant continued exemption on a temporary basis.

State's Exhibit 89: Two children over the age of 16 were exempted on
the basis that they were attending school. The State included only the
caseworker's form and nothing showing verification of school enrollment.
The auditors' conclusion is justified.

State's Exhibit 96: The recipient was exempted for "medical
reasons". The auditors concluded that the recipient was not exempt
because there was no physician's certificate. The State submitted a
physician's certificate, but the patient's name is not on the form. If
the State can show that the certificate is for the recipient, the
auditors' conclusions can be reversed.

State's Exhibit 98: This recipient was exempted for "medical
reasons." The evidence submitted shows that the recipient was treated
for diabetes and that a child over the age of 6 was treated for asthma.
There is no evidence that indicates that the recipient's diabetes
rendered her unable to work, nor did the caseworker exempt the recipient
in order to care for the child, who may or may not have suffered from
asthma to such a degree that the mother's presence was required on a
"substantially continuous basis." Thus, the auditors conclusions are
justified.

(3) State's Exhibit 100: The caseworker exempted two recipients
because they were attending school, but the auditors indicated that the
case files contained no verification of enrollment. The State did not
submit evidence that it veriufied enrollment, as required by federal
regulations. Thus, we cannot conclude that these recipients were
exempt.

State's Exhibit 105: The auditors indicated that a decision about
the WIN status was omitted, but that the IMU agreed with them that WIN
registration was mandatory for the recipient. The State submitted the
caseworker's form, which shows that the recipient had a child under the
age of 6. This would appear sufficient to exempt the recipient and we
reverse the auditors' findings for this recipient. /1/ The same
conclusion was reached by the Ninth Circuit in a case involving
similar language in Title XIX of the Act. California v. Settle, No.
82-4481 (9th Cir. June 17, 1983). /2/ The legislative history of
section 403(j) shows that "for at least the last 25 years there has been
recognition at the Federal level of the need for a program to reduce
errors in the Federal-State public assistance programs." S. Rep. 95-572,
95th Cong., 1st Sess. 42 (1977). /3/ In an argument focusing on
the propriety of the WIN registration requirement, the State contended
that it was a "paper" requirement, and that compliance with the
requirement would not have resulted in any more persons participating in
a job-oriented activity. The State also argued that the requirement was
meaningless because Congress and the Agency had failed to appropriate
sufficient funds to implement the employment-related services portion of
the program. The State contended that this insufficient funding
justified the State's noncompliance with the requirement. The
requirement to register, however, is clearly provided for in Title IV-A
and is a mandatory condition of eligibility. Thus, we do not believe it
is necessary to consider whether breach of that requirement is justified
by the level of funding for the WIN program. The State also made a
related argument that the State's failure to register recipients was not
a material breach of the contract between the State and the federal
government. However, the State offered no arguments about why it was
not a material breach or why we should even consider such a defense.
SSA argued that there is no ambiguity in the Act about the requirement
to register and that the concept of material breach is not relevant to
this situation. Without further elaboration by the State about its
position, we do not believe we need address the issue further.
/4/ Our references to SSA and Agency here also include its predecessor
in administering Title IV-A, the Social and Rehabilitation Service
(SRS). /5/ The State pointed out that the five counties sampled
represented 86% of the AFDC caseload. Reply Brief, p. 14. /6/
SSA argued that the errors identified in this disallowance were due to
systemic error on the part of the State and that the Board, in Decision
No. 319, had indicated that extrapolation might be appropriate where
systemic errors were made. SSA pointed out that the auditors had
concluded that the errors were made because the State failed to
emphasize the WIN registration requirement. In the Reconsideration, p.
7, we stated that we could not conclude there was evidence of systemic
error simply because some social workers did not provide full
information to eligibility workers. Here, the evidence shows that some
caseworkers did not include in the case files evidence supporting their
decisions to exempt recipients and that the State did not verify the
existence of the circumstances leading to the exemptions. We also
conclude here, however, that this is not enough to show that there was a
systemic failure in the State's administrative processes. /7/
Indeed, in most of these cases it was impossible to determine whether
the recipient was exempted on the basis of temporary illness or for
incapacitation. (Incapacitation requires that a physician certify the
existence of the condition.)

NOVEMBER 14, 1984