DEPARTMENTAL APPEALS BOARD
Department of Health and Human Services
SUBJECT: Missouri Department DATE: February 21 1989 of
Social Services Docket No. 87-188 Decision No. 1021
DECISION
The Missouri Department of Social Services (State) appealed a
disallowance by the Administration for Children, Youth and Families
(ACYF, Agency) of federal financial participation (FFP) claimed under
Title IV-E (Foster Care) of the Social Security Act (Act). ACYF
disallowed $5,511,265 in FFP for fiscal years (FYs) 1983 and 1984 for
claims for Title IV-E administration and training costs because the
costs were claimed under a Cost Allocation Plan (CAP) amendment that did
not become effective until July 1, 1984 and because the costs were
determined by an improper estimate that was not sanctioned by the
amended CAP.
In order for the State to receive the FFP in question, it must show both
that its CAP amendment, which became effective July 1, 1984, can be
given retroactive application to January 1, 1983 and that the amendment
authorized the specific methodology that the State used in computing the
claim. For reasons described below, we uphold the disallowance in full.
The State's CAP amendment became effective July 1, 1984 (consistent with
what the State itself originally proposed) and cannot support the claims
in question here since they arise from an earlier period. Furthermore,
the applicable regulations would not authorize an earlier effective date
for the State's CAP amendment under the circumstances here. Finally,
even if the State's CAP had been effective January 1, 1983 as the State
now proposes, the methodology the State used here in computing its claim
is not authorized by the amended CAP provisions and could not reasonably
be expected to reach the same result as those provisions.
Procedural Background.
This disallowance is part of a process that began when the State,
already having an approved CAP for administrative and training costs
associated with the provision of Title IV-E services, submitted, on
September 25, 1984, an amendment to its CAP to the Region VII Division
of Cost Allocation (DCA) of the Department of Health and Human Services
( Department, DHHS). The proposed CAP amendment concerned the time
study coding of certain administrative activities performed by social
workers so that the activities could be charged to the IV-E program.
While the proposed CAP amendment was being reviewed by DCA, the State,
realizing that the two-year deadline for filing claims imposed by
section 1132(a) of the Act was approaching, submitted in early 1985
claims for IV-E administrative and training services for the quarters
ending March 31, 1983 through December 31, 1984. The claims for the
period January 1, 1983 through September 30, 1984 were based on a sample
using random day logs the State took from August 15 through September
28, 1984. Among the IV-E services claimed by the State were those that
would have been allowable under the proposed CAP amendment.
On May 29, 1985, the Director, DCA, rejected the proposed amendment,
finding that the definitions used for the time study codes charged to
the IV-E program included unallowable social services which should be
allocated to other programs under the Act. While the DHHS Regional
Director was reviewing the DCA's rejection of the original CAP
amendment, the State submitted a revised amendment to the DCA Director,
who approved it with revised time study codes; this plan amendment was
given the same effective date as the effective date of the original
proposed CAP amendment, July 1, 1984.
When the DHHS Regional Director sustained the rejection of the original
CAP amendment, the State appealed that disapproval to this Board. In
Missouri Dept. of Social Services, DAB No. 844 (1987), the Board held
that the activities contemplated by the original amendment, if properly
defined by the State, could be reimbursable under the IV-E program as
administrative costs.
Before the Board had issued DAB No. 844, ACYF disallowed the claims
submitted by the State because the proposed CAP amendment had not been
approved, finding that under 45 C.F.R. 95.519 claims for FFP must be in
accordance with an approved CAP.
The State then appealed that ACYF disallowance determination to the
Board. That appeal raised the following issues: whether the State
could revoke its election to transfer IV-E funds to the Title IV-B
program; whether the State had documentation to support its claims for
FYs 1983-1985; and the effective date of the original CAP amendment
(which the Board found acceptable in the DAB No. 844 during the course
of the appeal of the disallowance). With the agreement of the parties,
the Board delayed its consideration of the latter two issues to decide
the question of the State's ability to revoke its election to transfer
funds. The parties agreed that a Board decision on that issue in ACYF's
favor would render moot the other two issues. The Board subsequently
found, however, in Missouri Dept. of Social Services, DAB No. 902
(1987), that the State could revoke a prior election to transfer IV-E
funds to its IV-B program in order to avoid a funding ceiling.
Consequently, the two issues left unresolved by DAB No. 902 -- the
effective date of the CAP amendment and the reliability of the data
acquired from August 15 to September 28, 1984 to support claims for
prior quarters --remain to be decided. After settlement discussions
between the parties proved unsuccessful, the parties submitted briefs
and participated in a hearing on these issues.
Discussion
I. The State's claim was based on an allocation methodology which was
not authorized in its cost allocation plan.
The short answer to the State's claim for the period January 1, 1983
through July 1, 1984 is that it was not computed based on a methodology
in an approved cost allocation plan. In its claim for that period, the
State attempted to apply a methodology set out in an amendment to the
State's cost allocation plan that became effective July 1, 1984. Since
that amendment was not in effect for the period covered by the claim,
the State was not authorized to use the methodology, and its claim must
be disallowed. Moreover, the regulations do not authorize the Agency to
approve an amendment to the cost allocation plan with an earlier
effective date under the circumstances here.
A. The Cost Allocation Plan Process
A state participating in the various categorical programs under the Act,
including Title IV-E, is required to make determinations as to the
amount of expenditures benefitting more than one program, such as staff
time, that are allocable to each program the state administers. A state
is required to submit a plan for cost allocation to the Director, DCA,
in the appropriate DHHS regional office. 45 C.F.R. 95.507(a). This
cost allocation plan is defined as "a narrative description of the
procedures that the State agency will use in identifying, measuring, and
allocating all State agency costs incurred in support of all programs
administered by the State agency." 45 C.F.R. 95.505. The CAP must
contain sufficient information to permit the DCA Director to make an
informed judgment on the correctness and fairness of the state's
procedures for identifying, measuring, and allocating all costs to each
of the programs administered by the state agency. 45 C.F.R.
95.507(a)(4). Finally, 45 C.F.R. 95.517 provides that a state must
claim FFP for costs associated with a program only in accordance with
its approved cost allocation plan.
The general rule establishing the effective date of a CAP amendment is
set forth at 45 C.F.R 95.515:
[T]he effective date of a cost allocation plan amendment shall
be the first day of the calendar quarter following the date of
the event that required the amendment (See section 95.509).
Thus, this regulation clearly indicates the prospective nature of a CAP
amendment. A CAP amendment may entail significant changes in how costs
are measured, recorded, and allocated to different programs, with these
affecting the amount of FFP a state receives under each of the programs.
The prospective natures of an amendment facilitates the Department's
ability to budget and allocate its resources among the various programs.
Establishing program costs is complicated when a state seeks to have a
CAP amendment made effective to an earlier date. A retroactive CAP
forces the parties to rely on data that inherently is less reliable than
a prospective CAP. Instead of information being gathered concurrently
as the expenditures are being made, in a retrospective plan data must be
retrieved from sources in the past. It is reasonable therefore to infer
that CAP is to be given retroactive effect in only exceptional
circumstances and that the exceptions to the general rule of section
95.519 should accordingly be applied sparingly.
Yet section 95.515 continues, however, to list limited circumstances
where the effective date of an amendment may be earlier:
(a) An earlier date is needed to avoid a significant
inequity to either the State or the Federal Government.
(b) The information provided by the State which was used to
approve a previous plan or plan amendment is later found to
be materially incomplete or inaccurate, or the previously
approved plan is later found to violate a Federal statute or
regulation. In either situation, the effective date of any
required modification to the plan will be the same as the
effective date of the plan or plan amendment that contained
the defect.
The fact that these exceptions are to be granted reluctantly is
illustrated by the fact that the regulation specifically limits the type
of events, listed at section 95.509, that may trigger the submission of
a CAP amendment.
These events are:
-- procedures in the existing CAP become outdated because of
organizational changes, changes in federal law or regulation, or
significant changes in program levels;
-- a material defect is discovered in the CAP;
-- a state's plan for public assistance is amended so as to
affect the allocation of costs; or
-- other changes occur which make the CAP invalid.
B. Analysis.
In support of its position that the CAP amendment's effective date
should be January 1, 1983 and not July 1, 1984, the State argued that
the above exceptions are relevant to this appeal. The State contended
that if the CAP amendment were found not to be retroactive to January 1,
1983, the State would suffer a significant inequity in being denied
millions of dollars to which it was undisputedly entitled under the Act.
According to the State, its original CAP was materially incomplete
because it failed to allocate allowable IV-E administrative and training
costs to the IV-E program. The State claimed that the discovery that
the original CAP did not fail identify and allocate all IV-E
administrative and training costs prompted the State to seek a CAP
amendment. The State pointed out that the Act, at section 474(a),
expressly provides that a state shall be entitled to federal
reimbursement for expenditures incurred in the administration of the
IV-E program; furthermore, the Act, at section 1132(a), permits a state
a two-year period to identify and claim FFP for IV-E expenditures.
Accordingly, the State reasoned that the combination of these two
provisions gave it the right, when it filed its claims in 1985, to have
the defects in the CAP corrected to an effective date of January 1,
1983.
The Agency's position is that the regulations do not authorize an
effective date for the CAP amendment earlier than July 1, 1984 because
none of the exceptions listed at 45 C.F.R. 95.519 allowing for an
earlier effective date are applicable under the circumstances of this
appeal. The Agency asserted that the State's original CAP was not
defective in any way and the mere fact that the State did not claim as
much FFP as it could have does not constitute a "substantial inequity"
allowing an earlier date for the CAP amendment.
The primary basis for the Agency's disallowance of the claimed costs is
that they were based on a CAP methodology that was not in effect for the
period prior to July 1, 1984. It is undisputed that the CAP amendment
in question only became effective on July 1, 1984. Consequently, this
amendment cannot serve as the basis for the State's claim here, at least
insofar as it covers the period prior to July 1, 1984. Section 95.517
described above clearly only authorize FFP for costs claimed in
accordance with an approved CAP. This regulation alone, therefore,
authorizes the disallowance for the period prior to July 1, 1984.
Furthermore, the regulations concerning the effective date of a CAP
would not have authorized the Agency to give the State an earlier
effective date for its amendment under the circumstances here even if
the State had made such a proposal. (The record here does not indicate
that the State ever proposed to the Agency that its CAP amendment be
effective January 1, 1983.)
We consider it evident that the regulations do not contemplate a
situation like the one before us where a state, quite openly, is seeking
retroactive effect of an amendment just to increase its FFP. No
triggering event listed in section 95.509 is present here. Under the
framework of the regulations, where a state is merely seeking more FFP
through a CAP amendment, that amendment can only be given prospective
effect. Otherwise, the exception to the general rule of section 95.515
would engulf the rule itself since virtually every amendment to a CAP
would presumably affect the amount of FFP the State received.
Therefore, as we discuss below, the exceptions as exceptions should be
granted in only extraordinary circumstances, for example, circumstances
beyond a state's control such a change in law or regulation.
Here it is undisputed that the State sought a CAP amendment because it
discovered that under its original CAP it was not claiming all its IV-E
administrative and training costs. Hovis Affidavit, paragraph 4, State
Appeal File (SAF) in Docket No. 86-136, Exhibit (Ex.) G. The question
then is whether 45 C.F.R. 95.519 permits the CAP amendment to be
effective earlier because of this discovery. Under the State's view,
the original CAP's failure to identify and allocate IV-E costs showed
that the CAP was "materially incomplete or inaccurate" (45 C.F.R.
95.515(b)) and resulted in a "substantial inequity" (45 C.F.R.
95.515(a)); thus the CAP was materially defective so as to warrant the
amendment's effective date being set back to January 1, 1983.
We are not prepared to interpret section 95.515 as broadly as the State
suggests. As noted above, the section's general rule is that the
effective date of a CAP amendment "shall be the first day of the quarter
following the date of the event that required the amendment."
Acceptable "events" are listed in section 95.509(a). The only "event"
that could be related to the facts of this case is subsection (2), the
State's allegation that it discovered a material defect in the original
CAP.
Apart from the possible failure of the original CAP to claim all IV-E
administrative and training costs incurred by the State, the State has
not challenged the Agency's assertion that the original CAP was fully in
accord with the Act and applicable regulations. Section 95.515 speaks
of an event that required the submission of an amendment. No change in
the Act, the regulations, or Agency policy required the State to submit
a CAP amendment. As the Agency pointed out, the submission of the
amendment was a voluntary action by the State. Under these
circumstances, we find that the State's discovery of a failure in its
CAP to claim every possible IV-E administrative and training cost is not
the type of event in section 95.509 that would trigger one of the
exceptions in section 95.515.
The loss of FFP, even here when it may reach into the millions of
dollars, does not in itself necessarily constitute a "substantial
inequity." If the State's interpretation of section 95.519 that the
mere failure to receive all available FFP constitutes an inequity were
to prevail, the implication would be that any state would be able to
have an amendment to its CAP made retroactive if it discovered that it
was not claiming all possible FFP under its original CAP. Such an
outcome would create chaos in the Department's ability to effectively
administer the programs under the Act and to plan budgets for the costs
of the program. When Congress created the two-year filing deadline in
section 1132(a) of the Act, it did not do so with the intention of
allowing states to retroactively amend their CAPs; the purpose of
section 1132(a) was to prevent states from coming in many years after
expenditures were made and claiming FFP and to set a time limit on
states' claims so that the Department could determine its obligations
towards the states under the Act and budget its funds accordingly. See
New York State Dept. of Social Services, DAB No. 521 (1984), p. 8.
Furthermore, while the State is correct that section 474(a) of the Act
declares that a State shall be entitled to federal reimbursement for
IV-E expenditures, that provision is limited by section 471's
requirement that any such expenditures must be made in accordance with
an approved CAP. The State did have an approved CAP for the period in
question, it made expenditures and submitted claims, and the Agency paid
FFP. It is only the amended claims, made according to untimely CAP
amendment, that the Agency rejected. Section 474(a) does not require
that the Agency pay FFP for all the claims that the State could have
made under a more artfully worded CAP. Furthermore, the mere assertion
of a claim does not qualify the claim for FFP; the claim must be
substantiated. We find no support that section 474(a) of the Act
requires approval of a retroactive CAP amendment. As discussed below in
Part III, the State has to be able to establish, through an accepted
methodology, that it incurred costs attributable to the IV-E program.
We additionally note that the State submitted the CAP amendment to DCA
in September 25, 1984, with a proposed effective date of July 1, 1984;
the revised CAP approved by DCA also bore an effective date of July 1,
1984. Thus, at the time it submitted and resubmitted the amendment, the
State did not even contemplate the amendment having effect prior to July
1, 1984.
The Board cases cited by the State for the proposition that a
significant inequity can result in an earlier effective date of an
amendment can be readily distinguished from the facts of this case. In
Iowa Dept. of Human Services, DAB No. 624 (1985), the Board declared
that the Health Care Financing Administration's (HCFA's) refusal to
grant retroactive approval to Iowa's amendment of its CAP for allocating
Medicaid costs imposed the type of inequity upon Iowa which section
95.515 seeks to avoid. In that case, however, the Board found that the
circumstances which led Iowa to fail to include a methodology for
allocating Medicaid costs in its CAP were caused in part by HCFA's
failure to issue guidelines to Iowa on the scope of the particular costs
in question. Here there was no action or inaction on the Agency's part
that caused the State not to include all possible IV-E administrative
and training costs in its original CAP. In Oklahoma Dept. of Human
Services, DAB No. 963 (1988), the Board examined whether there were
circumstances sufficiently compelling to warrant retroactively charging
Oklahoma's CAP's methodology for claiming various IV-E costs, finding
that a CAP could not be binding upon the parties if the CAP conflicted
with the Act or regulations. There is nothing in the facts of this
appeal to warrant a finding that the original CAP was in conflict with
the Act or applicable regulations.
In conclusion, we find that the State has not presented any convincing
arguments that the exceptions listed at section 95.515 allowing an
earlier date of a CAP amendment are available to the State under the
circumstances of this appeal. The mere assertion of a loss of FFP,
absent any Agency involvement in the failure to claim IV-E
administrative costs, does not constitute a "significant inequity"
within the context of section 95.515.
II. The State did not following the sampling methodology specified in
its cost allocation plan.
Even if we were to find that the CAP amendment could have an effective
date of January 1, 1983, the State would not be able to receive FFP for
that period because it failed to comply with the provision of its CAP
regarding the statistical methodology to support those IV-E claims. For
example, the CAP called for the State to perform random daily samples of
its IV-E administrative and training costs for each quarter, yet the
State admittedly did not begin to perform samples until August 1984.
The CAP also called for a sample to be of sufficient size to insure
meeting a confidence level requirement of 95 percent. Obviously the
State, by not performing any samples until August 1984, failed to meet
this standard set forth in its CAP. Unless the State can show some
other method of meeting the provisions of its CAP, which we discuss
below in Part III of this decision, the State's failure to apply during
the period in question the statistical methodology provided for in the
CAP amendment would invalidate the State's claims for that period.
III. The sample results were not shown to be representative of prior
quarters.
A. Background.
Even if we were to find that the amendment to the CAP became effective
on January 1, 1983, that does not automatically mean that all the
State's claims for IV-E administrative costs from that date merit FFP.
It is a fundamental principle. of grants management that a grantee has
the obligation to provide documentation to support its claims. New York
State Dept. of Social Services, DAB No. 204 (1981); see also 45 C.F.R.
74.61(b), (f), and (g). The Agency has challenged here the
documentation which the State alleges supports its claims.
Section 474(a)(3) of the Act provides that FFP is available for a
state's expenditures for IV-E training and administrative costs. The
Agency based its decision to disallow the State's claims for the period
January 1, 1983 through September 30, 1984. In addition to the failure
of the CAP to provide for such claims, on a finding that the State's
claims were based on estimates of prior year costs rather than actual
expenditures.
Because of the large number of services provided and the extensive
amount of time and manpower that would be required to verify each
individual claim, the use of statistical sampling as a means of
supporting claims has received wide approval. See California Dept. of
Social Services, DAB No. 816 (1986), and decision cited therein. Here
the sampling methodology proposed and approved in the CAP called for a
minimum number of random day logs (RDLs) to be selected in a random
manner on a monthly basis during each quarter, and that the distribution
of time justified by those results be applied to the pool of training
and administrative costs for each quarter. The State did not begin
sampling for IV-E administrative and training costs until August 15,
1984. When, in 1985, the State submitted claims for IV-E administrative
and training costs for the period January 1, 1983 through September 30,
1984, it used as a basis for its claims the sample results taken from
August 15 through September 28, 1984.
B. The parties' arguments.
The Agency argued that the data gathered from the State's sample is
unreliable. The Agency contended that the State's sample violates one
of the basic tenets of statistical sampling: sampling is valid only
when it is drawn from a known universe of data; or, in other words, a
sample taken form a population applies only to that population which has
been sampled. According to the Agency, the State is attempting to a
population that was not subject to the sample to apply the sample
results from one period retroactively. The Agency argued that what the
State did--gathering data from half of one quarter and "backcasting"
that data into prior quarters--amounts to nothing but a subjective
guess. Given the millions of dollars at stake here, the Agency urged
the Board to reject such guesswork, adding that the CAP's approved
methodology does not even authorize such "backcasting."
The State replied that the Agency, in rejecting the proffered results
from the sample, is acting in a totally arbitrary, capricious, and
unreasonable manner. The State asserted that it is undisputed that the
State did not receive a substantial portion of the IV-E funds to which
it was entitled during the period in question. According to the State,
the results from the August 15 - September 28, 1984 are inherently
reliable in determining costs from earlier quarters. The State
contended that the results were based upon known data obtained in a
State agency and program which have not changed in their constituency,
organization goals, and objectives over the years. State personnel
testified that from quarter to quarter the State's IV-E program was free
of any significant changes so as to allow the sampled results to be
reliable for prior quarters. According to the State, subsequent samples
taken for quarters after September 30, 1984 have verified that the
results of the August 15 - September 28, 1984 sample have been constant
over time. The State referred to several states where the Department
has permitted time studies to be applied retroactively and emphasized
that the Agency, by refusing to propose any acceptable alternative
method for determining the State's IV-E costs during the questioned
periods, was acting particularly unreasonably.
C. Analysis.
At the hearing held in this appeal both parties produced experts in
statistics to support their respective positions. The Agency's two
expert witnesses were adamant in their assertions that under proper
statistical sampling methods results from one time period can not be
accurately applied to a population from another time period that was
never sampled. The Agency's witnesses faulted the State for its failure
to take into account in the sample results a non-response factor (the
failure of State employees to turn in RDLs) and to calculate a sampling
error to give credibility to the sample results. The State's witness,
on the other hand, testified that retroactive application of the date
from the August 15 - September 28 sample was accurate and equitable
under the circumstances of this appeal. The State's witness cited the
proximity of the questioned quarters to the sample period and the high
probability that there few significant variations between the periods.
In several appeals raising the question of whether sample results can be
applied retroactively to earlier, unsampled periods, the Board adopted a
view on the use of such results that is less rigid than that espoused by
the Agency's expert witnesses. In California Dept. of Health Services,
DAB No. 666 (1985), the Board permitted California to claim enhanced
funding for abortions paid for between 1972 and 1977 based on data
acquired in 1977 and 1978. The Board noted that the parties in that
appeal had concluded that the 1977-1978 period would be the best
available evidence for identifying what services in the earlier period
were for family planning and agreed that there were no significant
differences between the periods to make use of the later period
inappropriate. P. 2.
In Ohio Dept. of Human Services, DAB No. 900 (1987), the Board declared,
at page 11:
While sampling in its purest form envisions samples from the same
period in question, common sense would dictate that samples from
another period may be used if it can be established that no
substantial change has occurred so as to invalidate the procedure.
Unlike California, the Agency in Ohio did not agree that there were no
significant difference between the periods in dispute.
The Board went on to find that Ohio had not produced sufficient evidence
to support its position that its social service program in question had
remained so constant that there were not significant differences between
the data from the audited period and other periods.
The Board's position may therefore be stated as permitting sample
results from one period to be used to support claims from other periods
provided that it can be shown that there are no significant differences
between the periods. The party asserting the use of data for unsampled
periods has the burden of showing that circumstances relating to the
sampled and unsampled periods are constant. We are not prepared to
state what degree of similarity in circumstances is necessary to support
the retroactive application of sampling results; each case must be
judged by its particular circumstances. The uncontested soundness of
the data provided during the sample period is not sufficient in itself
to justify expenditures made in earlier periods; the conditions
surrounding the expenditures must approach those in the sampled period.
Here the State claimed that its IV-E program during the period beginning
January 1, 1983 was consistent in all major respects to its program
during the sampled period. A State witness testified that the State's
calculation of its IV-E costs for the retroactive period was based on
three components: 1) actual costs; 2) actual percentage of IV-E
children in care; and 3) percentage of workers' time spent on IV-E
activities. According to the witness, only the third component was an
estimate based on data from the sampled period. Tr. 9. Another State
witness testified that there were no significant changes in the IV-E
program between the two periods at issue. Tr. 41-42. Under these
circumstances, the State's expert witness testified, the State's
"backcasting" of the sampled period results was "accurate"
and
"equitable."
While the Agency responded with a list of reasons--all contested or
explained by the State--why the sampled period results were defective
and how circumstances between the periods had changed (e.g., percentage
variations in incidents of child abuse or neglect, children in
alternative care placements, children receiving home protective
services, children receiving adoption assistance services, and the
number of children served under the IV-E program) we find one
particular, uncontested circumstance fatal to the State's position. In
testifying about the IV-E program's constancy between the periods at
issue, a State witness declared that seasonal variations do occur in the
number of clients receiving assistance. The witness specifically
mentioned that when children return to school after summer vacations,
reports of child abuse increase as teachers notice children who appear
to have suffered abuse or neglect. Yet the six-week period, August 15
to September 28, 1984, that the State purports to show as representative
for the period beginning January 1, 1983, is precisely within that
seasonal variation. As an Agency witness testified, those six weeks are
a unique period unlike any other in the year. Tr. 154. As teachers
notice potential cases of abuse and neglect and notify State
authorities, the State's IV-E program is bound to increase in activity
accordingly. More administrative costs are going to occur as more
reports are filed and more cases are investigated. We consider it
highly reasonable to conclude, as the Agency insists, that this six-week
period is atypical. The State may argue that over time IV-E activities
average out, but the fact remains that it is basing all its IV-E
administrative and training costs from January 1, 1983 on a period when
IV-E activities, and consequent claims for FFP, are likely to be at
their highest level.
We are not persuaded that the Agency's position in this matter is, as
argued by the State, unreasonable. As we said earlier, it is the
State's responsibility to justify its claims for FFP. It is not the
Agency's responsibility to devise a method, when the State cannot, for
maximizing the State's claims for FFP.
The parties' positions on the retroactive use of sample results also
involved the question of whether other agencies of the Department had
approved other state CAPs employing such sampling procedures. The State
supplied plans from several states which it alleged showed that the
Department had no prohibition against the retroactive application of
data. State Appeal File Ex. M. The Agency questioned whether the cited
CAPs in fact allowed retroactive application of data. At the hearing an
Agency witness testified that any approval of such plans by regional
offices was in contradiction to ACYF central office policy. The witness
admitted, however, that ACYF had not formally announced its policy
absolutely prohibiting retroactive application of sampling results until
a June 30, 1987 memorandum. Agency Hearing Ex. 1. The witness further
testified that the approval of the cited CAPs by regional offices was
erroneous and currently being re-examined.
In light of our finding above that significant differences existed
between the periods, we do not therefore consider it relevant to the
facts of this appeal whether, as the State alleged, other State's CAPs
utilizing retroactive application have been approved in the past,
especially in light of the Agency's vigorous denials of the correctness
of any such approval and the lack of information in the record about the
particular details of each of those other states' CAPs.
As we said above, the party advocating the use of retroactive sampling
has the burden of showing its validity. The six-week period proposed to
be used as the norm was admittedly an aberration, a seasonal variation.
Under that circumstance, we find that the sampled results from that
period are not only unrepresentative of the period January 1, 1983 to
June 30, 1984, but also unrepresentative of the quarter ending September
30, 1984 in which the sample was taken. The State has not met the
burden of showing that the circumstances between the periods were so
constant as to justify the retroactive application of the sampled period
results.
Conclusion
For the reasons stated above, we find that the amendment to the State's
CAP may not have an effective date earlier than July 1, 1984, and that
the State's application of data from period to claim IV-E costs for
other periods is not justified under the circumstances of this case.
Accordingly, we uphold its disallowance of $5,511,265 in FFP for FYs
1983 and 1984.
________________________________ Cecilia Sparks Ford
________________________________ Norval D. (John) Settle
________________________________ Donald F. Garrett Presiding
Board