Department of Health and Human Services
DEPARTMENTAL APPEALS BOARD
Appellate Division
SUBJECT:
Mid-Kansas Community
Action Program, Inc.
Docket No. A-09-39
Decision No. 2257
DATE: June 25, 2009
DECISION
Mid-Kansas Community Action Program, Inc. (Mid-CAP), a Head
Start grantee, appeals the December 15, 2008 decision of the
Administration for Children and Families (ACF) disallowing
$248,565 in Head Start funds. Of this amount, Mid-CAP appeals
$240,314, which ACF disallowed as “Cash Disbursements not
supported by source documentation as required by 45 C.F.R.
§ 74.21(b)(7).” ACF Ex. 10, at 1. ACF calculated this portion
of the disallowance by extrapolating audit findings for a review
of 19 checks to the total amount of Head Start funds expended by
Mid-CAP during the audit period ending June 30, 2007.
We uphold ACF’s disallowance of $500 charged to Head Start
because Mid-CAP failed to submit adequate documentation for a
check in this amount. However, for the following reasons, we
are unable to uphold the remainder of the disallowance. First,
there is no evidence indicating that the auditor intended to
select a statistically valid random sample for the purpose of
extrapolating the findings to the universe of Mid-CAP’s Head
Start expenditures. Second, there is no evidence for ACF’s
claim that it was impractical to individually review all of MidCAPs non-payroll Head Start cash disbursements. Third, there is
no evidence that the extrapolation methodology used here was
scientifically valid.
Background
Mid-CAP is a “community-based, not-for-profit organization
serving a thirteen county, rural area in south-central Kansas”
2
that operates multiple programs, including Head Start.
Appellate Brief (App. Br.) at 1. Head Start is a national
program that provides comprehensive developmental services to
economically disadvantaged preschool children and their
families. 42 U.S.C. ' 9831.
Mid-CAP is subject to the Single Audit Act, which requires nonfederal entities (such as non-profit organizations like Mid-CAP)
that spend more than $500,000 in federal grant funds during a
fiscal year to conduct a single, comprehensive financial and
compliance audit of their programs for that year. 31 U.S.C.
§ 7502(a)(1)(A); Office of Management and Budget Circular A-133
(OMB A-133) (made applicable to non-profits by 45 C.F.R.
§ 74.26(a)). 1 Head Start grantees’ A-133 audits are reviewed by
the Department of Health and Human Services (HHS) Office of the
Inspector General (OIG) and ACF. ACF Br. at 4.
Mid-CAP’s A-133 audit for the year ending June 30, 2007
identified multiple deficiencies and material weaknesses in MidCAP’s efforts to comply with federal fiscal requirements. 2 At
issue in this case is the material weakness involving cash
disbursements, i.e., non-payroll checks, which the auditor found
were issued “without proper support and internally required
authorization.” Appellant Attachment (App. Att.) 3, at 48. The
auditor based this finding on its review of 19 non-payroll
checks from which it concluded that two (10.53%) of the checks
did not have supporting documentation. ACF Ex. 4, at 6. Only
one of the two undocumented checks was a Head Start check. The
auditor recommended, “Procedures should be implemented
prohibiting the issuance of checks without all supporting
documentation and required authorization.” App. Att. 3, at 49.
The auditor did not question any costs related to these checks.
1
The $500,000 threshold was established by regulation.
Fed. Reg. 38,401 (June 27, 2003).
2
68
The auditor defined “material weakness” as “a significant
deficiency, or combination of significant deficiencies, that
results in more than a remote likelihood that material
noncompliance with a type of compliance requirement of a federal
program will not be prevented or detected by the entity’s
internal control.” App. Att. 3, at 45.
3
In March 2008, Mid-CAP submitted its A-133 audit for the year
ending June 30, 2007 to the federal audit clearinghouse. ACF
Br. at 4, citing App. Att. 3. On August 15, 2008, the OIG
notified Mid-CAP that it had reviewed the audit and found that
the audit itself “met federal audit requirements.” ACF Ex. 4,
at 1. It informed Mid-CAP that “final determinations with
respect to actions to be taken” based on the audit would be made
by a “resolution official of HHS.” Id. The OIG recommended
generally that, based on the “serious nature of weaknesses
identified” in the audit, HHS initiate “closer monitoring and
increased attention by grants management staff to protect the
Federal interest.” Id. at 2. In Attachment A to the letter,
the OIG identified material weaknesses and recommendations for
each weakness. For the category “cash disbursements,” the OIG
stated that it “recommends procedures be strengthened to ensure
cash disbursements are properly authorized and supported by
adequate documentation.” Id. at 3. The OIG did not identify
questioned costs as to cash disbursements or recommend that such
costs be disallowed on the basis of the audit.
By letter dated September 2, 2008, ACF placed Mid-CAP on “high
risk status,” which resulted in special award conditions being
placed on its grant. ACF Ex. 5. Thereafter, ACF and Mid-CAP
corresponded and met about the problems identified in the audit
and Mid-CAP’s efforts to improve its oversight of federal funds.
ACF Exs. 6-9. During this process, Mid-CAP supplied, at ACF’s
request, “all available back-up documentation” for the 19 checks
reviewed by the auditors in evaluating Mid-CAP’s control
mechanisms for cash disbursements. App. Br. at 4; see also ACF
Br. at 9, citing ACF Ex. 11.
By letter dated December 15, 2008, ACF disallowed $240,314 for
inadequately documented cash disbursements, $3,301 for bonus
checks, and $4,950 for Christmas gifts to employees. ACF Ex.
10, at 1. As to the cash disbursements, ACF stated:
The Office of Head Start directs the grantee to repay
$240,314 (the total amount of Head Start expenditures
reported[,] $2,403,140[,] x 10%) due to the lack of
supporting documentation as required by 45 C.F.R.
§ 74.21(b)(7).
Id. at 5. Mid-CAP appeals only the disallowance of cash
disbursements.
4
Analysis
1. We reverse the disallowance to the extent it is based
on statistical sampling.
ACF calculated this disallowance by extrapolating the results of
the auditor’s review of 19 checks to the universe of Head Start
expenditures of $2,403,140. ACF Ex. 10, at 5. Mid-CAP
challenges ACF’s extrapolation of the results of this review as
improper use of statistical sampling. App. Br. at 5-6. Where a
grantee challenges the use of statistical sampling, the agency
has the burden of showing that sampling is appropriate to use in
the context of the particular disallowance and that the sampling
and extrapolation methodology used to calculate that
disallowance is scientifically valid. 3 For the following
reasons, we conclude that ACF has not met its burden of showing
either that the use of sampling was appropriate in this case or
that its methodology was scientifically valid.
ACF asserts that statistical sampling was appropriate here
because it was not “practical to undertake a review of every
transaction.” ACF Br. at 11. The only support ACF cites for
this assertion is the fact that “Mid-CAP indicates that it
averages 235 monthly checking transactions.” Id. at n.11,
citing ACF Ex. 9, at 8. For the following reasons, we disagree
with ACF’s position.
First, nothing in the audit suggests that the auditor was in
fact selecting a sample in order to statistically extrapolate
totals of unallowable costs. On the contrary, the auditor was
merely reviewing sufficient transactions to determine whether
mid-CAP’s “financial statements . . . present fairly, in all
3
See, e.g., Illinois Dept. of Children and Family
Services, DAB No. 1564 (1996) (use of sampling to calculate the
title IV-E disallowance was not contrary to title IV-E;
extrapolation of sample results was appropriate); New York Dept.
of Social Services, DAB No. 1531 (1995) (federal agency
demonstrated that its statistical sampling methodology was
scientifically valid); Washington State University, DAB No. 145
(1981) (federal agency failed to identify why it would be
reasonable to apply the findings for one year to the prior
year).
5
material respects, the financial position of [Mid-CAP] as of
June 30, 2007” and to report on its review of Mid-CAP’s
“internal control over financial reporting.” App. Att. 3, at
1. Because nothing indicates that the auditor sought to
generate a scientifically valid random sample for identifying
unallowable costs, it is not appropriate to use the audit as a
basis for calculating a disallowance by extrapolating the
auditor’s findings for these 19 checks.
Second, Board cases upholding the use of sampling have looked to
whether “individual review of the underlying records would be
impractical due to volume and cost.” New York State Dept. of
Social Services, DAB No. 1394, at 22 (1993). “Typically
sampling is used when a claim for federal funds is based on the
sum of numerous cost items (each subject to proof of
allowability), because it is impossible, or at least costly and
impractical, to examine each item . . . .” New York State Dept.
of Social Services, DAB No. 1235, at 8-9 (1991) (review of
claims for mental health services for 117,002 clients over oneyear period). 4 In such situations, the Board has relied on an
extrapolated finding because “[i]f done in accordance with
accepted rules . . . [it] has a high degree of probability of
being close to the finding which would have resulted from
consideration of all the cost items.” Id. at 9.
Because ACF has not shown how many transactions would have
required review, ACF has provided no basis for concluding, as it
asserts, that actual review of the checks at issue was not
“practical.” ACF’s justification for the use of statistical
sampling – 235 monthly checking transactions – overstates the
number of transactions that would require review in at least two
ways. First, Mid-CAP operates and issues checks for multiple
programs other than Head Start. App. Att. 3, at 17-36. (In
fact, of the 19 cash disbursement checks the auditors reviewed,
only six were Head Start checks. See ACF Ex. 11, at 1; App.
Att. 3, at 17.) Consequently, the figure of 235 monthly checks
overstates the number of Mid-Cap’s Head Start checks that would
4
See, e.g., California Dept. of Social Services, DAB No.
816 (1986) (review of over one million federal assistance
payments); California Dept. of Social Services, DAB No. 524
(1986) (review of the costs of federal child care services over
two and a half years in four counties).
6
require review. Second, the auditor excluded payroll checks
from the category of cash disbursements. ACF Ex. 11, at 1
(auditor worksheet showing that auditors selected 25 checks for
review but then eliminated the six payroll checks from the cash
disbursement category). Therefore, the figure of 235 monthly
checks also overstates the number of checks in the category of
Head Start cash disbursements. Because ACF did not take either
of these factors into account, ACF failed to accurately
determine the number of monthly check transactions that would
have required review here and, therefore, had no reliable basis
for asserting that review of the individual Head Start nonpayroll transactions was not practical.
In addition, ACF has failed to show that the methodology used to
calculate the disallowance was scientifically valid. For
example, ACF does not provide any evidence from a statistician
or even assert that the 19 checks reviewed by the auditor
constitute a statistically valid random sample or were otherwise
selected pursuant to a scientifically valid sampling
methodology. ACF also fails to cite anything in the record that
would support a finding that ACF used a scientifically valid
methodology to extrapolate its finding that 10% of Mid-CAP’s
expenditures should be disallowed. Moreover, ACF’s
extrapolation methodology is facially invalid in the following
way. The auditor’s finding on which ACF relied was based on a
review of cash disbursements (a category that did not include
payroll checks), but ACF applied the resulting 10% disallowance
rate to Mid-CAP’s total Head Start expenditures, including
payroll expenditures. Thus, the universe of sampled
transactions was smaller than the universe of transactions to
which the sampling result was applied. Because payroll
expenditures were over a third of Mid-CAP’s Head Start
expenditures (App. Att. 3, at 17-18), this aspect of the
methodology would cause the disallowance to be significantly
overstated.
Finally, as discussed below, the cases on which ACF relies do
not support its position here.
ACF cites Dallas County Community Action Agency, DAB No. 1265
(1991), for the proposition that “[r]eviews of audit reports by
OIG and the resulting recommendations including recommendations
based on sampling have been recognized as reliable bases upon
which ACF might ‘take’ a disallowance” against a Head Start
grantee for inadequately documented costs. ACF Br. at 8. ACF’s
7
statement mischaracterizes both the holding in Dallas County and
its application to the facts of this case.
•
First, while the Dallas County disallowance was based on
the OIG’s recommendations resulting from the OIG’s review
of independent audits, nothing in that decision indicates
that the OIG used statistical sampling in calculating the
questioned costs or that ACF used statistical sampling in
calculating the disallowance. 5 Therefore, ACF is incorrect
when it asserts that, in Dallas County, the Board upheld
the use of statistical sampling as a basis for disallowing
inadequately documented Head Start expenditures.
•
Second, unlike in Dallas County, the OIG here did not
question specific cash disbursements or recommend that ACF
take a disallowance, much less suggest that ACF should
calculate a disallowance by extrapolating the auditor’s
findings about the 19 checks to the universe of Head Start
expenditures. See ACF Ex. 4, at 1-4; ACF Ex. 10, at 10-12.
Thus, it is misleading for ACF to state that “the
disallowance was specifically tied to an OIG recommendation
with respect to the absence of documentation for 10.53% of
the checks tested by Mid-CAP’s independent auditor.” ACF
Br. at 8.
ACF also relies on the following language from the Board’s
December 27, 1989 Ruling on Request for Reconsideration (Ruling)
of West Virginia Dept. of Human Services, DAB No. 1107 (1989).
ACF Br. at 8. “The key requirement for a disallowance is that
‘the federal agency must articulate a reasonable basis for the
disallowance with sufficient detail so that the grantee may
respond.’” Ruling, citing West Virginia, DAB No. 1107, at 6.
5
While the first audit of the three audits at issue in
Dallas County speaks of “costs for a particular sampled employee
and non-payroll transactions,” the costs disallowed under the
first audit were $4,118.12. Absent information to the contrary,
the reference to “particular costs” and the small amount of
costs disallowed for that audit indicates that ACF disallowed
specific costs identified in the sampled transactions as opposed
to extrapolating the sample results to the universe of
expenditures.
8
ACF asserts that it met this requirement because “Mid-CAP was
sufficiently apprised as to which costs were questioned and why,
and how the disallowed portion of its aggregate claim was
determined.” ACF Br. at 8.
Neither West Virginia nor the subsequent Ruling supports ACF’s
position here. In the Ruling, the Board noted in dicta that
that “sampling methodology . . . may be the most appropriate
technique when the federal agency is examining the propriety of
a large number of individual claims.” Ruling at 5 (emphasis
added). As discussed above, ACF failed to set forth any basis
for concluding that statistical sampling was an “appropriate
technique” under the circumstances of this case. Moreover,
while ACF may have “apprised” Mid-CAP of how it calculated the
disallowance (as required in West Virginia), clarity about the
methodology alone does not cure the flaws in the methodology or
make it scientifically valid.
For the preceding reasons, we reverse the disallowance to the
extent that it is based on sampling and extrapolation.
2. We uphold ACF’s disallowance of $500 because the
expenditure was not supported by adequate documentation.
The auditors found that two of the 19 checks did not have
adequate supporting documentation. ACF Ex. 11, at 1. ACF does
not dispute Mid-CAP’s assertion that only one of these checks
was charged to the Head Start grant. This check was for $500
and was paid to a local public school. ACF Ex. 11, at 1, 47.
In its letter brief, Mid-CAP represents the payment was for
“utilities consumed by Mid-CAP’s operation of a Head Start
Center in the vendor’s building.” App. Br. at 3. It states,
“This routine, monthly payment is made to the vendor for the
costs of gas, electricity, water, trash, and janitorial services
necessary for the health and safety of the 17 children served by
Mid-CAP’s Head Start Program.” Id.
For costs to be allowable under federal cost principles, they
must be reasonable for the performance of the grant award,
allocable thereto, and adequately documented. OMB A-122,
9
Attachment A, && A.2.a. and g. 6 As to documentation, the
Uniform Administrative Requirements for Awards and Subawards to
. . . Nonprofit Organizations at 45 C.F.R. Part 74 require nonprofit recipients of federal funds to have a financial
management system that provides for “[r]ecords that identify
adequately the source and application of funds for HHS-sponsored
activities,” including “information pertaining to . . .
obligations . . . .” and “[a]ccounting records . . . that are
supported by source documentation.” 45 C.F.R. '' 74.21(b)(2),
74.21(b)(7). Section 74.2 defines “obligations” as “. . .
contracts . . . and similar transactions during a given period
that require payments by the recipient during the same or a
future period.” Under the cost principles and uniform
administrative requirements, the Board has consistently held
that HHS grantees bear the burden of documenting the
allowability of all questioned costs charged to federal funds.
See, e.g., Northeast Louisiana Community Development
Corporation, DAB No. 2165 (2008); Action for a Better Community,
DAB No. 2104 (2007); Marie Detty Youth and Family Services
Center, Inc., DAB No. 2024 (2006).
The record contains the following documentary evidence related
to the payment: a cancelled check dated December 26, 2006 and
written for $500 to Halstead Middle School (ACF Ex. 11, at 47);
a document titled “Payment Voucher” listing payment of $500 to
Halstead Middle School for “HEADSTART-HALSTEAD” “utilities” (id.
at 48); a document titled “Voucher” stamped “PAID” and dated
December 27 for $500 that identifies the “vendor” as Halstead
Middle School for “utilities” (id. at 49); and a copy of a
contract titled “AGREEMENT Involving [Halstead Middle School]
and Mid-Kansas Community Action Head Start 2007-08 Program Year”
calling, inter alia, for the payment of $500 per month for the
“use, maintenance, and cleaning of classroom and playground
space” (App. Att. 2). 7 The contract is dated February 2007.
6
OMB A-122, “Cost Principles for Nonprofit Organizations,”
is codified at 2 C.F.R. Part 230, and Attachments A and B are
designated “Appendix A” and “Appendix B,” respectively. OMB A122 is made applicable to HHS grants to non-profit
organizations, including Head Start grantees, by 45 C.F.R.
'' 74.27(a) and 1301.10(a).
7
The first three documents were provided to ACF in
December 2008. ACF Br. at 9. Mid-CAP does not dispute ACF’s
(Continued . . .)
10
We uphold the disallowance of $500 because the evidence on which
Mid-CAP relies does not document any obligation to pay this $500
on behalf of its Head Start program in December 2006.
Specifically, the contract on which Mid-CAP relies is not
adequate documentation of an obligation because it postdates the
2006 check and is for the “2007-08 Program Year.” App. Att. 2.
While ACF pointed out this fact in its response brief (ACF Br.
at 10), Mid-CAP did not thereafter seek to present a contract
for the relevant period or any other basis for concluding that
it was obligated to make this payment. Therefore, Mid-CAP has
failed to present adequate source documentation for this
obligation as required by 45 C.F.R. '' 74.21(b)(2), 74.21(b)(7).
Conclusion
For the reasons stated above, we uphold the disallowance of $500
and reverse the remainder of the disallowance.
________ /s/___________
Sheila Ann Hegy
_______ /s/___________
Leslie A. Sussan
_______ /s/___________
Stephen M. Godek
Presiding Board Member
_______________________
(Continued . . .)
assertion that the contract “was not among the documents
originally submitted to ACF.” ACF Br. at 10, n.9.
