Department of Health and Human Services
DEPARTMENTAL APPEALS BOARD
Appellate Division
SUBJECT: Mississippi Department of Human Services
DATE: July 25, 1991
Docket No. 89-3
Decision No. 1267
DECISION
The Mississippi Department of Human Services (State) appealed a
funding
reduction imposed under section 403(h) of the Social Security Act
(Act)
by the Office of Child Support Enforcement (OCSE). 1/ Based on
audits
of the State's child support enforcement and paternity
establishment
program, OCSE determined that the State did not comply
substantially
with the requirements of Title IV-D of the Act. OCSE
proposed a one
percent reduction of the amount otherwise payable to the State
for Aid
to Families with Dependent Children (AFDC) during the period July
1,
1987 through June 30, 1988 (a $746,477 reduction).
The State challenged the jurisdiction of the Departmental Appeals Board
to
hear this appeal, the conduct of OCSE during this proceeding, the
regulations
that governed Title IV-D audits, and the statistical
sampling methodology
used by OCSE as a basis for its findings. The
State did not challenge
any of OCSE's findings that the State had not
provided establishment of
paternity services in specific cases reviewed
by OCSE. In fact, the
State admitted early in this proceeding that its
performance did not meet the
regulatory standard for substantial
compliance. Tape of 2/15/89
Conference Call. Resolution of this appeal
was delayed significantly by
settlement negotiations between the parties
and by OCSE's adjustments to its
statistical methodology late in the
proceeding.
For the reasons stated below, we uphold OCSE's decision to reduce by
one
percent the State's AFDC funding for the one-year period beginning
July
1, 1987. Specifically, we conclude that--
o the Board has jurisdiction to review,
reconsider and decide the
State's challenge to this disallowance;
o none of the allegedly prejudicial OCSE
actions during the
disallowance and appeal process harmed the State, and none
provide a
basis for overturning this disallowance;
o OCSE properly applied its
interpretation of the statutory term
"substantial compliance" to the time
periods at issue here;
o OCSE reasonably interpreted the
statutory requirement for
"substantial compliance" to mean that a state must
be taking action to
provide basic child support services (required under the
Act) in at
least 75% of the cases requiring those services;
o OCSE's adoption of a regulation
specifying a one-year limit on
the period permitted by the statute for
corrective action is not
arbitrary, capricious, or contrary to the purpose of
the statute; and
o the statistical sampling evidence submitted here reliably shows
that
the State failed to meet the OCSE audit criterion related
to
establishing paternity in Title IV-D cases.
Statutory and regulatory provisions
Each state that operates an AFDC program under Title IV-A of the Act
is
required to have a child support enforcement and paternity
establishment
program under Title IV-D of the Act. Section 402(a)(27)
of the Act.
The Title IV-D program has been in existence since July
1975. OCSE has
the responsibility for auditing state Title IV-D
programs, pursuant to
section 452(a)(4) of the Act, and evaluating whether
the actual
operation of such programs conforms to statutory and
regulatory
requirements. Following adoption of Title IV-D, the
participating
states were given 18 months by Congress -- until December 31,
1976 -- to
establish and begin operating their programs before compliance
audits
actually began. Under the applicable statute, a state was
subject to a
five percent reduction of its Title IV-A funds if the audit
found that
the state was not in compliance. Congress, however,
continuously
extended the initial moratorium on imposition of this funding
reduction,
so that no reduction was ever imposed during the first eight years
of
the program's operation, although OCSE did continue its annual audits.
On August 16, 1984, Congress adopted the Child Support
Enforcement
Amendments of 1984, section 9 of Public Law 98-378 (the
1984
Amendments). As amended, section 403(h)(1) of the Act provides
that--
if a State's program operated under Part D is found as a
result
of a review conducted under section 452(a)(4) not to
have
complied substantially with the requirements of such part
for
any quarter beginning after September 30, 1983, and
the
Secretary determines that the State's program is not
complying
substantially with such requirements . . ., the
amounts
otherwise payable to the State under this part [A] for
such
quarter and each subsequent quarter, prior to the first
quarter
throughout which the State program is found to be in
substantial
compliance with such requirements, shall be reduced . . .
.
(Emphasis added.)
The amended section then provides for graduated reductions, starting
with
a reduction of "not less than one nor more than two percent" and
increasing
to a maximum of five percent with each consecutive finding
that a state is
not complying substantially with Title IV-D
requirements.
The 1984 Amendments provided for the continuation of compliance
audits,
which could in appropriate cases be scheduled as infrequently as
once
every three years. 2/ Rather than directing immediate reduction
of
funding for a state which failed an audit, the Amendments provided
that
a reduction could be suspended while the state was given an
opportunity
to bring itself into compliance through a corrective action
plan
approved by OCSE. Section 403(h)(2)(A)-(C) of the Act, as
amended. If
a follow-up review of a state's performance showed that the
state still
did not achieve substantial compliance, a reduction would be
imposed.
Section 403(h)(2)(B)(iii) of the Act.
Section 9(c) of the 1984 Amendments provides that they "shall be
effective
on and after October 1, 1983."
OCSE proposed regulations implementing the Amendments on October 5,
1984,
49 Fed. Reg. 39488 (1984), and issued final regulations on October
1,
1985. 50 Fed. Reg. 40120 (1985). (We refer to these regulations
as
the "1985 regulations.") The 1985 regulations amended parts, but
not
all, of the audit regulations at 45 C.F.R. Part 305. Section
305.20(a),
as amended by the 1985 regulations, provided that, for the fiscal
year
(FY) 1984 audit period, certain listed audit criteria (related
primarily
to administrative or fiscal matters) "must be met." This
section also
provided that the procedures required by nine audit criteria
"must be
used in 75 percent of the cases reviewed for each criterion . . .
."
These criteria relate to performance of basic services provided under
a
Title IV-D state plan; one of these is the criterion at issue in
this
appeal. All the service-related audit criteria are based on
sections of
45 C.F.R. Part 305 which (with minor exceptions not relevant
here) were
originally published in 1976, with minor amendments in 1982.
(We refer
to these provisions, as amended in 1982, as the "existing
regulations"
since they were in effect during FY 1984.)
Thus, under the 1985 regulations, substantial compliance for FY
1984
audits was measured by audit criteria from the existing regulations,
but
a state had to be providing the required services in 75% of the
cases
requiring them. 3/ In follow-up reviews after a corrective
action
period (limited by regulation to one year, 45 C.F.R. 305.99(c)),
OCSE
would examine only the audit criteria that the state had
previously
failed or had complied with only marginally (that is, in 75 to 80%
of
the cases reviewed for that criterion). 45 C.F.R. 305.10(b) and
305.99,
as amended.
Background
OCSE's audit for FY 1984 (October 1, 1983 through September 30,
1984)
resulted in a September 24, 1986 notice to the State that it had
been
found to have failed to comply "substantially with the requirements
of
Title IV-D of the Act" in the following areas: (1) reports
and
maintenance of records, 45 C.F.R. 305.35(a), and (2)
establishing
paternity, 45 C.F.R. 305.24(c). 9/24/86 Letter to State
from OCSE
Director (State Exhibit (Ex.) N). 4/ OCSE found that 290 of
811 case
files it initially selected could not be located, and that the
State
took action in only 62 of 131 sample cases requiring action to
establish
paternity (47% of sampled cases). 5/
Rather than appealing OCSE's findings, the State opted to propose
a
corrective action plan (State Ex. O) that was accepted by OCSE
on
December 18, 1986, and the funding reduction was suspended.
12/2/88
Disallowance letter at 1. The corrective action period ended
on
September 23, 1987. The follow-up review by OCSE of the
State's
performance for calendar year 1987 resulted in the December 2,
1988
notice of a proposed funding reduction that is the subject of
this
appeal. OCSE found that the State had achieved substantial
compliance
with the reports and maintenance of records criterion, but had
failed
the establishing paternity criterion. Specifically, OCSE found
that the
State's performance had deteriorated during the follow-up review
period
-- the State had taken appropriate action in paternity cases in only
20
of 71 sample cases (28% of those cases). See State Ex. P.
The State filed an appeal of the disallowance that contained
numerous,
often duplicative arguments. In this appeal proceeding, the
State
styled these arguments as "Proposition I," "Proposition II," etc.
6/ In
our analysis below, we review each of these propositions (which
we will
identify in parentheses), but do not necessarily follow the order
in
which the State raised them. 7/
Analysis
I. The State's procedural challenges to this proceeding are
without
merit.
In its notice of appeal, the State argued that the regulations at
45
C.F.R. 205.146(e) and 305.100(g), which refer to "reconsideration" of
a
penalty disallowance, provide the State with the right to
a
reconsideration by the entity which issued the disallowance, prior
to
review by the Board. The State therefore contended that the
Board
should remand this case to OCSE for such a
reconsideration.
Notwithstanding the Board Chair's January 5, 1989 ruling
that the case
was properly before the Board, the State continued to press
this
argument in its briefs (Proposition I). Despite numerous
subsequent
reaffirmations of this ruling throughout this proceeding (see
8/2/89 and
1/23/90 rulings), we briefly discuss it here.
As the Board Chair noted in his ruling, this disallowance was issued
by
the top official of OCSE, who expressly directed the State to appeal
to
the Board. Consequently, absent a rule requiring such a review,
it
would have been unreasonable for the Board to remand this matter to
that
official for "reconsideration." 8/ Moreover, both regulations
cited by
the State refer back to section 1116(d) of the Act, which
encompasses
reconsiderations that have been delegated to the Board in 45
C.F.R. Part
16, Appendix A. See 1/5/89 Ruling at 2. The Board
Chair therefore
concluded that the State would be afforded its full right
to
reconsideration through an appeal to the Board.
The State argued in its appeal brief that this jurisdictional ruling
is
incorrect, citing to Words and Phrases that "As normally used in
context
of administrative adjudication `reconsideration' implies
reexamination,
. . . by the entity which initially decided . . . ." 36A
Words and
Phrases, "Reconsideration" (West Supp. 1991). Section
1116(d), however
refers to reconsideration by the Secretary, HHS. The
Secretary has
clearly delegated this Board authority to reconsider such
disallowances,
on his behalf. 45 C.F.R. Part 16, App. A, section B; 43
Fed. Reg. 9264
(March 6, 1978). The fact that the Secretary delegated
the authority to
make an initial decision to OCSE does not mean that review
by this Board
is any less a reconsideration by the Secretary or a decision by
the
Secretary. 9/
The State has been afforded its full right to reconsideration through
this
two-and-a-half year long proceeding; it can show no harm to its
interests
after having had a full and fair opportunity to present its
appeal to this
Board. We therefore reaffirm that this appeal is within
the Board's
jurisdiction. 10/
The State also complained about OCSE's introduction late in
this
proceeding of a correction to its statistical calculations
concerning
the data underlying OCSE's finding for the FY 1984 audit that the
State
failed to meet the establishing paternity criterion (Proposition
XIV).
The State variously argued that: this was an attempt by OCSE to
reap
the benefits of a remand for reconsideration while opposing an
actual
remand (10/12/90 br. at Proposition I); was a violation of
"fundamental
fairness" to the State (10/12/90 br. at Proposition III); was
an
improper attempt to issue new findings at the close of the
briefing
process (10/12/90 br. at Proposition XIV); and effectively
rendered the
original penalty disallowance's calculations "void" (10/12/90
br. at
Proposition XV).
We disagree with the State that there was anything untoward or
unfair
about OCSE's proffered corrections. (We note that the State did
not
challenge them procedurally until well after they were made,
presumably
because there was a chance that the end result might have been
favorable
to the State.) OCSE's changes in this case involved only
calculations
concerning data already in the record; they did not involve new
evidence
or a new theory of liability. While of course it would have
been easier
for all concerned if these recalculations were not necessary, so
long as
the State had sufficient notice to enable it to understand the
issues
and OCSE's position, the disallowance letter was adequate for us
to
proceed. See 45 C.F.R. 74.304(c). The State was given
ample
opportunity, including two in-person hearings, to review and
comment
thoroughly upon these changes, so that "fundamental fairness"
was
clearly provided. Board procedures give the Board
considerable
discretion to enable the parties to clarify or amend their
positions as
long as the opposing party has the opportunity to respond to any
change.
West Central Wisconsin Community Action Agency, DAB No. 861
(1987) "In
describing the disallowance notice that gives rise to Board
proceedings,
the regulations nowhere suggest that the Board must dismiss
a
disallowance merely because the Agency wished to modify the
disallowance
rationale based on a changed understanding of the circumstances
of the
case." Id. at n.2. Since the State was given a full
opportunity to
challenge the changes, we conclude that it was not harmed by
their
introduction.
The State also argued in its notice of appeal that the Board should
review
OCSE's conditional stay of the penalty's imposition because the
State was
allegedly thereby denied the full 30 days permitted by
regulation for filing
its notice of appeal. The Chair's 1/5/89 ruling
noted that the
practical effect of OCSE's action was to influence the
State to file its
notice four days early. Moreover, no harm was done,
since the
regulations contemplate only a rather summary statement of an
appellant's
grounds for appeal at the notice of appeal stage (45 C.F.R.
16.7(a)), and the
Board gave the State several generous extensions of
time in which to prepare
its appeal brief. See 5/26/89 Letter to the
State. att. 1.
Ultimately, the State was given six months in which to
prepare its appeal
brief.
The State continued to challenge OCSE's actions in subsequent
briefs,
asking the Board to reconsider its ruling because it set a
precedent
that would allow OCSE to require an appellant to file a notice of
appeal
"in a few hours." Appeal br. at 9 (Proposition II). We
fail to see how
this could set a precedent, however, when the Board's
regulations
clearly allow a 30-day period, and OCSE's actions cannot override
the
regulations. 45 C.F.R. 16.3(b); see also 45 C.F.R. 74.304(d).
The
State did not allege that it was harmed, moreover, and the remedy
it
sought from the Board was a remonstration against OCSE, not the
reversal
of this disallowance. The State also requested an amendment to
the
Board's regulations to prohibit such actions (reply br. at 19);
however,
the means for requesting a rulemaking are a petition under 5
U.S.C.
553(e), not an appeal before the Board. 11/ Consequently, we
conclude
that the Chair's 1/5/89 ruling that the State was not harmed is
still
valid and we affirm it.
The State also maintained that it was disadvantaged during this
proceeding
by the nondisclosure agreement that OCSE required the State's
counsel to sign
prior to reviewing the audit workpapers (Proposition
XI). That
agreement provided:
SAFEGUARDING INFORMATION AND DISCLOSURE RESTRICTIONS
The use or disclosure of information concerning applicants
or
recipients of support enforcement services must be
safeguarded
in accordance with 45 C.F.R. 303.21 and may not be used for
any
unapproved purpose or be redisclosed to any other Federal,
State
or local body.
Unauthorized use or disclosure of information obtained
through
the Federal Parent Locator Service may result in civil
or
criminal penalties under 26 U.S.C. 7213(a)(2).
Tax information must be safeguarded in accordance with 26
U.S.C.
6103(p)(4).
I understand the above noted restrictions and affirm that I
am
authorized as an official of my State to have access to,
review
and transmit copies of information, and the State is bound
to
safeguard and protect against wrongful use. I agree to abide
by
all applicable Federal laws and regulations, and I
expressly
agree that any information disclosed to me which may
be
protected under the Privacy Act, the above cited requirements
or
any other law or Federal regulation will not be
redisclosed
except to authorized officials of my State.
Att. to State's 6/30/89 Letter to Board.
The State interpreted this agreement to mean that the State was unable
to
share the contents of these workpapers with an outside consultant,
with other
states, or even with the Board during this proceeding. OCSE
explained
that this agreement was standard and was meant to protect
client-specific
information. Thus, all the State had to do to satisfy
the legal
requirements of the Privacy Act was to remove identifying
information from
any documents submitted to the Board, using a code (or
initials) to identify
cases discussed, and to limit sharing of those
documents to those persons
with a legitimate need for access who also
sign an agreement to keep this
material confidential. See OCSE 7/6/89
Motion to Compel Mississippi to
File a Complete Brief at 3; OCSE 8/25/89
Letter to the Board at 1-2. In
an August 28, 1989 telephone conference
call, the Presiding Board Member
discussed this issue with the parties
and specifically ruled that OCSE could
not restrict access to the
information and that State counsel could comply
with the nondisclosure
agreement by blocking out identifying
information. 9/13/89 Summary of
Telephone Conference Call at 2.
At that time, State counsel "stated
that he was specifically waiving any
right to be furnished or to inspect
Agency workpapers." Id.
State's counsel, however, reiterated his
complaints in the State's reply
brief (Proposition XI) and at the
hearing. Transcript of 4/4/91 Hearing
("Tr. I") at 12-41.
We are unable to understand why State counsel clings to this
outlandish
reading of the standard disclosure agreement to his client's
detriment.
In any event, however, the State clearly was given access to all
of the
information it needed (most of which was information from the
State's
own files) and given permission to use it to defend this case.
Since
State counsel waived the right to access, we see no reason to
discuss
this further. Consequently, we reject this proposition as a
basis for
overturning the disallowance.
The State also raised (as Proposition IV) a contention that the
Board
should make an in camera inspection of documents withheld by OCSE
in
connection with a Freedom of Information Act (FOIA) request. Appeal
br.
at 12-14. In addition to access to the audit workpapers, OCSE
had
provided hundreds of documents to the State in response to a
FOIA
request, but had withheld eight documents containing 33 pages as
exempt
from disclosure. See 9/26/89 Family Support Administration's
Response
to Mississippi's Question Regarding Withheld Documents at 1.
The
Presiding Board Member directed OCSE to answer questions about
the
contents of those documents (9/13/89 Summary of Telephone
Conference
Call), but denied the request for in camera inspection. See
10/11/89
Summary of Telephone Conference Call; 11/20/89 Summary of
Telephone
Conference Call (denying reconsideration of 10/11/89 ruling).
12/ We
agree with the Presiding Board Member's ruling that the
documents as
described do not appear to be relevant to the issues raised by
the State
in its appeal. Generally, these documents are predecisional
discussions
of proposed statistical sampling plans and calculations; the
final
methodology actually used by OCSE is the relevant document for
this
proceeding, and it was made available to the State. Consequently,
we
reaffirm the Presiding Board Member's ruling and reject the
State's
Proposition IV as a basis for overturning this disallowance.
II.
The State's challenges to the 1985 regulations are
without merit.
The State challenged the 1985 regulations that OCSE used in
concluding
that the State was not in substantial compliance.
Specifically, the
State argued that--
o the regulations are impermissibly retroactive under Bowen
v.
Georgetown University Hospital, 488 U.S. 204 (1988)
(hereafter
Georgetown), since OCSE lacked express statutory authorization to
apply
these regulations retroactively (discussed in Proposition V);
o despite the statutory provision setting
an effective date of
October 1, 1983, the legislative history of the 1984
Amendments
indicates that Congress expected any resulting regulations to
be
prospective only in effect (also discussed in Proposition V);
o the 75% standard as applied to the State here is not
avalid
interpretation of the statutory term "substantial compliance" since
the
State is being penalized for failing to meet only one
criterion
(Proposition X);
o the 75% standard in the regulations had
no empirical basis and
therefore was established in an arbitrary and
capricious manner under
Maryland v. Mathews, 415 F. Supp. 1206 (D.D.C. 1976),
and according to
the State's expert (9/12/90 Affidavit (Appeal record at
1045))
(Proposition VI); and
o the regulations were invalid because
they did not include a
definition of "violations of a technical nature,"
based on section
403(h)(3), as amended (Proposition VII).
OCSE disputed the State's position, but also pointed out that the Board
is
bound by all applicable laws and regulations under 45 C.F.R. 16.14.
The
regulations at issue were "effective" on the date of final
publication
(October 1, 1985). However, section 305.20(a), which sets
out the 75%
standard for service-related audit criteria, states that it
is to be applied
"[f]or the fiscal year 1984 audit period." The
preamble to the
regulations confirmed that OCSE intended to apply this
section to FY 1984
audits, based on the October 1, 1983 effective date
of the 1984
Amendments. 50 Fed. Reg. at 40126, 40138.
We are, of course, bound by the Department's regulations, even if
invalid
under a constitutional analysis, if those regulations are
applicable.
13/ While some of the issues here clearly would be
controlled by 45
C.F.R. 16.14, the State's arguments also raise
interrelated questions of
applicability. We do not need to sort out
these issues precisely,
however, since we conclude that all of the
State's arguments concerning the
regulations are completely without
merit. 14/ Our reasons are:
o Section 403(h)(1) of the Act, as
amended, requires reductions
for states not found to be in substantial
compliance in audits "for any
quarter beginning after September 30, 1983,"
and Congress explicitly
made the 1984 Amendments effective on October 1,
1983. The
circumstances here are therefore distinguishable from those
in
Georgetown, where the agency published cost-limit rules for
Medicare
providers in 1984 and attempted to apply the rules to 1981 costs, in
the
absence of any statutory authority to do so. Here, the
statute
expressly made the change in the standard retroactive.
o In support of its argument that the
statutory language setting a
1984 effective date was not an express
authorization for retroactive
rulemaking, the State argued that the
legislative history of the 1984
Amendments shows that Congress intended that
OCSE's implementing
regulations would have prospective effect only.
Reply br. at 25-26.
The legislative history on which the State relied,
however, does not
refer to OCSE's implementation of the substantial
compliance standard;
instead, it refers to the expectation by Congress that
OCSE would issue
new regulations focusing on whether states were effectively
attaining
program objectives (in addition to meeting the existing state
plan
requirements). 15/ S. REP. No. 387, 98th Cong., 2d Sess. 32-33
(1984).
o The effect of the 1985 regulations here is also
significantly
different from the effect of the cost-limit rules considered
in
Georgetown. There, Medicare providers were entitled to a specific
level
of reimbursement under the regulations in effect in 1981, and the
1984
rules would have retroactively reduced that level. Here, the
AFDC
funding reduction applies to periods after the 1985 regulations
were
published.
o The audit criteria at issue here were
in the existing
regulations, had been in effect without substantial change
since 1976,
and were based on IV-D state plan requirements. 16/ The 75%
standard is
more lenient than the standard in the existing regulations,
which
provided that the State must "meet" the criteria. Even if the
State is
correct that OCSE could not reasonably have implemented this
by
requiring action in 100% of the cases, the existing regulations
clearly
contemplated a compliance level greater than 75%. 17/ Certainly
the
1984 Amendments evidence a strong intention to finally hold the
states
accountable for providing services 18/, and we see no basis in
the
language or history to support the State's apparent interpretation
that
Congress meant "substantial compliance" to be something less than
the
75% standard adopted by OCSE.
o More important, the 1985 regulations
afforded the State a
corrective action period. The State had notice of
the 75% standard
prior to this period, and more than a year to adjust its
administrative
practices before the follow-up review period began.
o The regulations here merely interpret the statutory term
"substantial
compliance." Obviously, the range of compliance levels
OCSE could adopt
is limited by this term, particularly when it is read
together with
section 403(h)(3) of the Act (which permits a finding of
substantial
compliance only when any noncompliance is of a technical
nature). A
level lower than 75% would have been subject to challenge
as
inconsistent with statutory intent.
o Since the 75% standard reasonably
interprets the statutory term
"substantial compliance," the circumstances
here are distinguishable
from those considered in Maryland, where the court
found that
regulations setting "tolerance levels" for AFDC
eligibility
determination errors were not reasonably related to the purposes
of the
statute. Moreover, unlike the "tolerance levels" in Maryland,
the 75%
standard here had an empirical basis in past performance levels
measured
through OCSE's audits. While audit results from FYs 1980 and
1981
showed that some states were not yet achieving 75% levels, other
states
were achieving 100% levels at that time (see appeal record, vol. VII
at
936-948), and OCSE could reasonably expect all states to be
achieving
75% levels by FY 1984. 19/ Moreover, while the State implied
that this
factual basis had to be published as an integral part of the
rulemaking
proceeding (State reply br. at 24), OCSE stated in the preamble
that
this level was based on its experience with past audits. 50 Fed.
Reg.
40121.
o Even in the absence of the 1985
regulations, we would reject the
State's position that it should be found to
meet the substantial
compliance standard because it failed only a single
criterion. The
record here supports a finding that, in a substantial
number of the
cases entrusted to it where paternity had not been established,
the
State did not take any action to provide this service.
Yet,
establishing paternity was absolutely essential to the overall
child
support program under Title IV-D of the Act. If the State's
argument
were accepted, then thousands of children would not receive
the
assistance Congress has provided them and the AFDC program
would
continue to bear costs that should be borne by the absent parent.
Thus,
we conclude that the State did not achieve substantial compliance
under
any reasonable reading of that term.
o Finally, we reject the State's
arguments based on section
403(h)(3) of the Act. That section permits
OCSE to find substantial
compliance only where any noncompliance is "of a
technical nature which
does not adversely affect the performance of the child
support
enforcement program." OCSE implemented this provision through
its
regulations, determining that failure to meet the
critical
service-related audit criteria in its regulations is not
simply
technical since the required activities are essential to an
effective
program. 50 Fed. Reg. at 40130. We find that
interpretation to be
reasonable as applied here since the State's failures
under a
service-related criterion would adversely affect program
performance;
the State took no action whatsoever to provide a basic child
support
service in a significant number of cases. 20/
For example, the
follow-up reviewers found that in two of the larger counties
visited,
average IV-D case worker case load was 3,459 per case worker.
State Ex.
Q, Att. at 5.
Thus, we conclude that application of the 1985 regulations here
was
clearly proper, and that those regulations are consistent with the
1984
Amendments.
III. The State's arguments concerning the validity of the
one-year
corrective action period must also be rejected.
The State also argued that OCSE's regulation limiting corrective
action
periods to one year was arbitrarily and capriciously established,
and
was inconsistent with the intent of the statute (Proposition
VIII).
Section 403(h)(2)(A) of the Act, as amended, provides that --
the reductions required under paragraph (1) shall be
suspended
for any quarter if-- (i) the State submits a corrective
action
plan, within a period prescribed by the Secretary
following
notice of the finding under paragraph (1), which contains
steps
necessary to achieve substantial compliance within a time
period
which the Secretary finds to be appropriate; . . . .
(Emphasis added.) The Secretary's implementing regulations at 45
C.F.R.
305.99(c) provide that:
The penalty will be suspended for a period not to exceed
one
year from the date of the notice [of failure to meet
the
substantial compliance standard] if the following conditions
are
met:
(1) Within 60 days of the date of the notice,
the State
submits a corrective action plan to the
appropriate Regional
Office which contains a
corrective action period not to
exceed one year from
the date of the notice and which
contains steps
necessary to achieve substantial compliance
with the
requirements of title IV-D of the Act; . . . .
(Emphasis added.) The State argued that OCSE's blanket
determination
that one year was an appropriate length of time for a
corrective action
period thwarted congressional intent that a state be given
sufficient
time to bring itself into compliance. According to the
State, the
legislative history indicated that a state was to be penalized
only if
it refused "to undertake the necessary changes to correct
that
situation." S. REP. No. 387, 98th Cong., 2d Sess. 33 (1984).
In
particular, the State argued that necessary program changes
requiring
increased expenditures could only be implemented through
state
legislative action, and that in its own case the legislature met
only
once a year. Appeal br. at 28-29.
In response to the State's contentions, OCSE argued that its
regulation
was reasonable and fully consistent with the statute and
legislative
history. OCSE contended that audits had been conducted
beginning with
the second quarter of FY 1977 and the State, through prior
audit
reports, "had been repeatedly warned of its program deficiencies and
was
admonished to take action to improve IV-D services." OCSE Br. at
25.
OCSE noted that the State was implementing a reduction in force of
its
attorneys at the same time it was developing its corrective action
plan.
OCSE Br. at 9; see also State Ex. O at 5. Thus, the State's
failure to
meet the standard during the one-year corrective action period (in
fact,
its performance declined) was attributable more to the State's
staffing
decisions than to the time available for corrective action.
OCSE
maintained that its regulation was entirely consistent with
the
legislative history of the 1984 Amendments.
We agree with OCSE that its conclusion that a one-year period
was
appropriate for corrective action was reasonable, particularly since
the
state plan requirements at issue here had been in effect as of
1976.
Moreover, most states had had notice of the particular shortcomings
of
their programs from earlier years' audits (see, e.g., New Mexico, n.
3).
In fact, the State did not dispute OCSE's assertion that for
fiscal
years 1980 and 1981, the State took action to establish paternity
in
only 28% and 14% of the applicable cases respectively. OCSE Br. at
29;
appeal record, vol. VII at 938 and 944. Although the State argued
that
these previous warnings were ineffectual because there was never
any
penalty attached, the State at least knew as of April 1986 (17
months
before its corrective action period began in September 1987),
that
OCSE's preliminary audit had found the State's paternity
establishment
services to be far below the regulatory standard and that a
penalty was
all but imminent. State Ex. I. Under these
circumstances, the State's
reduction in staffing could reasonably be said to
be a refusal to
undertake the necessary changes to correct its
deficiency.
Given OCSE's responsibility to audit compliance for 54 jurisdictions,
its
decision to adopt an outer limit for the appropriate corrective
action period
is well within its discretion. As we noted above, the
1984 Amendments
certainly evidence a strong intention to finally hold
the states accountable
for providing services. The Senate Finance
Committee Report for the
1984 Amendments, which the State quoted out of
context, said:
In view of the changes proposed . . ., the penalty provisions
of
the law will apply only in cases where States not only
fail
substantially to carry out the requirements of the law but
also
refuse to undertake the necessary changes to correct
that
situation. For this reason, the Committee cannot foresee
any
situation in which legislative action to suspend these
revised
penalties would be appropriate.
S. REP. No. 387, 98th Cong., 2d Sess. 33 (1984).
We therefore reject the State's contention that the regulation
is
arbitrary, capricious or contrary to statutory intent in setting
a
one-year corrective action period. We would be bound by it in
any
event.
IV. The State's statistical sampling arguments do not provide a
basis
for overturning the funding reduction.
We next turn to the State's arguments about OCSE's statistical
sampling
methodologies. In both the program results audit and the
follow-up
review, OCSE evaluated through the use of statistical
sampling
techniques the audit criteria which required that the State be using
its
procedures in 75% of its Title IV-D cases. The State did not
challenge
the audit findings in individual sample cases, nor did the
State
challenge generally the use of statistical sampling methods as a
basis
for a decision here. The State raised two issues, however, with
respect
to the specific sampling methods OCSE used.
The first issue related to the fact that OCSE had used in the
follow-up
review a method called "systematic random sampling." The
Board has in
other cases upheld the use of this method in the absence of any
showing
that drawing the sample cases in a systematic way introduced a bias
into
the sample results. New Mexico at 17-18. The Board thus
asked the
State to address whether there was any basis in the data sampled
for the
follow-up review for finding that there was any bias of the
type
recognized by statisticians for invalidating a systematic random
sample.
1/31/91 Questions to Parties, Question No. 9. The State failed
to
respond to this question, and presented no evidence of such a bias.
The
State raised no other contention regarding the sampling methodology
used
in the follow-up review. Thus, we affirm the findings of that
review
without further discussion.
The second sampling issue related to the question of how to
evaluate,
using statistically valid methods, the data from the sample used in
the
program results audit. Below, we first discuss the sampling
procedures
and results for this audit. We then explain how the issues
have
developed and narrowed during the course of Board proceedings. We
then
discuss the evidence presented and explain why we find that the
State
did not achieve the appropriate standard for the establishing
paternity
criterion.
A. The sampling procedures and results
The following facts are undisputed.
In the program results audit, OCSE drew a stratified, cluster sample.
The
State had ten political subdivisions (referred to as either regional
offices
or counties). The sample was divided into 6 strata: strata 1,
2,
and 6 each consisted of one political subdivision; strata 3 and 4
each had
two political subdivisions; and stratum 5 had three
political
subdivisions. OCSE randomly drew one political subdivision
from each of
the three strata with more than one political subdivision and
then drew
a number of Title IV-D cases from that political subdivision, with
a
probability proportional to the size of the stratum.
After selecting the sample, OCSE examined each sample case to
determine
what action, if any, was required in the case (in other words,
which
audit criteria applied). For example, if the whereabouts of an
absent
parent who owed support was unknown, the case would be classified as
a
"locate" case, requiring review to see if the State took any action
to
locate the absent parent, as required by 45 C.F.R. 305.33(g). OCSE
then
examined the case files and other records to determine whether the
State
had, in fact, taken any required action during the relevant time
period,
finding either "action" or "no action" for each sample case
reviewed.
For the key criterion at issue here, OCSE found that out of 131
cases
requiring action to establish paternity, the State had taken action
in
only 62 cases.
OCSE then used the sample findings to calculate an "efficiency rate"
and
an "efficiency range" for each criterion. The "efficiency rate" is
the
single most likely estimate of the percentage of cases requiring
review
under the audit criterion which were "action" cases. The
"efficiency
range" is what is known in statistical parlance as a
"confidence
interval." A confidence interval is a statistician's
calculation of the
range of values within which the statistician can say with
a specified
degree of certainty the true value occurs.
Under OCSE's audit procedures, a criterion was considered "unmet" if
the
"high range" of the "efficiency range" was less than 75% and
only
"marginally met" if the "high range" was 75% to 80%. The "high
range"
figure is also referred to as the "upper limit" or "upper bound" of
the
confidence interval. The effect of OCSE using this figure is that
OCSE
assumes the risk associated with potential sampling error. (In
other
words, there is more risk that OCSE will pass a state which in
fact
failed than that it will fail a state which in fact passed.)
OCSE calculated the upper limit of the confidence interval by
multiplying
the "standard error" of the sample times 1.96, and adding
the result to the
efficiency rate. (As discussed below, 1.96 is a
multiplier which is
usually used to give a two-sided 95% confidence
interval. One of the
issues was whether use of this multiplier was
appropriate here.)
In a FY 1984 program results audit, OCSE examines all audit
criteria
listed in section 305.20(a) of the 1985 regulations. In a
follow-up
review, however, OCSE examines only those criteria which are
either
"unmet" or which are only "marginally met" in the program results
audit.
A state is subject to a penalty after a corrective action period if
the
state fails in the follow-up review to meet any criterion which
was
"unmet" during the original audit period or to maintain compliance
with
any criterion which was only "marginally met." See 45
C.F.R.
305.100(a).
Here, OCSE found that the State failed to meet only one
criterion,
"establishing paternity," in the follow-up review. The State
had failed
this criterion and another criterion, "reports and maintenance
of
records" under 45 C.F.R. 305.35, in the program results audit. Thus,
if
we find the State either failed to meet or only "marginally met" (75%
to
80%) the "establishing paternity" criterion in the program
results
audit, we would find that OCSE would have properly included
that
criterion in the follow-up review. This is important because
the
State's evidence on the statistical sampling method presumed that
the
State only had to meet a 75% standard. In fact, as we pointed out
to
the State prior to the hearing (1/15/91 Notice of Hearing at 3-4),
if
the State achieved only 80%, it still would have been subject to
review
for this criterion in the follow-up review, and a penalty would
be
appropriate since the State failed this criterion in the
follow-up
review (achieving only 28% as a raw score, leading the reviewers
to
conclude with 99.9% confidence that the 75% standard had not
been
achieved. See State Ex. P, att. at 4.) See 45 C.F.R.
305.100.
B. How the issues developed
The State did not contest OCSE's findings in individual sample cases
from
the program results audit. Moreover, the State did not
initially
challenge the sampling methodology. As a result of
developments in the
related Title IV-D cases, however, OCSE recognized that
it had
calculated the "efficiency rate" and "efficiency range" for
the
"establishing paternity" criterion as though a simple random sample
had
been drawn, rather than the stratified, cluster sample that was in
fact
drawn. In an effort to correct this mistake, OCSE developed
several
alternative methodologies, partly in response to comments by
a
statistical sampling expert who had appeared for several other
states
with similar cases and who ultimately appeared as a witness
for
Mississippi. (OCSE has stipulated that this witness qualifies as
an
expert, and we refer to him as "the State's expert.")
Under any of OCSE's methodologies, the State failed to meet the
75%
standard (and therefore did not have more than 80% compliance).
See
OCSE 11/21/90 Submission, att. at Ex. 3.
The percentage figure in question here may be expressed as a ratio of
the
number of cases in which action was taken to establish paternity to
the
number of cases in which such action was required. It is
undisputed
that there are two generally accepted methods for estimating a
ratio
from the results of a stratified sample and calculating the
confidence
interval associated with it. The "separate ratio estimator"
method uses
the ratios estimated for each stratum and then calculates an
overall
estimate by weighing these ratios according to the relevant
strata
population. The "combined ratio estimator" method pools the
total
number of cases where action was taken; pools the total number of
cases
where action was required; and takes an average of the ratio of
these
numbers. Tr. I at 63; 1 M. Hansen, Stratified Simple Random
Sampling
189 (1953 ed.).
The State's expert had originally suggested use of the separate
ratio
estimator method (see Tr. I at 88), and one of OCSE's
methodologies
(Methodology #3) is based on this method. In using this
method, OCSE
"collapsed" each of the strata with more than one political
subdivision
(strata 3, 4, and 5) in order to calculate what is called a
"within
strata variance." 21/ (The term "variance" is used for the
square of
the "standard error." The formulas at issue here calculate
the
variance, then take the square root in order to obtain the
"standard
error," which is necessary in order to calculate the
confidence
interval.) The State's expert then raised a number of
questions about
OCSE Methodology #3. While he had acknowledged in other
cases that OCSE
was "moving in the right direction" with this method, the
State's expert
said further refinements were needed because the particular
sample drawn
here had complications not addressed in the statistical
treatises. He
had also expressed his opinion that until a "correct
analysis" was done,
one could not say with a 95% degree of certainty that the
State had
failed. Ohio at 15; see State 12/4/90 submission, att.
Rather than having a series of counter affidavits from the
parties'
experts, as we had in some of the related cases, the Board set a
hearing
here, limited to the sampling issues. In its notice of hearing,
the
Board specifically noted that the State would have a burden to do
more
than challenge OCSE's method, given that OCSE had presented an
expert's
affidavit asserting it was reliable; the Board clearly indicated
that
the State should present what it considered a "correct analysis" or
the
Board would presume that, even with modifications to OCSE
Methodology
#3, the State would not achieve the correct standard (which we
noted was
not 75% but over 80%). 1/15/91 Notice of Hearing at 2-3.
At the hearing, the State's expert did not present any
alternative
methodology. Thus, the State failed to show that there was
any
statistically valid analysis of the sample data which would
indicate
that the State had passed the review. The State's expert did,
however,
explain more fully what his concerns were about the use,
without
refinements for the particular sample here, of the "separate
ratio
estimator" method. OCSE presented some evidence that
statisticians
would ordinarily rely on use of Methodology #3, as well as some
rebuttal
to the concerns expressed by the State's expert regarding the
"separate
ratio estimator" as applied here.
The validity of OCSE's Methodology #3 largely became a moot question
in
this proceeding, however. 22/ At the hearing, OCSE also presented
an
opinion, by a highly qualified expert (OCSE's expert), that use of
a
"combined ratio estimator" was more appropriate here than the
"separate
ratio estimator" (Tr. I at 113-115), and the State's expert
ultimately
agreed with this opinion. Transcript of 6/18/91 Hearing (Tr.
II) at 20.
23/ The results of using the "combined ratio estimator" were
presented
as OCSE Methodology #4. The State asked for and was given
time to have
its expert more carefully examine OCSE Methodology #4, and we
continued
the hearing on a later date.
Before the continued hearing, OCSE made a slight modification to
its
calculations to address one problem raised by the State's
expert
regarding Methodology #4. The State's expert agreed that
this
modification eliminated this concern. Tr. II at 22. He also
stated
that he did not question the mathematical accuracy of
OCSE's
recalculations. Tr. II at 20. The recalculated "efficiency
rate" for
the "establishing paternity" criterion is 48.9% and the
recalculated
upper limit of the confidence interval is 69.4%. OCSE
5/23/91
submission, att. 2 (Tepping comments) at 1. 24/
The remaining concerns expressed by the State's expert all went to
the
issue of whether an "exact" two-sided 95% confidence interval had
been
attained in Methodology #4, as modified. Tr. II at 22-23.
The State's
expert offered no further modifications to the methodology which
would
show that the State had achieved 75% (much less over 80%, as
required
here), nor did he express any opinion that the State had achieved
75%.
OCSE's expert, on the other hand, expressed his professional
opinion
that it was "extremely unlikely" that Mississippi did have a
compliance
rate of 75% or higher. Tr. I at 147; Tr. II at 51.
This evidence
would, by itself, be a sufficient basis on which we could
uphold the
penalty since (1) OCSE's expert was highly qualified in dealing
with
complicated statistical data (see OCSE 3/13/91 letter, att.); (2)
we
found him to be both knowledgeable and credible; and (3) while
the
State's expert was also highly qualified (see appeal record, Vol. IX
at
1057-1067), he did not appear as experienced in such
complicated
sampling situations and, in any event, expressed no opinion to
the
contrary.
We nonetheless proceed to address the concerns expressed by the
State's
expert, because we find that they are based on an erroneous premise
and
that OCSE's evidence rebutted those concerns. Thus, our findings
are
based not only on OCSE's expert's opinion that the State did not
achieve
the 75% standard, but on the evidence taken as a whole.
C. Analysis
1. OCSE did not adopt an "exact" confidence interval.
The concerns expressed by the State's expert were all premised on
the
State's view that OCSE had adopted a rule that it would pass a
state
unless an "exact" 95% confidence interval could be calculated
showing
that the state had not achieved the 75% standard. Tr. II at 22-
23. In
Ohio, we noted that OCSE had presented evidence that
statistical
validity does not depend on use of a 95% confidence
interval. We
further said that it would appear to be impermissibly
arbitrary for OCSE
(once having adopted the 95% level) to use it for some
states and not
for others. Ohio at 17, n. 8. The State's position
that OCSE had a
rule requiring use of an exact 95% confidence interval was
partly based
on this statement in Ohio.
In Ohio, however, it was undisputed that the upper limit of a
95%
confidence interval would be obtained by multiplying the
"standard
error" by 1.96 and adding that amount to the efficiency rate.
Ohio at
10. The Board had concluded that OCSE had adopted use of the
95%
confidence interval from OCSE's Program Results Audit Guide.
That
Guide, however, does not specifically say that OCSE would use the
upper
limit of a 95% confidence interval. Rather, it simply sets out
a
formula for calculating the confidence interval which uses
the
multiplier 1.96. Appeal record, Vol. IV at 607. The Board
equated this
with the 95% confidence interval because the experts in the
Ohio
proceeding did.
The State's expert said here, however, that the multiplier 1.96 does
not
always give an exact 95% confidence interval. He pointed out that
some
courts had required a 95% degree of confidence in order to find
sample
results reliable. He did not dispute OCSE's expert testimony
(see Tr.
II at 49) that 1.96 is a standard amount generally used for samples
with
a size of more than 30 units, and is a multiplier which gives
a
"nominal" 95% confidence interval. The State's expert said,
however,
that he had found one article where the authors suggested a
multiplier
of 2.00 even though the sample size was more than 500 units.
When the
Board pointed out that multiplying the standard error by 2.00
rather
than 1.96 would not make a difference here (the upper limit of
the
confidence interval would still be less than 75%), the State's
expert
said that the situation in the article he had referred to was
different
from that at issue here since the article was discussing a simple
random
sample rather than a stratified sample. Tr. II at 35.
In any event, we find that OCSE's use of the multiplier 1.96 to achieve
a
nominal 95% confidence interval is reasonable here. First,
OCSE's
expert testified that it is impossible to get an "exact" figure, as
the
State's expert was demanding (Tr. II at 49), and the State's
expert
essentially conceded this. Tr. II at 26. Use of the 1.96
multiplier is
"conventional" and statisticians would normally rely on the
nominal
confidence interval obtained. Tr. II at 54. Indeed,
OCSE's policy was
to multiply the "standard error" by 1.96 and even the
State's expert had
initially assumed that this was done to establish a 95%
confidence
interval.
Moreover, the 1.96 is the standard multiplier ordinarily used for a
95%
confidence interval which is two-sided. What this means is that
there
is only 2.5% probability that the true value is greater than the
upper
limit. (This is graphically illustrated in State's figures 1 and
2
discussed at the continued hearing. State's Hearing Ex. 4, figure
1.)
In other words, there is 97.5% certainty that the true value is
not
greater than the calculated upper limit. In light of this, OCSE
could
reasonably adopt the 1.96 multiplier, even if it might not in
every
instance achieve an exact two-sided 95% confidence interval.
We also note that the State's expert's concern with use of the 1.96
was
partly based on his concern about the sample size here. This
concern
was based on the fact that not all of the sample cases required
review
for the criterion at issue here, and that, in some strata, the number
of
cases requiring review was small. (For a graphic illustration of
the
breakdown of the sample, see State's Hearing Ex. 1.) OCSE's
expert
testified that, in the combined ratio estimator method, the
effective
sample size for calculating both the numerator and denominator of
the
ratio was the total sample size of 501. Tr. I at 120-121.
This is
consistent with the formulas, and the State's expert did not
directly
dispute it. 25/
Finally, the State's expert's opinion that use of the 1.96 might not
be
appropriate was partly based on his concern about whether there was
a
"normal distribution" here. Tr. II at 13. As we discuss next,
OCSE
showed that this concern was not warranted.
Thus, we conclude that OCSE was not required to determine an
"exact"
confidence interval and that, in any event, use of the 1.96
multiplier
was appropriate.
2. The distribution is not a concern.
OCSE's expert acknowledged that, for the 1.96 to produce an
exact
confidence interval, one would have to know, among other things,
that
the estimated efficiency rate has a "normal distribution" over
all
possible samples. Tr. II at 49. OCSE's expert presented an
analysis
showing that the distribution here was in fact close enough to
normal to
act as if it were. OCSE Hearing Ex. D; see also Tr. II at
64-65.
The State's expert did not question the validity of this analysis,
and
indeed acknowledged that it appeared from this analysis that
the
efficiency rates were normally distributed. Tr. II at 72. The
State's
expert said, however, that this evidence was undercut by the results
of
some "simulation" experiments OCSE's expert had performed. As
we
discuss next, we find no merit to this argument, and therefore
conclude
that the State's expert's concern about the distribution does
not
indicate that the results of Methodology #4 are unreliable.
E. The simulation experiments support the result.
In order to verify the validity of the methodology here, OCSE's
expert
conducted some simulation experiments. He described these
experiments
as follows:
Each experiment consisted of assuming that the true
efficiency
rate is 75% and assuming varying efficiency rates for
primary
sampling units in the several strata. Having defined such
a
hypothetical population, a sample using the sample
sizes
actually used was drawn and the estimated efficiency rate
and
the upper bound were computed. I repeated the process
1000
times for the assumed population. In addition I defined
two
other hypothetical populations with a 75% efficiency rate
and
drew 1000 replicates of the sample from each. I found that,
for
one population, only 33 of the 1000 replicates produced
upper
bounds below 75% and that, for each of the other
two
hypothetical populations, only 44 of the 1000
replicates
produced upper bounds below 75%. . . . Moreover, in no
instance
of the 3000 replicates was the upper bound as low as
that
calculated from the Mississippi audit survey for the
Paternity
criterion.
OCSE 5/23/91 submission, 2d att. at 2.
The State's expert noted that, if there were normal distribution,
one
would expect that there is only a 2.5% probability that the true
value
is greater than the upper limit calculated from the sample. He
said
that the simulation experiments undercut this assumption because
they
showed either a 3.3% probability (33 out of 1000 replicates) or a
4.4%
probability (44 out of 1000 replicates).
The assumption of a normal distribution here means that we are saying
that
there is only a 2.5% probability that the State achieved a rate
higher than
the upper limit of 69.4%. The results of the simulation
experiments
show only that the methods of analysis used by OCSE could
result in nominal
upper limits lower than 75% (even if that were the
true value) as many as 44
out of 1000 times (or, 4.4% of the time).
Thus, we do not view the simulation
results as undercutting the
assumption that there is only a 2.5% probability
that the State in fact
achieved higher than 69.4%. Even if the State's
expert is correct that
the simulation results somehow indicate the
distribution is not normal,
other evidence admittedly does indicate such a
distribution, and the
State's expert merely said the simulation results
raised a concern about
this. He did not say that the simulation
definitively resolved this
issue.
Moreover, OCSE's expert testified that the significant result of
his
simulations was that in no instance in all 3000 replicates was the
upper
limit as low as 69.4% and that, in his opinion, this made it
"extremely
unlikely" that the State in fact achieved 75%. Tr. II at
47. The
State's expert simply said that he could express no opinion on
this.
Tr. II, pp. 76-77.
Thus, we conclude that the simulation experiments support a finding
that
the State did not achieve 75%, rather than undercutting the result
here.
3. The other concerns also lack merit.
Another concern about the combined ratio estimator which the
State's
expert had expressed had to do with whether Methodology #4
sufficiently
took account of the "within strata" variances.
Subsequently, he dropped
this as a "substantial concern." Appeal
record, Vol. XI at 1072.
We fail to see how it is a concern at all. OCSE's expert responded
that
the variance calculated not only contains a "within strata"
component
but, in addition, there is a variance contribution from the
"collapsed"
stratum (original strata 3, 4, and 5). Thus, he said that
the estimator
is expected to provide an overestimate of the true variance
(which would
favor the State). OCSE 5/23/91 submission, att. 2; Tr. I
at 116,
128-129. The State's expert agreed that, looked at
mathematically, the
formula tends to overestimate. Tr. II at 36-37.
The State's expert also noted that in calculating the
variance,
Methodology #4 used an approximation. OCSE's expert testified
that he
had used what is called the "Taylor's series" approximation and that,
in
his view, this was a "pretty good" approximation. Tr. II at
50. The
State's expert agreed that any error in this approximation "in
and of
itself, was tolerable." Tr. II at 40.
He then explained that his concern was that more than one
approximation
was being used here and that he was not sure of the cumulative
effect.
He did not dispute OCSE's expert's assertion (Tr. II at 50),
however,
that the other approximations were both phases of the
same
approximation, depending on the assumption that we have a
normal
distribution and that the numbers would act as if we had drawn a
simple
random sample from a normal distribution. As discussed above,
there is
evidence that these assumptions are warranted here.
Moreover, the State's expert agreed that it was a matter of
judgment
whether to tolerate any approximations in calculating the
confidence
interval, but suggested a conservative outlook given the
consequences
for the State here. Tr. II at 26. In our view, OCSE
did take a very
conservative approach on the whole here and any error in
calculating the
upper limit most likely favored the State (due to the
overestimation of
the variance from collapsing the strata and the bias in the
estimator in
favor of the State). In any event, we think any such error
would be
immaterial. The upper limit OCSE relied on was 69.4%.
The State would
have had to achieve 75% to meet the criterion and over 80% to
have more
than marginally met the criterion, and there is absolutely no
evidence
which would support an affirmative conclusion here that the
State
achieved either level.
In sum, we find that the overwhelming weight of the evidence here is
that
the State did not achieve a 75% standard for the "establishing
paternity"
criterion in the program results audit period. In our view,
we can say
this with at least a 95% degree of certainty, and can say
with an even higher
degree of certainty that the State did not achieve
over 80%, which it would
have to achieve in order not to have this
criterion subject to review after
the corrective action period. 26/
Since the State also failed to achieve 75%
in the follow-up review, OCSE
properly imposed a funding reduction.
Conclusion
For the reasons stated above, we uphold OCSE's decision to reduce by
one
percent the State's AFDC funding for the one-year period beginning
July
1, 1987.
_____________________________ Judith A. Ballard
_____________________________ Donald F. Garrett
_____________________________ Alexander G. Teitz Presiding
Board
Member. 1. The Mississippi agency responsible for the
Title
IV-D program changed its name from Department of Public Welfare
to
Department of Human Services while this appeal was pending.
2. The State alleged that the statute authorized OCSE to do an
annual
audit only if a previous penalty had been imposed. See, e.g.,
Reply
brief (br.) at 5-6. This is erroneous -- the statute provides
for
audits to be conducted "not less often" than once every three
years,
unless there has been a penalty or corrective action period
implemented,
in which case an annual audit is required. Section
452(a)(4) of the
Act. The purpose of the provision was to give the
Secretary more
flexibility in allocating audit resources than had existed
before, when
annual audits of every jurisdiction were absolutely
required. See
Section 452(a)(4) of the Act (before the 1984
Amendments). Thus, OCSE
had discretion to audit this State for FY
1984.
3. The State noted that in its October 1984 notice of
proposed
rulemaking OCSE had assured states that it would continue to
apply
current audit regulations for fiscal years beginning prior to FY
1985;
the State argued that that assurance was violated by allegedly
"adding"
criteria. Initial br. at 30 (Proposition IX). As is
obvious from the
text, we do not agree with this view and, since the State
did not
explain its interpretation after having received notice of our
contrary
interpretation from other Title IV-D decisions (see n. 14 below), we
do
not discuss it further here.
4. This was not the State's first notice of problems with its
program.
OCSE began compliance audits in 1977, and the State did not
dispute
OCSE's claim that the State "had been repeatedly warned of its
program
deficiencies and was admonished to take action to improve Title
IV-D
services." OCSE Br. at 25. The State also did not dispute
that in FYs
1980 and 1981 its rough scores for the establishing paternity
criterion
were 28% and 14% respectively. See Appeal record vol. VII at
938, 944.
No penalty was previously imposed, however, due to
congressional
moratoria.
5. We note that the auditors examined whether any efforts were
made by
the State in these cases to furnish these services. The success
of
these efforts, while noted for statistical purposes, was
not
determinative as to whether the State was found to be in
substantial
compliance; the State received credit for "action" so long as it
took
some action towards these goals. See 50 Fed. Reg. at 40132.
6. We are obliged to adopt the State's labelling of its
arguments
since it failed to number the pages in several of its briefs.
See,
e.g., 1/23/90 Ruling at 2.
7. Proposition XII stated that an evidentiary hearing should be
held
in this case. The State ultimately withdrew this request (10/12/90
br.
at Proposition XII), but then the Board convened a hearing on its
own
motion to hear testimony on the statistical calculation issue
discussed
in Section IV of this decision (see Board's 12/6/90 and 1/15/91
letters
to parties).
8. Furthermore, even though the State had filed an appeal,
OCSE
retained the option of settling the case, i.e., reconsidering pursuit
of
this matter. In fact, the record shows that the Board gave the
parties
several extensions of time before requiring briefing in order to
permit
them to pursue possible settlement. See, e.g., 4/27/89 Letter
to
Parties.
9. The State also raised for the first time in its reply brief
the
question whether this proceeding was a "plan conformity matter"
rather
than a disallowance case, in which case it should be remanded to
OCSE
under regulations applicable to such issues. See 45 C.F.R.
201.6.
Introducing this issue violated the Presiding Board Member's 2/17/89
and
6/12/89 rulings that, since the State was given an extraordinarily
long
time to prepare its initial brief, all arguments not raised in
that
brief were waived. In any event, the issue here is not whether
the
State's child support enforcement plan met federal requirements,
but
whether its implementation of that plan was so faulty that a
penalty
disallowance should be imposed.
10. The State asked for en banc reconsideration of this
issue. The
Board's regulations clearly put jurisdictional questions
under the
authority of the Chair. See 45 C.F.R. 16.7(b) and Part 16,
App. A,
section G. There is no provision in the Board's regulations for
en banc
reconsideration. The three Board members assigned to this case
agree
with the Chair's ruling.
11. The State asked for similar remedies under its
Proposition III,
in which it complained of a violation of "due process" due
to OCSE's
notification after the disallowance letter that interest would
be
charged. (We had previously noted to the State (in our 1/23/90
Ruling
at n.2) that "due process" per se does not apply to a state.
South
Carolina v. Katzenbach, 383 U.S. 301 (1966)). The interest
dispute was
evidently settled (appeal record, Vol. VII at 956-959), however,
so we
have no controversy before us to resolve on this point (nor,
as
discussed above, do we need to address the State's request for
a
regulatory amendment made in this fashion).
12. The Presiding Board Member noted in his 10/11/89 ruling that
the
appropriate forum for the State to pursue its appeal of this FOIA
matter
was before a federal court. The State did file such an appeal,
which
was apparently settled by the parties. Reply br. at 24;
State's
10/12/90 br. at Proposition IV. We do not know the terms of
the
settlement, but the State always could have sought and been
granted
permission to supplement the record with any documents obtained
through
FOIA. 45 C.F.R. 16.13.
13. The State argued (Proposition V) that the Board had
jurisdiction
to apply Georgetown in this case, even though that decision was
issued
ten days after the disallowance at issue here. Appeal br. at
20. OCSE
did not argue that the fact that Georgetown was issued after
the
disallowance precluded us from applying it, and, in any event, since
we
find Georgetown distinguishable, the State's argument is not material.
14. Our conclusion here closely parallels our analysis of
virtually
identical arguments made by the parties in Board decisions on
other
Title IV-D appeals: Ohio Dept. of Human Services, DAB No. 1202
(1990);
Oklahoma Dept. of Human Services, DAB No. 1223 (1991); New Mexico
Human
Services Dept., DAB No. 1224 (1991); District of Columbia Dept. of
Human
Services, DAB No. 1228 (1991); Arizona Dept. of Economic Security,
DAB
No. 1255 (1991). Copies of all these decisions were furnished to
the
parties in this case for comment on any issues that were
applicable.
Neither party filed any such comments.
15. The 1985 regulations added new performance-related indicators
for
use beginning with the FY 1988 audit period.
16. The State seemed to argue that since there had previously
been no
teeth to the statute, due to the moratoria, there was no
legal
obligation for the State to improve its performance (Reply br. at
5),
and that the compliance standard was 0%, not 100% (Reply br. at
28-29).
The moratoria only delayed imposition of the penalty, however.
Congress
did not repeal this requirement or guarantee that no penalty would
ever
be imposed. Moreover, the statute at all times required the State
to
have in effect a child support enforcement plan, as a condition
for
receiving AFDC payments and as a condition for receiving funding
under
Title IV-D. Sections 402(a)(27) and 455(a) of the Act (1983).
17. The existing regulations required the states to have and
be
utilizing written procedures detailing step by step actions to be
taken.
45 C.F.R. 305.1, 305.24; 45 C.F.R. Part 303 (1983). Although
no
reduction had actually been imposed based on the existing
audit
criteria, this was due to the moratoria.
18. The Senate Finance Committee Report stated:
In view of the changes proposed . . ., the penalty provisions
of
the law will apply only in cases where States not only
fail
substantially to carry out the requirements of the law but
also
refuse to undertake the necessary changes to correct
that
situation. For this reason, the Committee cannot foresee
any
situation in which legislative action to suspend these
revised
penalties would be appropriate.
S. REP. No. 387, 98th Cong., 2d Sess. 33 (1984).
19. We note that the percentages given in the draft analyses by
OCSE
of 1980 and 1981 audit results (appeal record, Vol. VII at 936-948)
are
derived simply by dividing the number of complying sample cases by
the
total number reviewed. If OCSE had instead used the same method
for
estimating compliance levels it used in the 1984 and 1985 audits for
all
states (see our discussion below), the compliance percentages shown
on
the draft analyses for the earlier years would have been
higher.
Moreover, in Maryland, the Secretary had acknowledged that some
errors
in making eligibility determinations were unavoidable due to the
complex
nature of the requirements. Here, the State did not argue that
the
service-related requirements were complex or that there was any
barrier
to meeting those requirements which could not be overcome.
20. The State suggested that, if it were permitted
reconsideration by
OCSE, it might show that its violations were of a
"technical nature,"
and it speculated on some possible circumstances that
would fit that
definition. Appeal br. at 25. However, the State
never offered any
evidence that any of the cases found to be "no action" in
either review
met these hypothetical circumstances. The State also
maintained that
"the mere placing of a piece of paper in the case files
pulled in a year
already completed, was a purely mechanical activity in
paternity
establishment cases and amounted to failures of a technical
nature."
Reply br. at 38. We are not certain what the State meant by
this. If
the State meant that it had, in fact, taken action in the
sample cases
and simply failed to document it in the case files, this is a
mere
assertion for which the State provided no support. The State
should
have been aware, moreover, from its reading of the District of
Columbia
case, DAB No. 1228, that OCSE did not strictly require a piece of
paper
in each file, but has been willing to accept as evidence of
action,
reliable computer lists which include a case. If, on the other
hand,
the State meant that OCSE would accept the placing of a piece of
paper
in a file as an "action," this is not true, and, in any event, the
State
can hardly be heard to argue that failure to do even this much
amounted
to violations of a technical nature.
21. While it would have been better from the standpoint of
obtaining a
smaller sampling error if more than one political subdivision had
been
selected, apparently this was not feasible under the
circumstances. Tr.
I at 115-118. As discussed below, however, the
expected effect of
collapsing the strata favored the State.
22. As a result, we do not address the validity of OCSE
Methodology #3
here. This does not, however, undercut our decisions in
Ohio and New
Mexico where OCSE had relied on Methodology #3. In those
cases, we
found that the States' evidence was insufficient to rebut
OCSE's
evidence about the validity of the method and that, in any event,
the
States had not shown that any further refinements would change
the
ultimate results. We also found that further refinements were
unlikely
to change the ultimate results given the raw sample data. We
note that,
despite the State's expert's implication that adjustments or a
more
precise alternative to OCSE Methodology #3 would favor the States,
the
separate ratio estimator method actually resulted in a higher
upper
limit for Mississippi than the combined ratio estimator method.
23. The State's expert originally determined that the combined
ratio
estimator was not appropriate here because the ratios were not
constant
from stratum to stratum. Tr. I at 81-82. OCSE's expert
stated a strong
opinion that this was erroneous. Tr. I at
114-115. The State's expert
ultimately dropped this objection.
24. Even the State's expert described the efficiency rate (48.9%)
as
the "best guess of the true efficiency rate." Tr. II at 10.
While he
originally responded to OCSE's expert's presentation of OCSE
Methodology
#4 by indicating that there might be a question about whether
the
estimate was biased (but noting that OCSE's expert had determined it
was
not), the State's expert did not indicate any concern about this
after
further consideration. See Tr. I at 195; see also Tr. I at
161. OCSE's
expert said the estimator was biased 5% in favor of the
State. Tr. II
at 161.
25. If we agreed with the State's expert that the sample size was
too
small here, the appropriate result would be to remand to OCSE
to
increase the sample size (and therefore the reliability of the
results).
Yet, OCSE's expert testified that it would take a "considerable
increase
in the sample size" to get a better estimate and that he did not see
how
this would benefit the State, given that an increase would both
reduce
the length of the confidence interval and the bias of the
estimate
(which he said was 5% in favor of the State). Tr. I at
160-161. The
State's expert did not state an opinion to the contrary
and acknowledged
that the smaller the sample size the greater the confidence
interval.
Tr. I at 52. While the State said it was willing to bear the
burden of
an increase in its costs for OCSE's review of an increased sample
size,
we think that a remand here would be a total waste of time and money
for
both parties.
26. Before the hearing, we suggested that the State had a burden
to
show that its alleged "correct analysis" would make a difference
here.
The State questioned whether it should have this burden. We need
not
decide, however, whether the State had to meet such a burden, since
we
find that, even if OCSE properly had the burden to establish
the
reliability of its results, it clearly met that
burden