Department of Health and Human Services
DEPARTMENTAL APPEALS BOARD
Appellate Division
SUBJECT: Kentucky Department for Medicaid Services
DATE: July 31, 1995
Docket No. A-95-68
Control No. KY-95-001-MAP
Decision No. 1524
DECISION
The Kentucky Department for Medicaid Services (Kentucky)
appealed a determination of the Health Care Financing
Administration (HCFA) disallowing $4,528,066 in federal
financial participation (FFP) for the State fiscal year
(SFY) 1993. The disallowance was based on HCFA's
calculation that Kentucky had collected $10,985,385
(related FFP of $7,906,182) more in provider tax receipts
than permitted by the Medicaid Voluntary Contribution and
Provider-Specific Tax Amendments of 1991, Public Law No.
102-234 (1991 Amendments) and implementing regulations,
but had returned only $3,378,116 of the FFP due.
Kentucky argued that HCFA's method of calculating the
limit on tax receipts was inconsistent with the statute
and based on regulations which were ambiguous and
unclear. The statute permitted states to continue to
collect otherwise impermissible provider-specific taxes
during a transition period without reduction in FFP up to
a cap determined by a statutory formula. The issue here
is what Medicaid expenditures are included in part of the
formula (referred to by the parties as the multiplication
factor). HCFA contended that the expenditures include
only medical assistance expenditures, whereas Kentucky
argued that Congress intended to include all amounts
expended under the State plan, including both
administrative costs and medical assistance expenditures.
For the reasons explained below, we conclude that HCFA's
regulation reasonably interpreted the language
establishing the multiplication factor to include only
medical assistance expenditures, that Kentucky had timely
and adequate notice of HCFA's interpretation, and that
Kentucky did not establish that it reasonably relied on
any contrary interpretation.
Background
The use of health care related taxes (and provider
donations) as a revenue source by states to fund the non-
federal share of Medicaid costs has been a contentious
subject for a number of years. The 1991 Amendments
restricted the use of these sources of funding by states.
Recognizing that the abrupt loss of this source of
funding would create problems for states which had come
to rely heavily on it, the 1991 Amendments contained a
formula permitting collection of certain health care
related taxes up to a cap during a transition period for
SFYs beginning after January 1, 1992 and before October
1, 1995. The formula for calculating the annual cap is
codified at section 1903(w) of the Act. Section
1903(w)(5)(A) sets out the calculation of the
multiplication factor, as follows:
For purposes of this subsection, the limit under
this subparagraph with respect to a State is an
amount equal to 25 percent (or, if greater, the
State base percentage as defined in subparagraph
(b)) of the non-Federal share of the total amount
expended under the State plan during a State fiscal
year (or portion thereof), as it would be determined
pursuant to paragraph (1)(A) without regard to
paragraph (1)(A)(iv).
Paragraph (1)(A) of section 1903(w) provides that "the
total amount expended during such fiscal year as medical
assistance under the State plan (as determined without
regard to this subsection) shall be reduced by the sum"
of the amounts of any revenues received by the state from
four sources. The listed sources are certain provider-
related donations, health care related taxes other than
broad-based health care related taxes, certain broad-
based health care related taxes, and, during the
transition years, broad-based health care related taxes
to the extent they exceed the collection cap.
Section 1903(w)(5)(B)(i) provides that the base
percentage to be applied (if greater than 25%) is equal
to:
(I) the total of the amount of health care related
taxes (whether or not broad-based) and the amount of
provider-related donations (whether or not bona
fide) projected to be collected . . . during State
fiscal year 1992, divided by
(II) the non-Federal share of the total amount
estimated to be expended under the State plan during
such State fiscal year.
Regulations implementing the 1991 Amendments were first
published as an interim final rule in November 1992 and
then adopted without relevant change as a final rule in
August 1993. 57 Fed. Reg. 55,118 (November 24, 1992)
(interim final rule); 58 Fed. Reg. 43,156 (August 13,
1993) (final rule). The regulations provide that the
maximum amount of health care related taxes and provider-
related donations which a state may receive without a
reduction in FFP--
is calculated by multiplying--
(1) The State's total medical assistance
expenditures for the fiscal year, by
(2) The greater of:
(i) 25 percent; or
(ii) The "State base percentage" (as defined in
paragraph (b) of this section).
42 C.F.R. 433.60(a) (emphasis added).
Paragraph (b) of section 433.60 defines the State base
percentage as the amount of donations and taxes
identified in the regulation and estimated by HCFA to be
received by the State in SFY 1992 divided by "the total
non-Federal share of medical assistance expenditures
(including administrative costs) in that fiscal year
based on the best available HCFA data."
In the preamble to the interim final rule, HCFA provided
an example of how the methodology described in the
regulation would work. HCFA admitted before us that the
example was erroneous and, in fact, corrected it in the
preamble to the final rule (see underlined portions
below). The interim rule preamble stated:
The specific percentage to be applied for a State in
any fiscal year is the greater of 25 percent or the
"State base percentage." The State base percentage
is calculated by dividing the amount of all
provider-related donations and health care-related
taxes (whether or not they are permissible)
estimated to be received in State fiscal year 1992
by the State's share of the total amount estimated
to be expended under the State plan during such
State fiscal year. This percentage is multiplied by
the total non-Federal share of Medicaid expenditures
(including all of the administrative costs) in that
fiscal year to determine the actual dollar limit.
57 Fed. Reg. at 55,124 (emphasis added).
The preamble to the final rule noted that the explanatory
example was inconsistent with the regulation:
We are clarifying that, due to an inadvertent
editorial error, the formula contained in the
preamble of the interim final rule for determining
the maximum amount of provider-related donations and
health care-related taxes a State may receive
without a reduction in FFP is incorrect. The State
base percentage is calculated by dividing the amount
of the provider-related donations and health care-
related taxes to be received in State fiscal year
1992 by the total non-Federal share of medical
assistance expenditures (including administrative
costs) in that fiscal year based on the best
available HCFA data. This percentage is then
multiplied by the State's total medical assistance
expenditures for the fiscal year to determine the
actual dollar limit. This formula is consistent
with the statute and 433.60 of the regulations.
The preamble, however, included administrative costs
in the multiplication factor and not in the State
base percentage determination.
58 Fed. Reg. at 43,169 (emphasis added).
The method of calculating the "State base percentage" and
the specific percentage to be applied in Kentucky's case
(31.61%) were not disputed here.
Analysis
Kentucky argued that administrative costs should be
included in the multiplication factor to which the base
percentage is applied to calculate the cap on permissible
tax collections because (1) HCFA took conflicting
positions on the question and the regulations are at best
ambiguous; (2) a proper reading of the statute supports
inclusion of administrative costs; and (3) inclusion of
administrative costs would make the calculation of the
multiplication factor consistent with the calculation of
the base percentage and thereby carry out the statutory
purpose of preserving base year revenues. Finally,
Kentucky asserted that it had relied on representations
from HCFA officials in calculating the collection cap as
it did.
1. The regulations are clear and are not made ambiguous
by the error in the preamble.
Kentucky asserted that the inconsistent information in
the interim and final rule preambles evidenced ambiguity
in the regulations. Kentucky Br. at 12. Kentucky
pointed to the fact that the preamble to the interim rule
stated that the base percentage should be multiplied by
total Medicaid expenditures including administrative
costs while the preamble to the final rule took the
position that the percentage should be multiplied only by
the medical assistance expenditures.
HCFA argued that Kentucky at most alleged ambiguity in
the regulations, but that Kentucky failed to make any
showing that the interpretation favored by Kentucky was
the only or even the best reading of the regulatory
language. HCFA argued that the regulation should not be
considered ambiguous simply because of "inadvertent
editorial error" in the first preamble. HCFA pointed out
that the regulation itself was unchanged and at all times
defined the multiplication factor as "total medical
assistance expenditures" without a reference to
administrative costs while expressly defining the base
percentage to include administrative costs. HCFA Br.
at 4; see 42 C.F.R. 433.60(a)(1), 42 C.F.R.
433.60(b)(1).
Both parties agreed before us that the term "medical
assistance" has had a long-established meaning as a term
of art in relation to the Medicaid program. HCFA Br.
at 1-2; Kentucky Br. at 8-10; Kentucky Reply Br. at 5.
"Medical assistance" is defined for Medicaid purposes as
"payment for part or all of the cost" of a list of
specific covered "care and services." Section 1905(a) of
the Act. The list made part of the definition of
"medical assistance" does not include administrative
costs. The distinction is important for many purposes,
particularly because medical assistance expenditures are
shared at the rate of the federal medical assistance
percentage for a particular state (as described in
section 1905(b) of the Act). Section 1903(a)(1) of the
Act. By contrast, payment to the states for a portion of
the "sums expended for the proper and efficient
administration" of the state plan is provided for
elsewhere in the Act, at rates set according to the type
of administrative cost incurred, most often 50%. See
sections 1903(a)(2)-(7). Medical assistance expenditures
and administrative costs are thus two distinct components
of the total amounts expended under the Medicaid program,
and the federal government share is differently
calculated for each component.
In order for us to accept Kentucky's argument, however,
we would have to find that the phrase "total medical
assistance expenditures" at 42 C.F.R. 433.60(a)(1) was
ambiguous and should be read to mean "administrative
costs" as well. Such a reading would fly in the face of
the long-established and undisputed separation of the two
categories of medical assistance expenditures and
administrative costs throughout Medicaid law and
regulation. The history of the strong distinction made
between these phrases means that Kentucky could not
reasonably read the phrase "medical assistance
expenditures" in the regulation as including
administrative costs without an express indication that
the phrase was being used in other than its traditional
meaning.
It is true that elsewhere in the regulation and in the
preamble to the final rule inartful language appears
referring to "total non-Federal share of medical
assistance expenditures (including administrative costs)"
as meaning total amounts expended including both
administrative and medical assistance expenditures. This
usage of the word "including" may create a false
impression that administrative costs could sometimes be a
subset of medical assistance expenditures. It would have
been preferable perhaps to refer to total expenditures
including both medical assistance expenditures and
administrative costs.
While the phrasing might be problematic in a case in
which a portion of the regulation containing it were at
issue, it is difficult to see why that should lead us to
prefer a reading of the section involved here, which
refers only to "medical assistance," that is inconsistent
with the express statutory definition of that term. If
anything, the omission of the parenthetical "(including
administrative costs)," when it had been used elsewhere
(however inelegantly) when administrative costs were to
be added to a calculation, implies that those costs were
not intended to be included in the multiplication factor,
rather than the reverse.
It is also unfortunate that the original preamble
contained the mistaken example of calculating the
collection cap with administrative costs included in the
multiplication factor. However, the regulation itself
clearly required only medical assistance expenditures to
be included for that purpose. The regulatory language
must prevail over the explanatory material in the
preamble where they are inconsistent. Cf. Sutherland,
Statutes and Statutory Construction (5th ed.) at 47.13.
In sum, we conclude that, in light of the overarching
definition of medical assistance in the statute and its
long-standing distinction from administrative costs,
Kentucky could not reasonably have believed that the
phrase "medical assistance expenditures" in 42 C.F.R.
433.60(a)(1) was meant to encompass administrative costs.
2. HCFA's interpretation of the statutory language is
reasonable and entitled to deference.
Kentucky acknowledged that the language of the regulation
itself was not altered by the final rule. Kentucky
nevertheless argued that, "[r]egardless of the actual
meaning of these regulations," Kentucky's position is
supported by the "perfectly clear" statutory language
which must prevail. Id. at 13. Kentucky relied on the
statutory language directing that the base percentage be
applied to the state share of "the total amount expended
. . . as it would be determined pursuant to paragraph
(1)(A) without regard to paragraph (1)(A)(iv)." Section
1903(w)(5)(A) of the Act. Paragraph (1)(A) provides that
"the total amount expended . . . as medical assistance
shall be reduced by the sum" of the amounts of revenues
from four sources received by the state. The fourth
source, which is in paragraph (1)(A)(iv) and which is not
to be considered in determining the multiplication
factor, is the amount of taxes collected during the
transition years beyond the limits established in section
1903(w)(5)(A) itself.
Kentucky argued that, "[p]roperly construed," this
language means only that states should subtract
impermissible taxes and donations from the total amount
expended before calculating the tax collections cap, in
order to prevent states from inflating the basis of the
multiplier by including impermissible taxes and donations
so as to increase the amount of permissible tax
collections. Kentucky denied that the reference to
paragraph (1)(A) was intended to incorporate its
limitation of amounts expended to medical assistance
payments. Kentucky Br. at 6-8.
HCFA contended the statutory language on its face states
that the multiplication factor is determined as it would
be under paragraph (1)(A) and that the determination
under (1)(A) begins with only medical assistance
expenditures from which the listed reductions are then
taken. HCFA Br. at 6-8. HCFA portrayed Kentucky's
construction of the possible purpose of the reference to
paragraph (1)(A) as irrelevant, since congressional
intent need not be explored when the statutory language
is not ambiguous on its face. Finally, HCFA argued that,
even if another reading of the statute were possible, its
interpretation is reasonable and should receive
deference. HCFA Br. at 9.
Kentucky's explanation of the purpose of including a
reference to the reduction of expenditures by the amounts
of revenue from impermissible sources is plausible.
However, this purpose is in no way defeated by HCFA's
interpretation that the reduction is to be applied to the
smaller sum of medical assistance expenditures alone.
Furthermore, in interpreting a statute, it is best to
favor an interpretation which gives meaning to all the
terms of the law. Yet, Kentucky's interpretation makes
meaningless the reference to paragraph (1)(A) containing
the limitation to medical assistance expenditures.
The legislators could have achieved the purpose
attributed to them by Kentucky without referring to the
language in paragraph (1)(A) had they simply defined the
multiplication factor as the total amount expended under
the state plan less the sums listed in subparagraphs
(1)(A)(i) through (1)(A)(iii). Since they chose instead
to define the multiplication factor by reference to an
entire paragraph limited to medical assistance
expenditures, we conclude that HCFA's interpretation is
preferable in that it gives effect to all the terms of
the statute.
In any event, we need not determine the best possible
interpretation of the statute. Where HCFA has issued a
regulation which sets forth an interpretation of the
statute which was not unreasonable and was sufficiently
clear to give notice to the states (as we have found the
phrase "medical assistance expenditures" to be), we are
bound to apply that regulation. 45 C.F.R. 16.14; see,
e.g., Illinois Dept. of Public Aid, DAB No. 440, at 3
(1983); Oklahoma Dept. of Human Services, DAB No. 1436,
at 16-17 (1993).
Kentucky argued that we are bound by both law and
regulation and that, where they conflict (as it contended
was the case here), we must give overriding weight to the
statute. However, we need not reach that question here
since we conclude that the statute is not in conflict
with the plain language of the regulation.
We thus conclude that the statute and the regulations are
in accord in limiting the multiplication factor to
medical assistance expenditures only and not including
administrative costs.
3. The statute does not require inclusion of
administrative costs in the multiplication factor even
though they are included in the base percentage
calculation nor does it guarantee preservation of base
year revenues.
Kentucky argued further that inclusion of administrative
costs in the multiplication factor would be consistent
with the companion provision on calculation of the base
percentage, which uses "virtually the same language."
Kentucky Br. at 11. Kentucky argued that similar
language in companion provisions should be interpreted in
the same way.
The language at section 1903(w)(5)(B)(i) defines the base
percentage as the total amount of health care related
taxes projected for collection in SFY 1992 divided by
"the non-Federal share of the total amount estimated to
be expended under the State plan during such State fiscal
year." This language plainly differs in two significant
ways from the language at issue in section 1903(w)(5)(A),
which instructs that the base percentage be multiplied by
the "non-Federal share of the total amount expended under
the State plan during a State fiscal year (or portion
thereof), as it would be determined pursuant to paragraph
(1)(A) without regard to paragraph (1)(A)(iv)." First,
the base percentage calculation uses an estimated amount
for a specific base SFY, whereas the multiplication
factor uses actual expenditures for determining the cap
for each fiscal year. Second, the base percentage
calculation refers to total amounts expended, whereas the
multiplication factor contains the limiting language that
the amount should be determined as in paragraph (1)(A)
(which contains the reference to medical assistance
expenditures reduced by several amounts, as discussed
above). While it is axiomatic in statutory construction
that the same language occurring in more than one place
in a law should generally be treated as having the same
meaning, the corollary is that where the legislators have
chosen different language in companion provisions, they
should be assumed to have intended a distinction.
Sutherland, Statutes and Statutory Construction (5th ed.)
at 46.06 (1995 Supp.).
Kentucky further asserted that the use of a base year as
a reference point in the base percentage calculation
implied that Congress intended that states be permitted
to collect as much in health care related taxes in SFY
1993 as they did in SFY 1992. However, Kentucky argued
that if the base percentage is multiplied only by the
medical assistance expenditures, this intention would be
undermined since its tax collection cap would be lower in
SFY 1993 even if it had the same total expenditures as in
SFY 1992 and did not collect any impermissible taxes or
donations in SFY 1993. Kentucky described the purpose of
the transition cap on tax collections as "grandfathering
of the 1992 amount" of tax collections, "when those
collections exceeded 25 per cent" of a state's total
Medicaid expenditures. Kentucky Reply Br. at 1.
It is plain that Congress intended to cushion the impact
of the loss of revenue from tax collections by permitting
some time for a phase-out period during which some pre-
existing schemes would not result in loss of FFP for
states which had depended on those revenues for much of
their share of Medicaid costs. However, Kentucky did not
demonstrate that Congress necessarily intended to
guarantee that the formula which it adopted would
generate precisely the same amount of funding a state had
previously enjoyed. (Certainly, states which previously
derived less than 25% of their expenditures from these
taxes and donations were not grandfathered at all.) It
is not irrational to conclude that Congress may have had
particular concern for preserving from precipitous
decrease those revenues directed to care and services, as
opposed to program administration. As we noted above,
Congress could have written section 1903(w)(5)(A) in a
manner which expressly included administrative costs if
it wished to achieve that result. The numerous examples
cited by both parties of language throughout the Medicaid
program law distinguishing among medical assistance
expenditures, administrative costs, and total sums
expended make abundantly clear that Congress knew how to
differentiate among them when it chose to do so. Cf.
Kentucky Br. at 8-10; HCFA Br. at 4-5. Even if
Kentucky's interpretation of the statute is a reasonable
one, Kentucky has not shown that it is the only or the
best one. We find that HCFA's interpretation in its
regulation to be a reasonable reading of the statute to
distinguish between calculation of the base percentage
and of the multiplication factor. 1/
4. Kentucky did not establish that it relied on
misleading information from HCFA officials or lacked
notice of HCFA's interpretation of the statute.
We have already concluded that the language of the
regulation was sufficiently clear, despite the erroneous
explanatory material in the preamble to the interim rule,
to provide Kentucky with notice of HCFA's interpretation
of the statute to exclude administrative costs from the
multiplication factor. However, Kentucky also claimed to
have had communications with HCFA regional officials who
"agreed" with Kentucky's calculations about the amount of
the collection cap. Kentucky Br. at 3. In its brief,
Kentucky described the communications as occurring during
the process of determining the amount which Kentucky had
overcollected in SFY 1993, which casts doubt on whether
Kentucky could have been relying on them prospectively in
structuring its SFY 1993 tax collections. In any case,
Kentucky provided no documentation of the allegedly
misleading communications. Instead, Kentucky provided a
later letter from the Associate Regional Administrator to
Kentucky (dated July 30, 1994) to show that HCFA had
taken a "new position." Kentucky Br. at 5. The letter
stated that "additional clarification" has been received
from the Medicaid Bureau on how to calculate the
collection cap for SFY 1993. Kentucky Ex. C. In regard
to whether to use medical assistance expenditures alone
or to add administrative costs, the letter states that
the "statute is relatively clear" and that "the statute
and regulations are consistent in providing that MAP
[medical assistance payments] only be included in
calculating" the multiplication factor. Nothing in this
letter establishes that HCFA regional officials had
previously misled Kentucky as to the calculation of this
factor. In any case, discussions about the amount that
Kentucky should repay as a result of its having already
collected excess taxes do not demonstrate prospective
reliance by Kentucky on confusing information from
regional officials.
Kentucky also submitted a letter from its Commissioner to
HCFA dated July 25, 1994, in which he asserted that the
"amount of provider tax to be collected during SFY 1993
was adjusted based upon information supplied by Region IV
regarding computation of the tax cap." Kentucky Ex. D.
This assertion is not persuasive for two reasons. First,
Kentucky admitted collecting excess taxes resulting in
having to repay $3,378,116 in FFP even under the
interpretation under which it now claims to have been
operating. If Kentucky designed its tax collection
program prospectively based on the theory it has raised
here, presumably it would have collected only the
approximate amount of taxes permissible under that
theory. It therefore seems more likely that Kentucky's
interpretation was arrived at after-the-fact to justify
some portion of its excess collections. Second, Kentucky
offered no details whatsoever of the purported
"information supplied by Region IV," such as when the
information was provided, by whom, for what purpose, or
what its content was. We have already concluded that the
erroneous example contained in the preamble to the
interim rule was insufficient to undercut the contrary
language in the rule itself. Kentucky's vague reference
to information obtained from the region adds nothing and
is clearly insufficient to justify disregarding the rule.
Conclusion
For the reasons explained above, we conclude that HCFA's
calculation of the tax collection cap was proper under
the statute and regulations, and therefore we sustain the
disallowance.
___________________________
Cecilia Sparks Ford
___________________________
Norval D. (John) Settle
___________________________
Judith A. Ballard
Presiding Board Member
1. Kentucky's position that the denominator in the
formula for calculating the "State base percentage"
includes both administrative costs and medical assistance
expenditures and that therefore the amount to which the
resulting percentage is applied should include equivalent
costs is undercut by the following:
o The numerator in the "State base percentage" formula
includes the total amount of health care related
taxes (whether or not broad based) and the total
amount of donations (whether or not bona fide).
Section 1903(w)(5)(B) of the Act.
o The cap in subparagraph 1903(w)(1)(A)(iv), which
results from the calculation at issue here, applies
only to broad based health care related taxes, not
to other taxes or to donations.
o Paragraph 1903(w)(1)(A) as a whole provides for
reductions in expenditures under subsection (a)(1)
(that is, medical assistance expenditures) for both
taxes and donations.
o Paragraph 1903(w)(1)(B) provides for reduction of
expenditures under subsection 1903(a)(7) (i.e.,
other administrative costs) to the extent that
donations described in paragraph 1903(w)(2)(C)
exceed a specified limit. These donations are used
for activities (Medicaid eligibility determinations)
that are administrative in nature.
Both administrative costs and medical assistance
expenditures were likely included in the denominator of
the "State base percentage" because all taxes and
donations were included in the numerator, and some
donations might offset administrative costs rather than
medical assistance expenditures. The "multiplication
factor" at issue here, however, is not being used to
establish a total amount of allowable taxes and
donations. Instead, it is being used to calculate a cap
applicable only to a limited subset of taxes and
ultimately to determine reductions only in payments for
"medical assistance" expenditures, not for total
expenditures. Thus, calculation of the multiplication
factor does not have the type of direct correlation to
the "State base percentage" calculation that would
support Kentucky's position.