SUBJECT:

Graduate Hospital

Petitioner,

DECISION

Graduate Hospital (Graduate) appealed a decision by the National Institutes of Health (NIH) Grant Appeals Board upholding a determination by the NIH Division of Financial Advisory Services (DFAS) to disallow $791,214 charged to NIH grants by Graduate for its fiscal years 1986 through 1993. DFAS had determined that Graduate had improperly charged employer payments under the Federal Insurance Contributions Act (FICA) as a direct cost to its NIH research grants, even though FICA was reimbursed as an indirect cost under the applicable indirect cost rate agreements Graduate had negotiated with the Department's Division of Cost Allocation (DCA). The disallowed amount included $441,029 for FICA payments charged as direct costs, $317,666 for indirect costs associated with those direct costs, and $32,519 for indirect costs charged in excess of the negotiated indirect cost rates.

On appeal, Graduate argued that the original negotiated agreement governing indirect costs during the disallowance period was ambiguous regarding the characterization of fringe benefits, including FICA, and that the parties understood that FICA would be charged as a wage-related direct cost. According to Graduate, the NIH Grant Appeals Board improperly failed to construe ambiguities in the rate agreement against NIH and failed to give appropriate weight to extrinsic evidence of the parties' intent. Graduate argued, alternatively, that there was a mutual mistake, so the rate agreement should be rescinded. Graduate also attacked the procedures used by the NIH Grant Appeals Board, arguing that Graduate never had the opportunity to review or contribute to the record or to submit arguments because the notice letter was not addressed properly. According to Graduate, if we uphold the NIH Grant Appeals Board decision, we should allow Graduate to submit an audit to demonstrate that Graduate did not receive excessive reimbursement for its costs during the period in question.

For the reasons explained below, we conclude that --

• The wording of the indirect cost rate agreement governing costs for fiscal years ending 1988 through 1991 evidences a clear intent to reimburse all fringe benefits other than certain paid absences through use of the indirect cost rates set in the agreement.



• In the context of the applicable cost principles, the only reasonable reading of the term "fringe benefits" in this agreement is to include employer contributions to FICA. Contrary to what Graduate argued, the agreement cannot reasonably be read to treat FICA, like paid absences, as a direct cost considered part of the normal charge for salaries and wages.



• The extrinsic evidence on which Graduate relied supports a conclusion that Graduate's accounting system at the time of the negotiation treated FICA as a fringe benefit, rather than as a wage-related cost, and does not demonstrate any intent to exclude FICA from the fringe benefits reimbursed through use of the indirect cost rates.



• The indirect cost rate agreement governing costs for fiscal years 1991 and 1993 specifically identifies FICA as a fringe benefit and treats fringe benefits (other than certain paid absences) as indirect costs, rather than as part of a direct cost base.



• Graduate's reliance on a provision from the manual setting out Medicare principles of reimbursement is misplaced. That provision does not establish that Medicare treated FICA as a wage-related cost and not a fringe benefit during the time period at issue here. In any event, hospital grantees are warned that the Medicare reimbursement principles differ in some respects from the cost principles applicable under NIH grants.



• The NIH Grant Appeals Board process was not flawed, and, in any event, Graduate had ample opportunity to present its case to us, and we provided a de novo review of the issues.



• Rescinding the indirect cost agreements would not be appropriate under the circumstances here, but Graduate should be provided a limited opportunity, consistent with our decision, to show that the disallowance amount should be reduced.

General legal background

Uniform requirements for the administration of HHS grants, including those awarded by the Public Health Service (PHS) and its subagencies such as NIH, are set out at 45 C.F.R. Part 74. Part 74 provides that grant funds may be used only for allowable costs. § 74.170.⁽¹⁾ Principles for determining the allowable costs of research and development work performed by hospitals are in Appendix E to Part 74. 45 C.F.R. § 74.173.

The cost of a research agreement is comprised of the allowable direct costs incident to its performance plus the allocable portion of the allowable indirect costs of the hospital, less applicable credits. Appendix (App.) E., § III.A. General standards for determining allowability of costs are in Appendix E, Part IX, section A, and standards for selected items are in section B.

Appendix E defines "direct costs" as--

those that can be identified specifically with a particular cost center. For this purpose, the term cost center refers not only to the ultimate centers against which costs are finally lodged such as research agreements, but also to other established cost centers such as the individual accounts for recording particular objects or items of expense, and the separate account groupings designed to record the expenses incurred by individual organizational units, functions, projects, and the like. [Certain administrative functions and service activities] are identifiable as separate cost centers, and the expenses associated with such centers become eligible in due course for distribution as indirect costs of research agreements and other ultimate cost centers.

App. E, § IV.A.

"Indirect costs" are--

those that have been incurred for common or joint objectives and thus are not readily subject to treatment as direct costs of research agreements or other ultimate or revenue producing cost centers.

App. E, § V.A. The primary method of allocating costs between research, patient care, and other major hospital activities is what is called the "step down" method of cost finding. Costs within certain functional categories are grouped with items of expense of a like character and then distributed to related cost centers using an appropriate distribution base or method, for example, square footage for space-related costs. App. E, § V.B.⁽²⁾

Generally, indirect costs allocated to organized research through the step down method are treated as a "common cost pool" and distributed to individual research agreements benefitting from the cost pool by use of a single rate. This rate, referred to as the research indirect cost rate, is determined by dividing the total indirect costs in this cost pool by a distribution base, generally total direct salaries and wages or modified total direct costs. The result, stated as a percentage, may then be applied to a comparable direct cost base for each research project to determine indirect costs, which are allowable unless subject to limitations. In some circumstances, more than one cost pool and distribution base are used, if appropriate. App. E, § VII.

Responsibility for negotiating indirect cost rates with grantees that receive most of their funding from HHS is with the Division of Cost Allocation (DCA). Guidance on how hospitals should prepare an indirect cost rate proposal and negotiate an indirect cost rate agreement was provided, as relevant here, in a publication generally referred to as OASC-3.⁽³⁾

There are several types of indirect cost rates, including as relevant here, "provisional," "final," and "predetermined." Basically, a provisional rate is a temporary rate based on an estimate, rather than actual costs. Charges based on a provisional rate must be adjusted after a final rate is set based on actual costs incurred during the accounting period (normally the hospital's fiscal year). Once established, a final rate is not subject to adjustment. A predetermined rate is a fixed rate, based on experience and a reliable projection of a hospital's level of activity, approximating the actual rate. This type of rate is not subject to adjustment except in very unusual circumstances. OASC-3, at 1.

Each hospital claiming reimbursement for indirect costs must submit an indirect cost proposal, generally within six months after the close of the fiscal year. If an agreement is reached through negotiations between the hospital and DCA, the result is formalized in a negotiation agreement.

Case background

During the period at issue here, 6/30/86 through 7/1/93, Graduate received grant funds from NIH to conduct various research projects. Graduate submitted an indirect cost proposal to DCA on 10/20/88, proposing a final rate for fiscal year 1987. NIH Ex. 1, at 86.⁽⁴⁾ Graduate proposed a rate of 104%. NIH Ex. 1, at 87. After negotiations, DCA transmitted to Graduate an indirect cost rate agreement dated 12/15/88 (1988 Agreement) that was signed by Graduate's President on 1/24/89. NIH Ex. 1, at 61. The 1988 Agreement established the following research indirect cost rates:

Final 7/1/84 to 6/30/87 90.0%
Final 7/1/87 to 6/30/88 85.0%
Pred 7/1/88 to 6/30/89 85.0%
Pred 7/1/89 to 6/30/90 77.0%
Pred 7/1/90 to 6/30/91 72.0%
Prov 7/1/91 Until Amended 72.0%

NIH Ex. 1, at 57. The distribution base for these rates was described as follows:

Total direct costs including the first $25,000, for subgrants and subcontracts, and excluding individual items of equipment in excess of $1,000, alterations and renovations, patient care costs, stipends and tuition payments.

Id.

The following provision, included in the 1988 Agreement, is the key provision at issue here:

Treatment of Fringe Benefits: Vacation, holiday, sick pay and other paid absences are included in salaries and wages and are charged to grants and contracts as part of the normal charge for salaries and wages. Separate charges for the cost of these absences are not made. All fringe benefits are included in the indirect costs and are not charged as direct costs.

Id.(emphasis in the original).⁽⁵⁾

The 1988 Agreement also contained an attachment setting out certain limitations on acceptance of the rates, including that the same costs treated as indirect costs have not been claimed as direct costs and that similar types of costs have been accorded consistent accounting treatment. Id. at 61.

The attachment also noted that the proposal for fiscal year 1990 would be due no later than 12/31/90. NIH Ex. 1, at 58. Yet, the next proposal submitted by Graduate was one for the fiscal year 1992. That proposal was for an indirect cost rate of 131.03% including "NON FICA FRINGE AS INDIRECT" or, alternatively, for an indirect cost rate of 127.72% "EXCLUDING ALL FRINGE" and a research fringe benefit rate of 20.48%. NIH Ex. 1, at 144.

Subsequent negotiations led to an agreement dated 8/20/93 (1993 Agreement). The 1993 Agreement set a final research indirect cost rate of 72.0% for fiscal years 1992 and 1993. NIH Ex. 1, at 215. The direct cost base for this rate was described in the "SPECIAL REMARKS" section of the Agreement as follows:

Total direct costs, excluding Capital expenditures; that portion of subawards in excess of $25,000 and patient care costs.

NIH Ex. 1, at 218. This section also contained the following statement:

Fringe Benefits include: FICA, Retirement, Disability Insurance, Tuition Remission, Workers' Compensation, Unemployment Insurance, Health Insurance and Dental Insurance.

Id.

The 1993 Agreement also set a provisional fringe benefit rate of 21.6% for research, on a base of salaries and wages, effective starting on 7/01/93, after the period at issue here.

The 1993 Agreement contained separate provisions on the treatment of fringe benefits and the treatment of paid absences, which we set out and discuss below. NIH Ex. 1, at 215.

Graduate appealed this determination to the NIH Grant Appeals Board under the procedures at 42 C.F.R. Part 50, Subpart D. After a lengthy period during which the parties engaged in negotiations to try to resolve the dispute, the NIH Grant Appeals Board sent a notice of resumption of the appeals process to the individual who was Graduate's President at the time of the appeal. After receiving no response from Graduate, the NIH Grant Appeals Board issued its decision in June 1998, upholding DFAS.

On appeal to us, Graduate argued that the NIH Grant Appeals Board had "failed to conduct an adequate analysis of the intent of the parties when they entered into the original indirect rate agreement." Graduate argued:

Well-established principles of contract construction dictate that where a contract is ambiguous, the fact finder considers (1) the actual words of the agreement; (2) any alternative meanings offered by the parties; and (3) extrinsic evidence offered in support of those alternative meanings. St. Paul Fire and Marine Ins. Co. v Lewis et al., 935 F.2d 1428 (3d Cir. 1991); Restatement (Second) Contracts, § 20 (1981). Any ambiguity in a contract will be construed against the drafter. Kiewit Eastern Co., Inc. v. L&R Construction Co., Inc., 44 F.3d 1194 (3d Cir. 1995); St. Landry Parish School Board, Opelousas, Louisiana, Decision No. 17, Dkt No. 75-4 (DAB 1976). See also University of California, Dec. No. 763, Dkt No. 85-244 (DAB 1986)(Board considered language of agreement, interpretations offered by parties, and extrinsic evidence to overturn decision by PHS to disallow University's indirect costs).

GR br. at 8-9. Graduate argued that the 1988 Agreement was ambiguous and that the parties understood that FICA would be charged as a wage-related direct cost. According to Graduate, the NIH Grant Appeals Board improperly failed to construe ambiguities in the rate agreement against NIH and failed to give appropriate weight to extrinsic evidence of the parties' intent.

Alternatively, Graduate argued that the Agreement should be rescinded since there was a mutual mistake.

Graduate also attacked the process used by the NIH Grant Appeals Board. Graduate argued that the NIH Grant Appeals Board should have sent notice that it was resuming consideration of the appeal to the person who had been negotiating, rather than to Graduate's former President, who was no longer with Graduate at that time, and that Graduate did not have an opportunity to supplement the record before the NIH Grant Appeals Board issued its decision.

Graduate also asserted that it had not received excessive reimbursement for its grant costs. Graduate requested that, if we upheld the NIH Appeals Board, we allow Graduate an opportunity to present its own audit in response to the DFAS audit.

Conclusion

For the reasons stated above, we uphold the disallowance in principle, subject to reduction if Graduate makes the showing described above, consistent with our decision. If Graduate timely produces documentation that is rejected by NIH, Graduate may appeal to us, within 30 days of receiving NIH's determination, on the limited issue of whether the documentation timely produced to NIH establishes that the disallowance amount should be reduced.

Graduate misstates somewhat the principle enunciated in the St. Paul Fire and Marine case. That case was applying Pennsylvania law on insurance contracts and used the three listed factors in determining whether an ambiguity existed in the insurance contract at issue there. As the court said in Kiewit, however, where an agreement is unambiguous on its face, resort is not generally made to extrinsic evidence. 44 F.3d at 1194. In DAB No. 763, this Board examined the extrinsic evidence presented regarding how the rate was in fact calculated, but relied primarily on the wording and structure of the agreement provision at issue, considered in context, to reject the interpretation offered by DCA.

In this section, we first examine the wording of the key provision at issue here. We conclude that, in context, that provision is not ambiguous and cannot reasonably be interpreted to mean what Graduate now says it means. We nonetheless go on to examine the extrinsic evidence relied on by Graduate, concluding that it does not establish a different meaning than what the plain language suggests. We also examine the other arguments raised by Graduate.

In context, the wording of the 1988 Agreement indicates an intent that FICA is treated as an indirect, not a direct cost.

As mentioned above, the 1988 Agreement stated:

Treatment of Fringe Benefits: Vacation, holiday, sick leave pay and other paid absences are included in salaries and wages and are charged to grants and contracts as part of the normal charge for salaries and wages. Separate charges for the cost of these absences are not made. All fringe benefits are included in the indirect costs and are not charged as direct costs.

GR Ex. 1. NIH acknowledged that, since vacation, holiday, sick leave pay and other paid absences are fringe benefits, there is an inconsistency between the statement that all fringe benefits are included in indirect costs and the statement in the first sentence of the provision that vacation, holiday, sick leave pay and other paid absences are included in salaries and wages. As NIH pointed out, however, the issue here is whether the agreement was ambiguous with respect to treating FICA as an indirect cost, not whether it was ambiguous with respect to the costs referred to in the first sentence.

Moreover, the fact that FICA is not specifically mentioned does not automatically render the agreement ambiguous with respect to treatment of FICA if the term "fringe benefits" includes FICA. Graduate's original appeal of the DFAS determination appeared to acknowledge that FICA is a "fringe benefit." GR Ex. 7. Before us, Graduate pointed out that there is no definition of "fringe benefits" in the agreement and asserted that its accounting system treated FICA as a "wage-related" direct cost. We conclude that Graduate could not have reasonably interpreted the term "fringe benefits" in the third sentence of the agreement statement to not include employer contributions for FICA, for several reasons.

First, the accounting documents which Graduate submitted with its 1988 indirect cost proposal show an account (100) for "Social Security" under the heading "Benefits." GR Ex. 13.D. (Graduate itself equated "Social Security" with FICA. See, e.g., GR br. at 1.) There is no separate grouping for "Fringe Benefits" and, contrary to what Graduate's arguments imply, there is nothing in the accounting documents produced by Graduate for the period in question that instead identifies FICA or Social Security as a "wage-related" cost. In contrast, accounts for sick pay (052), vacation pay (053), and holiday pay (054) are included under the grouping "Salaries."

Second, the applicable cost principles in Appendix E to Part 74 specifically address two types of "Staff and/or employee benefits," as follows:

a. Staff and/or employee benefits in the form of regular compensation paid to employees during periods of authorized absences from the job such as annual leave, sick leave, military leave and the like are allowable. . . .

b. Staff benefits in the form of employer contributions or expenses for Social Security taxes, employee insurance, Workman's Compensation Insurance, the Pension Plan . . . , and the like are allowable provided such benefits are granted in accordance with established policies, and provided such contributions and other expenses whether treated as indirect costs or an increment of direct labor costs are distributed to particular research agreements and other activities in a manner consistent with the pattern of benefits accruing to the individuals or groups of employees whose salaries are chargeable to such research agreements and other activities.

App. E, § IX.B.40; see OASC-3, IX.B.40.⁽⁶⁾ Since the first sentence of the statement in the rate agreement refers to the types of benefits addressed in subsection a. of this principle, it would be illogical not to consider Social Security taxes (FICA) to be a fringe benefit addressed in the second sentence of the rate agreement provision -- that is, as a cost to be treated as an indirect, not a direct cost.

Graduate asserted that the NIH Grant Appeals Board had erred in relying on this provision, arguing: "Nowhere in the Rate Agreement...do the parties indicate their intent to rely on Appendix E to supply an explanation for undefined terms in the agreement." GR br. at 10. This assertion ignores the fact that the 1988 Agreement does state that the rates in the agreement were approved in accordance with the cost principles promulgated by the Department and refers to OASC-3, which in turn incorporates the Appendix E cost principles. NIH Ex. 1, at 57-61. Even without this reference, however, Appendix E would be relevant in interpreting the agreement since the regulations at Part 74 inform HHS grantees that the Appendix E cost principles apply to research grants to hospitals.

The alternative meaning offered by Graduate is unsupported and illogical.

Graduate did not offer any contemporaneous alternative definition of the term "fringe benefits" that would exclude FICA. Nor did Graduate offer evidence that it in fact interpreted the term "fringe benefits" at the time to exclude FICA (although Graduate did offer an affidavit by someone in a responsible position in the organization with which Graduate was affiliated to support other assertions Graduate made on appeal). Instead, Graduate argued that the 1988 Agreement provision is ambiguous, not because Graduate did not at the time consider FICA to be a "fringe benefit," but because an "alternative meaning" is that FICA was to be included in "the normal charge for salaries and wages" like "[v]acation, holiday, sick leave pay and other paid absences," pursuant to the first sentence of the 1988 Agreement provision on fringe benefits. GR br. at 9-10.

This alternative reading is an unreasonable one, however. The plain language of the first sentence of the 1988 Agreement provision refers only to a specific subset of fringe benefits: vacation, holiday, and sick pay, and other "paid absences." Graduate pointed to nothing to support an interpretation of the term "paid absences" to include FICA. The applicable cost principle on staff and employee benefits recognizes a difference between regular compensation for absences and FICA by treating paid absences in one paragraph and FICA and other benefits in another paragraph. Moreover, as noted above, the computer printouts Graduate submitted with its indirect cost rate proposals do not list FICA under the grouping "Salaries" although they do so list sick pay, vacation pay, and holiday pay. Thus, it would be unreasonable in context to interpret the rate agreement provision as including FICA with paid absences in the normal charge for salaries and wages, rather than with other fringe benefits as an indirect cost.

To support its position that the 1988 Agreement should be interpreted as treating FICA as a direct cost that is a "normal part of the charge for salaries and wages" like paid absences, Graduate relied on wording from Appendix E. In describing direct costs, Appendix E, section IV.B. states:

Typical of transactions chargeable to a research agreement as direct costs are the compensation of employees for the time or effort devoted to the performance of work under the research agreement, including related staff benefit and pension plan costs to the extent that such items are treated by the hospital as direct rather than indirect costs.

(Emphasis supplied by Graduate.)

This provides no meaningful support for Graduate's position, however, since it is a qualified statement, which clearly recognizes that staff benefit and pension plan costs may be treated as either direct or indirect costs. Indeed, provisions on indirect costs contain a similar statement that costs "normally" classified as indirect include "Administrative and General (including fringe benefits if not charged directly)." App. E, § V.A.⁽⁷⁾ Thus, contrary to what Graduate implied, there is no presumption that employer contributions to FICA will be treated as a direct, rather than an indirect cost. Instead, as NIH pointed out, fringe benefits are one of the types of costs that grantees are cautioned they should address in advance to avoid disallowances later. App. E, § IX.A.6.

Graduate apparently understood this because the letter transmitting Graduate's indirect cost proposal that led to the 1988 Agreement indicates that one enclosure was a "Fringe Benefits questionnaire." GR Ex. 13.B. While other enclosures were with the materials obtained from Graduate on a site visit, this questionnaire was not provided, even though it might have been relevant to how Graduate understood the term "fringe benefits" at the time the 1988 Agreement was negotiated.

The indirect cost proposal documents leading to the 1988 Agreement do not support Graduate.

Graduate relied on its indirect cost proposal documents as extrinsic evidence that it treated FICA as a wage-related direct cost. First, we note that such documents cannot reasonably be considered determinative. Often, changes are made to the treatment of costs as part of the negotiation process. Graduate provided no evidence here of the actual calculation of the rates included in the 1988 Agreement and what costs were ultimately included in direct versus indirect costs for this purpose. Since the indirect cost rate proposed by Graduate was 104% and the rates agreed to in the 1998 Agreement were lower, it is clear that adjustments were made in the course of the negotiations.

Second, even if we considered proposal documents to be probative of the intent of the agreement, the documents relied on by Graduate are themselves ambiguous on the treatment of FICA. First, the documents show that Social Security (FICA) was included in the "Benefits" grouping of accounts separate from the "Salaries" grouping that included salaries and wages, as well as sick pay, vacation pay, and holiday pay. See, e.g., NIH Ex. 1, at 94. Thus, the documents do not support the alternative reading of the Agreement provision on fringe benefits advanced by Graduate. Second, there is nothing in the documents which clearly states that the amounts identified as FICA and directly associated with research accounts were employer contributions to FICA, rather than employee contributions withheld from salaries and wages. Moreover, while the proposal documents indicate that some FICA costs were included with direct costs in the proposal, there is also evidence in the documents that some FICA was included in indirect costs. The handwritten document labeled "Indirect costs" includes more than $4.6 million identified as "Benefits." NIH Ex. 1, at 93. This amount corresponds to a cost center (983) included under "Employee Health and Welfare" elsewhere in the documents. NIH Ex. 1, at 105. An amount for Social Security (account 100) is included in the computer printout for the cost center (983) as are other accounts that appear under the grouping "Benefits" in other documents. NIH Ex. 1, at 101. Although this amount is too low to account for all FICA, it undercuts Graduate's position that its accounting system consistently treated FICA as a direct cost.⁽⁸⁾

In any event, the indirect cost proposal documents evidence an inconsistent treatment of costs in the FICA account, and an inconsistency between treatment of FICA and other costs in the "Benefits" grouping. As noted above, Appendix E refers to grouping like costs together and ensuring that costs are not duplicative. In this context, the most logical conclusion is that the ultimate calculation of the rate included an adjustment to treat all FICA as indirect costs, so the treatment of employer contributions to FICA would be consistent with treatment of other benefits that are not paid absences and that were grouped in the accounting system under the heading "Benefits." This conclusion is reinforced by the fact that the third sentence in the fringe benefits provisions in the 1988 Agreement contains underlining to emphasize that benefits are not treated as direct costs.

We also note that the indirect cost proposal submitted in 1992 provides more support for the NIH position than for Graduate's position. That proposal described two alternatives, one of which refers to "non-FICA fringe benefits" as being included in indirect costs and one of which accumulates all fringe benefits, including FICA, in a separate fringe benefit rate. Thus, the 1992 proposal shows that Graduate understood, at least as of that time, that FICA was a fringe benefit. Yet, Graduate provided no evidence that it sought at that time to disavow or to qualify the statement in the 1988 Agreement treating fringe benefits (other than paid absences) as indirect costs.

The other extrinsic evidence does not show that NIH understood that FICA was included in direct costs and not indirect costs.

Graduate also asserted that the inclusion of a line item for "Fringe Benefits" in the direct cost budgets on its Notices of Grant Award evidenced the parties' understanding that FICA would be treated as a direct cost. First, we note that the rate agreements were negotiated between Graduate and DCA, not between Graduate and NIH. The applicable requirements and policies make an award of indirect costs contingent on negotiation of an applicable indirect cost rate and allowability of costs contingent on consistent treatment of costs. Second, we note that the best evidence that the parties intended to include FICA in direct costs would have been some statement to that effect in the agreement. Yet, in spite of the fact that the agreement specifically includes paid absences in direct costs and delineates other costs treated as direct costs and included or excluded from the distribution base, no specific mention is made of FICA as a direct cost.

In any event, however, Graduate did not point to anything in the grant award documents that would support a conclusion that NIH should have known that the direct cost budget amounts for "Fringe Benefits" included amounts for FICA.

Moreover, Graduate ignored the fact that most of the Notices had an attachment referring to the 1988 Agreement and stating:

FRINGE BENEFITS: VACATION, HOLIDAY, SICK LEAVE PAY AND OTHER PAID ABSENCES ARE INCLUDED IN SALARIES AND WAGES AND ARE CHARGED TO GRANTS AND CONTRACTS AS PART OF THE NORMAL CHARGE FOR SALARIES AND WAGES. SEPARATE CHARGES FOR THE COSTS OF THESE ABSENCES ARE NOT MADE.

GR Ex. 5. Thus, even assuming that NIH had an obligation to examine whether Graduate's budgets were consistent with the rate agreement that Graduate had negotiated with DCA, NIH could have reasonably thought that the fringe benefits referred to in the direct cost budgets were only those fringe benefits that the rate agreement said would be treated as direct costs -- vacation, holiday, sick leave pay and other paid absences.

We recognize that if the award notices had included the complete fringe benefits provision, questions might have been raised sooner about what fringe benefits Graduate was including in its direct cost budgets. Apparently, the NIH "rate file" did not contain the complete provision negotiated between DCA and Graduate. GR Ex. 16 (NIH Appeals Board Decision at 5).

Since Graduate pointed to nothing in either the proposed or approved budgets that specifically identified FICA (or Social Security taxes) as a direct cost, however, we agree with NIH that its actions cannot reasonably be considered implicit approval for Graduate charging FICA as a direct cost.

Thus, we reject Graduate's position that approval of the budgets is evidence that the negotiated indirect cost rate was intended to include reimbursement for FICA as a direct, rather than an indirect cost.

Graduate did submit some expense summaries that appear to identify FICA (Social Security) as a direct cost of particular grant awards, but Graduate's reliance on these documents is misplaced as well. Graduate did not allege that those documents were submitted to NIH prior to the disallowance and some are related to fiscal year 1988, prior to the 1988 Agreement. GR Exs. 4 and 5. Under the applicable procedures, Graduate was responsible for making any adjustments required after the provisional rate for this period was made final. OASC-3, 3-5; see, also, NIH Ex. 1, at 58. The Financial Status Reports Graduate submitted for some of the grants after the 1988 Agreement was negotiated are too general to be a basis for charging NIH with knowledge of how Graduate was charging FICA. GR Ex. 4 and 5.

Graduate did not establish that the 1988 Agreement should be construed against NIH.

We reject Graduate's argument that we should construe the 1988 Agreement against NIH as the drafter of the agreement. We first note that NIH was not a party to the negotiations; rather, the agreement was negotiated between DCA and Graduate. The record indicates that the 1988 Agreement was transmitted to Graduate from DCA, but this does not necessarily establish that DCA drafted the language at issue. The transmittal from the Acting Director, DCA, states that the documents "reflect an understanding reached between your institution and a member of my staff. . . ." NIH Ex. 1, at 56. Nothing in this transmittal establishes who drafted what part of the agreement, and Graduate presented no evidence that, in fact, DCA drafted the provision at issue, rather than Graduate.⁽⁹⁾

Moreover, the principle of construction on which Graduate relied applies where a provision is susceptible of more than one reasonable interpretation. Graduate simply has not established that the provision at issue here was ambiguous with respect to treating employer contributions to FICA as an indirect cost.

The 1993 Agreement specifically identifies FICA as a fringe benefit and treats fringe benefits other than paid absences as indirect costs.

As noted above, the Special Remarks section of the 1993 Agreement specifically defines fringe benefits to include FICA and other benefits, such as insurance, listed in paragraph b. of the Appendix E cost principle on staff/employee benefits. The 1993 Agreement also contains the following provisions:

TREATMENT OF FRINGE BENEFITS:

The fringe benefits are charged using the rate(s) listed in the Fringe Benefits Section of this Agreement. The fringe benefits included in the rate(s) are listed in the Special Remarks Section of this Agreement.

TREATMENT OF PAID ABSENCES:

Vacation, holiday, sick leave pay and other paid absences are included in salaries and wages and are claimed on grants, contracts and other agreements as part of the normal cost for salaries and wages. Separate claims for the costs of these paid absences are not made.

NIH Ex. 1, at 215. While the first sentence of the provision on fringe benefits refers to the fringe benefit rates that did not apply until after the disallowance period at issue here, the second sentence can be read more broadly to apply to all of the rates set in the Agreement, including the 72% rate.

Like the statement on paid absences in the 1988 Agreement, the provision on paid absences in the 1993 Agreement cannot reasonably be read to treat employer contributions for FICA as a wage-related direct cost. Indeed, nothing in the 1993 Agreement can reasonably be read to distinguish FICA from other costs included in the specific definition of fringe benefits or to treat FICA as a direct cost.

The 1993 Agreement does not specifically state that the final research indirect cost rate set for fiscal years 1992 and l993 treats fringe benefits as indirect. This is the most logical conclusion, however, given the following facts: 1) the final rate was set at 72.0%, the same percentage as the provisional rate set for this period in the 1988 Agreement; 2) the 1988 Agreement specified indirect cost treatment for fringe benefits other than paid absences; 3) nothing in the 1993 Agreement specifies that FICA is treated as a direct cost; and 4) provisional fringe benefit rates were set for future periods.⁽¹⁰⁾

The indirect cost proposal documents that led to the 1993 Agreement do not provide evidence that the parties intended to treat FICA as a direct cost.

As we discussed above, evidence of what was proposed as an indirect cost rate is not determinative of how the rate was in fact calculated, particularly where the proposed rate differs from the negotiated rate. In its 1992 proposal, Graduate proposed two alternatives, either a research indirect cost rate of 131.03% or a research indirect cost rate of 127.72% and a fringe benefit rate of 20.48%. The proposed 131.03% rate was described as including non-FICA fringe as indirect and the 127.72% rate was described as excluding all fringe. NIH Ex. 1, at 144. The Agreement did not adopt either of these alternatives. Instead, it appears that, in exchange for obtaining a fringe benefit rate for future periods, Graduate agreed to adopt as final the 72.0% provisional rate for the period at issue here.

Like the earlier proposal documents, the documents related to the 1992 proposals that were relied on by Graduate are incomplete and ambiguous. A printout of various cost centers shows that the $78,598 identified in the proposal as total FICA for research is listed under "Fringe Benefits," rather than under "Salaries." While this printout appears to assign the FICA amounts directly to various grants or to individual salaries, a worksheet showed some costs under "Employee Benefits" being allocated to research as part of General Service Costs and treated as indirect costs.

Overall, given that the proposal recognized FICA as a fringe benefit and made it explicit that the 131.03% included non-FICA fringe, the proposal supports a conclusion that Graduate was aware at the time that the treatment of FICA as direct or indirect was an issue for negotiation. Yet, there is no evidence that Graduate objected to rates that treated the listed fringe benefits, explicitly including FICA, as indirect, rather than direct costs. Since Graduate proposed a direct cost treatment for FICA and then signed an agreement which does not specifically provide that treatment, it is unreasonable to conclude that Graduate's proposed intent on the treatment of FICA became the parties' intent in the Agreement. If DCA had accepted Graduate's proposal, one would expect something in the Agreement distinguishing FICA from other fringe benefits, such as by including FICA with paid absences as a cost treated as part of the normal cost for salaries and wages. Instead, the Agreement specifically defines fringe benefits to include FICA -- a clear indication that DCA had rejected Graduate's proposal.

Graduate did not show that the Medicare cost reimbursement principles applicable at the time the rate agreements were negotiated required it to treat FICA as a "wage-related direct cost."

Graduate also argued that whether a cost is allocated as direct or indirect depends on the grantee's accounting system and that, "as a provider in the Medicare program . . . and as consistent with Generally Accepted Accounting Principles . . . , Graduate treated FICA as a wage-related direct cost in its organization's accounting procedures." GR br. at 14-15. Graduate asserted that, on its Medicare cost reports, Graduate followed the Provider Reimbursement Manual in reporting FICA, a payroll tax, as a wage-related direct cost." Id. at 15. Two of the exhibits cited in support of this assertion are simply vague, unsworn statements made by Graduate's representatives in response to the disallowance and are therefore unpersuasive. One of the exhibits (Exhibit 19) is part of Chapter 11 of the Provider Reimbursement Manual (referred to as HIM-15) published by the Health Care Financing Administration, the HHS agency that administers the Medicare program. The version of Chapter 11 which Graduate submitted is the version in effect in 2000. Graduate did not assert that this version was in effect during the relevant time period, however, and our research indicates that an earlier version of Chapter 11 was deleted in 1998. See Transmittal 406, HIM-15.

Moreover, the version of Chapter 11 submitted by Graduate is not part of the definition of fringe benefits for Medicare reimbursement purposes. That definition is and has been broad enough to include employer contributions to FICA.⁽¹¹⁾ Instead, the Chapter 11 provision on which Graduate relied deals with an Exhibit 7, a reporting form for "wage-related" costs, and states:

Wage related costs may be different from fringe benefits allowed under Medicare as Generally Accepted Accounting Principles (GAAP) will be used in reporting wage related costs. In addition, some costs such as payroll taxes, which are reported as a wage related cost on Exhibit 7, are not considered fringe benefits.

This provision is not entirely clear since it does not specify under what principles payroll taxes are not considered a fringe benefit. Moreover, contrary to what Graduate's arguments imply, nothing in this statement specifically requires treatment of FICA, or any other wage-related cost, as a direct, rather than an indirect cost.

Graduate relied on the court decision in Sarasota Memorial Hospital v. Shalala, 60 F.3d 1507, 1512-13 (11^th Cir. 1995) in support of its position that Medicare treated FICA as a wage-related cost and not a fringe benefit. In fact, the decision supports a conclusion that the definition of fringe benefits in HIM-15 was historically interpreted as including both the employer contributions to FICA and any employer payment of the employee contribution on the employee's behalf. The decision, issued in 1995, struck down treatment as a fringe benefit of payment by a government employer of the employee contribution on the employee's behalf, for purposes of calculating a wage index used to reimburse hospitals under the prospective payment system, since this treatment was inconsistent with how employee contributions withheld by an employer from gross wages were treated for this purpose. The decision did not address treatment of the employer contribution as a fringe benefit.

In any event, Appendix E is what governs allowability of costs of research, not the Medicare principles. While Appendix E was based in part on the Medicare principles, grantees are warned that there are some differences, and Appendix E groups employer contributions to FICA with other staff/employee benefits that are clearly considered fringe benefits. OASC-3, at 2.

Finally, and most important, Graduate's after-the-fact assertion that it treated FICA as a wage-related cost and not as a fringe benefit is contradicted by the contemporaneous documents showing FICA being grouped with "Benefits" or labeled a "Fringe Benefit" (like all of the other costs mentioned in the same paragraph in Appendix E), rather than being grouped with "Salaries" (like paid absences).

Graduate's claims about the NIH Grant Appeals Board process do not provide a basis for reversing the disallowance.

Graduate attacked the fairness of the process used by the NIH Grant Appeals Board, primarily on the ground that, after negotiations broke down, the NIH Grant Appeals Board notice that the informal appeal process would be resumed was sent to Graduate's former President when it should have been sent to the person designated to represent Graduate in the negotiations. Graduate also said that it was not made aware of what was in the record before the NIH Grant Appeals Board, nor given an opportunity to supplement that record before the decision was issued.

NIH argued that it acted reasonably in sending the notice to the former President, since he was the person who had filed the appeal on Graduate's behalf. NIH also argued that any flaws in the NIH process were irrelevant, given that our review is a de novo review and that Graduate has had ample opportunity to submit any additional evidence to us. We agree.

While Graduate did inform NIH that it intended to be represented by accounting/auditing experts "during negotiations" and that its Assistant Comptroller would coordinate negotiations, the letter in which Graduate did this does not address who would represent Graduate in the informal appeals process if the negotiations broke down. GR Ex. 9. Although a later letter from Graduate's Director of Sponsored Project Accounting providing information to DCA stated that he should be contacted about further questions about the "attached material," this letter also is insufficient to inform either DCA or NIH that the President was no longer Graduate's official representative for purposes of the appeal. GR Ex. 14. Graduate did not allege that it had timely notified NIH that it had a new President. Under these circumstances, NIH was reasonable in sending the notice to the person who had originally filed the appeal on Graduate's behalf, as its President.

In any event, any flaws in a preliminary review process are not a basis for reversing a disallowance. The purpose of the informal, preliminary appeal process in 42 C.F.R. Part 50, Subpart D, is to preclude submission of cases under the more formal Departmental Appeals Board process at 45 C.F.R. Part 16 before the grantor agency has had an opportunity to review the decisions of its officials and to settle disputes with grantees. 42 C.F.R.§ 50.403. The informal process does contemplate an opportunity for a grantee to submit additional information to the review committee beyond what was submitted with the request for review. Graduate's loss of this opportunity appears to be attributable to its own failures, rather than to NIH's. Even if we determined that NIH had improperly failed to provide this opportunity by sending the notice to the wrong person, however, such a failure would not be fatal to NIH's case. Graduate had ample opportunity before us to submit any relevant documentation that Graduate thought should have been considered by NIH, but was not.

The disallowance is upheld in principle, but Graduate should be offered a limited opportunity to provide documents that might reduce the disallowed amount.

Graduate argued that it should have the opportunity to provide an audit that would show that it did not claim reimbursement under its NIH grants in excess of its actual allowable costs. The NIH Grant Appeals Board denied this request on the ground that the documents needed to substantiate this assertion were no longer available or reliable.

To the extent that Graduate is seeking to reopen its indirect cost rate negotiations, we determine that this would not be appropriate. Graduate signed agreements for each fiscal year of the disallowance period that set rates that were either final rates or predetermined rates. Those agreements treated employer contributions to FICA as indirect, rather than direct, costs and Graduate's failure to make the appropriate adjustments to its claims resulted in duplicate reimbursement for those costs. Graduate failed to establish that there was a mutual mistake that should result in recission of the agreements. The circumstances here are not the type of unusual circumstance that would justify reopening the rate negotiations.

On the other hand, the disallowance calculated by DFAS was apparently based at least in part on budgeted amounts, rather than on actual costs from Graduate's cost reports or records. For example, the DFAS determination that Graduate had for some grants for some years claimed indirect costs in excess of the negotiated rates was apparently based on budgeted, rather than actual, amounts, yet there are some financial status reports in the record showing the rates used in those reports. If either Graduate or NIH has the financial status reports relevant to the determination that Graduate claimed in excess of the applicable rate, those reports should be examined to determine whether any mistake by Graduate in budgeting for indirect costs at an excessive rate was later corrected.

Moreover, the record contains what appears to be expense summaries indicating that Graduate had direct costs in the salary category or other cost categories in excess of the budgeted amounts and overall greater costs than the amount awarded. If it was within Graduate's authority to charge those direct costs to NIH grant funds in place of the FICA costs that were improperly charged as direct costs and those unreimbursed costs were allowable, some reduction of the disallowance of direct costs and associated indirect costs might be warranted.

We understand NIH's concern that Graduate may not have reliable records available to meet Graduate's burden to document that it incurred allowable costs that should reduce the disallowance amount. We determine, however, that Graduate should be given a limited opportunity to provide documentation satisfactory to NIH to show that a reduction is warranted. Thus, we uphold the disallowance in principle, subject to reduction if Graduate produces documentation, within a reasonable time period set by NIH, that establishes an appropriate reduction.

Cecilia Sparks Ford

Marc R. Hillson

1. We cite here to the version of Part 74 in effect during the relevant time period. We note that the record contains Notices of Grant Award for some of the grants at issue here. Each of these notices references Part 74, as well as PHS policy issuances. See, e.g., Graduate Exhibits (GR Exs.) 3, 4.

2. Appendix E states:

Where a distribution can be made by assignment of a cost grouping directly to the area benefited, the distribution should be made in that manner. Care should be given, however, to eliminate similar or duplicative costs from any other distribution made to this area.

App. E, § V.B.3. General considerations on cost groupings include the following:

Where the hospital elects to treat as indirect charges the costs of pension plans and other staff benefits, such costs should be set aside as a separate cost grouping for selective distribution to related cost centers, including organized research.

App. E, § V.B.4.e.

3. OASC-3 is titled: "A Guide for Hospitals: Cost Principles and Procedures for Establishing Indirect Cost and Patient Care Cost Rates for Grants and Contracts with the Department of Health Education and Welfare." Graduate's original appeal letter in 1996 referred to OASC-3 as "still in effect." GR Ex. 7.

4. NIH Exhibit 1 is the record for the NIH Grant Appeals Board decision. Although many of these documents were also submitted by Graduate, we refer to the NIH exhibit for ease of reference since the pages are numbered sequentially on the bottom .

5. Graduate in effect acknowledged that a 1987 Agreement contained this same provision, as DFAS had found. GR br. at 9. The 1987 Agreement is not included in the record, although it is referred to in several places. In any event, since the 1988 Agreement established "final" rates for periods covering 7/1/84 through 6/30/88, "predetermined" rates for periods covering 7/1/88 through 6/30/91, and provisional rates for "7/1/91 Until Amended," the 1988 Agreement governs all of the disallowance period, except the period from 7/1/91 to 6/30/93. That period is governed by the 1993 Agreement, which set the "final" rate for that period and which we discuss below.

6. Paragraphs with the exact same wording appear under the heading "Fringe benefits" in the cost principles for non-profit organizations. See Office of Management and Budget Circular A-122, Attachment B, § 7.f.

7. Expenses under the heading "administrative and general" are those incurred for the administrative offices of the hospital that do not relate solely to any major division of the institution. App. E, § VI.B.

8. We also note that an amount for "Benefits" is included as an indirect cost for "overhead" related to Research Administration. No explanation for that amount is apparent from the documents, and Graduate did not offer one. How this amount was allocated to Research Administration might have been explained by Medicare Cost Report Worksheets B and B-1, generally used to show the step down of costs. Like the "Fringe Benefits Questionnaire," however, these Worksheets are referred to as having been included with the indirect cost proposal, but were not included with the materials Graduate provided in response to the disallowance.

9. We note that the language on fringe benefits in an agreement in an appendix to OASC-3 is different from the provision at issue here. OASC-3 at 54 ("Fringe benefits applicable to direct salaries and wages are treated as direct costs. All other fringe benefits are treated as indirect costs"). Unlike the provision at issue here, the language in the example neither spells out what wage-related costs are treated as direct nor contains the emphasis that other fringe benefits are not direct costs.

10. Fringe benefit rates are in effect a type of indirect cost rate since they pool fringe benefits and distribute them on a direct cost base, here salaries and wages.

11. Section 2144.1 of the HIM-15 manual defines fringe benefits as "amounts paid to, or on behalf of, an employee, in addition to direct salaries or wages, and from which the employee, his/her dependent . . ., or his/her beneficiary derives a personal benefit before or after the employee's retirement or death . . . ." Section 2122.3 deals specifically with employment-related taxes of provider-based physicians, including FICA, workers' compensation and unemployment compensation, and states that such taxes are considered business expenses of the employer and not fringe benefits. The reason for this is that the provider-based physician costs are reimbursable as physician services, not as hospital services under Medicare. The fact that this statement is limited to taxes for physicians may indicate a different rule for non-physician employees, moreover.