370 F3d 1153 Nvt Technologies Inc v. United States

370 F.3d 1153

NVT TECHNOLOGIES, INC., Plaintiff-Appellant,
v.
UNITED STATES, Defendant-Appellee.

No. 03-5045.

United States Court of Appeals, Federal Circuit.

June 3, 2004.

William T. Welch, Barton, Baker, McMahon & Tolle, LLP, of McLean, VA, argued for plaintiff-appellant. With him on the brief was John R. Tolle.

Richard S. Ewing, Attorney, Commercial Litigation Branch, Civil Division, United States Department of Justice, of Washington, DC, argued for defendant-appellee. With him on the brief were David M. Cohen, Director; Bryant G. Snee, Assistant Director; and Wanda Rubianes-Collazo, Attorney.

Before GAJARSA, LINN, and PROST, Circuit Judges.

Opinion for the court filed by Circuit Judge LINN. Dissenting opinion filed by Circuit Judge PROST.

LINN, Circuit Judge.

1

NVT Technologies, Inc. ("NVT"), appeals from a final decision of the United States Court of Federal Claims, denying NVT's motion for summary judgment and granting the government's cross-motion for summary judgment on the administrative record, in NVT's protest over the final cost comparison determination of a bid evaluation process. NVT Techs., Inc. v. United States, 54 Fed.Cl. 330 (2002). The Court of Federal Claims concluded that the solicitation at issue in this case was not ambiguous, that the government's interpretation of the solicitation was correct, and that the agency's bid evaluation determination would stand. The trial court further opined that, in the alternative, if the solicitation was ambiguous, the ambiguity was patent and NVT could not prevail for that reason. Because we hold that the solicitation was ambiguous and the ambiguity was patent, we affirm the judgment of the Court of Federal Claims.

BACKGROUND

2

On November 15, 2000, the United States Department of the Navy ("Navy" or "agency") issued solicitation number N68711-01-R-5101 to obtain bids for facility maintenance and utility services, including facilities maintenance and repair, utilities operations and maintenance, facilities planning, engineering, and environmental support services, for the Marine Corps Recruit Depot in San Diego, California. The Depot is used by a population of up to 9,600 people, including 2,600 permanent personnel and between 3,500 and 7,000 Marine recruits. The appearance of the facility is key, and subject to a policy that "the unique buildings/facilities... be maintained in a professional manner as a projection of Marine Corps pride that is world renowned."

3

The solicitation was issued as part of a government cost comparison effort to determine whether it was more efficient to accomplish the specified maintenance and services for the Depot under an outside contract or by government performance of the tasks. Office of Management and Budget ("OMB") Circular No. A-76 sets forth an elaborate mandatory process for comparing the costs of in-house and private production. In performing the cost comparison under OMB A-76, the agency is to compare the bid of the private sector source with the cost of providing the good or service with government facilities and personnel, or the agency's "Most Efficient Organization" or "MEO." See Am. Fed'n of Gov't Employees v. United States, 258 F.3d 1294, 1296 (Fed.Cir.2001). The MEO is not necessarily an existing organization, but the organization the agency would establish if it were competing for the work. Id. at 1296 n. 1. If government performance is determined to be more economical, no award is made under the request, and the solicitation may be canceled.

4

The bid process was to consist of three steps. First, interested bidders from the private sector would submit "statements of qualifications." These submissions would be evaluated and the agency would inform each offeror, based on that evaluation, whether it was considered to be a "viable competitor." Bidders, however, would be able to continue to the second step of the bid process regardless of the evaluation results. Next, each offeror would submit technical and price proposals. These proposals would be reviewed, and a competitive and acceptable offer would be selected as the private-sector offer to compete against the agency's in-house proposal. Finally, the selected private-sector offer would be compared to the in-house proposal to ensure that both offers offered the same level of performance and the proposed costs would be compared. The initial award would then be given to the lowest cost bidder.

5

In response to the solicitation, the Navy received statements of qualifications from three offerors, including NVT. After evaluation, only NVT was determined to be a viable competitor, and NVT alone elected to continue in the remainder of the competition. Prior to the closing date for receipt of NVT's technical and price proposals, the Naval Audit Service certified that the agency's management plan satisfied the Performance Work Statement requirements and that the in-house cost estimate was reasonable. The agency's bid was then sealed.

6

After NVT's technical and price proposals were submitted, the agency conducted two rounds of discussions with the company. The agency determined that NVT's final revised proposal was acceptable, and a price evaluation board further found NVT's proposed price to be realistic, complete, and reasonable. The agency thus selected NVT's proposal as the private-sector offer to compete against the government's in-house plan. A Navy quality comparison team then evaluated the agency's in-house plan "to ensure that the technical proposals of NVT and the government offered the same level of performance and performance quality." The team found that the in-house plan was "balanced in performance and quality to that proposed by [NVT]." The Navy then proceeded to a cost comparison and concluded that in-house performance of the specified work would cost less than performance by NVT. Specifically, NVT's adjusted cost at $43,025,120 was found to be $3,937,980 higher than the agency's proposed cost of $39,087,140. The initial award went to the in-house MEO on July 26, 2001.

7

On August 27, 2001, NVT appealed the cost comparison decision to the administrative appeal authority ("AAA"), alleging inter alia that the in-house estimate for materials was too low, but identifying no particular provision it believed was misinterpreted. On August 30, 2001, the agency revised its estimate in response to other allegations in NVT's appeal, increasing its cost proposal. However, the agency's revised estimate was still less than NVT's proposal. On September 27, 2001, the AAA issued a final determination, accepting the agency's amended estimate and denying the remainder of NVT's allegations, including the low materials estimate. On October 4, 2001, NVT protested the AAA's final decision to the General Accounting Office ("GAO"). In addition to raising several issues it had raised before the AAA, NVT also argued that the government's estimate was too low with respect to roof leak repairs, raw labor estimates, and performance of ceramic tile repair. On January 3, 2002, the GAO issued a decision denying NVT's protest.

8

On January 24, 2002, NVT filed a complaint in the Court of Federal Claims, alleging that the government had unlawfully concluded that it should continue base operations and maintenance services in-house. In particular, NVT contended that the government's interpretation of the volume of ceramic tile repair specified by the solicitation was wrong, and that if the proper interpretation had been used in the in-house estimate, NVT's bid would have been lower, entitling NVT to the award. The government responded that NVT's reading of the solicitation is not supported, that the government's interpretation is reasonable, and that the award to the MEO should be affirmed. On June 5, 2002, the Court of Federal Claims heard oral argument on these issues. At argument, the trial court questioned whether NVT had made a mistake of bid and whether the AAA had an obligation to "ferret out that mistake" and allow NVT to correct it. The trial court then ordered supplemental briefing on the bid mistake issue and on the government's obligation to inform NVT of NVT's misreading of the required specifications.

9

At the heart of the proceedings before the trial court, and on appeal to this court, is Solicitation Attachment J-C 2(a), a table containing hundreds of line-item work-load estimates. The table includes a "Paragraph Number" reference to the text of the solicitation; a "Title," identifying the item; and three columns, headed: "Number," "Unit of Measure," and "Frequency," reflecting the scope of work for that line-item. An exemplary entry related to Preventative Maintenance ("PM") for the Motor Vehicle Wash Facility is reflected in the following line item:

10

Paragraph                                                     Unit of
Number          Title                                Number   Measure   Frequency
------          -----                                ------   -------   ---------
C-5.2.3.2.8.5   PM Spray Nozzles                     41       EA        Monthly
11

This line item requires the contractor to assign sufficient staff at the appropriate skill level and pay grade to perform preventative maintenance on each of 41 spray nozzles once a month. The contractor also was required to include costs of materials to perform the stated tasks. Because the bids were to be based on a one-year cost proposal, the Frequency column provided an indication of how often, in a year, each of the listed tasks were to be performed. For example, based on the above line item, the offeror's price would be determined by multiplying the cost of spray nozzle maintenance times 41 nozzles, times the twelve "monthly" occurrences required. It is undisputed that this is the proper interpretation for all but thirteen line items in Attachment J C 2(a).

12

The specific table entries in dispute in this case are the first twelve of the thirteen line items in Attachment J-C 2(a), headed "MISC CERAMIC TILE REPAIR." All thirteen line items in this category are reproduced below:

13

Paragraph                                                     Unit of
Number          Title                                Number   Measure   Frequency
------          -----                                ------   -------   ---------
C-5.2.3.2       MISC CERAMIC TILE REPAIR
C-5.2.3.2       Form, Place & Finish Concrete        119      SF        12
C-5.2.3.2       Lay 8"×8"×16" Concrete Cinder
                  Block                               52      SF         6
C-5.2.3.2       Lay Brick                              7      SF         5
C-5.2.3.2       Miscellaneous Concrete Repairs        60      SF        16
C-5.2.3.2       Prepare, Place & Finish Plaster       48      SF        19
C-5.2.3.2       Prepare, Place & Finish Stucco        48      SF        18
C-5.2.3.2       Replace 1"×1" Ceramic Wall Tile       59      SF         4
C-5.2.3.2       Replace 12"×12" Ceramic Marbleized
                  Floor Tile                          98      SF         9
C-5.2.3.2       Replace 2"×2" Ceramic Wall Tile       59      SF         8
C-5.2.3.2       Replace 4"×4" Ceramic Wall Tile      387      SF        89
C-5.2.3.2       Replace 6"×6" Ceramic Quarry
                  Floor Tile                         365      SF        97
C-5.2.3.2       Replace 6"×6" Ceramic Wall Tile       52      SF         5
C-5.2.3.2       Cyclic Maint to Repair Ceramic
                  Tile in Messhalls, etc.              6      HRS       Daily
14

The last of the thirteen items listed in this category is like the example noted above, and its interpretation is not in question.

15

In calculating its estimate for the twelve disputed line items, NVT multiplied the "Number" column by the "Frequency" column and prepared its technical and price proposals based on a total of 76,209 square feet of tile and 28,647.85 man-hours of labor. MEO's bid instead took the figures in the "Number" column to represent the total square feet of tile for each line item. The MEO did not use the "Frequency" column as a multiplier, and its cost estimate was thus based on 1,354 square feet of tile and 1,354 man-hours of labor.

16

Before the Court of Federal Claims, both parties argued that the solicitation plainly supported their respective interpretations. The trial court concluded that the solicitation was not amenable to two reasonable interpretations, that only the government's reading was reasonable, and that the solicitation was therefore unambiguous. The trial court went on to conclude that, even if NVT's reading of the provisions at issue was also reasonable — making the solicitation ambiguous — any ambiguity was patent, and NVT could not prevail. On the issue of mistake in bid, the trial court's analysis focused on whether a different rule for mistake in bid should apply in A-76 cases if the agency had notice of the error. The trial court rejected NVT's contention that the solicitation was defective, which would require a rebid of the solicitation, concluding that NVT should have instead sought clarification about the solicitation. For these reasons, the Court of Federal Claims granted the government's motion for summary judgment on the administrative record and denied NVT's motion for summary judgment.

17

NVT timely appealed to this court, asserting that the Court of Federal Claims erred in concluding that: 1) the government's interpretation was reasonable; 2) that NVT's interpretation was not reasonable; 3) that the solicitation was not ambiguous; and 4) that the ambiguity, if any, was patent. NVT also appeals the mistake in bid issue. Jurisdiction over this A-76 bid protest case was proper before the Court of Federal Claims. See Am. Fed'n of Gov't Employees, 258 F.3d at 1297-1298. We have jurisdiction over final decisions of the Court of Federal Claims under 28 U.S.C. § 1295(a)(3).

ANALYSIS

A. Standard of Review

18

Bid protest actions are subject to the standard of review established under section 706 of title 5 of the Administrative Procedure Act ("APA"), 28 U.S.C. § 1491(b)(4) (2000), by which an agency's decision is to be set aside only if it is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law," 5 U.S.C. § 706(2)(A) (2000). See Impresa Construzioni Geom. Domenico Garufi v. United States, 238 F.3d 1324, 1332 (Fed.Cir.2001); Advanced Data Concepts, Inc. v. United States, 216 F.3d 1054, 1057 (Fed.Cir.2000).

19

We review a trial court's determinations on summary judgment de novo, Impresa, 238 F.3d at 1330, independently examining the administrative record to determine whether any "genuine issue[s] as to any material fact" are present and whether the "moving party is entitled to a judgment as a matter of law," Fed.R.Civ.P. 56(c). Because no genuine issues of material fact exist here, we review the trial court's decision as a matter of law.

20

The interpretation of a contract or solicitation is a question of law which we review de novo. Wickham Contracting Co. v. Fischer, 12 F.3d 1574, 1577 (Fed.Cir.1994); Fortec Constructors v. United States, 760 F.2d 1288, 1291 (Fed.Cir.1985). Whether a contract provision is ambiguous is also a question of law. Cmty. Heating & Plumbing Co. v. Kelso, 987 F.2d 1575, 1579 (Fed.Cir.1993). Whether an ambiguity is patent or latent is likewise a question of law. Interwest Constr. v. Brown, 29 F.3d 611, 614 (Fed.Cir.1994). This determination is made on a case-by-case basis. Interstate Gen. Gov't Contractors, Inc. v. Stone, 980 F.2d 1433, 1435 (Fed.Cir.1992).

21

B. Was the solicitation ambiguous?

22

The threshold question in this appeal is whether the solicitation plainly supports only one reading or supports more than one reading and is ambiguous. "To show an ambiguity it is not enough that the parties differ in their respective interpretations of a contract term. Rather, both interpretations must fall within a `zone of reasonableness.'" Metric Constructors, Inc. v. NASA, 169 F.3d 747, 751 (Fed.Cir.1999) (citations omitted). Before this court, each party argues that there is no ambiguity because its respective interpretation of the solicitation is plainly supported, and thus reasonable, while the other party's interpretation is unreasonable.

23

Contract interpretation begins with the language of the written agreement. Foley Co. v. United States, 11 F.3d 1032, 1034 (Fed.Cir.1993). When interpreting the contract, the document must be considered as a whole and interpreted so as to harmonize and give reasonable meaning to all of its parts. McAbee Constr., Inc. v. United States, 97 F.3d 1431, 1434-35 (Fed.Cir.1996). An interpretation that gives meaning to all parts of the contract is to be preferred over one that leaves a portion of the contract useless, inexplicable, void, or superfluous. Gould, Inc. v. United States, 935 F.2d 1271, 1274 (Fed.Cir.1991).

I.

24

In preparing its estimate, NVT multiplied the figures in the "Number" column by the corresponding figures in the "Frequency" column for all of the hundreds of items in the Solicitation Attachment J C 2(a), including each of the twelve disputed line items in the "MISC CERAMIC TILE REPAIR" portion. For example, NVT interpreted the first disputed line item, "Form, Place & Finish Concrete," to call for 1,428 square feet of concrete, or 119 square feet of concrete poured and finished twelve times per year (119 × 12 = 1428).

25

NVT argues that its interpretation was reasonable. NVT notes that the words "monthly," "quarterly," or "weekly" appeared in the "Frequency" column for most of the hundreds of line items in Attachment J-C 2(a) and that to prepare the estimate for each of these line items, the value in the "Number" column needed to be multiplied by the number 12, 4, or 52, respectively, representing the number of times in a calendar year that the work was to be performed. NVT used the same approach for the twelve line items in question, using the number in the Frequency column as a multiplier as it had for the numerical equivalent of the words in the Frequency column for all of the other line items. NVT argues that its interpretation, applying the same logic to the line items in question as to the other line items in the schedule, is a harmonious reading of all of the provisions of the contract. NVT further notes that nothing in the specification suggests that a different methodology be used for the line items in dispute.

26

Finally, NVT argues that its price estimate is reasonable. Although NVT calculated a total of 76,209 square feet of tile to be replaced, NVT contends, over the government's objections, that while that figure was large, the total floor space of the Depot is over 3 million square feet. Thus, NVT's figure is only 2% of the total floor space. Moreover, NVT notes that the solicitation emphasized the importance of the appearance of the Depot and the heavy volume of traffic. For these reasons, although NVT's bid of over $6 million for the miscellaneous ceramic tile work alone is a significant portion of the $43 million total bid, NVT contends that there is nothing to indicate or suggest that this was not what was called for in the solicitation.

27

The government argues that NVT's interpretation is incorrect and unreasonable because it is inconsistent with other portions of the solicitation. In particular, the government notes that NVT's construction results in a disproportionate amount of tile, given that the corpus of the solicitation only refers incidentally to ceramic tile. Further, the government asserts that NVT's interpretation is contrary to the terms "maintenance" and "preventative maintenance," as defined in the solicitation. "Maintenance" is defined as "[t]he recurrent day-to-day, periodic or scheduled work required to preserve or restore a facility to such condition that it may be effectively utilized for its designed purpose." "Preventative maintenance" means "[s]ervicing accomplished on a scheduled or recurring basis to lengthen the equipment life. The effort involves visual inspection, operational tests, lubrication, adjustments and the repair or replacement of minor defective component parts." Given these definitions, the government argues that NVT's interpretation, resulting in the installation of 1.3 football fields worth of replacement tile per year, is unreasonable, while the government's estimate, resulting in much less replacement tile, is reasonable. The government further contends that NVT's interpretation is unreasonable given the information available to each bidder in the solicitation in the form of explanatory facility history files, technical library materials, and a software database file. The government argues that this information would have informed NVT's bid and that NVT failed to take advantage of these resources. For these reasons, the government asserts that NVT's interpretation is inconsistent with the corpus of the solicitation and therefore is unreasonable.

II.

28

The government interprets the line items in question as not requiring multiplication of the "Number" and "Frequency" columns as was done by NVT. Rather the government's interpretation considers only the Number column and does not use the numbers in the Frequency column as multipliers. The government contends that this was reasonable because it was apparent that these line items should be treated differently, based on the presence of numbers rather than words in those line items, as well as the fact that the numbers bear no relationship to the regular and periodic repetitions signaled by the words. The government asserts that its reading is consistent with the rest of the solicitation, because the ceramic tile repair is denoted as "Miscellaneous" and because if the disputed line items were intended to represent such a large portion of the contract, more information would have been given about these tasks in the specification.

29

NVT contends that the government's interpretation, where the figures in the Frequency column for the disputed line items are simply ignored, violates the contract interpretation canon that a reading that renders terms superfluous is disfavored. NVT argues that for the government's interpretation to be consistent with the remainder of the document, the Frequency column should have read "Annually" or been left blank, which is not the case. NVT also argues that the government's interpretation renders meaningless the thirteenth line item "Cyclic Maint to Repair Ceramic Tile in Messhalls, etc."

III.

30

We conclude that both NVT's and the government's interpretations are within the "zone of reasonableness." Metric Constructors, 169 F.3d at 751. NVT's interpretation is reasonable, because it applies the same logic to the thirteen disputed line items as applied to the hundreds of other line items present in the schedule. The schedule nowhere indicated that the data in the "Number" and "Frequency" columns were to be treated differently for the thirteen items in question or that the frequency numbers for those items should be ignored. Moreover, although NVT's interpretation resulted in an extensive amount of tile replacement, the amount was not wholly unreasonable: (a) given that the amount only represents a small percentage of the total floor space of the facility; (b) given the extensive use of the facility; and (c) given the mandate in the solicitation that the appearance of the Depot is of great importance. Further, despite the emphasis that the government places on the fact that NVT failed to avail itself of the external resources provided by the Navy, the government points to no particular portion of the external resources that would be inconsistent with NVT's interpretation.

31

We also find the government's interpretation to be reasonable. The government's interpretation is consistent with the notion of incidental repair and maintenance, as noted in the corpus of the solicitation. Moreover, the portion of the price attributable to tile repair and maintenance in the government's estimate is proportional to what would be expected for a small set of line items that received no special treatment in the solicitation. There was no evidence in the record illuminating the government's understanding of the significance of the numbers in the "Frequency" column for the disputed line items, and in fact, at oral argument, government's counsel admitted that he was unaware of the significance of those numbers; however, the point is not what the numbers in that column meant, but rather whether it was unreasonable for the government to treat them differently from the words used in the Frequency column for all of the other line items. Recognizing that on their face, the numbers in the Frequency column are different from the words used elsewhere in the attachment, and further recognizing that those numbers do not have any apparent correlation to the regular and periodic repetitions signaled by the words used for the other line items, we do not think it was unreasonable for the government not to use them as multipliers, especially because the use of the numbers as multipliers results in such seemingly disproportionate totals and the non-use of the numbers as multipliers yields totals that are entirely consistent with the corpus of the solicitation and the attention given in the solicitation to the line items in dispute.

32

Because "both interpretations ... fall within a zone of reasonableness," Metric Constructors, 169 F.3d at 751, we conclude that the solicitation, with respect to these disputed line items, was ambiguous.

33

C. Was the ambiguity patent?

34

Because we hold that the solicitation was ambiguous, we turn now to the question of whether the ambiguity was patent. An ambiguity will only be construed against the government if it was not obvious on the face of the solicitation and reliance is shown. See, e.g., Edward R. Marden Corp. v. United States, 803 F.2d 701, 705 (Fed.Cir.1986). If the ambiguity is patent, it triggers a duty to inquire. A patent ambiguity is one that is "obvious, gross, [or] glaring, so that plaintiff contractor had a duty to inquire about it at the start." H & M Moving, Inc. v. United States, 204 Ct.Cl. 696, 499 F.2d 660, 671 (1974). If an ambiguity is obvious and a bidder fails to inquire with regard to the provision, his interpretation will fail. Triax Pac., Inc. v. West, 130 F.3d 1469, 1475 (Fed.Cir.1997).

35

NVT argues that while the line items in question used numbers unrelated to the calendar in the Frequency column, and differed in that respect from the remaining hundreds of line items that made reference to regular and periodic calendar occurrences as Frequency entries, these differences do not create a patent ambiguity. Rather, NVT contends that instead of focusing on the uniqueness of the disputed line items, it is more important to understand that there is nothing about those line items that prohibits a reasonable reading consistent with the remainder of the solicitation. The government argues that the fact that numbers were used in the Frequency column for these twelve line item entries, and not for the others, should have raised a red flag for NVT, requiring NVT to inquire. The government also argues that in tallying its estimate and noting that these thirteen line items accounted for a large portion of its total price estimate — approximately $6 million out of a $43 million bid, or some 14% — despite their being treated as incidental in the solicitation, NVT was also put on notice that it should have inquired. Where, as here, a certain set of line items is expressed in a manner so different from hundreds of other line items, yielding totals disproportionate to the remainder of the solicitation, we find the differences to be "obvious, gross, [or] glaring," H & M Moving, 499 F.2d at 671, requiring NVT to inquire. Because the solicitation was ambiguous, but any ambiguity was patent, and because it is undisputed that NVT did not inquire into such ambiguity, NVT cannot prevail.

D. Mistake in Bid

36

Finally, NVT argues that even if we find the solicitation ambiguous, its bid mistake argument may still prevail, based on In re IT Corporation, 2002 WL 1476286 (Comp.Gen. July 10, 2002). The government counters that the trial court correctly denied NVT's mistake in bid claim because NVT failed to avail itself of the full information available and referenced in the specification. Moreover, the government notes that it was unaware of any solicitation defect at a time when a correction would be appropriate — i.e., before the selection of the private sector's best proposal — because the significance of the disputed line items was not raised until NVT's appeal before the GAO. The government also distinguishes the IT Corporation case, because in that case, the agency concluded that all of the information available to the offerors did not adequately identify the work required to meet the needs of the contract.

37

IT Corporation, which does not bind this court, stands for the proposition that if a solicitation is defective, such that bidders cannot know on what they are bidding, the solicitation must be rebid. After receiving supplemental briefing from the parties, the trial court carefully analyzed this issue and concluded that the fact that this was an A-76 solicitation did not affect the analysis of this issue, because the associated FAR provisions give no additional rights to bidders of A-76 offers. The trial court further concluded that because NVT failed to inquire, its contention that the solicitation was defective must fail, citing Comtrol, Inc. v. United States, 294 F.3d 1357, 1363-65 (Fed.Cir.2002) (stating that a contractor cannot rely on a defective specification where the contractor failed to access all of the information available and incorporated into the specification). The trial court made no findings as to whether the external information referenced in the solicitation would have provided NVT with complete knowledge, unlike the situation in IT Corporation.

38

We agree with the trial court's conclusion, if not its basis, and hold that NVT's mistake in bid argument must fail. In particular, "[i]f a contractor enters into a contract aware of the fact of defective specifications, it is not entitled to recover on a claim based on these defective specifications." Robins Maint., Inc. v. United States, 265 F.3d 1254, 1258 (Fed.Cir.2001) (quoting Johnson Controls, Inc. v. United States, 229 Ct.Cl. 445, 671 F.2d 1312, 1320 (Ct.Cl.1982)). Here, we have concluded that the ambiguity in the solicitation, which is one and the same with the defect alleged by NVT, was patent, and the contractor should have thus inquired. NVT, in failing to inquire as to an ambiguity, cannot then claim that the specification was defective to absolve itself of that failure. Therefore, we affirm the trial court's denial of NVT's mistake in bid claim.

CONCLUSION

39

Because both NVT and the government have proffered interpretations of the disputed line items that are reasonable, we conclude that the solicitation was ambiguous. However, because the ambiguity in the disputed line items was patent, and should have prompted an inquiry by NVT that it failed to make, we conclude that NVT cannot prevail. Further, due to NVT's failure to inquire, NVT's mistake in bid claim must fail. Accordingly, the decision of the Court of Federal Claims is

40

AFFIRMED.

41

PROST, Circuit Judge, dissenting.

42

The majority concludes that the solicitation was ambiguous and that this ambiguity was patent. I respectfully dissent from both of these conclusions. Instead, I find the solicitation unambiguous because only NVT's interpretation thereof falls within a zone of reasonableness. In the alternative, I find any ambiguity to be latent. Thus, because NVT's bid would have been lower than the government's bid under the proper interpretation, I believe NVT should have been awarded the contract. I would therefore reverse the decision of the Court of Federal Claims.

43

* The majority's conclusion that the solicitation was ambiguous is based on its finding that both NVT's and the government's interpretations were reasonable. While I agree as to NVT, I disagree with respect to the government. As a general matter, each of the hundreds of lines in the solicitation gave details for the listed task under columns titled "Number," "Unit of Measure," and "Frequency." In addition, these bids were to be based on a one-year cost proposal, meaning that the "Frequency" column provides an "indication of how often, in a year, each of the listed tasks were to be performed." Opinion at 1157. Accordingly, for each line item — including the twelve line items in question — NVT performed the same calculation, multiplying the value in the "Number" column by the value in the "Frequency" column to arrive at a yearly total. This, I believe, was reasonable for a number of reasons. To begin, NVT's reading of the twelve disputed line items was consistent with its reading of the more than 800 other line items. In addition, NVT's interpretation does not read out material parts of the solicitation, i.e., the column headings. NVT used the frequency values for these twelve line items just as it did for all the other line items; it did not ignore them.

44

Further, as the majority notes, the end result of NVT's bid calculation method was also reasonable. The 287 identified buildings at the depot were built between 1918 and 1984, cover approximately 3,221,472 square feet of floor space on 433 acres, and are used by a population of up to 9,600 people. Thus, I agree with the majority that NVT reasonably interpreted the solicitation to require 76,209 square feet of tile replacement and 28,647.85 man-hours of labor. The estimated amount of tile replacement only amounted to a small percentage — two percent — of the total floor space. Opinion at 1161. This included 36,287 square feet of floor tile, approximately one percent of the total facility floor space, and 37,187 square feet of wall tile, plaster, or stucco. Moreover, these numbers also appear reasonable in light of the high maintenance standards stressed in the solicitation. Marine Corps policy dictated that "the unique buildings/facilities ... be maintained in a professional manner as a projection of Marine Corps pride that is world renowned." To that effect, the Performance Work Statement required contractors to ensure that "all buildings, surfaces, and component parts remain in a serviceable condition and are free of warping, bucking, cracks, ... loose, missing or chipped tiles; noticeable surface imperfections, ... and other damage or defects." Given the contract's high standards, the majority correctly concluded that NVT's bid did not call for a disproportionate amount of tile repair and was therefore reasonable.

45

I disagree, however, with the majority's conclusion that the government's interpretation of the solicitation was also reasonable (and therefore demonstrates the existence of an ambiguity). According to the majority, because "the numbers in the Frequency column are different from the words used elsewhere," id. at 1161-62, and because they bear no relationship to any regular and periodic repetition, it was reasonable for the government to treat these twelve disputed line items differently. Specifically, in contrast to the more than 800 other line items that appear under identical column headings, the government did not use the value under the "Frequency" column as a multiplier for the value in the "Number" column. Id. In addition, the majority finds the government's interpretation reasonable because the relevant estimate was consistent with the notion of incidental repair as well as proportional to what would be expected for a small set of line items that received no special treatment in the solicitation. Id.

46

I disagree for several reasons. First, given that the solicitation uses many different frequency indicators in a broad spectrum of formats, the fact that the disputed line items contained solely numbers did not justify the government's (and the majority's) dissimilar treatment thereof. This variety includes different primarily verbal, periodic indicators such as the following: every # years, Annually, # X Year, Semi-Annually, Quarterly, Bi-Monthly, Monthly, 2X Monthly, Bi-Weekly, Weekly, Daily, 2X Daily. But some line items give no specifics whatsoever, simply noting that the work needs to be performed "As needed" or "As-Req." Moreover, the solicitation includes line items that express frequency in terms of the number of "treatments a year" and the number of "times a year." In addition, the solicitation contains line items that give frequency in terms of hours and days, without any indication of periodicity. For example, with respect to hours, the solicitation contains line items calling for "20,000 hours," "6000 hrs," and "2500 hrs." With respect to days, the solicitation contains line items given in terms of "18 days," "60 Days," "90 Days," and "180 days." It even includes a line item with the "Frequency" given as "7/30 Days." Finally, in the twelve disputed line items, the solicitation uses purely numerical figures in the "Frequency" column.

47

No matter what format the line item may have been in, the solicitation clearly required the multiplication of the "Number" by the "Frequency" to arrive at the estimated annual cost. The fact that the twelve disputed line items do not contain any words does not make them any different. As demonstrated above, the line items come in many different formats, including words, a combination of words and numerals, and solely numerals. Moreover, the line items list frequency in both periodic and non-periodic terms. Accordingly, the twelve disputed line items do not differ from any standard format used throughout the solicitation because the solicitation has no such clear standard format. As with all these other line items, the government should have simply applied the default presumption that the value in the "Frequency" column represents a yearly (per year) number because the bid price was to be based on the cost for one year's worth of maintenance.

48

Second, as a legal matter, the government's interpretation violates the canons of contract interpretation that require (a) consistency, and (b) giving meaning to each contract term. First, the government's interpretation is unreasonable because it does not apply consistently to the entire solicitation. The solicitation called for the multiplication of the "Number" and the "Frequency" to arrive at the amount of work or materials required. For more than 800 line items, as the government acknowledges, bidders had to use this methodology to assemble their bid. But the government did not similarly treat the twelve disputed line items. In order to interpret the contract terms consistently, the government should have multiplied these "Frequency" figures by the "Number," applying the per-year presumption. As a corollary to the consistency issue, if the government did not intend the numbers given in the "Frequency" column to be a frequency multiplier, it could have drafted the solicitation to indicate such repairs were to be done "Annually," thereby requiring multiplication by one. This would have been consistent with the remainder of the solicitation, which has hundreds of "Annually" line items. By using a different value in the "Frequency" column for these twelve line items, however, the government clearly meant something different.

49

Further, by ignoring the "Frequency" value with respect to the disputed line items, the government also violated the canon of contract interpretation that disfavors rendering terms superfluous. If these numbers had no purpose, as government counsel suggested at oral argument, they should not have been listed in the "Frequency" column. Moreover, by ignoring these numbers, the government not only unreasonably ignored the values given in the individual line items, but also the column heading, which clearly indicates that the numbers themselves represent a "Frequency" to be used for purposes of assembling a bid.

50

In sum, the solicitation unambiguously required the multiplication of the "Number" and "Frequency" columns for each line item to arrive at a yearly figure for purposes of bid assembly, including the twelve disputed line items. As such, I find NVT's interpretation of the solicitation the only reasonable interpretation, and, for the reasons given above, disagree with the majority that the government's interpretation was also reasonable.

II

51

Even if I agreed with the majority that an ambiguity existed here, I do not find that ambiguity patent. The majority concludes that the ambiguity is patent because the line items are "expressed in a manner so different from hundreds of other line items, yielding totals disproportionate to the remainder of the solicitation." Opinion at 1162. Yet in reaching this conclusion, the majority abandons the reasons why it concluded an ambiguity existed in the first place. In particular, the majority disregards, in my view, the reasons it found NVT's interpretation reasonable, all of which point in the direction of finding a latent rather than patent ambiguity.

52

In my opinion, the ambiguity was latent. It was not, as the majority concludes, obvious, gross or glaring. As discussed above, the disputed line items do not differ dramatically from the other more than 800 line items; there are many other formats in which the solicitation lists "Frequency." NVT likely entered these line items, as with the hundreds of others, into a spreadsheet program, multiplying them out to arrive at a bid price. A look at NVT's proposal reveals that only two percent of the total floor space requires retiling every year. Yet the facility includes at least 287 buildings built between 1918 and 1984 on 433 acres with 3,221,472 square feet of floor space heavily used by up to 9,600 people at a time. Furthermore, the mandate in the solicitation requires that contractors maintain the facilities at the highest standards in order to project Marine Corps pride. Given this context, it is the government's and not NVT's estimates that seem disproportionately low, calling for replacement of only 1,354 square feet of tile. This represents only 0.04% of the total square footage of the facility. Assuming an ambiguity existed, therefore, NVT's interpretation was far more reasonable, reflecting the fact that any defect in the solicitation was latent.

53

The government drafted the solicitation and should bear the burden of any mistakes contained therein. In this case, the solicitation unambiguously called for bidders to multiply the "Number" and "Frequency" columns for each line item. Even if there was an ambiguity, the law requires that it be resolved against the drafter. As discussed, the patent ambiguity exception to this rule does not apply here. Because NVT's bid would have been lower than the government's bid had they been calculated using the same methodology, I respectfully dissent and would reverse the Court of Federal Claims decision granting the government's motion for summary judgment.