PROST, Circuit Judge. This bid-protest case arises from a peculiar procure- ment mechanism. Contracting officers often must discuss deficiencies and significant weaknesses in proposals with offerors before proposals are final. And so when an offeror proposes a price that is unreasonably high (so as to pre- clude an award), the government must discuss that unrea- sonableness with the offeror, potentially giving it a chance to revise its proposal to fix what may have went wrong. If the price is too high yet not unreasonable, the government

Case: 20-2041 Document: 99 Page: 3 Filed: 08/25/2021

DYNCORP INTERNATIONAL, LLC v. US 3 need not discuss it and the offeror need not get another try. The upshot is that an offeror whose initial proposal is un- reasonably priced may fare better than one whose isn’t. Here, six firms vied for spots to perform logistics work for the Army across the globe. DynCorp lost. Its prices were higher than its competitors’; its proposed technical approach was worse. After balancing four proposal-evalu- ation factors, none of which DynCorp was best on, the Army went with other offerors. Now DynCorp argues that the price it gave the Army was so high as to be unreasonable—and that the Army should have concluded as much and given it the oppor- tunity to revise its proposed approach. DynCorp takes is- sue with the Army’s price-reasonableness analysis, which it says skirted regulatory requirements and was irrational besides. The Court of Federal Claims dismissed DynCorp’s bid protest, finding no error in the Army’s analysis. As we ex- plain below, we agree. Accordingly, we affirm. BACKGROUND I. THE SOLICITATION This case involves the fifth iteration of the Army’s Lo- gistics Civil Augmentation Program—i.e., “LOGCAP V”— a procurement for logistics support services. Appellant, DynCorp International, LLC (“DynCorp”), was an unsuc- cessful offeror. Four of the appellees—Kellogg, Brown & Root Services, Inc. (“KBR”), Vectrus Systems Corpora- tion (“Vectrus”), Fluor Intercontinental, Inc. (“Fluor”), and PAE-Parsons Global Logistics Services, LLC (“P2GLS”)— were successful. By way of background, in November 2017 the Army is- sued the LOGCAP V solicitation under Request for Pro- posal No. W52P1J-16-R-0001. DynCorp Int’l LLC v. United States, 148 Fed. Cl. 568, 572 (2020) (“DynCorp I”). Case: 20-2041 Document: 99 Page: 4 Filed: 08/25/2021 4 DYNCORP INTERNATIONAL, LLC v. US Generally, LOGCAP is a procurement program for civilian logistics support services to the United States Army and related Department of Defense components throughout the world. In this iteration, the Army was to award four to six indefinite-delivery, indefinite-quantity (“IDIQ”) con- tracts—each covering services among six geographic com- mands, plus Afghanistan. ^[1] Concurrent with the IDIQ awards, the Army was also to award seven sets of task or- ders—one set for each command and another for Afghani- stan. Id. The services are broad: for instance, “supply operations, transportation services, engineering services, base camp services, and other logistics and sustainment support services,” including “minor construction[,] food ser- vices[,] laundry[,] morale, welfare, and recreation ser- vices[,] billeting[, and] facility management.” J.A. 1130447. Under the solicitation, proposals were to be evaluated as a best-value tradeoff between four factors: (1) “tech- nical/management,” (2) “past performance,” (3) “small business participation,” and (4) “cost/price.” DynCorp I, 148 Fed. Cl. at 572 (capitalization normalized); see also J.A. 1002511, 1002624. For the first three factors, each proposal was to be assigned a qualitative adjectival rat- ing—“good,” “acceptable,” or the like. See DynCorp I, 148 Fed. Cl. at 572–73. The technical/management factor As an initial matter, we reiterate that “contracting of- ficers are entitled to exercise discretion upon a broad range of issues confronting them in the procurement process.” Impresa, 238 F.3d at 1332 (cleaned up). “Contracting offic- ers are given broad discretion in their evaluation of bids,” and when a decision within the scope of that discretion “is reasonable[,] a court may not substitute its judgment for that of the agency.” R & W Flammann GmbH v. United States, 339 F.3d 1320, 1322 (Fed. Cir. 2003). Naturally, an agency has no discretion to disregard binding regulations. E.g., 5 U.S.C. § 706(2). But proposal price-reasonableness analysis otherwise sits comfortably amid this discretionary background. In Agile Defense, this court considered discretion in cost-realism analysis—a facet of proposal evaluation simi- lar to price-reasonableness and also based in FAR 15.404-1. And there we endorsed the Court of Federal Claims’ view that “contracting agencies enjoy wide latitude in conducting the cost realism analysis.” Agile Def., 959 F.3d at 1385–86 (cautioning against “unduly circum- scrib[ing] a contracting officer’s discretion and ham- string[ing] a contracting agency’s efforts”). These principles extend to price reasonableness too. Indeed, the Court of Federal Claims has observed as much. See infra section II.A.^[4]. After all, the point of a price-rea- sonableness analysis is to protect the public from paying a price that is too high for what is being procured. See Agile Def., 959 F.3d at 1384. And what exactly is “reasonable” is a judgment based on the specifics of the government’s needs.

Case: 20-2041 Document: 99 Page: 16 Filed: 08/25/2021

16 DYNCORP INTERNATIONAL, LLC v. US Against this backdrop, we consider the interpretive question before us. We agree with the Court of Federal Claims: FAR 15.404-1(b)(2) and (3) are permissive, not pro- hibitive. The FAR states that the government “may use various price analysis techniques and procedures to ensure a fair and reasonable price.” FAR 15.404-1(b)(2) (emphasis added); see FAR 2.101 (“May denotes the permissive.”). It explains that the listed techniques are nonlimiting “exam- ples.” FAR 15.404-1(b)(2). Further, the provision states that the first two techniques are simply “preferred.” FAR 15.404-1(b)(3). “Preferred” expresses here a sugges- tion, not a strict hierarchy. In this context, “preferred” does not mean “required” but instead “encouraged.” See also, e.g., Webster’s Third New International Dictionary (1961) (defining “prefer” as “like better : value more highly”). It is more “should” than “shall.” See AT&T v. United States, 307 F.3d 1374, 1379–80 (Fed. Cir. 2002) (interpreting reg- ulations as a “caution,” “not a prohibition,” where text pro- vided that a preferred type of contract conditionally “should be considered”); Qwest Corp. v. FCC, 258 F.3d 1191, 1200 (10th Cir. 2001) (explaining that “should” con- veys “recommended course of action” but does not imply ob- ligation of “shall”). DynCorp points to what it views as limiting prohibitory language: namely, that “if the contracting officer deter- mines that information on competitive proposed prices or previous contract prices is not available or is insufficient to determine that the price is fair and reasonable, the con- tracting officer may use any of the remaining techniques as appropriate.” FAR 15.404-1(b)(3). Here we do not read “if” to mean “only if.” And at any rate (assuming that the in- formation is “available”), the “if” clause at most asks for a

Case: 20-2041 Document: 99 Page: 17 Filed: 08/25/2021

DYNCORP INTERNATIONAL, LLC v. US 17 determination that the preferred technique is “insufficient to determine that the price is fair and reasonable.” 8 The nature of this supposed “determination” confirms our view. “Insufficient” in this context is a judgment call. The yardstick by which sufficiency is measured here is not some specific rule, formula, calculation, or detailed fact- finding. Rather, it is the reasonable-discretion-informed appropriateness of the technique under the circumstances. In exercising her fact-specific judgment to select a price- analysis technique, a contracting officer will reasonably view some to be suitable and some not—in a manner that may be implicit. FAR 15.404-1(b)(3) imposes no further re- quirement than that. The “if” provision, accordingly, oper- ates not to cabin a contracting officer’s discretion but to give the preference whatever weight is due under the cir- cumstances of a particular procurement. Of course, a con- tracting officer’s price analysis must be rational: the choice is committed to discretion, not whim. See 5 U.S.C. § 706. But the plain language of the provision itself is not prohib- itive and does not conditionally restrict a contracting of- ficer’s discretion.

We turn next to the FAR as a whole. See Antonin Scalia & Bryan A. Garner, Reading Law 167 (2012) (“The 8 As the government observes, the text refers to the technique being “insufficient to determine that the price is fair and reasonable,” not “whether the price is fair and rea- sonable” or “that the price is unreasonable.” See United States Br. 35–38. The government’s argument is that even if FAR 15.404-1(b)(3) were limiting, it would only be limit- ing if the contracting officer could declare an offeror’s price to be reasonable using the preferred techniques, preventing the government in that instance from using the other tech- niques to instead find the price unreasonable. Case: 20-2041 Document: 99 Page: 18 Filed: 08/25/2021 18 DYNCORP INTERNATIONAL, LLC v. US text must be construed as a whole. . . . Context is a primary determinant of meaning. A legal instrument typically con- tains many interrelated parts that make up the whole.”). FAR 15.404-1(b)(3) is part of a larger set of related rules— the Federal Acquisition Regulation. And the broader text of the FAR confirms that this language isn’t prohibitive. To this end, it’s useful to look at the language that the FAR uses where it does prohibit or condition certain ac- tions. For instance: “no person may.” See, e.g., FAR 2.101 (“[T]he words ‘no person may . . .’ mean that no person is required, authorized, or permitted to do the act de- scribed.”). Or “shall not.” See, e.g., FAR 6.302-1(b) (“This authority shall be used, if appropriate, in preference to the authority in 6.302-7; it shall not be used when any of the other circumstances is applicable.”). It’s also revealing to look at the language the FAR uses when requiring action. For one: “shall.” See, e.g., FAR 2.101 (“Shall denotes the imperative.”); FAR 15.404-1(a)(2) (“Price analysis shall be used when cer- tified cost or pricing data are not required . . . .” (emphasis added)); FAR 15.404-4(b)(1)(i) (“[The government] [s]hall use a structured approach for determining the profit or fee objective in those acquisitions that require cost analysis . . . .” (emphasis added)). Or “must.” See, e.g., FAR 2.101 (“Must (see ‘shall’).”). Similarly, the FAR at various points within a single provision uses both “shall” (a command) and “should” or “may” (a preference or permission). See, e.g., FAR 15.404-1(a); cf. Huston v. United States, 956 F.2d 259, 262 (Fed. Cir. 1992) (“When, within the same statute, Congress uses both ‘shall’ and ‘may,’ it is differentiating be- tween mandatory and discretionary tasks.”). What’s more, FAR 15.404-1(b) lacks a documentation requirement. Generally, “[c]ontracting officers are not ob- ligated by the APA to provide written explanations for their actions.” Impresa, 238 F.3d at 1337. And elsewhere the FAR details where documentation of a determination is Case: 20-2041 Document: 99 Page: 19 Filed: 08/25/2021 DYNCORP INTERNATIONAL, LLC v. US 19 required. See, e.g., FAR 15.304(c)(3)(iii) (“Past perfor- mance need not be evaluated if the contracting officer doc- uments the reason past performance is not an appropriate evaluation factor for the acquisition.”); FAR 15.208(e) (“The contracting officer must document the contract file when oral withdrawals are made.”); FAR 15.407-1(d) (“The con- tracting officer shall prepare a memorandum documenting both the determination and any corrective action taken as a result.”); see also Vectrus Br. 32–33 & n.8 (collecting ex- amples). Such a requirement is absent here. Indeed, elsewhere the FAR uses far stronger language for the exact formulation that DynCorp argues here: a de- fault rule prohibiting departure absent satisfying certain conditions. For example, the FAR instructs in source se- lections that past performance “shall be evaluated” as a factor, but “need not be evaluated if the contracting officer documents the reason past performance is not an appropri- ate evaluation factor for the acquisition.” FAR 15.304(c)(3)(i), (iii). Accordingly, the FAR drafters’ use of “preferred” and “may” here—without an express documentation require- ment and against a discretionary proposal-evaluation backdrop—is consistent with the choice of technique being committed to a contracting officer’s discretion.

Finally, we observe that other tribunals to encounter FAR 15.404-1 have viewed it the same way we do. Indeed, DynCorp has not identified any decision holding that FAR 15.404-1(b)(3) restricts agency discretion in the way it suggests. The Court of Federal Claims in Survival Sys- tems, USA v. United States, in contrast, explained that FAR 15.404-1 simply “provides guidance” in a way that nonetheless “permits the agency broad discretion.” 102 Fed. Cl. at 267. Specifically, FAR 15.404-1(b)(2) “per- mits the government discretion in its choice of method to determine price reasonableness.” Id. at 269. And in

Case: 20-2041 Document: 99 Page: 20 Filed: 08/25/2021

20 DYNCORP INTERNATIONAL, LLC v. US Overstreet Electric Co. v. United States, the court noted that the regulation “indicates that the [g]overnment has discretion in choosing which of several acceptable price analysis methods to employ.” 47 Fed. Cl. 728, 733 n.8 (2000). So too in Alabama Aircraft Industries. 83 Fed. Cl. at 696 (“[The] FAR lacks an explicit directive to contracting agencies mandating the use of any particular analytical tool in evaluating the reasonableness and realism of an of- feror’s price.”). And the Court of Federal Claims in this case agreed too. DynCorp II, 2020 U.S. Claims LEXIS 959, at *8–10. Though these views are not binding, they are consistent with ours. B In the alternative, even if the language of FAR 15.404-1(b)(3) were read to require a specific determi- nation that the two “preferred” techniques were inade- quate, in our view the contracting officer made that determination. Although it was not documented, it is clear from the record and was within the scope of the contracting officer’s discretion. Accordingly, the record is “sufficient to permit meaningful judicial review,” see Palantir USG, Inc. v. United States, 904 F.3d 980, 994 (Fed. Cir. 2018), and “the agency’s decisional path is reasonably discernible,” see Wheatland Tube Co. v. United States, 161 F.3d 1365, 1369–70 (Fed. Cir. 1998).

FAR 15.404-1(b)(3) is not particularly demanding. It explains that “if the contracting officer determines that in- formation on competitive proposed prices or previous con- tract prices is not available or is insufficient to determine that the price is fair and reasonable, the contracting officer may use any of the remaining techniques as appropriate to the circumstances applicable to the acquisition.” FAR 15.404-1(b)(3) (emphasis added).

Case: 20-2041 Document: 99 Page: 21 Filed: 08/25/2021

DYNCORP INTERNATIONAL, LLC v. US 21 As an initial matter, DynCorp argues that the suppos- edly required “determination” concerns whether the under- lying information is “insufficient” to compare prices. See Appellant’s Br. 35–36. And it contends that the infor- mation cannot be “insufficient” to compare prices here, as the record is full of competitive pricing information. See Appellant’s Br. 36. The Army used this information pre- protest to determine that the lowest-priced offers were rea- sonable. Accordingly, the argument goes, the Army could have compared proposals and therefore could not have de- termined the underlying information to be insufficient to do so. This misses the point. “Insufficient” in this context means insufficient for determining price reasonableness overall. See FAR 15.404-1(b)(3). And that insufficiency is a value-laden consideration: for some procurements, an overall price comparison between proposals or a historical comparison might make sense; for some, it might not. Therefore, the question becomes whether it is discernible from the record that the Army itself viewed competitive or historical price comparisons to be insufficient to assess rea- sonableness.

DynCorp argues that the record does not reflect the Army having determined that the two preferred price-anal- ysis techniques were insufficient. We disagree. In February 2019, while proposals were being evalu- ated the first time around, the Army issued a memoran- dum for each region separately comparing the cost-plus- fixed-fee portions to each other and the firm-fixed-price portions to each other. J.A. 1069866–94. The Army found adequate price competition in each region and that the low- est-priced offers were reasonable. J.A. 1069866–94. But it noted at the time that it might need further information to evaluate the other, higher-priced offers. See, e.g., J.A. 1069877; see also J.A. 1755951 (noting that it “re- served the right to request this information if needed to

Case: 20-2041 Document: 99 Page: 22 Filed: 08/25/2021

22 DYNCORP INTERNATIONAL, LLC v. US assess reasonableness of the proposed prices”). And in- deed, when later analyzing all prices in December 2019, the Army did request further cost and pricing data from offerors and chose to analyze “data other than certified cost and pricing data”—i.e., data falling under FAR 15.404- 1(b)(2)(vii), a non-preferred technique. See J.A. 1753885, 1755618, 1755715, 1755844, 1755951, 1772629. In doing so, it wrote that this other information was “needed to as- sess reasonableness.” J.A. 1753885, 1755618, 1755715, 1755844, 1755951, 1772629. The decisional basis is clear. This was a complicated procurement in which each offeror proposed a different technical approach. Upon receiving that information, the agency documented its analysis at great length, specifically invoking FAR 15.404-1(b)(2)(vii). Accordingly, it is clear from the record that the contracting officer, in its discre- tion, viewed the “preferred” techniques to be insufficient to assess reasonableness. III DynCorp’s second argument is that the Army’s pro- curement decisions here were arbitrary and capricious for purportedly failing to address what DynCorp views as a fundamental problem: “dramatic disparities that sepa- rate[d] the prices of the offerors.” Appellant’s Br. 36–37. DynCorp argues that the Army failed to “examine the rel- evant data and articulate a satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made.’” Appellant’s Br. 37 (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)). DynCorp accuses the Army of a “critical logical gap”— namely, a supposed failure to “square its conclusion that all offerors’ prices were reasonable with the known price disparities.” Reply Br. 13. DynCorp’s view is that it is an “obvious problem” that all the proposed prices could be rea- sonable despite wide variation among them. Appellant’s Case: 20-2041 Document: 99 Page: 23 Filed: 08/25/2021 DYNCORP INTERNATIONAL, LLC v. US 23 Br. 29. And it insists that the Army “cannot coherently ex- plain” how that can be. Id. at 4. As for the consequences of this alleged error, DynCorp insists that the consequence of accounting for that price disparity would likely have been a finding that DynCorp’s prices were unreasonable. From that finding, DynCorp says, would stem discussions and an opportunity to revise its proposal. Appellees, for their part, suggest that “competing offe- rors’ prices in a best value procurement always vary”—a “banal fact” that means that, inevitably, “some proposals are more expensive than others.” E.g., Fluor Br. 37. There is simply no “fundamental problem” to address, they say. And they suggest that the disparities in price here all fol- low from differences in proposals’ technical approaches (differences all reasonable, all unchallenged, and all con- sidered in the Army’s analysis). Review in bid protests under the arbitrary-and-capri- cious standard is “highly deferential.” Glenn Def., 720 F.3d at 907 (quoting Advanced Data Concepts, Inc. v. United States, 216 F.3d 1054, 1058 (Fed. Cir. 2000)). Under it, we must “sustain an agency action evincing rational reasoning and consideration of relevant factors.” Advanced Data Concepts, 216 F.3d at 1058 (citing Bowman Transp., Inc. v. Ark.-Best Freight Sys., Inc., 419 U.S. 281, 285 (1974)). And an “explicit explanation is not necessary . . . where the agency’s decisional path is reasonably discernible.” Wheat- land Tube, 161 F.3d at 1369–70. The Army did not ignore the price differences. Indeed, it acknowledged that DynCorp had “the highest total pro- posed price when compared [with] the total proposed prices of the other offerors.” See J.A. 1124863–67 (memorandum discussing DynCorp’s total proposed prices). But it ac- counted for the differences—for instance, by noting differ- ential labor hours and rates among offerors. Id.; see also, e.g., KBR Br. 38–39 (noting that the Army “reviewed the unique aspects of each offeror’s firm-fixed-price [portions] Case: 20-2041 Document: 99 Page: 24 Filed: 08/25/2021 24 DYNCORP INTERNATIONAL, LLC v. US and assessed related differences in pricing” (discussing J.A. 796–850, 975–1028, 1130–80, 1304–53, 1473–530, 1650–86)). In some areas DynCorp proposed using more hours and higher labor rates—including, for example, a higher proportion of more-expensive expatriate hours. See J.A. 1124863–64. DynCorp’s technical choices were Dyn- Corp’s prerogative. And DynCorp points to nothing specific in its pricing that was unreasonable given those choices. Overall, the Army determined that each offeror’s price was reasonable for its chosen technical approach. E.g., J.A. 975–83 (examining reasonableness of pricing for AFRICOM task order firm-fixed-price portion for DynCorp, taking into account direct labor, various direct costs, direct labor overhead, global-solutions overhead, administrative costs, and proposed profit, etc.); id. at 983 (concluding that the “proposed price . . . for the [firm-fixed-price] portion of the AFRICOM effort is reasonable for DynCorp’s proposed approach”). It also determined that each offeror’s technical approach was appropriate, and DynCorp does not chal- lenge those findings. At bottom, DynCorp’s challenge amounts to an argument that the technical approach it chose was unreasonable, but it fails to identify any error in the Army’s view that its proposed technical approach was acceptable. The different (but reasonable) prices resulted from different (but reasonable) technical approaches. There was therefore no problematic aspect of the price dis- parity to address, and therefore no “important aspect of the problem” left unconsidered. Cf. Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43. To the extent that DynCorp challenges the contracting officer’s choice of particular price-analysis techniques as arbitrary or capricious, we are not convinced. DynCorp also complains that the Army irrationally did not explain why it didn’t employ a simple proposal-to-pro- posal price comparison. But we don’t see why it must (or even should) have. As we explained above, the regulations didn’t require that. And the context is clear: This was a sweeping procurement to establish an involved logistics Case: 20-2041 Document: 99 Page: 25 Filed: 08/25/2021 DYNCORP INTERNATIONAL, LLC v. US 25 support program in various operational theaters around the globe, with complex technical requirements and sub- stantial latitude afforded to the offerors to figure out how they would accomplish them. Naturally, prices would vary. And the Army had discretion to view a crude price-tag com- parison as insufficient and instead to pick from among other price-analysis techniques as appropriate to the spe- cifics. See, e.g., J.A. 983 (picking FAR 15.404-1(b)(2)(vii)). In the end, the Army explained its analysis in detail in a way that permits meaningful review on the merits. Dyn- Corp alleges no errors in that price analysis other than its favored technique not having been used. And that doesn’t clear the high hurdle to declaring the Army’s approach un- tethered to rationality. Finally, DynCorp insists that the Army’s price-reason- ableness analysis on remand was merely an impermissible post hoc rationalization. It’s true that an agency cannot rely on a new rationale for an old decision. See, e.g., Citi- zens to Pres. Overton Park v. Volpe, 401 U.S. 402, 419 (1971). But that isn’t what the Army did. Rather, on re- mand it made new determinations about price reasonable- ness—ones that the Court of Federal Claims had concluded it hadn’t made before. It declined to revisit the overall award, but in the context of this remand, the only basis for doing so would have been the unreasonableness of one of the offerors’ prices. Accordingly, based on the record and the agency’s dis- cretion in this area, we are not persuaded that the Army’s procurement was arbitrary or capricious. CONCLUSION We have considered DynCorp’s remaining arguments but find them unpersuasive. DynCorp’s proposed price was high. But it was not unreasonably high for the technical approach it proposed. In so finding, the government did Case: 20-2041 Document: 99 Page: 26 Filed: 08/25/2021 26 DYNCORP INTERNATIONAL, LLC v. US not violate FAR part 15 and did not act irrationally. We therefore affirm. AFFIRMED