Consolidated Safety Services, Inc. v. United States

In the United States Court of Federal Claims

No. 23-521C

(Filed: September 22, 2023)

Stephen P. Ramaley, Miles & Stockbridge P.C., Washington, D.C., for Plaintiff. With him on the briefs were Roger V. Abbott and Lauren S. Fleming.

Vijaya S. Surampudi, Commercial Litigation Branch, Civil Division, United States Department of Justice, Washington, D.C., for Defendant. With her on the briefs were Brian M. Boynton, Principal Deputy Assistant Attorney General, Patricia M. McCarthy, Director, and Corinne A. Niosi, Assistant Director.

SOLOMSON, Judge.

This pre-award bid protest action centers on whether the government reasonably assigned the North American Industrial Classification System (“NAICS”) code that best fits the solicitation at issue. This case is significant because it illustrates the limits of the government’s oft-deployed deference defense. As part of this defense — typical in bid protest-type actions — the government argues that the arbitrary and capricious standard of review means that if the agency decision at issue exhibits some thought, or contains some articulated rationale, reviewing courts are required to leave it alone. But frequently, a statute or regulation provides substantive standards for the challenged government action, to which an agency must adhere and against which its decisions are judged.

Where the government fails to consider those standards, its decision is in violation of law. Indeed, even where an agency applies the correct standards its decision nevertheless may be found arbitrary and capricious when evaluated against them, even if the agency’s decision may be characterized as reasonable in a general sense.

In this case, the National Centers for Coastal Ocean Science (“NCCOS”), a division of the National Oceanic and Atmospheric Administration (“NOAA”) (collectively “NCCOS” or the “Agency”), classified the solicitation as primarily seeking Environmental Consulting Services, NAICS code 541620. Plaintiff, Consolidated Safety Services, Inc. (“CSS”), asserts that the solicitation is primarily for research and development (“R&D”), such that the Agency should have assigned NAICS code 541715, Research and Development in the Physical, Engineering, and Life Sciences (except Nanotechnology and Biotechnology).

This arcane NAICS classification question is of critical import here because the procurement at issue has been set aside for small business. Whether CSS is considered a small business for the purpose of this procurement depends on the NAICS designation. If the solicitation is classified as Environmental Consulting Services, CSS does not qualify as a small business and is thus ineligible to compete for the contract. If the solicitation is designated under the R&D NAICS code, however, CSS qualifies as a small business and may compete for the contract. CSS argues that the Agency improperly assigned the consulting NAICS code, which unlawfully precludes CSS from competing for the contract, worth over $72 million.

CSS first appealed the Agency’s NAICS designation to the United States Small Business Administration’s (“SBA”) Office of Hearing and Appeals (“OHA”). OHA affirmed the Agency’s NAICS selection. The question presented to this Court is whether the Agency’s selection of NAICS code 541620 — and OHA’s decision affirming it — was arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law. For the reasons explained below, the Court agrees with Plaintiff, enjoins the Agency from proceeding with the solicitation as currently classified, and remands this case to OHA for further action not inconsistent with this decision.

The Office of Management and Budget (“OMB”) assigns NAICS codes to all domestic industries as described and delineated in the NAICS Manual. Exec. Off. of the President, Off. of Mgmt. and Budget, North American Industry Classification System: United States, 2022, 14, [hereinafter NAICS Manual].¹ OMB revises the NAICS Manual every five years and released the latest revision in 2022. Id. at 13. NAICS is “an industry classification system that groups establishments into industries based on the similarity of their production processes.” Id. at 14. NAICS’s structure is “hierarchical,” grouping economic activities into twenty broad sectors, various subsectors, numerous industry groups, and 1,012 individual industries. Id. at 14, 16.

1The NAICS Manual is available at: https://www.census.gov/naics/reference_files_tools/2022 _NAICS_Manual.pdf.

The purpose of NAICS is to “provide common industry definitions” to “facilitate economic analyses.” NAICS Manual at 14. While NAICS was “designed for statistical purposes,” the classifications and associated codes “also may be used for various administrative, regulatory, and taxation purposes.” Id. The SBA utilizes NAICS codes for one such regulatory purpose — assigning “size standards” for specific industries. 13 C.F.R. § 121.101(a). These “size standards define whether a business entity is small and, thus, eligible for Government programs and preferences reserved for ‘small business’ concerns.” Id.; see also 13 C.F.R. § 121.201 (providing that “[t]he size standards described in this section apply to all SBA programs unless otherwise specified” and explaining that “[t]he size standards . . . are expressed either in number of employees or annual receipts,” which “indicate[] the maximum allowed for a concern and its affiliates to be considered small”).

Pursuant to SBA regulation, “[t]he procuring agency contracting officer, or authorized representative, designates the proper NAICS code and corresponding size standard in a solicitation, selecting the single NAICS code which best describes the principal purpose of the product or service being acquired.” 13 C.F.R. § 121.402(b) (emphasis added).² In determining the “principal purpose” of an acquisition:

Primary consideration is given to [1] the industry descriptions in the U.S. NAICS Manual, [2] the product or service description in the solicitation and any attachments to it, [3] the relative value and importance of the components of the procurement making up the end item being procured, and [4] the function of the goods or services being purchased.

13 C.F.R. § 121.402(b)(1). Additionally, “[a] procurement is generally classified according to the component which accounts for the greatest percentage of contract value.”

Id. § 121.402(b)(2); see RLB Contracting, Inc. v. United States, 118 Fed. Cl. 750, 756 (2014) (“The Federal Acquisition Regulation (‘FAR’) parrots these requirements, reaffirming that classification is normally a product of the ‘component which accounts for the greatest percentage of contract value.’” (citing 48 C.F.R. § 19.303(a)(2))), aff’d, 621 F. App’x 1026 (Fed. Cir. 2015). The NAICS code that a contracting officer (“CO”) assigns to a procurement and the corresponding size standard “is final unless timely appealed to SBA’s [OHA].” 13 C.F.R. § 121.402(d).

2OHA decisions regularly treat the NAICS Manual as a regulatory document, because it provides the industry definitions that a CO must consider to determine the single NAICS code that “best describes the principal purpose” of a procurement. 13 C.F.R. § 121.402(b); see also NAICS Appeal of Elevated Techs., Inc., SBA No. NAICS-6146, 2022 WL 1154844, at *4 (Apr. 12, 2022); SupplyCore, Inc. v. United States, 137 Fed. Cl. 753, 762–63 (2018) (analyzing the NAICS Manual in a bid protest case challenging an OHA decision involving a NAICS appeal). Indeed, the NAICS Manual is essentially incorporated by reference into the SBA regulations. See 13 C.F.R. § 121.402 (explaining that “[p]rimary consideration is given to the industry descriptions in the U.S. NAICS Manual” and that adversely affected parties may appeal to OHA).

Given the parties’ arguments, this case concerns just two specific NAICS codes: Environmental Consulting Services, NAICS code 541620; and Research and Development in the Physical, Engineering, and Life Sciences (except Nanotechnology and Biotechnology), NAICS code 541715. As both industries begin with the same two digits, both fall within Sector 54, “Professional, Scientific, and Technical Services.” NAICS Manual at 455. Activities within this sector consist of “performing professional, scientific, and technical services for others.” Id. Furthermore, as both industries begin with the same three digits (541), they fall within the same “Professional, Scientific, and Technical Services” subsector. Id. The “distinguishing feature” of this subsector is that “most of the industries grouped in it have production processes that are almost wholly dependent on worker skills”; these industries “sell expertise.” Id. As might be expected, individual industries within this subsector “are defined on the basis of the particular expertise and training of the services provider.” Id.

The two industries differ, however, in the fourth digit of their respective NAICS codes. As Environmental Consulting Services begins with 5416, it is part of the “Management, Scientific, and Technical Consulting Services” industry group. NAICS Manual at 466. “This industry group comprises establishments primarily engaged in providing advice and assistance to businesses and other organizations on management, environmental, scientific, and technical issues.” Id. (emphasis added). The Environmental Consulting Services industry is defined as:

Compris[ing] establishments primarily engaged in providing advice and assistance to businesses and other organizations on environmental issues, such as the control of environmental contamination from pollutants, toxic substances, and hazardous materials. These establishments identify problems (e.g., inspect buildings for hazardous materials), measure and evaluate risks, and recommend solutions. They employ a multidisciplined staff of scientists, engineers, and other technicians with expertise in areas, such as air and water quality, asbestos contamination, remediation, ecological restoration, and environmental law. Establishments providing sanitation or site remediation consulting services are included in this industry.

Id. at 469 (emphasis added) (defining NAICS code 541620).

On the other hand, R&D in the Physical, Engineering, and Life Sciences (except Nanotechnology and Biotechnology) begins with the four digits of 5417, indicating that this industry is part of the “Scientific Research and Development Services” industry group. NAICS Manual at 470. This industry group “comprises establishments engaged in conducting original investigation undertaken on a systematic basis to gain new knowledge (research) and/or the application of research findings or other scientific knowledge for the creation of new or significantly improved products or processes (experimental development).” Id. (emphasis added).³ The individual “industries within this industry group are defined on the basis of the domain of research; that is, on the scientific expertise of the establishment.” Id. (emphasis added). R&D in the Physical, Engineering, and Life Sciences (except Nanotechnology and Biotechnology) industry is defined as:

Compris[ing] establishments primarily engaged in conducting research and experimental development (except nanotechnology and biotechnology research and experimental development) in the physical, engineering, and life sciences, such as agriculture, electronics, environmental, biology, botany, computers, chemistry, food, fisheries, forests, geology, health, mathematics, medicine, oceanography, pharmacy, physics, veterinary, and other allied subjects.

Id. at 472 (emphasis added) (defining NAICS code 541715).

As the terms “research” and “experimental development” are defined in the industry group description, these terms do not necessarily hold their ordinary dictionary meanings. See NAICS Manual at 470. In accordance with the NAICS Manual, “research” means “original investigation undertaken on a systematic basis to gain new knowledge,” while “experimental development” means “the application of research findings or other scientific knowledge for the creation of new or significantly improved products or processes.” Id.

3As discussed below, neither the NAICS Manual nor OHA decisions specify a particular ratio of research tasks to development tasks to qualify for an R&D NAICS code.

With this background, the Court next turns to the procurement at issue and the procedural history of this case.

A. The Solicitation

The Agency issued Request for Quote No. 1305M223QNCNS0003 (“RFQ” or “Solicitation”) on February 17, 2023. AR 174. The Solicitation contemplates the award of a single blanket purchase agreement (“BPA”) against the awardee’s Federal Supply Schedule contract. AR 185. The Solicitation’s salient requirements are described in: (1) the Performance Work Statement (“PWS”), AR 273–303; (2) the Labor Category Crosswalk (“Labor Categories”), AR 304–05; and (3) the Price Schedule, AR 306–313.

The PWS expressly describes the general purpose of the Solicitation: to “obtain the necessary scientific, technical, and administrative services to efficiently and effectively support the mission and goals of [NCCOS].” AR 274 (PWS § 1.1, “Background”). NCCOS’s mission, in turn, is to “deliver[] ecosystem science solutions for stewardship of the nation’s ocean and coastal resources in direct support of . . . [NOAA] priorities, offices, and customers to sustain thriving coastal communities and economies.” Id. NCCOS is “a research organization with programmatic authority and a major, direct role in implementing the Harmful Algal Bloom and Hypoxia Research and Control Amendments Act.” AR 277 (PWS § 1.3, “NCCOS Research Enterprise”).

NCCOS also “provides science for a large customer base, including other NOAA offices, federal agencies, and state agencies.” Id. NCCOS’s “[r]esearch endeavors span all U.S. coastal states and territories, and may be field, laboratory, or computer-based.” Id.

The PWS further explains that NCCOS seeks, via the procurement at issue, to “obtain the necessary administrative, scientific and technical support services to ensure the delivery of timely, quality, and relevant research and coastal science products that sustain thriving coastal communities and economies.” AR 278 (PWS § 1.4, “Scope” (emphases added)). The PWS then explains the Solicitation’s “objective” as follows:

4This background section constitutes the Court’s findings of fact drawn from the administrative record. Rule 52.1 of the Rules of the United States Court of Federal Claims (“RCFC”), covering judgment on the administrative record, “is properly understood as intending to provide for an expedited trial on the record” and requires the Court to “make factual findings from the record evidence as if it were conducting a trial on the record.” Bannum, Inc. v. United States, 404 F.3d 1346, 1354, 1356 (Fed. Cir. 2005). Citations to the administrative record, ECF No. 7-1, are denoted as “AR” followed by the page number of the record bolded in the lower right-hand corner of the page. Additional findings of fact are made throughout Part Error! Reference source not found..

NCCOS requires support in conducting field, laboratory, and computer-based research, administering grants and cooperative agreements, and providing operational activities that support the NCCOS mission. The contractor shall furnish technical and scientific support services and equipment (where necessary) in disciplines such as marine biology and oceanography, environmental toxicology, social science, ecological modeling and statistics, program and project management, scientific and policy analysis, administrative support, technical writing, and graphical support.

Id. (PWS § 1.5, “Objective” (emphasis added)).

Section 2 of the PWS provides the Solicitation’s “Specific Requirements/Tasks.”

AR 279. These tasks are broken down into seven main categories: (2.1) Administration, Management, Planning, Analysis, and Coordination; (2.2) Marine Science Data Acquisition; (2.3) Scientific Data Enterprise; (2.4) Geospatial, Statistical, And Modeling Analyses; (2.5) Publications, Communications, Outreach; (2.6) Program Execution and Analysis; and (2.7) NOAA Mission Support. AR 279–90. Each category is further divided into numerous subtasks. Id.

The Solicitation’s Labor Categories attachment describes the types of employees the contract awardee must provide to accomplish the PWS requirements and defines each labor category’s “functional responsibilities.” AR 304. The labor categories include Environmental Scientists (Levels I–V), Geospatial Scientists (Levels II–V), Social Scientists (Levels II–V), Program Support Specialists (Levels II–IV), Program Analysts (Levels II– IV), Communications Analysts (Levels II–IV), and Program Managers (Level III). AR 304–05.

Finally, the Price Schedule provides estimates of how many hours the contract awardee will be required to supply within each labor category. AR 308. For instance, the Price Schedule estimates that the awardee will have to provide 78,960 work hours for staff meeting the requirements of an Environmental Scientist III, but only 1,880 hours for staff within the Program Manager III category. Id. Simple algebra yields that 66.99% of the estimated labor hours will be performed by Environmental Scientists, 18.45% will be performed by Geospatial Scientists, and 4.85% will be performed by Social Scientists. See id. The remaining 9.71% of estimated hours are divided amongst Program Specialists, Program Analysts, Communications Analysts, and Program Managers. See id.

The Agency classified the Solicitation under NAICS code 541620, Environmental Consulting Services. AR 182. Due to this NAICS designation, the Solicitation had a size standard of $19 million in average annual receipts. AR 174. As CSS does not qualify as a “small business” under this size standard, CSS is unable to compete for this procurement unless it prevails in this case. AR 5; ECF No. 1 (“Compl.” ¶ 15).

B. CSS’s Appeal to SBA OHA

On February 27, 2023, CSS filed a “NAICS Appeal” with OHA, contending that the Agency’s designation of the Solicitation as Environmental Consulting Services, NAICS code 541620, was clearly erroneous. AR 3–25. In that appeal, CSS argued that the work solicited under the RFQ “predominantly entail[s] performing research, not advising the agency about research.” AR 18. CSS thus argued that the Agency should have assigned NAICS code 541715 for R&D, and not a consulting code. AR 17–24. CSS is the incumbent contractor for at least some of the work the Solicitation covers. AR 5, 30. On March 9, 2023, the Agency CO submitted a statement to OHA, explaining her NAICS code selection decision. AR 341–68 (“CO Statement”). Therein, the CO opined that “[t]he primary purpose of the acquisition involves data collection and field surveys.” AR 344. The CO further explained her view that “[t]he principal purpose and greatest percentage of anticipated work are for marine science data acquisition, which comprises approximately 60% of the estimated value of this BPA.” AR 346–47 (referring to PWS § 2.2, “Marine Science Data Acquisition”). Additionally, according to the CO, “80% . . . of the estimated BPA ordering value, is comprised of . . . scientific professionals.” AR 347. Despite these assertions apparently supporting an R&D designation, the CO concluded that “[a]ll of the tasks [sic] areas in PWS Section 2.0, including those in marine science data acquisition, best align with the NAICS Code 541620, Environmental Consulting Services, as the tasks relate closely to providing advice and assistance to businesses and other organizations on environmental issues.” AR 347. The CO emphasized the Solicitation’s focus on “environmental issues, such as ‘the control of environmental contamination from pollutants, toxic substances, and hazardous materials’ that require the contractor to ‘measure and evaluate risks’ and recommend solutions for ‘ecological restoration’ and ‘remediation.’” AR 348 (quoting NAICS code 541620). The CO also relied on PWS Section 1.3, asserting that because “NCCOS has a wide variety of needs for business operation support, including administration, management, planning, analysis, and coordination,” that means the Solicitation’s “required services align with the NAICS code of 541620, which states that the industry is primarily engaged in providing advice and assistance to businesses and organizations on environmental issues.” AR 346 (quoting PWS § 1.3, “NCCOS Research Enterprise”).

The CO further opposed CSS’s position on the grounds that the Solicitation “does not require the contractor to provide Research and Development as defined in [FAR] Part 35.001, Research and Development Contracting,” because the Solicitation “is not focused on determining and exploiting the potential of scientific discoveries or improvements nor is the predominant focus of the work to create new processes or products.” AR 344. Relatedly, the CO argued that because “NCCOS’s mission does not include research and development within the meaning of FAR Part 35,” the RFQ should not be categorized under an R&D NAICS code. AR 346 (asserting that “NCCOS does not contract for research and development”).

In sum, the CO concluded “that NAICS code 541620 best describes the principal purpose of the support services being procured, in light of the industry descriptions in the NAICS code manual, the description in the solicitation, and the relative weight of each element in the solicitation.” AR 349.

On March 30, 2023, OHA issued a decision affirming the Agency’s selection of NAICS code 541620, Environmental Consulting Services, for the procurement at issue. AR 379–96 (“OHA Decision”). After a lengthy recitation of the Solicitation’s components and the parties’ arguments, AR 380–92, OHA’s analysis was exceedingly sparse, see AR 393–94. OHA explained that “NAICS code 541620 has covered procurements which involved research, analysis, evaluation of contaminated sites, development of documentation, and participation in strategy sessions with Government counsel.” AR 393 (citing Prudent Technologies, Inc., SBA No. NAICS-4710, at *5 (2005)). Analogizing that case to the Solicitation, OHA reasoned that “[h]ere, the Contractor will perform marine science data acquisition as 60% of the work and as the principal purpose of the RFQ.” Id. According to OHA, such data acquisition “involves collecting, organizing, processing and archiving samples, and collecting data from a number of platforms, to provide expert support, mentoring, and field investigation with NCCOS leadership, coastal management decision-making, laboratories, and programs to support social science research.” Id. (emphasis added).⁵

OHA also concluded that CSS’s preferred NAICS code was inapplicable to the Solicitation. First, OHA noted that “[w]hile [CSS] points to the RFQ key word tasks for ‘research’ and for ‘development’ or the [level of effort] being substantially performed by scientists in support of NAICS code 541715, such claim is farfetched and not within the meaning of the NAICS Manual.” AR 393. Listing out the seven task categories in PWS Section 2, OHA asserted that “[t]hese tasks do not call for research and experimental development of new processes as the principal purpose of the service being acquired.” Id. OHA also reasoned that “[w]hen assigning NAICS Code 541715, the project must include ‘research and experimental development.’” Id. (emphasis added) (citing NAICS Manual at 472–73). The R&D NAICS code would be improper here, according to OHA, because this RFQ “does not seek original research toward the development of new processes or products.” AR 394. Rather, OHA posited, “the Contractor will perform mostly data collection,” which “will be supportive of NCCOS’s mission and provide information on coastal ecosystems,” and will provide “support for NCCOS’s laboratories.” Id.

5As discussed infra, this Court finds that OHA’s characterization of this procurement as primarily focused on social science research constitutes a clear factual error.

Finally, OHA concluded that CSS’s “reliance on NAICS Appeal of [LJR] Sols., LLC, SBA No. NAICS-5790 (2016) [was] misplaced.” AR 394. Whereas the procurement in that case “was directly involved with the conduct of particular research,” the RFQ here is primarily for “data collection to support general social science research, which does not call for research and development.” Id. (discussing LJR Sols. at *6–7).

For these reasons, OHA “conclude[d] that the CO made the appropriate NAICS code designation for this procurement, and that [CSS] has failed to meet its burden of establishing there was clear error in the designation.” AR 394–95.

C. CSS’s Action in this Court

On April 14, 2023, CSS filed a complaint against Defendant, the United States — acting by and through NOAA NCCOS — in this Court. Compl. at 1. CSS alleges that: (1) the CO’s designation of NAICS code 541620 was arbitrary and capricious, id.

at 23 (Count I); (2) OHA’s failure to consider whether Environmental Consulting Services “best describes the actual services acquired under the Solicitation” renders OHA’s decision arbitrary and capricious, id. at 27 (Count II); and (3) OHA’s determination that the Solicitation does not involve “research” was factually unsupported, rendering OHA’s decision arbitrary and capricious, id. at 30 (Count III). CSS seeks a “declaratory ruling that the OHA Decision was arbitrary and capricious and not in accordance with law.” Id. at 1. CSS initially requested either a remand to OHA or a permanent injunction ordering the Agency to assign the R&D NAICS code to the Solicitation. Id. at 1, 40.

Pursuant to the Court’s scheduling order, ECF No. 6, the government filed the administrative record on May 5, 2023. ECF No. 7-1. On May 26, 2023, CSS filed its motion for judgment on the administrative record. ECF No. 8 (“Pl. MJAR”). On June 16, 2023, the government filed its cross-motion. ECF No. 9 (“Def. MJAR”). On June 27, 2023, CSS filed its response and reply brief. ECF No. 10 (“Pl. Resp.”). And on July 7, 2023, the government filed its reply. ECF No. 11 (“Def. Reply”). On July 18, 2023, the Court held oral argument on the parties’ cross-MJARs. ECF No. 15 (“Tr.”).

The Tucker Act provides that an “interested party” may file an “action” in this Court “objecting [1] to a solicitation by a Federal agency for bids or proposals for a proposed contract or [2] to a proposed award or [3] the award of a contract or [4] any alleged violation of statute or regulation in connection with a procurement or a proposed procurement.” 28 U.S.C. § 1491(b)(1); see also Aero Spray, Inc. v. United States, 156 Fed. Cl. 548, 559 & n.18 (2021) (“Section 1491(b) actions are typically referred to as ‘bid protests.’”). This Court’s Tucker Act jurisdiction includes challenges to a government agency’s selection of a NAICS code in conjunction with a subsequent OHA decision in a NAICS code appeal. Palladian Partners, Inc. v. United States, 783 F.3d 1243, 1254 (Fed. Cir. 2015) (“Because OHA’s NAICS code determination and the contracting officer’s amendment to the solicitation are actions ‘in connection with a proposed procurement,’ we conclude that they are within the scope of jurisdiction granted under the Tucker Act.”); see also 22nd Century Techs., Inc. v. United States, 57 F.4th 993, 999 (Fed. Cir. 2023) (“Where an SBA decision is made ‘in connection with a proposed procurement,’ the Claims Court would normally have jurisdiction to review that decision under § 1491(b)(1).”); Def. Integrated Sols., LLC v. United States, 165 Fed. Cl. 352, 367 (2023).⁶ “Standing is an integral part of jurisdiction,” Seventh Dimension, LLC v. United States, 160 Fed. Cl. 1, 14 (2022) (citing Lujan v. Defs. of Wildlife, 504 U.S. 555, 560 (1992)), and “[t]he party invoking federal jurisdiction bears the burden of establishing standing.” CliniComp Int’l, Inc. v. United States, 904 F.3d 1353, 1358 (Fed. Cir. 2018) (citing Myers Investigative & Sec. Servs., Inc. v. United States, 275 F.3d 1366, 1369 (Fed. Cir. 2002)). To establish “interested party” standing in a § 1491(b) action, a plaintiff must allege and ultimately demonstrate that it is “an actual or prospective bidder” with a “direct economic interest” in the procurement. CliniComp Int’l, 904 F.3d at 1358 (quoting Digitalis Educ. Solutions, Inc. v. United States, 664 F.3d 1380, 1384 (Fed. Cir. 2012)).⁷ In the pre-award context, it is sufficient to allege a “non-trivial competitive injury which can be addressed by judicial relief.” Weeks Marine, Inc. v. United States, 575 F.3d 1352, 1361–62 (Fed. Cir. 2009); but see Oracle America, Inc. v. United States, 975 F.3d 1279 (Fed. Cir. 2020) (explaining that a court should apply the “substantial chance” test if there is an “adequate factual foundation” to do so).

6The government agreed that it does not “make a difference in this case” whether the Court reviews the OHA Decision or the underlying decision of the CO. Tr. 10:15-17.

7But see CACI, Inc.-Fed. v. United States, 67 F.4th 1145, 1151 (Fed. Cir. 2023) (“Our prior caselaw treating the interested party issue as a jurisdictional issue . . . is no longer good law in this respect.” (citations omitted)).

In this case, CSS cannot compete for an award under the Solicitation due to the Agency’s having assigned NAICS code 541620, Environmental Consulting Services. Pl.

MJAR at 20. This NAICS designation has a corresponding size standard of $19 million in average annual receipts, which CSS exceeds. Id.; see also AR 174. On the other hand, CSS would qualify as a small business and thus would be able to compete if the Solicitation were classified under any R&D NAICS code. Pl. MJAR at 20. An R&D NAICS code has a corresponding size standard of 1,000 employees, which CSS meets. Id. Moreover, there is no dispute that CSS is the incumbent contractor for at least some of the work the Solicitation covers. AR 30. As CSS has “diligently pursued its rights” through the OHA protest, it “has prospective bidder status.” CGI Fed., Inc. v. United States, 779 F.3d 1346, 1351 (2015). In addition, CSS’s allegation that it is unable to bid for the Solicitation due the Agency’s erroneous NAICS designation — and that CSS would be able to submit a quote if the Agency assigned the correct NAICS code — constitutes a “non-trivial competitive injury which can be addressed by judicial relief.” Orion Tech., Inc. v. United States, 704 F.3d 1344, 1348 (Fed. Cir. 2013) (quoting Weeks Marine, 575 F.3d at 1361–62). Accordingly, CSS has standing to pursue its claims in this case. The government does not challenge CSS’s standing.

Pursuant to 28 U.S.C. § 1491(b)(4), this Court applies the standard of review contained in the Administrative Procedure Act (“APA”), 5 U.S.C. § 706. Nat’l Gov’t Servs., Inc. v. United States, 923 F.3d 977, 981 (Fed. Cir. 2019). In accordance with the APA, this Court reviews an agency’s procurement decisions to determine whether they are “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). A plaintiff succeeds on the merits where it demonstrates that either: “(1) the [agency]’s decision lacked a rational basis; or (2) the procurement procedure involved a violation of regulation or procedure.” Impresa Construzioni Geom. Domenico Garufi v. United States, 238 F.3d 1324, 1332 (Fed. Cir. 2001) (citations omitted). An agency’s decision is arbitrary and capricious if the agency “entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or [the decision] is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” Ala. Aircraft Indus., Inc.-Birmingham v. United States, 586 F.3d 1372, 1375 (Fed. Cir. 2009) (alteration in original) (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)).

When interpreting SBA regulations — including the NAICS Manual (as arguably incorporated by reference into 13 C.F.R. §§ 121.201, 121.402) — as when interpreting all regulations:

[W]e apply the same interpretive rules we use when analyzing the language of a statute. And it is well established that, when interpreting statutes or regulations, “[t]he plain meaning that we seek to discern is the plain meaning of the whole statute [or regulation], not of isolated sentences.”

Beecham v. United States, 511 U.S. 368, 372, 114 S. Ct. 1669, 128 L.Ed.2d 383 (1994) (citations omitted).

Guardado v. United States, 2023 WL 5426590, at *11 (Fed. Cl. Aug. 23, 2023) (quoting Boeing Co. v. Sec’y of Air Force, 983 F.3d 1321, 1327 (Fed. Cir. 2020)). Thus, as Judge Hertling recently explained:

“Interpretation of a regulation is a question of law” that is “reviewed without deference.” Cameron v. United States, 550

F. App’x 867, 872 (Fed. Cir. 2013). When “a term in a

regulation is ambiguous, the Supreme Court has accorded deference to the agency’s interpretation of its own regulation,” but there are limits. Id. Plainly erroneous interpretations receive no deference, and neither do those that are “nothing more than a convenient litigating position or a post hoc rationalization advanced by an agency seeking to defend past agency action.” Christopher v. SmithKline Beecham Corp., 567 U.S. 142, 155 (2012) (cleaned up).

Guardado, 2023 WL 5426590, at *11 & n.12 (citing Defense Integrated Sols., LLC v. United States, 165 Fed. Cl. 352, 368–72 (2023), with approval).

Analyzing those general principles, this Court previously has held that an OHA decision is not entitled to any “special” deference, but rather is entitled either to Auer or Seminole Rock deference⁸ where the regulation at issue is “hopelessly ambiguous,” or to the deference courts give to reasonable and persuasive agency interpretations under Skidmore, applying the typical APA standard of review. Def. Integrated Sols., 165 Fed. Cl. at 371 (quoting Animal Legal Def. Fund v. U.S. Dep’t of Agric., 223 F. Supp. 3d 1008, 1022– 23 (C.D. Cal. 2016)); see also id. n.27.

8In Defense Integrated Solutions, this Court explained: “[o]nly after exhausting all ‘traditional tools’ of regulatory interpretation and finding a regulation is ‘genuinely ambiguous’ may courts consider applying what has been termed Auer or Seminole Rock deference. Pursuant to such deference, courts must defer to an agency’s reasonable interpretation of the agency’s own ‘genuinely ambiguous’ regulations.” 165 Fed. Cl. at 368 (cleaned up) (quoting and discussing

The OHA Decision is arbitrary and capricious for two main reasons. First, OHA’s conclusion that the Solicitation does not seek R&D, see AR 393–94, flatly ignores scores of tasks in the Solicitation that naturally fit the definitions of “research” and “experimental development” in the NAICS Manual. The OHA Decision’s finding that an R&D designation would be improper is thus factually erroneous. Given that more than 85% of the Solicitation’s labor hours are concentrated in PWS Sections 2.2–2.4, which predominantly entail R&D tasks, the R&D NAICS code that CSS prefers is objectively reasonable. Second, both the Agency and OHA fail to explain how NAICS code 541620, covering Environmental Consulting Services, best describes the contemplated work. The Solicitation’s paucity of tasks focused on “providing advice” and “recommend[ing] solutions,” which are the key, defining characteristics of Environmental Consulting Services, confirms that this NAICS code is not the best choice, and thus the government’s selection of that code is contrary to SBA regulatory requirements. But before jumping into the details, the Court first must frame the proper inquiry.

A. SBA Regulations Do Not Provide the Government with Maximum Discretion to Select Any Reasonable NAICS Code

Pursuant to the SBA regulation governing an agency’s NAICS code selection, the CO must “designate[] the proper NAICS code and corresponding size standard in a solicitation, selecting the single NAICS code which best describes the principal purpose of the product or service being acquired.” 13 C.F.R. § 121.402(b) (emphasis added). This means that the Agency cannot justify its selection of a particular NAICS code on the grounds that it fits the procurement in some general sense or that the selection is “good enough for government work.” Brown v. Plata, 563 U.S. 493, 562 (2011) (Scalia, J., dissenting). The SBA regulations further cabin the government’s discretion by Kisor v. Wilkie, –– U.S. ––, 139 S. Ct. 2400, 2414–15 (2019), Chevron U.S.A., Inc. v. Nat’l Res. Def. Council, Inc., 467 U.S. 837, 843 & n.9 (1984), Auer v. Robbins, 519 U.S. 452 (1997), and Bowles v. Seminole Rock & Sand Co., 325 U.S. 410 (1945)).

delineating how the “single . . . best” NAICS code that describes the “principal purpose” of the acquisition shall be determined:

Primary consideration is given to [1] the industry descriptions in the U.S. NAICS Manual, [2] the product or service description in the solicitation and any attachments to it, [3] the relative value and importance of the components of the procurement making up the end item being procured, and [4] the function of the goods or services being purchased.

13 C.F.R. § 121.402(b)(1). Critically, “[a] procurement is generally classified according to the component which accounts for the greatest percentage of contract value.” 13 C.F.R. § 121.402(b)(2). These are the substantive yardsticks this Court must use to assess the Agency’s NAICS code selection and the subsequent OHA Decision.

The government asserts that “CSS must prove that the NAICS Code designation was based upon clear error of fact or law,” Def. MJAR at 19, and that “there is no need for the NAICS Code to be a perfect fit,” id. at 26 (discussing InGenesis, Inc. v. United States, 104 Fed. Cl. 43, 52 (2012)). But the government (and this Court) must be more precise about what is under review — it is the Agency’s decision, affirmed by OHA, that NAICS code 541620 “best describes the principal purpose of the . . . service being acquired.” 13 C.F.R. § 121.402(b) (emphasis added). Applying the APA standard of review, 28 U.S.C. § 1491(b)(4), the ultimate question for this Court is whether the government reasonably selected NAICS code 541620 as representing the “best” description of the Solicitation. 13 C.F.R. § 121.402(b) (emphasis added); see also Arcata Assocs., Inc. v. United States, 110 Fed. Cl. 290, 300 (2013) (“Selection of the applicable NAICS code is governed by 13 C.F.R. § 121.402(b).”).

Contrary to the government’s suggestion, Def. MJAR at 26, InGenesis is consistent with this Court’s formulation of the question presented here, 104 Fed. Cl. at 52. In that case, Judge Bruggink recognized that “the selected code [must] capture[] the principal purpose of the contract.” Id. Judge Bruggink did not reject the regulatory “best describes” standard; rather, he reached the common-sense conclusion that the selected NAICS code “need not be a perfect fit to every facet of the performance work statement.” Id. (emphasis added). Likewise, the OHA Decision begins from the premise that “SBA regulations do not require the CO to select the perfect NAICS code.” AR 391 (emphasis added). This Court takes no issue with that statement to the extent that OHA only means to express that, in some cases, reasonable minds may disagree as to the NAICS code that “best describes the principal purpose of the . . . service being acquired.” 13 C.F.R. § 121.402(b). In such a case, neither OHA nor this Court may substitute its judgment for the CO’s. On the other hand, OHA agrees, as it must, that the applicable SBA regulations do require the CO to “assign the NAICS code that best describes the principal purpose of the product or service being acquired . . . .” AR 391 (emphasis added). Accordingly, that substantive yardstick requires this Court to assess whether the CO (and, in turn, OHA) reasonably concluded that NAICS code 541620, Environmental Consulting Services, constituted the “best” description of the Solicitation.

During oral argument, the government agreed with the Court’s framing of the issue:

THE COURT: . . . So the question, again, is not kind of in a general sense whether or not a rational person could pick either of the two NAICS codes at issue here. It’s whether or not the CO and OHA reasonably selected the single NAICS code which best describes the principal purpose of the service being acquired.

[GOVERNMENT COUNSEL]: Yes, that would be correct.

Tr. 10:24 - 11:5.

B. The Government Erred as a Matter of Law in Concluding that the Solicitation Does Not Seek R&D

The “Analysis” section of the OHA Decision spans approximately two pages and essentially consists of two material findings: (1) that “[a]ny argument that this is a Research and Development procurement is simply not supported by the record or OHA case law”; and (2) “[w]hen the work of a contractor includes the identification of problems, the evaluation of risks, and the recommendation of solutions, . . . it fit[s] squarely within the definition of NAICS code 541620.” AR 393 (citing Prudent Technologies, Inc., SBA No. NAICS-4710, at *5 (2005)). The fatal defect in the first finding is that the administrative record flatly refutes it as a factual matter. The second finding is erroneous because it ignores the relevant legal question: whether NAICS code 541620 “best” describes the contemplated contract work. 13 C.F.R. § 121.402(b). The fact that OHA has found that certain types of work included in this procurement may be properly categorized under NAICS code 541620 does not mean that such work predominates here — or that the government properly selected that NAICS code as the best fit for this procurement. Accordingly, and as explained below, this Court concludes that both OHA findings are arbitrary, capricious, or otherwise contrary to law.

1. The Solicitation Primarily Seeks R&D Services As the Court noted above, the NAICS Manual defines “research” as “original investigation undertaken on a systematic basis to gain new knowledge,” and “experimental development” as “the application of research findings or other scientific knowledge for the creation of new or significantly improved products or processes.” NAICS Manual at 470. The OHA Decision’s repeated assertions that the Solicitation does not call for R&D in accordance with these definitions are factually erroneous. AR 393– 94. Indeed, the OHA Decision’s conclusion that “[t]he RFQ does not seek original research toward the development of new processes or products,” AR 394, is critically undermined by literally dozens of tasks in the Solicitation that naturally fit the definitions of “research” and “experimental development,” NAICS Manual at 470.

But before diving into the particular tasks and subtasks, the Court first notes that the OHA fails to address a host of PWS provisions generally supportive of CSS’s position. For example, the PWS explains that NCCOS seeks to “obtain the necessary administrative, scientific and technical support services to ensure the delivery of timely, quality, and relevant research and coastal science products that sustain thriving coastal communities and economies.” AR 278 (PWS § 1.4, “Scope” (emphasis added)). NCCOS’s goal to deliver “research and coastal science products,” id., plainly encompasses “research” and “experimental development,” NAICS Manual at 470. Moreover, the PWS’s “objective” is that the contractor will “conduct[] field, laboratory, and computer-based research,” and “shall furnish technical and scientific support services and equipment (where necessary) in disciplines such as marine biology and oceanography, environmental toxicology, social science, ecological modeling and statistics, program and project management, scientific and policy analysis, administrative support, technical writing, and graphical support.” AR 278 (PWS § 1.5, “Objective”). Contrary to OHA’s assertion, this is not mere “data collection.” AR 394. To be fair, the PWS’s “objective” also includes “administering grants and cooperative agreements, and providing operational activities that support the NCCOS mission,” AR 278, but the administrative record makes clear that less than 10% of the Solicitation’s estimated hours are allocated to support staff, see AR 308. In contrast, more than 85% of the estimated contract hours are to be performed by environmental and geospatial scientists. See id.

Turning to the particular tasks, after listing the seven task areas in PWS Section 2,9 the OHA Decision then asserted that “these tasks do not call for research and experimental development of new processes as the principal purpose of the service being acquired.” AR 393. But three of these task areas — Marine Science Data Acquisition, Scientific Data Enterprise, and Geospatial, Statistical, And Modeling Analyses — are directly research-oriented on their face. AR 282, 285–86. As demonstrated infra, each of those task areas include numerous “research” and “experimental development” subtasks. NAICS Manual at 470. More critically, however, OHA’s conclusion that two of the Solicitation’s subtasks do “not constitute original research,” AR 394, ignored literally scores of subtasks in PWS Sections 2.2–2.4 that do constitute “research” and “experimental development,” id. at 282–87; NAICS Manual at 470.

9These seven task areas are: (2.1) Administration, Management, Planning, Analysis, and Coordination; (2.2) Marine Science Data Acquisition; (2.3) Scientific Data Enterprise; (2.4) Geospatial, Statistical, And Modeling Analyses; (2.5) Publications, Communications, Outreach; (2.6) Program Execution and Analysis; and (2.7) NOAA Mission Support. AR 279–90.

For starters, PWS Section 2.2, Marine Science Data Acquisition — which encompasses “60% of the [Solicitation’s] work” according to the CO and OHA10 — includes the following subtasks:

•2.2.1. Assist in the development of experimental/survey design and selection of methods for data collection. Develop new protocols when needed.

•2.2.4. Conduct laboratory research in accordance with NOAA designs/protocols to provide data to inform coastal management decisions or product development . . . .

•2.2.5. Document, maintain, and input field, experimental, and laboratory data and observations, including any needed metadata.

•2.2.6. Organize, archive and process, and appropriately store/dispose of samples (biological tissue, environmental, etc.) from field and laboratory studies.

•2.2.7. Plan and conduct primary data collection to support social science research . . .

•2.2.10. Assist in logistical planning, development of experimental designs, and selection of methods for various research studies.

•2.2.11. Provide expert support in design of field collection programs for biological monitoring, ground validation, and accuracy assessment for mapping.

•2.2.12. Perform or assist with necropsies on dead, stranded marine mammals.

•2.2.13. Support NCCOS leadership in the development of a research strategy for sentinel habitats and organisms that can ultimately support human health risk assessments in marine ecosystems.

•2.2.14. Support with field and laboratory research activities including sentinel habitat and species monitoring and assessment leading to the development of an integrated ecosystem risk assessments and models.

•2.2.16. Conduct research with minimal guidance from [the Principal Investigator] to support ecosystem, environmental, socioeconomic, and/or marine animal health related research.

•2.2.17. Assist with planning, design, setup, and implementation of laboratory studies and complex experiments under the guidance of a PI.

•2.2.18. Participate in the collection, processing, archival, identification, and enumeration of marine benthic invertebrate, coral, fish and or marine mammal samples from a variety of sources including preserved specimens, digital photographs/3-D photogrammetry, digital imagery, video graphic images, and acoustic-profile recordings.

•2.2.19. Perform scientific analysis/research in support of chemical impacts and other water quality issues for environmental measurement.

•2.2.21. Perform a wide variety of laboratory procedures . . . . Design, develop and/or adapt new clinical diagnostic assays to assess the condition of marine organisms.

•2.2.23. Maintain laboratory cultures of marine and estuarine organisms.

Conduct lethal and sublethal bioassays to evaluate the effects of chemical contaminants in a variety of model organisms (fish, crustaceans, mollusks, algae, bacteria). Develop and evaluate procedures to measure biomarkers of contaminant exposure and effects in organisms.

•2.2.24. Maintains and conducts experiments on algal cell cultures, bacterial strains, cell lines and laboratory animals. Conducts harmful algae analysis including sample preparation and molecular probe assays, sequencing, microarrays, light, fluorescence and scanning electron microscopy and flow cytometry. Conducts toxin analyses including sample preparation, High Performance Liquid Chromatography, Liquid Chromatography-Mass Spectrometry, receptor, immunological, enzymatic assays.

•2.2.25. Conducts analyses of organic and inorganic contaminants in environmental samples, develop, and adapt new methods for the analysis of contaminants in environmental samples.

•2.2.26. Measure fecal coliform and multiple antibiotic resistant fecal coliform bacteria concentrations in marine and estuarine water samples.

Develop and employ real-time Polymerase Chain Reaction methods for the detection of viruses and bacterial pathogens of public health concern in marine and estuarine waters.

•2.2.27. Investigate, assess, present on new technologies and techniques for detecting, tracking, and predicting the occurrence of microbes of public health concern.

•2.2.28. Conduct . . . research to include proteomics, metabolomics, genomics, lipidomics etc.

•2.2.30. Lead or assist with logistical planning and associated execution of fieldwork necessary to support marine animal health and ecosystem health research including photo-identification and/or remote biopsy sampling and other efforts to support sentinel species and habitat assessments.

•2.2.33. Lead or participate in field operations requiring scuba diving on air or nitrox including visual observations and photographic documentation, placement and retrieval of monitoring devices, and collection of samples for laboratory studies at depths up to 130 feet.

•2.2.34. Programming/deploying and retrieving field instrumentation like data sondes, environmental sensor processors, in-situ automated submersible imaging flow cytometers, and other data sensors.

•2.2.35. Deployment and use of control/capture strategies and equipment for contaminants/HABs.

AR 282–85 (emphases added).

By the Court’s measure, at least 25 out of the 35 subtasks in PWS Section 2.2 entail “research” — “original investigation undertaken on a systematic basis to gain new knowledge,” NAICS Manual at 470 — in the fields of marine biology, oceanography, ecology, and environmental toxicology, AR 282–85. Although the OHA Decision characterized these tasks as “mostly data collection” as opposed to “original research,” AR 394, this distinction has no basis in the definition of “research,” NAICS Manual at 470, or, more importantly, in reality. By OHA’s reasoning, Charles Darwin’s research in the Galapagos might also be summarized as “mostly data collection.” AR 394. Or as Plaintiff’s counsel of record put it, “folks at the [L]arge [H]adron [C]ollider are also engaged in data collection.” Tr. 103:15–16. It is pure sophistry to draw a distinction between what is clearly systematic investigation of the natural world to collect data and something else called “research.” While some of the subtasks in PWS Section 2.2 are more mechanical than others, data collection is a necessary component of empirical research, as countless graduate students and seasoned PhDs alike can attest. OHA itself agrees. See Advanced Concepts Enters., Inc., SBA No. NAICS-5968, 2018 WL 6113389 at *14 (Oct. 24, 2018) (determining that data collection “constitutes original investigation, or research,” and that such data collection was necessary “to develop software and systems to support…tests and simulations” for missile defense).

Moreover, at least eight of the subtasks in PWS Section 2.2 call for “experimental development,” or “the application of research findings or other scientific knowledge for the creation of new or significantly improved products or processes.” NAICS Manual at 470. The “creation of new . . . processes” undoubtedly includes the design of novel experiments, protocols, research strategies, clinical diagnostic assays, procedures, methods, and technologies and techniques. Id. In other words, the ultimate contract awardee will be asked to develop new ways to study the ocean and assess its health. While the OHA Decision at one point refers to “experimental development” as the “application of research findings for the creation of new or significantly improved products,” AR 393 (emphasis added), OHA conspicuously (and inexplicably) omitted the phrase “or processes” from the NAICS Manual definition, NAICS Manual at 470. Even assuming that “experimental development” in its ordinary usage might connote a company’s creating a new product like a drug or a microchip, here it is a term of art that encompasses the “creation of new . . . processes.” NAICS Manual at 470. Given that NAICS code 541715 includes scientific fields like biology and oceanography, the phrase “experimental development” must be interpreted to extend beyond mere products. Id. The OHA Decision failed to comprehend this material point.

2. The Solicitation Includes Minimal Consulting Tasks Conversely, astonishingly few of the subtasks in PWS Section 2.2 can be fairly characterized as “providing advice” or “recommend[ing] solutions,” in accordance with the definition of Environmental Consulting Services. NAICS Manual at 469. The government argues that at least seven subtasks fall within the definition of Environmental Consulting Services, Def. Reply at 6, but at least five of them are straightforward research tasks (nos. 6, 10, 17, 19, 31), involving data collection, experimental design, laboratory studies, environmental measurement, and field investigations, AR 282–84. The other two subtasks are ambiguous at best. Although the purpose of subtask 4 is “to inform coastal management decisions or product development,” AR 282, it is not clear that the contractor is the one who is “providing advice” or “recommend[ing] solutions,” NAICS Manual at 469. What is clear, however, is that the contractor will “[c]onduct laboratory research.” AR 282. Likewise, while the goal of subtask 13 is to “ultimately support human health risk assessments in marine ecosystems,” the contractor’s concrete task is “the development of a research strategy for sentinel habitats and organisms.” Id. (emphasis added).

Although an environmental consultant can surely engage in research in the course of “identify[ing] problems” and “measur[ing] and evaluat[ing] risks” — such that research and consulting are not mutually exclusive activities¹¹ — an environmental consultant for NAICS purposes must “recommend solutions” and “provid[e] advice and assistance to businesses and other organizations on environmental issues, such as the control of environmental contamination from pollutants, toxic substances, and hazardous materials.” NAICS Manual at 469–70. The government’s few examples, Def. Reply at 6, tend to show that research predominates over consulting services such as recommending solutions or providing advice. These examples are thus unpersuasive and do not demonstrate that the selected NAICS code best describes the work the Solicitation contemplates.

11See Prudent Techs., Inc., SBA No. NAICS-4710, at *1, 5 (finding that the appropriate NAICS designation was Environmental Consulting Services when a solicitation sought to “procure consulting services in support of the Government’s legal counsel in preparation for actual or potential litigation concerning contaminated sites,” when the solicitation involved “research, analysis, evaluation of contaminated sites, development of documentation, and participation in strategy sessions with Government counsel.”); FAR 2.101 (defining “[a]dvisory and assistance services” to include “R&D activities”).

Moving on to PWS Section 2.3, the “Scientific Data Enterprise” task requires the contractor to collate, organize, verify, analyze, manage, and share the data the contractor collects. AR 285 (PWS § 2.3). Such data-related tasks are often necessary to perform an “original investigation undertaken on a systematic basis to gain new knowledge.”

NAICS Manual at 470. If empirical researchers neglected to manage and analyze the observations they collected, their “original investigation” would be incomplete, and they would be virtually guaranteed no “new knowledge.” Id. Moreover, while some of the subtasks are arguably advisory in nature, see AR 285 (PWS §§ 2.3.5-2.3.8), the fact that the words “recommend,” “solution,” “advise,” and “assistance” do not appear in this Section tellingly counsels against the consulting NAICS code.

Likewise, PWS Section 2.4, “Geospatial, Statistical, and Modeling Analyses,” contains a plethora of R&D tasks:

•2.4.1. Provide expertise in geospatial/geostatistical analysis, support data visualization and the formulation of geospatial products, including maps.

•2.4.2. Analyze data using cutting edge statistical and pattern analysis methods (such as regression trees, spatial pattern analyses) to shape and subset data; automate components for models or data classification using Python and/or other languages. Visually display the data [sic] and analytical results.

•2.4.3. Develop and modify integrated biophysical models and predictive ecological models. Evaluate model performance, and develop improvements as needed across the full modeling process (eg., precision, accuracy, efficiency).

•2.4.4. Conduct statistical analysis to analyze scientific data and to determine accuracy of derived products, such as maps and models.

•2.4.5. Develop biogeographic assessments, ecosystem assessments, socio-economic assessments, vulnerability assessments, and other products as needed to inform coastal management decision-making.

•2.4.6. Compile, process, and analyze of remote sensing data (passive and active sensors - (acoustic, optical, oceanographic and derived spatial predictors) in support of mapping programs; vertical and horizontal positioning, sound speed corrections, optical imagery corrections, data processing, image analysis, analytical aerotriangulation, digital cartographic data compilation, composition of data structures for Geographic Information System (GIS) application, and the completion of associated graphic, textual, and digital products required to meet program requirements.

•2.4.7. Document and maintain experimental data and observations and assist or lead compilation, summarization, verification, graphical representations, extraction, logging, analysis and interpretation of data.

•2.4.8. Develop and adapt new methods for the analysis of contaminants in environmental samples. Develop environmental and toxicological models and conduct statistical analysis of data including preparing graphical representations of data. Develop and maintain contaminant and risk assessment databases.

•2.4.9. Support the NCCOS research enterprise with the development of approaches and tools — including models, forecasts, and technical reports for forecasting the impacts to habitats.

•2.4.10. Researches and develops models of ecological and/or disease processes to support forecasts of ecosystem conditions in relation to environmental and/or anthropogenic stressors.

•2.4.12. Develop algorithms to process and classify remotely-sensed multispectral data into (e.g., clear, partly cloudy, and cloudy pixels) for developing Harmful Algal Blooms forecasting products . . . .

•2.4.13. Modify, improve, and evaluate algorithms producing currently operational products.

•2.4.14. Conduct statistical analysis to determine accuracy of products derived from remotely sensed imagery.

•2.4.15. Develop, modify, evaluate, and improve integrated biophysical models such as HABs and ecological predictive models of marine species distribution and habitat utilization.

•2.4.16. Develop ecological models to conduct impact assessments under various management strategies.

•2.4.17. Characterize and develop indexes of land-based sources of pollution including contaminants in coastal and marine ecosystems.

AR 286–87 (emphases added).

In this section, 16 out of the 17 subtasks are clearly “research” oriented. AR 286– 87; see NAICS Manual at 470 (“Techniques may include modeling and simulation.”).

Moreover, at least 12 of these subtasks entail “experimental development,” because they require the contractor to develop new algorithms, maps, forecasts, and predictive models using advanced geospatial and statistical techniques. AR 286–87; see NAICS Appeal of Applewood Eng’g, SBA No. NAICS-6119, 2021 WL 3857343 at *11 (Aug. 24, 2021) (determining that NAICS code 541715 was appropriate for “a procurement for research and development efforts into Earth’s atmosphere, the oceans and land surfaces, and cryosphere, including atmospheric composition and ocean biogeochemistry, as well as climate analysis for weather, climate and environmental prediction,” and the “development of climate models, data assimilation systems, and climate forecast systems” (emphasis added)). The OHA Decision does not explain why the development of such “new or significantly improved products or processes” fails to satisfy the definition of “experimental development.” NAICS Manual at 470. And although the government seems to suggest that most of the Solicitation’s subtasks must be considered “experimental development” for the Solicitation to receive an R&D NAICS code, Def.

Reply at 2, that would largely render the “R” in R&D superfluous. NAICS Manual at 472. In any event, nothing in the NAICS Manual requires a particular ratio of research tasks to development tasks to qualify for the R&D NAICS code that CSS seeks. And if OHA discerns such a ratio requirement, OHA fails to explain how or from where it is derived. The OHA Decision noted that “[w]hile [CSS] points to the RFQ key word tasks for ‘research’ and for ‘development’ or the [level of effort] being substantially performed by scientists in support of NAICS code 541715, such claim is farfetched and not within the meaning of the NAICS Manual.” AR 393. Here, OHA is referring to CSS’s argument that “the word ‘consult’ or ‘consulting’ appears 0 times throughout the entire PWS,” whereas “the word ‘research’ appears 118 times in the PWS.” AR 19; see AR at 386 (summarizing this argument). CSS also pointed out that 85.44% of the estimated labor hours in the Solicitation are to be performed by environmental and geospatial scientists. AR 13; see AR 386 (summarizing this argument). The OHA Decision, however, never explains why these arguments supporting an R&D NAICS code are “farfetched” or inconsistent with the NAICS Manual’s definition of R&D. AR 386. Indeed, they are not. Contrary to the government’s argument, CSS’s noting that there are 118 instances of the word “research” in the PWS is not “cherry-picking.” Def. MJAR at 12. Rather, cherry-picking is when OHA gives just two examples of ostensible consulting tasks and suggests that they are representative of all the tasks in the PWS or somehow reflect the principal purpose of the Solicitation. AR 394.

By ignoring that the lion’s share of the subtasks in PWS Sections 2.2–2.4 constitute either “research” or “experimental development,” the OHA Decision “entirely failed to consider an important aspect of the problem.” Ala. Aircraft, 586 F.3d at 1375 (quoting Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43). The OHA Decision’s finding that the R&D NAICS code designation would be improper is the very definition of arbitrary and capricious.

C. The OHA Decision Did Not (And Could Not) Find that NAICS Code 541620

Was the Single Best Choice for this Solicitation In light of the foregoing, the question is not whether the R&D NAICS Code CSS seeks (i.e., NAICS code 541715) may plausibly be assigned to the RFQ at issue — that such a code may apply is now obvious — but rather whether the Agency (and OHA) reasonably concluded that the Environment Consulting Services NAICS code is the “best” NAICS code given the RFQ’s anticipated work. 13 C.F.R. § 121.402(b). Moreover, while the OHA Decision trains the bulk of its analysis on trying to support the erroneous conclusion that the R&D NAICS code is entirely inapplicable, the OHA Decision did not evaluate the relevant regulatory factors to determine the Solicitation’s “principal purpose.” 13 C.F.R. § 121.402(b).

With respect to the question whether the Agency and OHA reasonably deemed the chosen NAICS code “best,” the OHA Decision’s analysis is sparse and unpersuasive.

The analysis was limited to (at most) two case analogies and a handful of tasks plucked from the Solicitation. AR 393–94. Its cursory comparison of the instant case to previous OHA decisions amounted to one short, conclusory paragraph, AR 393 (citing Prudent Technologies, SBA No. NAICS-4710), and a gesture towards a case proving a proposition that no party contends would be dispositive. AR 394 (citing Rollout Systems, SBA No. NAICS-5901, for the proposition that “work upon network administration functions ‘already in existence’ was not [R&D]”). Beyond the poor case comparisons, the OHA Decision explained that “NAICS code 541620 has covered procurements which involved research, analysis, evaluation of contaminated sites, development of documentation, and participation in strategy sessions with Government counsel.” Id. (citing Prudent Technologies, SBA No. NAICS-4710 at *5). But the fact that NAICS code 541620 has been used in such procurements does not remotely demonstrate that it is the “best” NAICS code for this Solicitation.

Other statements within the OHA Decision suffer from the same fallacy. For instance, the OHA Decision noted that “[w]hen the work of a contractor includes the identification of problems, the evaluation of risks, and the recommendation of solutions, OHA found that it fit squarely within the definition of NAICS code 541620.” AR 393. But the fact that OHA determined that NAICS code 541620 best described the work in some other procurement tells us nothing about whether that code — rather than the one CSS prefers — is the best fit here.

Attempting to analogize Prudent Technologies to this case, the OHA Decision explained that “[h]ere, the Contractor will perform marine science data acquisition as 60% of the work and as the principal purpose of the RFQ,” which “involves collecting, organizing, processing and archiving samples, and collecting data from a number of platforms, to provide expert support, mentoring, and field investigation with NCCOS leadership, coastal management decision-making, laboratories, and programs to support social science research.” AR 393. Further on, the OHA Decision notes that “[a]nalyzing contaminants in environmental samples and developing environmental and toxicological models work identified in this RFQ clearly fall under the NAICS Manual description for NAICS code 541620.” AR 394. Finally, the OHA Decision asserts that “[o]rganizing and preparing technical reports also falls under NAICS code 541620, as the Contractor will provide advice on environmental matters to the procuring agency.” Id. All those statements may be broadly accurate but none of them, singularly or collectively, demonstrate that NAICS code 541620, covering Environmental Consulting Services, “best describes the principal purpose” of the Solicitation, 13 C.F.R. § 121.402(b), particularly considering “the relative value and importance of the components of the procurement,” id. § 121.402(b)(1).

OHA’s failure to evaluate the PWS against the relevant regulatory factors illustrates how unreasonable OHA’s conclusion is here. First, the OHA Decision failed to thoroughly compare the industry description of Environmental Consulting Services in the NAICS Manual to the Solicitation’s tasks. See 13 C.F.R. § 121.402(b)(1) (providing that “[p]rimary consideration is given to the industry descriptions in the U.S. NAICS Manual, the product or service description in the solicitation and any attachments to it”).

Although the OHA Decision referred to PWS tasks involving “the identification of problems, the evaluation of risks, and the recommendation of solutions,” AR 394, which is part of the definition of Environmental Consulting Services, see NAICS Manual at 469, the OHA Decision did not identify where the Solicitation requires the contractor to perform such tasks. Indeed, a close read of the Solicitation yields the conclusion that very few anticipated tasks focus on “recommend[ing] solutions.” Compare NAICS Manual at 469, with AR 279–90 (PWS §§ 2.1–2.7). OHA’s evaluation of the PWS amounts to little more than a conclusory statement, asserted rather than supported with record evidence. The OHA Decision then ignored the bulk of the definition of Environmental Consulting Services. OHA did not consider whether the Solicitation primarily involves “providing advice and assistance to businesses and other organizations on environmental issues, such as the control of environmental contamination from pollutants, toxic substances, and hazardous materials.” NAICS Manual at 469 (emphasis added). Again, a close read of the Solicitation yields a paucity of tasks focused on “providing advice.” Compare NAICS Manual at 469, with AR 279–90. Nor did OHA assess whether the Solicitation calls for a “multidisciplined staff of scientists, engineers, and other technicians with expertise in areas, such as air and water quality, asbestos contamination, remediation, ecological restoration, and environmental law.” NAICS Manual at 469. In fact, the RFQ’s Labor Categories attachment includes zero engineers or legal experts. AR 304–05. Why the Solicitation hardly mentions “providing advice” or “recommend[ing] solutions,” NAICS Manual at 469, then, is no great mystery. And as for the handful of examples of purported “Environmental Consulting Services” tasks the OHA Decision does mention — “[a]nalyzing contaminants in environmental samples,” “developing environmental and toxicological models,” and “[o]rganizing and preparing technical reports” — whether these tasks constitute consulting services is debatable at best. AR 394. It is certainly not obvious that they are not R&D tasks.

Second, OHA violated 13 C.F.R. § 121.402(b) insofar as the OHA Decision failed to evaluate “the relative value and importance of [the Solicitation’s] components.” 13 C.F.R. § 121.402(b)(1). The OHA Decision asserted that “the Contractor will perform marine science data acquisition as 60% of the work and as the principal purpose of the RFQ.” AR 393 (referring to PWS § 2.2, “Marine Science Data Acquisition”). But this 60% estimate — which was taken directly from the CO’s Statement submitted to OHA, AR 347 — is baseless. Indeed, the Court asked counsel of record for the government whether “[t]he 60 percent calculation [is] anywhere in the record?” Tr. 54:2–5. Counsel for the government conceded, in response, that “[t]he calculation is not in the record.” Id.¹² The government then asserted that the CO’s 60% estimate stemmed from the Price Schedule — which estimates how many labor hours each type of contractor employee will work — and the Labor Categories, which describes the tasks that each type of employee will perform. Tr. 54:22–25. The government’s counsel of record, however, was unable to explain how such an estimate could be calculated based on this information. Tr. at 54– 65. This serious flaw in OHA’s analysis constitutes an independent ground upon which this Court holds that the OHA Decision is arbitrary, capricious, and contrary to law.

12Counsel for the government merely “assume[d]” the backup work papers exist even though they were not included in the administrative record. Tr. 56:3-13; see also Tr. 58:14-23 (government counsel agreeing that the administrative record does not contain support for the mapping of the labor categories to the PWS requirements to the particular, associated dollar value “that gets to 60 percent”). Failing to include such crucial calculation information is fatal particularly where, as here, the calculation is not easily replicable merely by analyzing the documents that are in the record and performing some simple algebra. This is “a problem decision makers could avoid by following the admonition they have no doubt heard since their grade-school math class: Show your work.” Tolliver Grp., Inc. v. United States, 151 Fed. Cl. 70, 109–10 (2020). Moreover, the 60% assessment, contrary to SBA regulatory requirements, was not even an assessment of the relative contribution of the different tasks to contract value. Tr. 67:14-18 (government counsel conceding that “[t]here’s no dollars and cents associated with this specific” estimate).

Even accepting for the sake of argument that “60% of the work” in the Solicitation really is “marine science data acquisition,” the OHA Decision proffered no explanation for why the Solicitation is then best classified as Environmental Consulting Services. AR 393. The OHA’s conclusion, in that regard, is a complete non-sequitur. See Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43 (explaining that an agency must articulate a ‘rational connection between the facts found and the choice made.’” (quoting Burlington Truck Lines v. United States, 371 U.S. 156, 168 (1962))).¹³ If anything, this estimate supports CSS’s position that the Solicitation is best described as R&D, as the Court discussed supra.

Third, the OHA Decision failed to adequately consider that “[a] procurement is generally classified according to the component which accounts for the greatest percentage of contract value.” 13 C.F.R. § 121.402(b)(2). Given that a solicitation is “generally classified” according to this factor, it is especially weighty. Id. But the OHA Decision entirely dodged CSS’s argument that 85.44% of the estimated labor hours in the Solicitation are to be performed by environmental and geospatial scientists, which is powerful evidence in favor of an R&D NAICS code. AR 385–86. The OHA Decision also overlooked a similar admission in the CO Statement — that “80% . . . of the estimated BPA ordering value, is comprised of . . . scientific professionals.” Id. at 347 (emphasis added). As OHA all but completely failed to consider the “relevant factors” before finding that the CO did not err in selecting the NAICS code for Environmental Consulting Services, the OHA Decision is arbitrary and capricious. Weyerhaeuser Co. v. United States Fish & Wildlife Serv., 139 S. Ct. 361, 371 (2018) (quoting Judulang v. Holder, 565 U.S. 42, 53 (2011)); see Impresa Construzioni, 238 F.3d at 1332–33 (explaining that an agency must “provide[] a coherent and reasonable explanation of its exercise of discretion”).

In sum, CSS is correct that the Solicitation “predominantly entail[s] performing research, not advising the agency about research.” AR 18. Given that more than 85% of the Solicitation’s estimated labor hours will be performed by environmental and geospatial scientists, and these labor hours are concentrated chiefly in the “research” and “experimental development” tasks in PWS Sections 2.2–2.4, CSS’s preferred R&D NAICS code is a far better choice for this Solicitation, per the SBA regulations. See 13 C.F.R. § 121.402(b); see also NAICS Manual at 472.¹⁴ The Solicitation’s estimated labor hours are the best proxy for “contract value,” and because these hours are predominantly devoted to R&D tasks, this is “the component which accounts for the greatest percentage” of the RFQ’s anticipated work. 13 C.F.R. § 121.402(b)(2); see CO Statement at AR 347 (estimating that “80% . . . of the estimated BPA ordering value, is comprised of . . . scientific professionals”).

13The government asserts that “OHA properly evaluated the Solicitation’s primary purpose,” but it does not identify where or how OHA did so. Def. MJAR at 18–20. Nor does the government explain the disconnect between the proposition that “Marine Science Data Acquisition contains 60% of the total work” and OHA’s conclusion that the Solicitation is best described as Environmental Consulting Services. Id. (quotations omitted).

14This conclusion holds even if “approximately 60% of the estimated value” of the Solicitation is concentrated in PWS Section 2.2, as the CO asserted without support, AR 347, and upon which

For all the reasons above, the Agency’s and OHA’s finding that the Environmental Consulting Services NAICS code “best describes the principal purpose” of the Solicitation is objectively unreasonable. 13 C.F.R. § 121.402(b). As the Court explained, while a NAICS designation need not perfectly describe the procurement at issue, the selected NAICS code must still qualify as “a fit that [best] describes the principal purpose of the services being acquired.” InGenesis, Inc., 104 Fed. Cl. at 52. The Solicitation contains very few tasks in the way of “providing advice” or “recommend[ing] solutions,” which is the whole point of hiring a consultant. NAICS Manual at 469. Thus, the OHA Decision’s affirmance of this irrational NAICS designation was arbitrary and capricious. See Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43 (explaining that an agency’s decision is arbitrary and capricious if it “runs counter to the evidence”).

D. The OHA Decision Is Not Entitled to Deference Because It Is Inconsistent With, and Fails to Reconcile, Earlier OHA Precedent

The government contends that if there is any conceivable basis on which to find that the NAICS code selection rationally covers the Solicitation, the Court should defer to the OHA Decision’s affirming the selection of the NAICS code 541620. See Def. MJAR at 16. Some deference would perhaps be warranted¹⁵ if this Court were confident that the OHA Decision had synthesized OHA’s prior case law delineating the borders between consulting services and R&D pursuant to the NAICS Manual. As a review of the OHA case law shows, however, the OHA Decision’s treatment of prior OHA cases ranges from inadequate to erroneous. This is yet another reason why the Court declines to defer to the OHA Decision under review.

the OHA Decision relied, AR 393. See also Def. MJAR at 22 (“[T]he majority of this solicitation’s work is to be performed under PWS Task 2: Marine Science Data Acquisition.”). As the Court demonstrated supra, PWS Section 2.2 primarily contains “research” and “experimental development” subtasks. AR 282–85; NAICS Manual at 470. In contrast, PWS Section 2.2 contains sparingly few subtasks focused on “providing advice” or “recommend[ing] solutions.” AR 282– 85; NAICS Manual at 469.

15See supra Section IV (discussing applicable deference principles and citing Def. Integrated Sols., 165 Fed. Cl. at 371, for the proposition that an OHA decision is not entitled to any special deference); see also supra note 2.

As a general matter, “OHA is bound to follow its own precedent absent limited exceptions inapplicable here.” Precise Sys., Inc. v. United States, 122 Fed. Cl. 263, 270 (2015) (citing 13 C.F.R. § 134.226(a)(2)); see also Master Boat Builders, Inc., SBA No. Siz-6198, 2023 WL 3611731 (April 27, 2023) (“OHA has the responsibility of adjudicating appeals from size determinations and its decisions set binding precedent.” (citing 15 U.S.C. § 634(i), 13 C.F.R. §§ 134.102(k); 134.226(a)(2)).¹⁶ OHA’s precedent rule has significant implications for the APA standard-of-review prism through which this Court reviews the OHA Decision. In particular, there is a “well-established principle that ‘all administrative agencies have an obligation to render consistent opinions and either follow, distinguish, or overrule their own precedent.’” Precise Sys., 122 Fed. Cl. at 271 (quoting N.L.R.B. v. Rhone–Poulenc, Inc., 789 F.2d 188, 193 (3d Cir. 1986)).¹⁷ Whatever deference may be accorded to OHA opinions is diminished where its decisions are inconsistent. Cf. I.N.S. v. Cardoza-Fonseca, 480 U.S. 421, 447 n.30 (1987) (“An additional reason for rejecting the [government’s] request for heightened deference to its position is the inconsistency of the positions the [Board of Immigration Appeals] has taken through the years.”).¹⁸ In this case, the OHA Decision asserts that, to assign an R&D NAICS code, “the project must include ‘research and experimental development.’” AR 393 (emphasis added) (citing NAICS Manual at 472-73). Thus, OHA “will overturn the assignment of a research and development NAICS code if the procurement does not call for both research and experimental development.” AR 393 (emphasis added) (citing Delphi Research, Inc., SBA No. NAICS-5377, 2012 WL 2950518 (July 10, 2012)). OHA does not even try to square that statement with the NAICS Manual, which, as OHA itself explains, “states that in order to be classified as a[n] [R&D] procurement, [it] must call for original investigation undertaken on a systemic basis to gain new knowledge and/or the application of research findings for the creation of new or significantly improved products.” AR 393 (quoting NAICS Manual at 470, and citing Dayton T. Brown, Inc., SBA No. NAICS-5164, 2010 WL 9012920 (Nov. 8, 2010) (emphasis added)).

1613 C.F.R. § 134.226(a)(2) provides that “[a]n OHA decision creates precedent, unless: (i) [a]nother regulation in this chapter [13 C.F.R. §§ 101.100 to 147.670] applicable to a specific type of appeal provides that the OHA decision does not create precedent; or (ii) the decision is designated as one not to be cited as precedent.” The provision “codif[ied] OHA’s longstanding . . . practice of citing its prior decisions as precedent.” Rules of Procedure Governing Cases Before the Office of Hearings and Appeals, 75 Fed. Reg. 47,435 (Aug. 6, 2010).

17This “consistency doctrine” serves several important purposes in administrative law. See Precise Sys., 122 Fed. Cl. at 271 n.4.

18At oral argument, the government agreed that “OHA is supposed to be consistent with its case precedent, and where they deviate from the case precedent, this Court has found that that rationale is unacceptable.” Tr. 87:24 – 88:2.

This Court acknowledges that OHA, at least in some cases, has found that NAICS code 541712 requires a solicitation to include both research and development. See, e.g., Dynamac Corp., SBA No. NAICS-5025, 2009 WL 1090337, *6 (Feb. 4, 2009) (concluding that “NAICS code 541712 applies to ‘research and development’ and not pure research or scientific investigation, which does not involve creating new or significantly improved products or processes”); Dayton T. Brown, 2010 WL 9012920 at *5 (holding that “this procurement does not call for experimental development, as the NAICS code designation for 541712 requires”).¹⁹

But the problem for the government in this case is that other, earlier OHA decisions unequivocally support CSS’s position. For example, in Information Ventures, Inc., SBA No. NAICS-4882, 2008 WL 276639 (Jan. 18, 2008), OHA affirmed the selection of an R&D NAICS code for a “solicitation for clinical research operations and management support.” 2008 WL 276639, *1. Nothing in that decision suggests that the eventual contractor would perform development tasks as defined (and required) in either Dynamac or Dayton T. Brown. Nevertheless, OHA concluded in Information Ventures that R&D was an appropriate classification where “the work to be performed . . . is all an integral part of the R&D studies” that a National Institute of Health agency “will conduct” down the line. Id. at *6 (“The tasks the contractor will perform are all vital parts of the conduct of [NIH] research.”). Yet the contractor’s work, “properly categorized as research” even without a distinct development facet, made “the research and development NAICS code…most appropriate.” Id. at *6-7. Thus, in Information Ventures, the presence of research work supported — and the lack of development work did not preclude — the assignment of the R&D NAICS code. Id. at *6. (concluding “after reviewing the RFP that research is precisely what this procurement seeks”).

In another case apparently involving the same company, Information Ventures, Inc., SBA No. NAICS-4945, 2008 WL 2416143 (April 18, 2008), the appellant argued for the application of a consulting services NAICS code instead of an R&D NAICS code. 2008 WL 2416143 at *2. The CO conceded that the contractor “will not be conducting laboratory or other physical research work” but maintained that the R&D NAICS code was appropriate because “the work that will be conducted under this contract is a necessary and essential prerequisite for the design of [government] toxicology studies.” 2008 WL 2416143 at *3 (emphasis added). OHA sided with the agency, affirming the R&D NAICS code based solely on contract work that can be characterized, at best, as tangentially related to research; OHA determined that the contractor’s “compiling information on [various] substances [and] preparing summaries of the literature and charts” is sufficient for an R&D classification. Id. at *5. OHA reasoned that such “work is the vital first step in the research process, determining what the current state of knowledge is on any subject to be studied.” Id. (emphasis added). Quite obviously, “determining what the current state of knowledge is” does not constitute “research” as the NAICS Manual defines that term by reference to the production of “new knowledge,” NAICS Manual at 470, let alone “development.” Indeed, lest there be any doubt, OHA affirmed the use of the R&D NAICS code notwithstanding OHA’s finding that “the contractor here will not actually perform any physical research itself[.]” 2008 WL 2416143 at *5. (emphasis added).²⁰

19In earlier editions of the NAICS Manual, NAICS code 541715 was numbered 541712.

The OHA Decision further found that CSS’s “reliance on NAICS Appeal of [LJR] Sols., LLC, SBA No. NAICS-5790 (2016) [was] misplaced.” AR 394.21 According to OHA, whereas the procurement in LJR “was directly involved with the conduct of particular research,” here “the primary work is the data collection to support general social science research, which does not call for research and development.” Id. (emphasis added). The OHA Decision refutes itself insofar as it previously noted that “60% of the work” concerns marine science, not social science. AR 393. Moreover, only 4.85% of the Solicitation’s estimated work-hours are allocated to social scientists, whereas 85.44% of the estimated hours are allocated to environmental and geospatial scientists. AR 308.

The government later conceded that this claim — included in both the OHA Decision and the government’s briefing, Def. MJAR at 11 — is indefensible.²² Nor does the NAICS Manual distinguish between “particular” and “general” research. AR 394; Tr. 77:4-8.23 But that is far from the only problem with the OHA Decision’s treatment of LJR in this case.

20Delphi Research, on which the OHA Decision relies in this case, AR 393, is also ambiguous as to OHA’s interpretation of the NAICS Manual’s definition of the R&D NAICS codes. In Delphi, OHA rejected an agency’s assignment of an R&D NAICS code to a procurement because the solicitation did “not call for the contractor to create new processes or products.” 2012 WL 2950518 at *8. But OHA acknowledged that there is another line of decisions in which OHA has approved the R&D NAICS code for a procurement that merely “delegate[s] an important part of the research process for the contractor to perform independently.” Id. at *9. The problem with an R&D NAICS code in Delphi was that the solicitation did “not clearly assign any research function to the contractor.” Id. (emphasis added). Here, in contrast, the government concedes that the contract is for research but focuses instead on the alleged lack of development tasks. CSS is correct, however, that the ultimate contract awardee will perform both research and development tasks, as this Court explained supra.

²¹The correct name of the referenced OHA decision is LJR Solutions, LLC, SBA No. NAICS-5790, 2016 WL 6916459 (Nov. 17, 2016). OHA mistakenly referred to the case as LRJ. AR 394.

The LJR OHA appeal involved a procurement for on-site animal husbandry support services for a Food and Drug Administration program. 2016 WL 6916459 at *1.

The appellant in LJR “argue[d] that the appropriate NAICS code for th[e] procurement is 561210, Facilities Support Services[.]” Id. at *3 (noting that “[t]hese services include providing quality care to research animals, including feeding, sanitation and technical procedures support”); see also id. at *6 (OHA’s finding that “[t]his procurement seeks a contractor to care for the animals used in the FDA’s research”). One would not know it from reading the OHA Decision, but LJR upheld the assignment of an R&D NAICS code merely because the animal care “activities are part of the research process.” Id. at *7. OHA in LJR further explained that “[w]ork which is an integral part of an agency’s research, and essential for the conduct of the research, is properly classified under a research NAICS code.” Id. (citing Information Ventures, SBA No. NAICS-4882, discussed supra). The icing on the cake is this gem: “[e]ven if the contractor will not perform any research itself, . . . but it will perform work which is a vital part of the research process, the contract is properly classified under a research NAICS code.” Id. (emphasis added) (citing Information Ventures, SBA No. NAICS-4945, discussed supra). OHA even concluded that an R&D NAICS code may be assigned where the procurement is only tangentially related to R&D — i.e., “to support competencies required for the success of the research effort.” Id.

The point by now is clear: LJR fully supports CSS’s position given LJR’s expansive treatment of the term “research” and its rejection of the idea that any development activities are required to qualify as R&D.24 The OHA Decision’s feeble and flawed attempt to distinguish LJR does not adequately deal with the complexity of prior case law, is clearly erroneous and, at a minimum, renders the OHA Decision arbitrary and capricious. See ANR Pipeline Co. v. FERC, 71 F.3d 897, 901 (D.C. Cir. 1995) (“[W]here an agency departs from established precedent without a reasoned explanation, its decision will be vacated as arbitrary and capricious.”); cf. Lyons Sec. Servs., Inc. v. United States, 38 Fed. Cl. 783, 785–86 (1997) (“Where GAO precedent can be viewed as inconsistent, however, we attach no deference to GAO’s ruling.” (citing Bowsher v. Merck & Co., 460 U.S. 824, 837–38 (1983))); Delta Chem. Corp. v. West, 33 F.3d 380, 382 (4th Cir. 1994) (“[T]he lack of consistency in the cited GAO opinions, both internally and between opinions, renders them of little value and undeserving of judicial deference”).²⁵ In sum, what is clear from the preceding discussion is that the challenge for OHA in the factual context of this case extends well past LJR. At best, the various lines of OHA authority are unclear regarding when an agency may assign an R&D NAICS code to a procurement, thus requiring OHA to engage with and reasonably explain the applicability of its precedent — again, something the OHA Decision did not do. And, at worst, the OHA precedent is hopelessly inconsistent.²⁶ Either way, the OHA Decision is arbitrary and capricious.

22See Tr. 22:13–17 (“THE COURT: Okay. But you’re not going to contend that this contract was primarily evaluating general social science research. [GOVERNMENT COUNSEL]: No, I am not going to contend that.”); Tr. 19:4-6 (government agreeing that “[m]arine data science acquisition” predominates in this procurement); Tr. 77:12 – 78:1-4 (government counsel agreeing that the OHA Decision was “flat wrong” in asserting that the Solicitation is focused on “general social science research”).

²³The government at oral argument also conceded that, contrary to its argument, the Solicitation was for a contract to “perform[] research,” and not merely to “evaluat[e] research.” Tr. 22:22-24. The challenged NAICS code (541620) does not mention the word “research.” Tr. 24:7-9.

²⁴Government counsel essentially conceded this point. Tr. 73:5-12 (“[GOVERNMENT COUNSEL]: . . . “I’ll concede that if this was . . . the only case law relevant, it could support the Plaintiff’s position.”); Tr. 76:1-15 (government counsel agreeing that the procurement at issue in

E. The Court Rejects the Government’s Reliance on FAR Part 35

The CO repeatedly asserted that the acquisition at issue “does not require the contractor to provide Research and Development as defined in [FAR] 35.001, Research LJR “was looking for research only”); Tr. 78:12-15 (government counsel agreeing that LJR “itself does not require development of products and services”).

25Indeed, even if OHA had confessed that its reasoning in LJR is erroneous, OHA cannot so easily ignore its own binding precedent. Fogo De Chao (Holdings) Inc. v. U.S. Dep’t of Homeland Sec., 769 F.3d 1127, 1144 (D.C. Cir. 2014) (holding that an agency must provide a reasoned explanation for treating similar situations differently “or at least something more reasoned than confessing a decade-long pattern of material and gross error” (internal quotation marks omitted)); see also BNP Paribas Energy Trading GP v. FERC, 743 F.3d 264, 269 (D.C. Cir. 2014) (“Such an opaque dismissal of an analogy falls well short of the APA’s requirement that the Commission ‘provide an adequate explanation to justify treating similarly situated parties differently[.]’” (quoting Comcast Corp. v. FCC, 526 F.3d 763, 769 (D.C. Cir. 2008))).

26In a more recent decision, for example, OHA affirmed an agency’s substituting NAICS code 541990 for NAICS code 541715 — the R&D NAICS code CSS prefers in this case. See Calspan Corp., SBA No. NAICS-6062, 2020 WL 5992052 (July 2, 2020). There, OHA noted that it “will overturn the assignment of a research and development NAICS code if the procurement does not call for both research and experimental development.” Id. at *9. But later in the very same opinion, OHA acknowledged that in LJR, OHA took a more lenient approach, concluding that an R&D “NAICS code is appropriate when the contractor is ‘involved in the research process itself’ or provides vital functions integral to an agency’s research process.” Id. (emphasis added) (quoting LJR Sols., SBA No.

and Development Contracting.” AR 344. According to the CO, “NCCOS’s mission does not include research and development within the meaning of FAR Part 35” and “[t]herefore, NCCOS does not contract for research and development.” AR 346.

Although the OHA Decision does not attempt to justify the selection of NAICS code 541620 with those assertions the CO made, the government’s MJAR does repeat this argument. See Def. MJAR at 5, 9 (“NOAA also demonstrated that the solicitation did not include any research and development clauses under FAR Part 35.”).

At oral argument, however, the government abandoned its position, conceding that “a basic research contract . . . can still be R&D under FAR 35.” Tr. 16:19-22. Indeed, the government agreed that its “argument hinges on the conclusion that there must be development for [a procurement] to qualify under the NAICS code that Plaintiff prefers,” as defined in the NAICS Manual. Tr. 17:1-11; see also Tr. 39:9-11 (government conceding that it does not maintain “that this contract isn’t about research” but rather “[i]t’s just that there’s no development”). But this Court has already rejected that argument, as explained supra (i.e., because it is unsupported by OHA precedent, it lacks clear textual support in the NAICS Manual, and because the Solicitation does include some development tasks and the NAICS Manual does not specify a particular ratio of research to development tasks).

In any event, the government’s argument does not prove that FAR Part 35 is inapplicable to this procurement, but instead assumes that conclusion on the basis of an underlying assumption that the procurement is not for R&D. It thus merely begs the question whether the CO should have applied FAR Part 35 (and any associated contract clauses) to this procurement. Relatedly, this Court notes that 15 U.S.C. § 638, which covers research and development assistance to small business, defines the terms “research” and “research and development” conterminously (and in the disjunctive) as:

[A]ny activity which is (A) a systematic, intensive study directed toward greater knowledge or understanding of the subject studied; (B) a systematic study directed specifically toward applying new knowledge to meet a recognized need; or (C) a systematic application of knowledge toward the production of useful materials, devices, and systems or methods, including design, development, and improvement

NAICS-5790, at 7). But there is zero evidence in LJR — or in Calspan’s discussion of that case — that the procurement at issue in the former involved any development work whatsoever. Indeed, OHA in Calspan distinguished LJR Solutions on the grounds that the Calspan procurement did not involve “collecting data or samples like those in LJR,” the implication of course being that such tasks would be sufficient to justify assigning the R&D NAICS code. Id. at *10.

of prototypes and new processes to meet specific requirements[.]

15 U.S.C. § 638(e)(5) (emphasis added).²⁷

Moreover, a contract clause from within the RFQ itself supports CSS’s position.

Contract clause NAM 1330-52.203-70 (“Scientific and Research Misconduct”) all but conclusively demonstrates that the RFQ is designed to solicit research and the production of new information. AR 214-16. The Court may deduce several things from the RFQ’s inclusion of this clause. First, this Solicitation is clearly for “scientific activities” and research. NAM 1330-52.203-70(a) (defining “scientific activities”). Second, that contract clause incorporates by reference NOAA Administrative Order (“NAO”) 202-735D, Scientific Integrity. See NAM 1330-52.203-70(b)(2). NAO 202-735D, in turn, defines research and development as follows:

Research is creative and systemic work undertaken in order to increase the stock of knowledge, including knowledge of humankind, culture and society, and to devise new applications of available knowledge.

. . . .

Development is the systematic work, drawing on knowledge gained from research and practical experience and producing additional knowledge, which is directed to producing new products or processes or to improving existing products or processes.”

NAO 202-735D.2, § 3.11 (Jan. 19, 2021) (emphasis added); see id. §§ 5.02, 5.03 (discussing “[c]overed individuals engaged in science and the development of scientific products” (emphasis added)).

But what does NAO 202-735D mean by “products”? The NAO provides the answer: a “[s]cientific product” comprises “[t]he results of scientific activities, including the analysis, synthesis, compilation, or translation of scientific information and data in electronic and hardcopy formats for the use of NOAA, the Department of Commerce, or the Nation.” NAO 202-735D.2, § 3.17. The category is expansive: “[t]hese products include, but are not limited to, experimental and operational models, forecasts, graphics, and verbal and written communications of all kinds relating to scientific activities . . . .” Id. (emphasis added). This definition of scientific products, effectively incorporated in the Solicitation, is surely broad enough to capture the research products the Solicitation anticipates the selected contactor will develop and provide to the Agency. See, e.g., AR 266 (RFQ providing that the government “will evaluate the project solutions and technical expertise proposed for the tasks listed in PWS 2.2 through 2.4, . . . and [the proposed] approach to ensure high quality products and services are delivered” (emphasis added)); AR 278 (RFQ providing that “NCCOS seeks . . . to ensure the delivery of timely, quality, and relevant research and coastal science products” (emphasis added)).²⁸

27Cf. 42 C.F.R. § 52h.2(t) (defining “Research and development contract project” as “an identified, circumscribed activity, involving a single contract or two or more similar, related, or interdependent contracts, intended and designed to acquire new or fuller knowledge and understanding in the areas of biomedical or behavioral research and/or to use such knowledge and understanding to develop useful materials, devices, systems, or methods” (emphasis added)).

Because CSS prevails on the merits of its complaint, the Court next considers the appropriate relief in this case. CSS’s complaint seeks a “declaratory ruling that the OHA Decision was arbitrary and capricious and not in accordance with law,” and either an order remanding to OHA, or a permanent injunction ordering the Agency to classify the Solicitation as R&D under NAICS code 541715. Compl. at 1. CSS later explained, however, that “CSS does not request the Court to affirmatively select a specific NAICS code.” Pl. Resp. at 19. Rather, CSS concedes that “the appropriate remedy would be to issue a declaratory judgment that NAICS code 541620 does not reflect the principal purpose of the RFQ and remand the matter to OHA with instructions to select a NAICS code in a manner consistent with 13 C.F.R. § 121.402(b).” Id. (citing RLB Contracting, Inc. v. United States, 118 Fed. Cl. 750, 762 (2014) (remanding the case and directing the government to make a new NAICS determination consistent with SBA regulations), aff’d, 621 F. App’x 1026 (Fed. Cir. 2015). In deciding whether to grant permanent injunctive relief, a court must consider whether “(1) the plaintiff has succeeded on the merits, (2) the plaintiff will suffer irreparable harm if the court withholds injunctive relief, (3) the balance of hardships to the respective parties favors the grant of injunctive relief, and (4) the public interest is served by a grant of injunctive relief.” Centech Grp., Inc. v. United States, 554 F.3d 1029, 1037 (Fed. Cir. 2009). CSS argues that all four factors favor injunctive relief. Pl. MJAR at 36–40. The government agrees that if CSS is successful on the merits, the government does not contest the other three factors, “given the status of this procurement.” Def. MJAR at 28. The government further “recognize[s]” that if it loses on the merits, then “an injunction could be an appropriate remedy,” though, “at most,” such an injunction “should consist of a remand to the agency.” Id.

28The Defense Federal Acquisition Regulation Supplement (“DFARS”), 48 C.F.R. § 235.001, although not applicable here, provides further confirmation that CSS is correct. The DFARS defines “Research and development” as “those efforts described by the Research, Development, Test, and Evaluation (RDT & E) budget activity definitions found in the DoD Financial Management Regulation (DoD 7000.14–R), Volume 2B, Chapter 5.” DFARS 235.001. The DoD Financial Management Regulation, in turn, defines those efforts as including “basic research.” DoD 7000.14-R, Vol. 2B, Ch. 5, § 1.5.1 (Sept. 2022), https://comptroller.defense.gov/portals/45/documents/fmr/current/02b/02b_05.pdf. The DFARS definition of R&D (i.e., to include basic research) is thus not only consistent with FAR Part 35 and the NAO (regarding scientific integrity), but also supports CSS’s position here.

The Agency’s NAICS code selection and the OHA Decision’s subsequent affirmance of that selection were arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law. Accordingly, Plaintiff’s motion for judgment on the administrative record is GRANTED; Defendant’s motion for judgment on the administrative record is DENIED.

Defendant is ENJOINED from proceeding with the procurement at issue with the currently assigned NAICS code, and this case is remanded to OHA for further consideration and action not inconsistent with this opinion. Whether the R&D NAICS code CSS prefers, or some other NAICS code, best describes the procurement at issue, this Court leaves for the Agency and OHA to decide.

IT IS SO ORDERED.

s/Matthew H. Solomson

Matthew H. Solomson

Judge