Association of American Universities v. Department of Defense

• Court: District Court, D. Massachusetts
• Decided: July 18, 2025
• Docket: 1:25-cv-11740
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT DISTRICT OF MASSACHUSETTS _______________________________________ ) ASSOCIATION OF AMERICAN ) UNIVERSITIES, et al., ) ) Plaintiffs, ) ) Civil Action No. v. ) 25-11740-BEM ) DEPARTMENT OF DEFENSE, et al., ) ) Defendants. ) _______________________________________)

MEMORANDUM AND ORDER ON PLAINTIFFS’ MOTION FOR PRELIMINARY RELIEF MURPHY, J. The Government has, for the fourth time, purported to announce a policy that has consistentlybeen deemed unlawful, without acknowledgment of its apparent illegality and without any attempt to structure the policy in a manner that fulfills the established requirements of law. All three prior cases in this District have enjoined or vacated similar attempts to impose a 15% cap on indirect cost rates for federal research grants. See Massachusetts v. Nat’l Insts. of Health, 1:25-cv-10338-AK (preliminary injunction enjoining National Institutes of Health; entered on March 5, 2025, and converted to permanent injunction and final judgment; appeal pending); Ass’n of Am. Univs. v. Dep’t of Energy, 1:25-cv-10912-ADB (preliminary injunction enjoining Department of Energy; entered on May 15, 2025, and converted to permanent injunction and final judgment); Ass’n of Am. Univs. v. Nat’l Sci. Found., 1:25-cv-11231-IT (granting summary judgment and entering final judgment on June 20, 2025, against National Science Foundation, vacatingthe rate cap). In each of those cases, the court has found not only that the policy at issue violates law on the books but that the Government has wholly failed its legal obligation to rationally explain why it has adopted these policies. Despite these successive losses—and despite the benefit of thorough, reasoned decisions from three different judges—the Government has imposed a nearly identical policy with respect to Department of Defense(“DOD”) research grants issued to universities. Before the Court is Plaintiffs’ motion for injunctive relief or a stay of the policy.1 For the reasons set forth below, Plaintiffs’ request for a preliminary injunction is

GRANTED, and Plaintiffs’ request for a stay is DENIED. I. Background A. Factual Background 1. The Parties Plaintiffs are the Association of American Universities (“AAU”), the Association of Public and Land-Grant Universities (“APLU”), the American Council on Education (“ACE”) (collectively, “Organizational Plaintiffs”), and 12 universities2 that house significant scientific study and innovation supported by DOD grants (collectively, “University Plaintiffs,” and with Organizational Plaintiffs, “Plaintiffs”). Defendants are DOD and Secretary of Defense Peter Hegseth (collectively, “Defendants”).

1 Plaintiffs originally filed a motion for a temporary restraining order, Dkts. 13, 39, which the Court granted, Dkt. 48, and subsequently extended, Dkts. 62, 70. As the motion has been fully briefed, and the parties appeared before this Court for a hearing on the issue, the motion is converted to one seeking a preliminary injunction. See Fed. R. Civ. P. 65(b)(3) (“If the [temporary restraining] order is issued without notice, the motion for a preliminary injunction must be set for hearing at the earliest possible time . . . At the hearing, the party who obtained the order must proceed with the motion; if the party does not, the court must dissolve the order.”). After receiving supplemental briefing on, inter alia, the potential remedy for the pending motion, see Dkt. 63 (clerk’s notes for hearing with timeline for additional briefing); Dkt. 67 (Plaintiffs’ supplemental briefing), the Court treats the motion as one seeking a stay or preliminary injunction. 2 Those universities are Arizona Board of Regents on Behalf of Arizona State University (“ASU”); Brown University (“Brown”); California Institute of Technology (“Caltech”); The Regents of the University of California (“UC”); Board of Governors of the Colorado State University System Acting by and through Colorado State University (“CSU”); Cornell University (“Cornell”); The Board of Trustees of the University of Illinois (“UIUC”); The Johns Hopkins University (“JHU”); University of Maryland, College Park (“UMCP”); Massachusetts Institute of Technology (“MIT”); University of Pittsburgh of the Commonwealth System of Higher Education (“Pitt”); and University of Washington (“UW”). 2. DOD Grants Congress has authorized DOD to issue grants supporting research that advances American defense capabilities, protects national security, and meets DOD’s other research needs.3 See 10 U.S.C. § 4001.4 As relevant here, DOD frequently engages institutions of higher education (“IHEs” or “universities”) to partner on those research efforts.5 It is undisputed that this

partnership with academia is “essential for the development of transformative research concepts that drive [DOD’s] future capabilities.” Dkt. 1-2 (“Michael Memo”) at 3. Federal research grants typically work through reimbursement. See 2 C.F.R. § 200.305(b)(3) (stating that reimbursement is the federal government’s preference unless certain conditions are met). IHEs are required to document their research costs, which are then “charged” back to the relevant grant. See 2 C.F.R. § 200.400(d). 3. Statutory and Regulatory Framework a. The Uniform Guidance The same statute that authorizes DOD to issue research grants provides that those grants will be awarded “in accordance with chapter 63 of title 31” of the U.S. Code. 10 U.S.C. § 4001(b)(1) (referring to 31 U.S.C. § 6301 et seq.). In turn, that chapter authorizes the Office of

Management and Budget (“OMB”) to issue guidelines “to promote consistent and efficient use of

3 The Court uses “grants” to refer to both grants and cooperative agreements. See 31 U.S.C. §§ 6304–05 (distinguishing between the two). 4 See also 10 U.S.C. § 4010 (directing Secretary of Defense to carry out “Defense Established Program to Stimulate Competitive Research (DEPSCoR) as part of the university research programs of the Department of Defense”); id. § 4141 (requiring university research grants to be issued through competitive processes). 5 See generally, e.g., National Center for Science and Engineering Statistics, NSF 25-314, Higher Education Research and Development (HERD) Survey: Fiscal Year 2023 (2024), Table 26, https://ncses.nsf.gov/pubs/nsf25314/assets/data-tables/tables/nsf25314-tab026.pdf [https://perma.cc/2HYB-W7UN] (Federally financed higher education R&D expenditures, ranked by all federal R&D expenditures, by federal agency: FY 2023) (showing that, in 2023, DOD awarded more than $9 billion to over 470 different universities). . . .grant agreements.” 31 U.S.C. § 6307; see also id. § 503 (granting OMB authority to “establish governmentwide financial management policies”). OMB has thereupon issued regulations that govern the fiscal administration of federal grants, namely the Uniform Guidance for Federal Awards. 2 C.F.R. pt. 200 (hereinafter, the “Uniform Guidance”).6 The Uniform Guidance establishes basic principles and procedures for

federal grant accounting. i. Allowable vs. UnallowableCosts Two such principles relevant to this case are “allowable” versus “unallowable” costs. Allowable costs—in other words, what the Government will pay for—must, among other requirements, be “necessary and reasonable”; must “conform to any limitations or exclusions”; must be consistently accounted for; and must be “adequately documented.” Id. § 200.403. Unallowable costs, by contrast, are what the Government will not pay for—for example, entertainment costs are generally unallowable. See id. § 200.438. Accordingly, the principle of allowability filters out costs that have been deemed undeserving of federal dollars, based on the nature of the expense.

ii. Direct vs. Indirect Costs Whereas allowability dictates what costs can be reimbursed using grant funds, the idea of direct and indirect costs informs how those costs can be tied back to the grants themselves. Simply, some allowable costs are easier to associate with a specific project than others. For example, if a government-sponsored research program uses some amount of a chemicalor other consumable as part of an experiment, the organization can relatively easily track that expense and attribute it to

6 Though styled as a “guidance,” grant-issuing federal agencies are required to implement the Uniform Guidance or seek an exemption. 2 C.F.R. § 200.106. Indeed, DOD has implemented the Uniform Guidance, id. pt. 1104, and Defendants agree “that the OMB uniform guidance is binding on the DOD,” Dkt. 64 (“Tr.”) at 33:22–25. the specific project. Such expenses are called “direct” costs because they are directly allocable. See id. § 200.413(a). By contrast, some allowable costs cannot easily beattributed to a particular project. For example, research is very often conducted in buildings, and buildings cost money. Maintenance of that building is explicitly an allowable cost, id. § 200.452, but it would make little sense to talk about how much of the cost to fix the front door of a research lab is attributable to

any given research project housed within that lab. Because these costs cannot be directly associated withany given project without resort to unreasonably fictitious methods of accounting, they are referred to as “indirect costs.” See id. § 200.414(a); cf. id. §200.413(a). “There is no universal rule for classifying certain costs as direct or indirect costs.” Id. §200.412. Unlike the distinction between allowable and unallowable costs, which isbased on the nature of the expense, a cost’s being treated asdirect or indirectis a function of accounting and of the principles set forth in the Uniform Guidance. For example, most administrative salaries are properly considered indirect costs because administrators usually do work that is relevant to multiple functions or activities of an institution, and so their salaries cannot reasonably be

attributed to one grant over another. Id. § 200.413(c). However, if an organization hires an administrator to work exclusively on a single government-sponsored grant project, that salary might be treated as a direct cost. Id. §§ 200.413(c)(1)–(3). Thus, salaries for two individuals doing the exact same type of work might appear as either direct or indirect on the final balance sheet depending on non-substantive factors. Likewise, “minor items”—such as pens and postage stamps—are typically treated as indirect costs. Id. § 200.413(d); id. pt. 200, App. III(B)(6)(b)(2). However, in circumstances where a substantial (and traceable) number of these minor items are used for a particular project, they may be treated as direct costs. See id. § 200.412 (limiting the consistency principle to “like circumstances”). Thus, there is no substantive difference between direct and indirect costs, and it is reductive (if not sometimes plainlyincorrect) to refer to indirect costs simply as “overhead.” Rather, an indirect cost may be as glamorous and scientific as a supercomputer, see id. pt. 200, App. III(B)(2)(a), (B)(4)(a); or it may be as pedestrian, if nonetheless essential, as a paperclip, id. pt. 200, App. III(B)(6)(a)(2)(b), (B)(6)(b)(2). iii. Indirect Cost Rates

Indirect costs are charged back to federal grants using indirect cost rates. Roughly speaking, an indirect cost rate is a ratio (expressed as a percentage) of indirect costs to direct costs. In other words, for every dollar that a university spends on research salaries and experiment materials, i.e., direct costs, how much does it cost to keep the lights on, i.e., indirect costs? The resulting percentage provides a basis for IHEs to seek reimbursement for allowable indirect costs—for instance, for the use of expensive lab equipment—proportionally with their allocable direct costs. Of course, it is much more complicated than that. Appendix III of the Uniform Guidance sets out a detailed methodology for IHEs to apportion their indirect research costs among several categories, which are then aggregated into two major categories (“facilities” and “administration”),

which are then ultimately used to calculate the indirect cost rates. See generally id. pt. 200, App. III. The formula has built-in restrictions. Most notably, OMB has capped administrative costs at 26%. Id. pt. 200, App. III(C)(8)(a). The overall process is designed and intended “to ensure that Federal funds are not used to subsidize industry and foreign government funded programs.” Id. pt. 200, App. III(C)(1)(a)(3). Indirect cost rates are typically predetermined, before grant issuance, based on past expenditures. See id. pt. 200, App. III(C)(4). Predetermined indirect cost rates “simplify the administration of cost-type research and development contracts (including grants) with educational institutions, . . . facilitate the preparation of their budgets, and . . . permit more expeditious closeout of such contracts when the work is completed.” Id. (citing 41 U.S.C. § 4708). Indeed, Congress has explicitly blessed their use. 41 U.S.C. § 4708 (“A cost-type research and development contract (including a grant) with a university, college, or other educational institution may provide for payment of reimbursable indirect costs on the basis of predetermined fixed-percentage rates.”). Just as indirect costs are not really “overhead,” indirect cost rates are not, at least from a

conceptual standpoint, proper measures of efficiency. For one example, it would likely decrease an organization’s efficiency to track how many pens each of its researchers uses vis-à-vis different research projects, but if an organization were to consistently undertake that exercise, it might be able to treat those pens as a direct expense, artificially deflating its indirect cost rate by taking on a burdensome (and, it would seem, pointless) exercise.7 Likewise, it would obviously be inefficient for a lab to hire a different janitor to take out the trash for each of its government-funded research projects, though that might theoretically allow the institution to treat those janitors’ salaries as direct costs.8 Thus, while lowering costs generally—direct or indirect—naturally correlates with increased efficiency, it does not follow that an organization’s having more direct

versus indirect costs necessarily indicates anything about its efficiency. Once an IHE goes through the lengthy process of determining its indirect cost rate under the Uniform Guidance, that rateisthenaudited and, if necessary,corrected by an assignedfederal agency, known as the IHE’s “cognizant agency.” 2 C.F.R. pt. 200, App. III(C)(11). For some

7 But see 2 C.F.R. pt. 200, App. III(C)(8)(b) (directing institutions not to change their accounting methods to convert indirect costs to direct costs). 8 However, as this would almost certainly be unnecessary and unreasonable, it would likely be unallowable. See 2 C.F.R. § 200.403; see also id. pt. 200, App. III(C)(8)(b). This example demonstrates how an institution’s indirect cost rate is substantially a product of the regulations themselves. Hypothetically, one could devise a system wherein a janitor’s salary is divided according to the number of rooms he cleans, cross-referenced against the research objectives pursued in those rooms, and allocated accordingly. Under that system, a janitor’s salary might reasonably be treated as a direct cost, decreasing the institution’s apparent indirect cost rate. Importantly, however, the Uniform Guidance does not provide for that system. IHEs, DOD is the cognizant agency, normally whereDOD provides the bulk of thatIHE’s federal grants. Id. pt. 200, App. III(C)(11)(a)(1). Once an IHE and its cognizant agency agree that the IHE’s indirect cost rate has been properly accounted, that becomes the IHE’s“Negotiated Indirect Cost Rate” (“NICRA”).9 See id. pt. 200, App. III(C)(11)(f); id. § 200.414(c). The Uniform Guidance suggests that an IHE’s NICRA should be renegotiated every two to four years. Id.

pt. 200, App. III(C)(4). A grant recipient may also apply to its cognizant agency for a one-time extension of up to four years. Id. § 200.414(g). iv. Applying and Deviating from NICRAs Much of this lawsuit turns on the correct reading of section 200.414(c) of the Uniform Guidance. That section states, as a baseline rule, that NICRAs “must be accepted by all Federal agencies.” Id. § 200.414(c)(1). If an agency wishes to deviate from this baseline rule and apply “a rate different from the negotiated rate,” it must establish alternate “policies, procedures and general decision-making criteria” that the agency “will follow to seek and justify deviations from negotiated rates.” Id. §§ 200.414(c)(1), (3). Moreover, it must publicize those deviating rules and procedures. Id. This

publicity rule tracks a more general requirement that agencies, in their notices of funding opportunities, include policies relating to indirect cost reimbursement and, as discussed in further detail below, cost sharing. Id. § 200.414(c)(4). The text of the regulation thus parallels the “policies, procedures and general decision-making criteria” that a deviating agency is required to establish with the Uniform Guidance accounting system that it seeks to replace. The unmistakable preference under the

9 AAU has alleged that its member universities generally have indirect cost rates between 50% and 60%. Dkt. 13-1 (“AAU Decl.”) ¶ 13. Of the University Plaintiffs that provided specific rates, they have alleged that they currently have NICRAs that range from 48.5% to 69.5%. Dkt. 1 (“Compl.”) ¶¶ 11–22. regulation is that federal agencies adopt the uniform accounting system for indirect costs. See id. § 200.414(c)(1) (“Negotiated indirect cost rates must be accepted by all Federal agencies.”). Nevertheless, insofar as one insists, the regulation does permit an agency to adopt a methodology with “justif[iable]” idiosyncrasy. Id. § 200.414(c)(3). v. De Minimis Rates and Cost Sharing

Finally, if a grant recipient has not already determined its indirect cost rate and negotiated its NICRA, it may elect to forego that process and instead charge a “de minimis” indirect cost rate of up to 15%. Id. § 200.414(f). Importantly, however, grant recipients “are not required to use the de minimis rate,” and a federal agency “may not require” a grant recipient “to use a de minimis rate lower than [its] negotiated indirect cost rate.” Id. This prohibition against imposed de minimis indirect cost rates resonates with the Uniform Guidance’s overall treatment of cost sharing. “Cost sharing” is the term used to refer to that part of a federally funded research project that is not paid for with federal funds despite being otherwise allowable. Id. § 200.1. Cost sharing thus provides an analytic complement to the principles already discussed: allowability establishes what can be reimbursed; whether a cost is direct or

indirect determines how it will be reimbursed; cost sharing fixes how much of the total will be reimbursed. If an agency wishes to mandate cost sharing, the Uniform Guidance prescribes a certain way of doing that. The agency must explicitly state that it is mandating cost sharing in its notice of funding opportunity. Id. § 200.414(c)(4); id. App. I(b)(2)(ii). The agency must then do so again in the actual award. Id. § 200.211(9). Cost sharing is subject to accounting and audit requirements. See id. §§ 200.306(b)–(k). Highlighting the degree to which the Uniform Guidance disallows mandatory cost sharing outside these bounds, agencies are forbidden from even considering an applicant’s voluntary cost sharing during grant application review, preventing the appearance of unspoken requirements. Id. § 200.306(a) (“Voluntary committed cost sharing is not expected under Federal research grants.”). The Uniform Guidance’s clearpresumption against (and specific procedures in the case of) mandatory cost sharing provides context for why agencies may not unilaterally impose de minimis indirect cost rates. See id. § 200.414(f). Indeed, the cost sharing rules would have little meaning or effect if they could be worked around so easily to achieve cost

sharing by another name. 4. Historical Limitations on Indirect Cost Funding a. By Congress Congress has had three notable interventions with respect to indirect cost rates. In 1962, Congress first authorized the use of “predetermined fixed-percentage rates” for “payment of reimbursable indirect costs,” i.e., predetermined indirect cost rates. See 41 U.S.C. § 4708.10 For approximately three years thereafter, Congress capped indirect cost rates for research grants at 20%.11 By 1965, however, Congress abandoned limits on indirect cost rates.12

10Codifying Act of Sept. 5, 1962, Pub. L. No. 87-638, 76 Stat. 437. 11See Department of Defense Appropriation Act, 1963, Pub. L. No. 87-577, § 540, 76 Stat. 318, 334 (1962); Department of Defense Appropriation Act, 1965, Pub. L. No. 88-446, § 538, 78 Stat. 465, 481 (1964). Around this time, Congress appears to have also enacted several other agency-specific indirect cost limits through appropriations riders. See, e.g., Departments of Labor, and Health, Education, and Welfare Appropriation Act, 1958, Pub. L. No. 85-67, § 208, 71 Stat. 210, 224 (1957) (“None of the funds provided herein shall be used to pay any recipient of a grant for the conduct of a research project an amount for indirect expenses in connection with such project in excess of 15 per centum of the direct costs.”). 12See Department of Defense Appropriation Act, 1966, Pub. L. No. 89-213, § 638, 79 Stat. 863, 879 (1965). At that time, however, Congress did require cost sharing. See id. (“None of the funds provided herein shall be used to pay any recipient of a grant for the conduct of a research project an amount equal to as much as the entire cost of such project.”). Cost sharing is, however, no longer mandated. See 2 C.F.R. § 200.306(a). More than fifty years passed before Congress acted again.13 In 2007, it imposed an effective 53.8% indirect cost rate limit specifically for DOD grants.14 However, again, Congress abandoned that requirement after only three years.15 Finally, in 2017, Congress took action againstan executive budget proposal that sought to cap indirect cost rates for National Institutes of Health (“NIH”) grants at 10%.16 Congress not

only rejected that proposal but enacted a separate appropriations rider specifically directing NIH not todeviate from the indirect cost reimbursement procedures in place.17 Both the House and Senate Reports on the appropriations rider identified serious problems with the proposal. The Senate Report observed that “[t]he methodology for negotiating indirect costs has been in place since 1965” and predicted that anindirect rate capwould “radically change the nature of the Federal Government’s relationship with the research community, abandoning the Government’s long-established responsibility for underwriting much of the Nation’s research

13 In the intervening years, there was certainly debate and discussion, including a 1991 House of Representatives investigation into university overcharging. See Marcy E. Gallo & Laurie Harris, Cong. Rsch. Serv., R48540, Univs. & Indirect Costs for Federally Funded Rsch. 8–9 (2025). 14 Department of Defense Appropriations Act, 2008, Pub. L. No. 110-116, § 8115, 121 Stat. 1295, 1340 (2007). The limit was expressed as 35% of the total award cost, or roughly 53.8% expressed as an indirect cost ratio. See id. To be clear, these are financial equivalents. For example, a recipient might spend $350 of a $1000 grant on indirect costs (35% of the total cost award), leaving $650 for direct costs. To calculate the indirect cost rate, one would then divide $350 by $650, yielding roughly 0.538 (53.8%). 15 See Consolidated Security, Disaster Assistance, and Continuing Appropriations Act, 2009, Pub. L. No. 110-329, § 8109, 122 Stat. 3574, 3645 (2008); Department of Defense Appropriations Act, 2010, Pub. L. No. 111-118, § 8101, 123 Stat. 3409, 3450–51 (2009). 16See Office of Management & Budget, Major Savings and Reforms: Budget of the U.S. Government Fiscal Year 2018, at 43 (2017), https://www.gov info.gov/content/pkg/BUDGET-2018-MSV/pdf/BUDGET-2018-MSV.pdf [https://perma.cc/M56V-5HFL]. 17See Consolidated Appropriations Act, 2018, Pub. L. No. 115-141, § 226, 132 Stat. 348, 740 (2017) (“[T]he provisions relating to indirect costs in part 75 of title 45, Code of Federal Regulations, including with respect to the approval of deviations from negotiated rates, shall continue to apply to the National Institutes of Health to the same extent and in the same manner as such provisions were applied in the third quarter of fiscal year 2017. None of the funds appropriated in this or prior Acts or otherwise made available to the Department of Health and Human Services or to any department or agency may be used to develop or implement a modified approach to such provisions, or to intentionally or substantially expand the fiscal effect of the approval of such deviations from negotiated rates beyond the proportional effect of such approvals in such quarter.”). infrastructure, and jeopardizing biomedical research nationwide.” S. Rep. No. 115-150, at 109 (2017). The House Report likewise called the proposal “misguided” and stated that it would “have a devastating impact on biomedical research across the country.” H.R. Rep. No. 115-244, at 50 (2017). Congress has reenacted the same appropriations rider each year since, and it remains in effect to this day.18

b. By Agencies As far as this Court is aware, except for OMB and its predecessor, no federal agency has taken any significant action impacting how indirect cost rates arecalculatedor applied since 1958. In that year, the Bureau of the Budget (later, OMB) issued Circular A-21 (later relocated to the Code of Federal Regulations as the Uniform Guidance), replacing individual agency polices previously in place.19 Among the revisions to Circular A-21 and the Uniform Guidance made by OMB over the past seven decades, two bear substantially on the present issues. In February 1986, OMB announced a new rule limiting reimbursement for administrative costs (a subset of indirect costs) to 26%.20 This proposal was based on the average rate for all major universities over the preceding

five years and was intended to “solv[e] the problem of rising indirect costs” while “at the same time reduce paperwork for the institutions.”21 After notice and comment, OMB adopted a more

18 See Further Consolidated Appropriations Act, 2024, Pub. L. No. 118-47, div. D, tit. II, § 224, 138 Stat. 460, 677. 19 See Marcy E. Gallo & Laurie Harris, Cong. Rsch. Serv., R48540, Univs. & Indirect Costs for Federally Funded Rsch. 4, 7–8 (2025); see also Dkt. 39 (“Mem.”) at 15–21 (“Prior Efforts to Contain Indirect Costs in Grants to IHEs”). 20Proposed Revision of Circular A-21, “Cost Principles for Educational Institutions,” 51 Fed. Reg. 5286-01, 1986 WL 87893 (Feb. 12, 1986). 21Id. targeted rule limiting administrative costs chargedfordepartment heads and faculty to3%.22 This figure was arrived at through consideration of multiple public reports analyzing such costs and was subsequently adjusted a few months later to 3.6% based onadditional data.23 This 3.6% limit remains in effect today.24 In June 1991, OMB again proposed a 26% limit on administrative costs, as part of a broader

effort to reform and clarify its indirect cost accounting rules.25 In the rule-change announcement, OMB Director Richard Darman was quoted: The Administration remains committed to funding a fair share of the costs of the research enterprise. But this share must not include excessive administrative costs. Abuses in cost recovery must be stopped. At the same time, since many of these abuses would not occur if there were agreed accounting standards, we must proceed immediately to define those standards.26 In October 1991, the26% administrative rate limit was adopted and remains in effect today.27 By contrast, the parties have not presented—nor has the Court found—any example of a grant-issuing agency, such as DOD, successfully enacting a policy that significantly limits indirect cost reimbursement or otherwise modifies OMB’s cost principles.28 Such agencies have primarily

22 Revision of OMB Circular A-21, “Cost Principles for Educational Institutions,” 51 Fed. Reg. 20908-01, 1986 WL 98846 (June 9, 1986). 23 Revision of OMB Circular A-21, “Cost Principles for Educational Institutions,” 51 Fed. Reg. 43487-02, 1986 WL 116982 (Dec. 2, 1986). 242 C.F.R. pt. 200, App. III(B)(6)(a)(2)(a). 25Proposed Revisions to Circular A-21, 56 Fed. Reg. 29530-01, 1991 WL 298704 (June 27, 1991). 26Id. 27 Revisions to Circular A-21, “Cost Principles for Educational Institutions,” 56 Fed. Reg. 50224-01, 1991 WL 195459 (Oct. 3, 1991); 2 C.F.R. pt. 200, App. III(C)(8)(a). 28 As discussed above, a similar indirect cost rate limit, specific to NIH, was introduced in 2017 through a budget proposal—i.e., through legislation rather than executive authority—and was rebuffed by Congress. See notes 16–18, supra, and accompanying text. Other agencies’ more recent attempts are discussed below. participated in the process only as cognizant agencies (effectively, as designated auditors for the OMB system). See 2 C.F.R. pt. 200, App. III(C)(11)(a).29 5. The Rate Cap Policy On May 14, 2025, Secretary Hegseth issued a memorandum entitled “Implementation of a 15% Indirect Cost Cap on Assistance Awards to Institutions of Higher Education,” which

announced that “[e]ffective immediately, [DOD] will pursue a lower cap on indirect cost rates for all new financial assistance awards to institutions of higher education.” Dkt. 1-1 (“Hegseth Memo,” and collectively with the Michael Memo, the “Rate Cap Policy” or “Policy”) at 2. As to new grants, the memorandum instructed the Under Secretary of Defense for Research and Engineering to: (i) “[n]otify the Office of Management and Budget of [DOD’s] intent to cap indirect cost rates for all new financial assistance awards to institutions of higher education at 15% or lower”; (ii) “[d]evelop and publish formal policy guidance—including procedures, decision criteria, and justifications—that will govern all DoD deviations from negotiated rates”; (iii) ensure that the “formal policy guidance” is made “public and integrated into all upcoming grant solicitations, including Notices of Funding Opportunity”; and (iv) “[e]nsure new awards to

institutions of higher education contain the newly established standard cap.” Id. at 2. Plaintiffs contend that, as early as May 21, 2025, grant managers began notifying researchers with proposals in progress that “[e]ffective immediately[,] all new submissions must comply with [the Hegseth Memo],” and that DOD “cannot process your proposal unless it complies with the memo,” i.e., includes indirect cost reimbursement of no more than 15%.

29 The Court also assumes that there are other instances of agencies’ invoking the negotiated cost rate exception provided for in section 200.414(c). Defendants have provided none. Plaintiffs briefly referenced several examples at oral argument but noted that they were “pretty hard to find.” Tr. at 39:14–40:6. Having done its own fruitless survey of funding notices, the Court agrees and therefore can only conclude that these exceptions have historically been rare and unnoteworthy. Dkt.13-5 (“BU Decl.”) ¶ 18.30 Plaintiffs further assert that some institutions were directed to resubmit previously submitted proposals after revising them to reflect a 15% indirect cost rate. Dkt. 13-14 (“JHU Decl.”) ¶ 27 & Ex. 1. On or about June 10, 2025, the Under Secretary of Defense, Emil Michael, issued a second memorandum “implement[ing] the Department’s policy deviating from the negotiated indirect cost

rate for [IHE grants].” Michael Memo at 2. For new awards, the Michael Memo requires DOD components to implement the cap immediately: components are “not to allow indirect cost rates above 15% in all new assistance awards to IHEs as of the date of publication of th[e Michael Memo].” Id. at 3. The Michael Memo applies this 15% limit on indirect cost rates only to universities, not other recipients of DOD grants. Defendants assert that this Rate Cap Policy will result in cost savings “up to $900M per year,” allowing DOD “to repurpose those funds — toward applied innovation, operational capability, and strategic deterrence.” Hegseth Memo at 2; see also Michael Memo at 2–3 (discussing how the Rate Cap Policy will lead to “significant cost savings”).

DOD’s Rate Cap Policy is one of at least three similar agency policies announced in the first six months of this year. In February 2025, NIH likewise announced a plan to limit indirect cost rates to 15%.31 Shortly thereafter, three sets of plaintiffs—a group of 22 states, a group of five medical associations, and a group of 17 associations and universities, including many of the

30 In support of their motion, Plaintiffs submitted declarations from an additional nine IHEs: Boston University (“BU”); University of Colorado Boulder (“CU Boulder”); Duke University (“Duke”); University of Kansas – Lawrence Campus (“KUL”); University of Kansas Medical Center (“KUMC”); Oregon State University (“OSU”); University of Pennsylvania (“Penn”); University of Southern California (“USC”); The University of Toledo (“UToledo”); and University of Wisconsin-Madison (“UW-Madison”). 31 NIH, Supplemental Guidance to the 2024 NIH Grants Policy Statement: Indirect Cost Rates, NOT-OD-25-068 (Feb. 7, 2025), https://grants.nih.gov/grants/guide/notice-files/NOT-OD-25-068.html [https://perma.cc/H4G6-X3YX]. plaintiffs in the present action—filed suit, arguing that the NIH policy was unlawful under the Administrative Procedure Act (“APA”). Another session of this court, Judge Kelley presiding, found that the policy not only flouted the appropriations rider that Congress enacted in the wake of the 2017 NIH budget proposal, but also that it violated the government-wide regulations governing indirect cost rates and the reasoned decision-making requirements of the APA, issuing

a temporary restraining order followed by a preliminary and then permanent injunction. See Massachusetts v. Nat’l Instits. of Health (“NIH I”), 770 F. Supp. 3d 277 (D. Mass. Mar. 5, 2025), judgment entered, 2025 WL 1063760 (D. Mass. Apr. 4, 2025), appeal docketed, No. 25-1343 (1st Cir. Apr. 9, 2025). Despite that court’s decision in NIH I, on April 11, 2025, the Department of Energy (“DOE”) announced a nearly identical policy capping indirect cost rates at 15%.32 Shortly thereafter, nine universities and three associations representing universities, including several plaintiffs in this case, filed suit to challenge the DOE policy. Yet another session of this court, Judge Burroughs presiding, found that the policy violated the APA and entered a temporary

restraining order followed by a preliminary and then permanent injunction. See Ass’n of Am. Univs. v. Dep’t of Energy (“DOE”), — F. Supp. 3d —, 2025 WL 1414135 (D. Mass. May15,2025), judgment entered,No. 25-cv-10912, Dkt.71(D.Mass. June30, 2025). While the DOE court considered whether to convert its temporary restraining order into a preliminary injunction,on May 2, 2025, the National Science Foundation (“NSF”)issued its own, nearly identical policy capping indirect costratesat 15%.33 Shortly thereafter, thirteen universities

32 DOE, Policy Flash 2025-22: Adjusting Department of Energy Grant Policy for Institutions of Higher Education (IHE) (Apr. 11, 2025), https://www.energy.gov/management/pf-2025-22-adjusting-department-energy- grant-policy-institutions-higher-education-ihe [https://perma.cc/F4G8-2HP5]. 33 NSF, Policy Notice: Implementation of Standard 15% Indirect Cost Rate (May 2, 2025), https://www.nsf.gov/policies/document/indirect-cost-rate [https://perma.cc/9L8J-AEJU]. and three associations representing universities, including several plaintiffs in this case, filed suit challenging the NSF policy.34 A third session of this court, Judge Talwani presiding, granted summary judgment against NSF, holding that the policy violated the APA, and vacated the policy entirely. See Ass’n of Am. Univs. v. Nat’l Sci. Found. (“NSF”), — F. Supp. 3d —, 2025 WL 1725857 (D. Mass. June 20, 2025).

B. Procedural Background On June 16, 2025, Plaintiffs commenced this action for declaratory and injunctive relief, alleging that the Rate Cap Policy violates (i) section 706(2)(A) of the APA because it is contrary to law (Counts I, II, IV, and V); (ii) section 706(2)(A) of the APA because it is arbitrary and capricious (Count III); and (iii) section 706(2)(C) of the APA because it is in excess of statutory authority (Count VI).35 That same day, Plaintiffs moved for a temporary restraining order, which the Court granted on June 17, 2025, and subsequently extended, first for fourteen days, and then for an additional three days. Defendants submitted an opposition on June 24, 2025, and Plaintiffs filed a further reply in support of their motion on June 27, 2025. The Court held a hearing on July 2, 2025, requested additional briefing, and took the matter under advisement. On July 9 and

11, 2025, the parties submitted briefing addressing, in relevant part, the proposed remedy and the impact of the Supreme Court’s decision in Trump v. CASA, Inc., 606 U.S. —, 2025 WL 1773631 (June 27, 2025).

34In response to that lawsuit, and in seeming, tacit acknowledgment of the growing consensus among judges regarding the illegality of the policy, NSF agreed to stay implementation of its policy through at least one week after the hearing on the parties’ cross-motions for summary judgment. 35Count VI applies only to existing grants and is not at issue in the present motion. II. Jurisdiction Defendants assertthat this Court lacks jurisdiction over Plaintiffs’ claimsbecause Plaintiffs have not shown that they have standing, and because DOD’s grant-related decisions are committed to discretion by law under the APA. Dkt. 58 (“Opp.”) at 12–13.36 The Court addresses each argument in turn.

A. Standing At the outset, Defendants contend that Plaintiffs have not established standing because they have failed to establish an injury in fact. More specifically, Defendants argue that because no Plaintiff has alleged that DOD has altered the terms of an existing grant by applying the 15% indirect cost rate; no Plaintiff has alleged that it has received a new grant award at the 15% indirect cost rate; and no Plaintiff has alleged that it is guaranteed or entitled to receive a new grant after June 12, 2025—“the only ‘injury’ a plaintiff can identify would stem from potentially being offered a grant award with a lower indirect cost rate than contained in that plaintiff’s NICRA . . . or having to submit (or decide not to submit) grant proposals using the new 15 percent indirect cost rate.” Opp. at 23 (emphasis removed). Defendants assert that “[a]n opportunity to accept or reject a federal grant on terms other than those requested by the offeree is not ‘an invasion of a

legally protected interest,’” id. (quoting Lujan v. Defs. of Wildlife, 504 U.S. 555, 560 (1992)), but rather constitutes a “generalized grievance” that is insufficient to confer standing, id. Defendants further assert that any alleged injury is speculative. Id. For the reasons set forth below, this Court disagrees.

36Defendants further argued in their brief that even if the Court has jurisdiction over Plaintiffs’ APA claims regarding the Michael Memo, the Hegseth Memo is not final agency action, and the Court therefore lacks jurisdiction under the APA to fashion a remedy for it. Opp. at 13. Defendants largely abandoned that argument at oral argument, as discussed further in Section III(B)(1)(a), supra. “A federal court possesses Article III jurisdiction to hear a case or controversy only if it alleges an injury in fact.” Aguasvivas v. Pompeo, 984 F.3d 1047, 1052 (1st Cir. 2021) (citing Susan B. Anthony List v. Driehaus (“SBA List”), 573 U.S. 149, 157–58(2014)). “An injury in fact must be ‘concrete and particularized’ and ‘actual or imminent,’ not ‘conjectural’ or ‘hypothetical.’” O’Neil v. Canton Police Dep’t, 116 F.4th 25, 30–31 (1st Cir. 2024)(quoting SBA

List, 573 U.S. at 158). “‘[A] future injury’ is imminent ‘if the threatened injury is certainly impending, or [if] there is a substantial risk that the harm will occur.’” New Hampshire Lottery Comm’n v. Rosen, 986 F.3d 38, 50 (1st Cir. 2021) (alterations in original) (quoting SBA List, 573 U.S. at 158). It is true, as Defendants contend, that “Plaintiffs are not certain to receive any specific grant award.”37 Opp. at 23 n.36. However, that does not convince this Court that the alleged injury is merely a speculative one.38 A recent decision by another session in this District is instructive. In NSF, 2025 WL 1725857, the court considered whether many of the same plaintiffs as those in this action had standing to challenge an analogous policy limiting indirect cost rates by NSF. There,

the court observed that, “[e]ven though the 15% Indirect Cost Rate only cuts indirect costs as to

37 But see ASU Decl. ¶ 9. Plaintiffs allege that DOD has designated some of their pending proposals as “award anticipated,” id., which they contend demonstrates a substantial likelihood that the awards will be granted. See Tr. at 29:10–13 (“Arizona State University which has I think 13 or 14 awards that have been designated by DOD as award anticipated. So, like, DOD has indicated they’re going to get these awards.”). The Court finds that Plaintiffs have standing based on their overall loss of funding eligibility. Nonetheless, the Court agrees that these “anticipated” awards further substantiate Plaintiffs’ likelihood of injury. 38Nor does Roberts v. Progressive Preferred Insurance Co., 2024 WL 2295482 (N.D. Ohio May 21, 2024), a case cited by Defendants, alter that conclusion. See Opp. at 23. In Roberts, the court held that the “[p]laintiffs’ claim for prospective relief based on these hypothetical, future grants—that Plaintiffs may or may not be eligible for—amounts to nothing more than a generalized grievance about Defendants’ programs.” Roberts, 2024 WL 2295482, at *8. In that case, however, the plaintiffs “d[id] not allege that when they filed th[eir] lawsuit, any grants were being offered by Defendants for which Plaintiffs were ‘able and ready’ to apply,” id. at *7, or that “they would otherwise be eligible or ‘able and ready’ to apply for any ‘similar’ grant,” id. at *8; see also id. at *8 (“Rather, the record before the Court suggests that the Grant, as challenged, was offered as a one-time opportunity.”). Under those circumstances—which are unlike the circumstances in this action—the court found that the plaintiffs lacked standing. Even if that case were not distinguishable, however, that decision is not binding on this Court. future grants that NSF has not yet awarded, any IHE that has negotiated an indirect cost rate with HHS or [DOD’s Office of Naval Research(“ONR”)] above 15% becomes eligible for less federal funding if NSF implements the 15% Indirect Cost Rate.” Id. at *7 (emphasis in original). In concluding that the plaintiffs there had suffered an injury in fact, the court reasoned that, “where Plaintiffs have shown that they are eligible for less federal funding of indirect costs under the 15%

Indirect Cost Rate such that their research programs will face significant disruption, they have demonstrated an injury that is sufficient to confer standing.” Id. at *9 (citing Murthy v. Missouri, 603 U.S. 43, 57 (2024)). Here, too, Plaintiffs have shown that they are eligible for less federal funding pursuant to the Rate Cap Policy, and that their research programs will face significant disruption as a result.39 See, e.g., Dkt. 13-1 (“AAU Decl.”) ¶¶ 11–13, 15–20. Defendants’ analogy to contract law supports Plaintiffs’ position. At oral argument, Defendants characterized Plaintiffs’ standing argument—“[w]hat plaintiffs are pointing to is a series of discrete contracts [grants] that have existed in the past and they’re saying, as a result of that, they have a legally cognizable interest in prospective grants that haven’t been awarded yet,

and that is contrary to black letter law.” Dkt. 64 (“Tr.”) at 27:2–7. But Plaintiffs have more than just their pastgrants; they also have their current NICRAs.40 In the language of contract, a NICRA is analogous to a “binding preliminary commitment”:

39Plaintiffs’ injuries resulting from the Rate Cap Policy are discussed in more detail in the Court’s analysis of irreparable harm. See infra Section III(B)(2). 40 To the extent that any Plaintiff’s NICRA lapses, now or in the future, that Plaintiff still has its right to negotiate a new NICRA consistent with the regulations. [A] binding preliminary commitment [] does not commit the parties to their ultimate contractual objective but rather to the obligation to negotiate the open issues in good faith in an attempt to reach the alternate objective within the agreed framework. . . . What [a party to a binding preliminary commitment]may demand . . . is that his counterparty negotiate the open terms in good faith toward a final contract incorporating the agreed terms. This obligation does not guarantee that the final contract will be concluded if both parties comport with their obligation, as good faith differences in the negotiation of the open issues may prevent a reaching of final contract. It is also possible that the parties will lose interest as circumstances change and will mutually abandon the negotiation. The obligation does, however, bar a party from renouncing the deal, abandoning the negotiations, or insisting on conditions that do not conform to the preliminary agreement. SeeTchrs. Ins. & Annuity Ass’n of Am. v. Trib. Co., 670 F. Supp. 491, 498 (S.D.N.Y. 1987). Here, the federal government has already negotiated Plaintiffs’ indirect cost rates—in some instances, DOD itself was the negotiating, cognizant agency.41 Now it seeks to violate those commitments by “insisting on conditions that do not conform to the preliminary agreement.” Id. Moving from contract to estoppel, an IHE’s NICRA reflects a concrete investment, undertaken on account of the Government’s own regulations, that risks invalidation under the Rate Cap Policy.42 Even a cursory review of the Uniform Guidance demonstrates that the burden associated with preparing an indirect cost rate proposal and negotiating a NICRA is substantial—so much so that IHEs may elect to forego the whole process and opt instead to charge only a “de minimis” rate. See 2 C.F.R. § 200.414(f); see also generally id. pt. 200, App. III. This is to say nothing of how universities have, in meaningful ways, organized themselves around the Uniform Guidance’s accounting requirementsand its representationthat“[n]egotiated indirect cost rates must be accepted by all Federal agencies,” id. § 200.414(c)(1). See, e.g., Dkt. 13-4 (“ASUDecl.”) ¶ 20 (“ASU has made building and equipment purchases, personnel and operational

41See, e.g., Dkt. 13-18 (“MIT Decl.”) ¶ 15. 42To be clear, the Court does not mean to suggest that this case is settled on principles of contract or estoppel, only that the injuries recognized here have “a close relationship to harms traditionally recognized as providing a basis for lawsuits in American courts.” See TransUnion LLC v. Ramirez, 594 U.S. 413, 425 (2021). plans and decisions, and entered into employment and other contracts, in reliance on [the indirect cost recovery system].”). The Court therefore concludes that, under the circumstances, “[q]ualifying for less federal funding is a ‘future injur[y]’ that is sufficient to confer standing.” NSF, 2025 WL 1725857, at *7 (second alteration in original) (quoting Dep’t of Com. v. New York, 588 U.S. 752, 767 (2019)).43

B. Sovereign Immunity Defendants next contend that this Court lacks jurisdiction to entertain this suit because “the actions at issue are ‘committed to agency discretion by law.’” Opp. at 24 (quoting 5 U.S.C. § 701(a)(1)). According to Defendants, “a court would have no meaningful standard against which to judge the agency’s exercise of discretion.” Id. (quoting Heckler v. Chaney, 470 U.S. 821, 830 (1985)). This argument ignores that DOD’s authority to issue grants is limited by OMB regulations governing the administration of those grants. See 10 U.S.C. § 4001(b)(1) (permitting issuance of grants “in accordance with chapter 63 of title 31,” 31 U.S.C. § 6301 et seq.); see also 31 U.S.C. § 503(a) (granting OMB authority to “establish governmentwide financial management policies

for executive agencies”); 2 C.F.R. pt. 1104 (implementing the Uniform Guidance).44 The Court therefore has “[a] relevant ‘statutory reference point’ . . . other than the decisionmaker’s own views of what is an ‘appropriate’ manner of distribution.” Cf. Milk Train, Inc. v. Veneman, 310 F.3d

43 Defendants challenge only the existence of an injury in fact, the first prong of the standing inquiry. See ITyX Sols. AG v. Kodak Alaris, Inc., 952 F.3d 1, 9 (1st Cir. 2020) (“Standing requires that a plaintiff satisfy ‘three elements: injury in fact, traceability, and redressability.’” (quoting Kerin v. Titeflex Corp., 770 F.3d 978, 981 (1st Cir. 2014))). “Regardless, where the 15% Indirect Cost Rate reduces . . . funding for indirect costs on grants for which . . . Plaintiffs are eligible, and where Plaintiffs seek an order vacating and enjoining implementation of that Policy, [the traceability and redressability] elements are satisfied.” NSF, 2025 WL 1725857, at *9. 44Defendants further agreed at oral argument “that the OMB uniform guidance is binding on the DOD.” Tr. at 33:22–25. 747, 751 (D.C. Cir. 2002) (quoting Drake v. Fed. Aviation Admin., 291 F.3d 59, 72 (D.C. Cir. 2002)); see also DOE, 2025 WL 1414135, at *10 n.2 (“Even assuming that this type of grant funding would be considered an ‘administrative decision traditionally regarded as committed to agency discretion,’ and the Court is not convinced that it is, the exception is typically limited to those ‘rare circumstances’ where ‘no meaningful standard’ exists by which a reviewing judge

could cabin that agency discretion. Here, ‘applicable regulations cabin the Department’s discretion as to when it can’ deviate from negotiated indirect costs, which ‘create meaningful standards by which to judge the agency’s action.’” (citations omitted) (quoting California v. U.S. Dep’t of Educ., 132 F.4th 92, 97 (1st Cir. 2025))). The Court sees no reason to depart from the well-reasoned decisions of other courts in this District characterizing a “decision to stop awarding indirect costs based on the rates . . . negotiated with IHEs” as “not of the type ‘traditionally regarded as committed to agency discretion.’” See, e.g., NSF, 2025 WL 1725857, at *10. The Court therefore concludes that Plaintiffs’ claims are justiciable under the APA. See id. at *11 (observing that the “challenge to the Policy on the

grounds that NSF lacked authority to implement it and that the Policy is arbitrary and capricious presents ‘the sort of routine dispute that federal courts regularly review’” (quoting Weyerhaeuser Co. v. U.S. Fish & Wildlife Serv., 586 U.S. 9, 23 (2018))). III. Preliminary Injunctionand Stay A. Legal Standard Plaintiffs seek a stay of the entire Rate Cap Policy under the APA, 5 U.S.C. § 705,45 or a preliminary injunction barring implementation against Plaintiffs and their university members of the immediately effective portionsof the Rate Cap Policy, i.e., those portions implementing a 15% cap for all awards issued on or after June 12, 2025. The same standard governs both forms of

relief. See, e.g., Massachusetts Fair Hous. Ctr. v. U.S. Dep’t of Hous. & Urb. Dev., 496 F. Supp. 3d 600, 609 (D. Mass. 2020)(applying same standard); Colorado v. Env’t Prot. Agency, 989 F.3d 874, 883 (10th Cir. 2021) (“These four [preliminary injunction] factors also determine when a court should grant a stay of agency action under section 705 of the APA.”). “[T]he issuance of preliminary injunctive relief is ‘an extraordinary and drastic remedy that is never awarded as of right.’” Howe v. U.S. Bank Nat’l Ass’n as Tr. for RMAC Tr. Series 2016-CTT, 440 F. Supp. 3d 99, 102 (D. Mass. 2020) (quoting Peoples Fed. Sav. Bank v. People’s United Bank, 672 F.3d 1, 8–9 (1st Cir. 2012)). “A plaintiff seeking a preliminary injunction must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction

is in the public interest.” Together Emps. v. Mass Gen. Brigham Inc., 32 F.4th 82, 85 (1st Cir. 2022) (quoting Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008)). “The movant’s likelihood of success on the merits weighs most heavily in the preliminary injunction calculus.” Ryan v. U.S. Immigr. & Customs Enf’t, 974 F.3d 9, 18 (1st Cir. 2020). “If the movant ‘cannot

45In relevant part, 5 U.S.C. § 705 provides: On such conditions as may be required and to the extent necessary to prevent irreparable injury, the reviewing court . . . may issue all necessary and appropriate process to postpone the effective date of an agency action or to preserve status or rights pending conclusion of the review proceedings. demonstrate that he is likely to succeed in his quest, the remaining factors become matters of idle curiosity.’” Id. (quoting New Comm Wireless Servs., Inc. v. SprintCom, Inc., 287 F.3d 1, 9(1st Cir. 2002)). “The court may accept as true ‘well-pleaded allegations [in the complaint] and uncontroverted affidavits.’” C&W Facility Servs. Inc. v. Mercado, 2018 WL 4854630, at *2

(D. Mass. Oct. 5, 2018) (quoting Rohm & Haas Elec. Materials, LLC v. Elec. Cirs. Supplies, Inc., 759 F. Supp. 2d 110, 114 n.2 (D. Mass. 2010)). “The Court may also rely on otherwise inadmissible evidence, including hearsay, in deciding a motion for preliminary injunction.” Bos. Taxi Owners Ass’n, Inc. v. City of Bos., 84 F. Supp. 3d 72, 78 (D. Mass. 2015)(citing Asseo v. Pan Am. Grain Co., 805 F.2d 23, 26 (1st Cir. 1986)). B. Analysis 1. Likelihood of Success on the Merits Plaintiffs seek preliminary relief only as to the immediately effective portions of the Rate Cap Policy, which concern new grants, arguing that the Policy violates section 706(2) of the APA.46 As relevant here, the APA provides that courts must “hold unlawful and set aside agency action” that is “in excess of statutory jurisdiction, authority, or limitations, or short of statutory

right”; “arbitrary” or “capricious”; or “otherwise not in accordance with law.” 5 U.S.C. § 706(2). Plaintiffs contend that they are likely to establish that Defendants violated section 706(2) of the APA because: (i) the Rate Cap Policyis a final agency action subject to the APA; (ii)the Rate Cap Policy conflicts with governing regulations; (iii) the Rate Cap Policy conflicts with duly enacted statutes; and (iv) the Rate Cap Policyis arbitrary and capricious. The Court addresses each in turn.

46Plaintiffs argue in the alternative that, should the Court instead grant a stay, the appropriate relief would be to stay the Rate Cap Policy as a whole. Pls.’ Suppl. Br. at 3. However, at this stage, Plaintiffs have argued only that the immediately effective portion of the Policy is unlawful, and so the Court will not analyze the portion of the Policy that applies to existing grants. a. Final Agency Action To be reviewable under the APA, a challenged action must constitute “final agency action.” 5 U.S.C. § 704. Agency action is considered “final” where it is “the ‘consummation’ of the agency’s decisionmaking process,” and an action “by which ‘rights or obligations have been determined,’ or from which ‘legal consequences will flow.’” Harper v. Werfel, 118 F.4th 100,

116 (1st Cir. 2024) (quoting Bennett v. Spear, 520 U.S. 154, 178 (1997)). Defendants do not dispute that the Michael Memo constitutes final agency action and do not dispute that the Court may review the Hegseth Memo as part of its administrative record.47 See Tr. at 32:19–33:14. The Rate Cap Policy is therefore ripe for review. b. Accordance with Regulation Defendants agree that DOD is bound to follow the Uniform Guidance. Tr. at 33:22–25; see also 10 U.S.C. § 4001(b);31 U.S.C. § 503; 2 C.F.R. pt. 1104; F.C.C. v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009) (“An agency may not . . . disregard rules that are still on the books.”). The central question is whether the Rate Cap Policy is a valid exercise of DOD’s authority to deviate from negotiated cost rates under section 200.414(c) of the Uniform Guidance. Courts “construe a regulation in light of the congressional objectives of its underlying

statute.” Montoya v. CRST Expedited, Inc., 88 F.4th 309, 321 (1st Cir. 2023) (quoting United States v. Lachman, 387 F.3d 42, 52 (1st Cir. 2004)). Here, the Uniform Guidance serves “to promote consistent and efficient use of . . . grant agreements[] and cooperative agreements,” 31 U.S.C. § 6307, as part of OMB’s broader mandate to “establish[] financial management policies

47Plaintiffs argue that the Hegseth Memo independently constitutes final agency action. Dkt. 61 (“Reply”) at 8. Defendants disagree. Opp. at 25–26. As discussed below, Section III(B)(1)(d), infra, the Court finds that the Rate Cap Policy is arbitrary and capricious, even taking the additional reasoning provided by the Michael Memo into account. Accordingly, the Court need not resolve the question of whether the Hegseth Memo is independently reviewable. and requirements,” id. § 503. Accordingly, the Court reads the Uniform Guidance primarily through afiscal lens, essentially as an accounting manual,and attempts to understandits provisions “in the broader context of the statutory and regulatory scheme.” Montoya, 88 F.4th at 321. Careful consideration of the Uniform Guidance reveals that the Rate Cap Policy is inconsistent with both its letter and its principle. In short, the Rate Cap Policy uses an improper

tool (indirect cost rates) to accomplish an impermissible goal (implicit cost sharing). In limited circumstances, and subject to certain procedural requirements, the Uniform Guidance permits federal agencies to establish and use alternate methods of accounting for indirect cost rates. See 2 C.F.R. §§ 200.414(c)(1)–(3).48 But the Rate Cap Policy does something fundamentally different—it “force[s] IHEs to fund greater shares of research themselves.” Opp. at 36. In the language of the Uniform Guidance, thatis called cost sharing, and it is an entirely separate part of the regulatory scheme.49 See 2 C.F.R. §§ 200.201, 200.306.50 Federal agencies are indeed permitted to mandatecost sharing in some limited circumstances, but not by hijacking the indirect cost rate determination.

This is Defendants’ fundamental problem: they are trying to shove a square peg in a round hole. Indirect cost rates are products of accounting; accountingsystemsseek to reflect and quantify

48See also supra Section I(A)(3)(iv). 49Tellingly, Defendants’ brief is replete with references to cost sharing. See Opp. at 17, 19, 36 & nn.26, 29, 43, 48. Nevertheless, Defendants labor to suggest that the Rate Cap Policy is something different. See id. at 36 (“Decreasing the indirect cost rate will force IHEs to fund greater shares of research themselves or from other sources—just as they did during earlier periods in which flat rates or cost-sharing were used.”). 50See also supra Section I(A)(3)(v). the world around us. But the Rate Cap Policy makes no claim to objectivity.51 Thus, it can never “justify” its result with reference to the realworld, as accounting systems do, and as contemplated by the regulation. See id. § 200.414(c)(3). An indirect cost rate produced by the Rate Cap Policy fails of its essential purpose, to “provide for payment of reimbursable indirect costs.” 41 U.S.C. §4708. Thus, it is an indirect cost rate in name only.52

At oral argument, Defendants pointed to the 26% administrative limit as an example of how indirect cost rates under the Uniform Guidance are already, in fact, “divorced” from actual cost accounting. Tr. at 24:17–25:22 (referring to 2 C.F.R. pt. 200, App. III(C)(8)(a)). That may, very well, be right. As Defendants correctly noted, the 26% administrative costs figure is based on a 40-year-old statistic that might no longer reflect reality.53 Nevertheless, that would make the figure itself unjustified rather than the whole system unjustifiable. An accounting system is capable of justification where it has logic based on evidence that can be subject to critique,

51 Defendants suggest that “actual” indirect costs are “[un]knowable” and therefore cannot be used as a measure for evaluating different methods for indirect cost reimbursement. Opp. at 30 (“[T]he actual amount of indirect costs that are associated with any specific federal award . . . is . . . fuzzy and leaves ample room for disagreement.”). This is wrong; indirect costs are knowable—OMB has issued lengthy guidance for how to know them. 2 C.F.R. pt. 200, App. III; see also id. § 501 et seq. (providing for post factum audits). Defendants might mean to say that indirect costs are a constructed idea—like most concepts in accounting, they do not exist in the abstract and therefore are susceptible to interpretation and reimagination. But by the same logic, a person’s “income” is unknowable because reasonable people argue about how to apply the tax code. 52At oral argument, Plaintiffs briefly referenced several examples of past exceptions. Tr. at 39:14–40:6. The Court has been unable to independently verify if these instances occurred. Nevertheless, they serve as useful illustrations: For example, you might have a set of awards that are about, like, commercializing existing discoveries. So you say, well, it doesn’t make sense to sort of defray the overall research infrastructure of the university, or you might have a class of awards that requires heavy use of the federal government’s own facilities. You say, well, the federal government is sort of putting in some skin through those facilities so we’re not going to give you the full negotiated rates. Id. In both examples, something exceptional about the project—its commercial nature, its use of government facilities—makes the previously negotiated rate (and the accounting model that produced it) irrespective of reality. A deviation corrects accordingly, bringing the result back in line with reality. 53See supra notes 20–27 and accompanying text. regardless of whether any such critiques are valid. The administrative limit is based on evidence, albeit now-questionable evidence. By contrast, the Rate Cap Policy is pure fiat, with nothing to “justify.”54 Cf. 2 C.F.R. § 200.414(c)(3). Unmoored from the general principles of the Uniform Guidance, it is unsurprising that the Rate Cap Policy likewise fails to satisfy the regulations’ technical requirements. Specifically,

deviating agencies are required to set forth the “policies, procedures and general decision-making criteria that their programs will follow to seek and justify deviations from negotiated rates.” Id. Defendants do not even attempt to argue that the Hegseth or Michael Memos set forth “procedures” or “decision-making criteria.” Opp. at 28–29 (“[T]he Policy need not set forth further ‘procedures and decision-making criteria.’”).55 Indeed, the only “procedure” either Memo sets forth is an instruction to follow the procedure that already exists—“[DOD] Components should consider the criteria for identifying and computing indirect cost rates as described in [the Uniform Guidance.]” Michael Memo at 3. Notwithstanding, in the very next sentence, the Michael Memo directs anon sequitur result—“However, [DOD] Components must not exceed the 15% indirect cost rate cap

on any award to IHEs.” Id. Likewise, “decision-making criteria” implies reasoned judgment, but the Rate Cap Policy squarely forecloses it. These failings independently and clearly put

54 The Court discusses Defendants’ lack of justification for the Rate Cap Policy in more detail, infra Section III(B)(1)(d). 55In a purely technical sense, the Memos can be said to set forth “policies.” However, that is true only insofar as one ignores the broader context, i.e., how indirect cost rates function in the overall scheme and how other sections of the regulation foreclose the kind of cost-sharing policy put forth here. Defendants out of step with the regulations, but they are also emblematic of a more fundamental disconnect between the Rate Cap Policy and the Uniform Guidance.56 It is therefore likewise unsurprising that Defendants’ reading of the regulation is in such tension with its broader text. The regulation states that an institution’s NICRA “must be accepted by all Federal agencies.” 2 C.F.R. § 200.414(c)(1). Defendants argue that the exception to this

rule requires only notice, see Opp. at 28, but that reading converts strong, mandatory language (“must be accepted”) into a mere presumption.57 Defendants further argue that subsection (f) is “irrelevant” because it merely “empowers grantees to opt out of negotiated rates,” id. at 29 (emphasis in original), but that makes little sense—under subsection (f), opt out is possible only for recipients “that do not have a current [NICRA],” 2 C.F.R. § 200.414(f).58 For recipients that

56Final agency guidance strongly supports a reading of the regulation that requires deviating agencies to put forth a meaningful accounting system with comprehensive justification and procedure: Language in paragraph (c) provides for the consistent application of negotiated indirect cost rates, and articulates the conditions under which a Federal awarding agency may use a different rate. These conditions include approval of the Federal awarding agency head (as delegated per standard delegations of authority) based on documented justification, the public availability of established policies for determinations to use other than negotiated rates, the inclusion of notice of such a decision in the announcement of funding opportunity, as well as in any pre-announcement outreach, and notification to OMB of the decision. Comments received regarding these proposals were mostly positive, and indicated that these provisions would likely lead to greater consistency, and transparency in the application of indirect cost rates governmentwide. Some commenters recommended that for even greater consistency decisions about the use of rates be subject to OMB approval rather than Federal agency approval. The [Council on Financial Assistance Reform, or “COFAR”] considered this, but ultimately recommends that responsibility for administering Federal financial assistance programs continue to rest with the Federal awarding agencies, and that the conditions set by OMB for these determinations are stringent enough to ensure that they do not occur without strong justification. The COFAR did not recommend the change. Uniform Administrative Requirements, Cost Principles, and Audit Requirements for Federal Awards, 78 Fed. Reg. 78,590, 2013 WL 6795572 (Dec. 26, 2013) (emphases added). 57Compare this to how the Uniform Guidance manages cost sharing. See 2 C.F.R. § 200.306(a); id. pt. 200, App. I(b)(2)(ii) (requiring only notice). 58 Recipients may charge less than allowed to satisfy any cost-sharing requirement, 2 C.F.R. § 200.306(c), but this does not change the recipient’s indirect cost rate. Rather, indirect costs are calculated as normal but then only partially reimbursed. See id. dohave a NICRA, subsection (f)’s relevance is that it clearly establishes that grant recipients “are not required to use the de minimis rate.” Id. That text would mean nothing if agencies were technically disallowed from imposing a de minimis rate but nevertheless free to unilaterally impose an indirect cost rate that is its literal equivalent.59 To be clear, it is not that the Government can never engage in “cost savings,” Michael

Memo at 2, only that it cannot do so through unauthorized sleight of hand.60 Section 200.414 provides no clever workaround for agencies to disallow already allowed costs or to impose de facto cost sharing by other means. See id. § 200.414(f) (“Recipients and subrecipients are not required to use the de minimis rate.”). c. Accordance with Statute “An agency . . . ‘literally has no power to act’—including under its regulations—unless and until Congress authorizes it to do so by statute.” Fed. Election Comm’n v. Cruz, 596 U.S. 289, 301 (2022) (quoting Louisiana Pub. Serv. Comm’n v. FCC, 476 U.S. 355, 374 (1986)). As discussed, Congress has not authorized DOD to issue grants inconsistent with OMB’s guidance. See 10 U.S.C. § 4001(b)(1); 31 U.S.C. §§ 503, 6307.61 Moreover, the Court has concluded that

the Rate Cap Policy is inconsistent with the Uniform Guidance.62 DOD therefore acts ultra vires in disregarding those regulations.

59 The Court also finds it telling that it is called a “de minimis” rate, 2 C.F.R. § 200.414(f), implicitly acknowledging that the rate is so insignificant as to “merit disregard.” See De minimis, Merriam-Webster, https://www.merriam-webster.com/dictionary/de%20minimis [https://perma.cc/6ZBH-QMUN] (last visited July 18, 2025) (defining “de minimis” as “lacking significance or importance” and “so minor as to merit disregard”). Indeed, a de minimis rate is minimal enough not to require adherence to normal accounting requirements. 2 C.F.R. § 200.414(f) (“The de minimis rate does not require documentation to justify its use and may be used indefinitely.”). 60 The Government might change which costs are allowable, thereby disclaiming certain expenses—for instance, it could decide to disallow reimbursement for pens. Of course, that would require changing the regulations. See 2 C.F.R. pt. 200, App. III(B)(3)(b). 61See also supra Section II(B). 62See supra Section III(B)(1)(b). Plaintiffs press further to argue that the Rate Cap Policy is inconsistent with statutes directly addressing the use of indirect cost rates. Plaintiffs point to 41 U.S.C. § 4708, whereby Congress first authorized the use of predetermined indirect cost rates to “provide for payment of reimbursable indirect costs” without imposing any limit. Dkt. 39 (“Mem.”) at 44–45; see also NSF, 2025 WL 1725857, at *12 (holding that a similar rate cap policy “is not authorized by the

text of Section 4708”). More generally, Plaintiffs review the history (or lack thereof) of comparable indirect cost limitations and invoke the major questions doctrine to suggest that the Court should “look with skepticism” on the Government’s “newly uncovered” claim of authority based on “broadly worded statutes.” Mem. at 44, 46–47 (citing, inter alia, West Virginia v. Env’t Prot. Agency, 597 U.S. 697, 721 (2022)).63 Because the Court does not agree that DOD’s authorization is so broadly worded, it need not decide whether any similar rate cap policy could theoretically be squared with Congress’s other statutes. Accordingly, the Court declines to weigh in on this issue. d. Arbitrary and Capricious Agency action “may be set aside if found to be ‘arbitrary, capricious, an abuse of discretion,

or otherwise not in accordance with law.’” NIH I, 770 F. Supp. 3d at 303–04 (quoting Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 41 (1983)). “An agency action qualifies as ‘arbitrary’ or ‘capricious’ if it is not ‘reasonable and reasonably

63The Court notes that the posture of Plaintiffs’ major questions argument is somewhat askew from the norm. Here, the regulations invoked by DOD are those of another agency. Cf. West Virginia, 597 U.S. 697 (reviewing Environmental Protection Agency rule in suit against Environmental Protection Agency). The relevant question might therefore be whether OMB could authorize this or a similar rate cap policy. On that question, the history is mixed. OMB has twice imposed fixed limits impacting indirect cost rates: a 3.6% limit on certain administrative personnel costs and a subsequent, 26% limit on administrative costs generally. See supra notes 20–27 and accompanying text. These examples, however, are not necessarily on point: each of those changes purported to reflect actual costs and was supported by specific data, making them cognizable as accounting principles within OMB’s mandate, rather than as the kind of fiat, budgetary policy at issue here and that Congress has arguably seen fit to address heretofore directly through legislation. See id. explained.’” Am.Pub.Health Assoc. v.Nat’l Insts.of Health(“NIH II”), —F. Supp. 3d —,2025 WL 1822487, at *15 (D. Mass. July 2, 2025) (quoting Ohio v. Env’t Prot. Agency, 603 U.S. 279, 292 (2024) (third-level quotation marks omitted)). “The task of a court reviewing agency action under the APA’s arbitrary and capricious standard is to determine whether the agency has examined the pertinent evidence, considered the relevant factors, and articulate[d] a satisfactory

explanation for its action including a rational connection between the facts found and the choice made.” DOE, 2025 WL 1414135, at *11 (alteration in original) (internal quotation marks omitted) (quoting Penobscot Air Servs., Ltd. v. FAA, 164 F.3d 713, 719 (1st Cir. 1999)). “With that said, ‘[t]he scope of review under the “arbitrary and capricious” standard is narrow and a court is not to substitute its judgment for that of the agency.’”64 NIHI, 770 F.Supp. 3d at 304 (alteration in original) (quoting State Farm, 463 U.S. at 43). “While this is a highly deferential standard of review, it is not a rubber stamp.” Penobscot, 164 F.3d at 720 (quoting Dubois v. U.S. Dep’t. of Agric., 102 F.3d 1273, 1285 (1st Cir. 1996)). “[T]he court must undertake ‘a thorough, probing, in-depth review’ and a ‘searching and careful’ inquiry into the record.” Id.

(quoting Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 415–16 (1971)). Plaintiffs argue that the Rate Cap Policy is arbitrary and capricious for two reasons: because DOD’s justifications lack the “reasoned basis” required and because DOD failed to consider multiple “important aspects of the problem.” Mem. at 38–44. Defendants contend that there is a reasoned basis for the Rate Cap Policy: it is to cut costs from DOD research, Opp. at 33–34, or as the Hegseth Memo explains, to “cut[] bureaucratic fat so that we can build muscle,”

64 Defendants argue that “a particularly lenient standard of review” applies “to agency action in the contracting context,” citing a case from the Court of Federal Claims. Opp. at 33 (citing Palantir USG, Inc. v. United States, 129 Fed. Cl. 218, 260 (Fed. Cl. 2016), aff’d, 904 F.3d 980 (Fed. Cir. 2018)). However, as another session of this court explained, “that case involved an agency’s exercise of discretion in soliciting and selecting bids during a particular procurement process” and “thus has no relevance here, where Plaintiffs challenge an agency’s decision to implement a new policy that applies across its entire grant-making program.” NSF, 2025 WL 1725857, at *16 n.7. Hegseth Memo at 3. Defendants further argue that Plaintiffs merely seek to “second guess” the agency’s policy choice. Opp. at 36. Under the APA, agencies must provide a reasoned basis for their actions, which requires at least a “rational connection between the facts found and the choice made.” State Farm, 463 U.S. at 43, 48–49 (citation omitted). Agencies must also consider “important aspect[s] of the

problem.” Dep’t of Homeland Sec. v. Regents of the Univ. of Cal., 591 U.S. 1, 30 (2020)(quoting State Farm, 463 U.S. at 43). In the context of a policy change, an agency must both “display awareness that it is changing position” and “‘show that there are good reasons for the new policy,’ although it need not demonstrate that ‘the reasons for the new policy are better than the reasons for the old one.’” DOE, 2025 WL 1414135, at *11 (quoting Fox Television Stations, 556 U.S. at 515). Further, “if an agency’s ‘new policy rests upon factual findings that contradict those which underlay its prior policy; or when its prior policy has engendered serious reliance interests’ an agency’s failure to consider such factors ‘would be arbitrary or capricious.’” NIH I, 770 F. Supp. 3d at 307 (quoting Fox Television Stations, 556 U.S. at 515). Thus, “when an agency rescinds a

prior [position,] its reasoned analysis must consider the ‘alternative[s]’ that are ‘within the ambit of the existing [position].’” Regents, 591 U.S. at 30 (quoting State Farm, 463 U.S. at 51). In determining whether the Rate Cap Policy is arbitrary and capricious, the Court “may consider only the agency’s explanation given at the time the relevant decision was made, as opposed to [a] post hoc rationale.” Melone v. Coit, 100 F.4th 21, 31 (1st Cir. 2024) (citing SEC v. Chenery Corp., 318 U.S. 80, 87–88 (1943)); see also Regents, 591 U.S. at 22–23 (explaining the importance of considering only contemporaneous explanations, including because it “instills confidence that the reasons given are not simply ‘convenient litigating position[s]’” and promotes “the orderly functioning of the process of review” (alteration in original) (citations omitted)). Here, the record is limited to the Hegseth Memo and Michael Memo. The Hegseth Memo explains that the objective of “pursu[ing] a lower cap on indirect cost rates” is “not only to save money, but to repurpose those funds — toward applied innovation, operational capability, and strategic deterrence.” Hegseth Memo at 2. The Michael Memo

provides as its rationale: [T]he Secretary of Defense has determined that significant cost savings would result from such deviations. In making the decision to limit IHE indirect cost rates to 15%, the Department compared the indirect costs on existing financial assistance awards to IHEs to the indirect cost rates allowed by several non-profit foundations that provide financial assistance to IHEs and determined that such a move would lead to cost savings.

Michael Memoat 2. It also explains that “[t]he academic community’s work with [DOD] remains essential for the development of transformative research concepts that drive our future capabilities.” Id. at 3. Like each court to consider similar attempts to cap indirect costs on the same or similar bases, this Court cannot hold that this is a “reasoned basis” for the change, let alone any real explanation whatsoever. The idea that “decreasing indirect costs will result in increased direct costs and more effective research outcomes,” Opp. at 34, fails under the APA because it is merely conclusory. See NIH I, 770 F. Supp. 3d at 305 (“[Defendants] claim[] that more funds will go to direct research but fail[] to address how the money will actually be directed to cover direct costs and how that research will be conducted absent the necessary indirect cost reimbursements provided by the federal government. This is particularly true considering the number of universities and associations that have made clear that research will have to be cut, as other funding sources will not be able to make up the shortfall.”); DOE, 2025 WL 1414135, at *12 (concluding that DOE’s basis for attempting to cap indirect costs at 15% was conclusory and noting that “even if the DOE had stated an intent to reinvest indirect costs in direct research funding, that explanation would not be particularly well-reasoned, given the amount of support in the record for the notion that Plaintiffs would not be able to accept further direct research funding (or, indeed, even utilize their current direct research funding) absent an indirect cost rate that continues to approximate actual expenditures”); NSF, 2025 WL 1725857, at *17–18 (describing similar policy justifications

as “goals andconclusions, not reasoning” and concluding that “the court cannot discern how NSF determined that the 15% Indirect Cost Rate will make NSF grant funding more effective”). Defendants themselves concede that this supposed justification represents “strategic goals.” Opp. at 34. But goals cannot substitute for the reasoned basis required under the APA. See DOE, 2025 WL 1414135, at *12 (“Statements of aspirational goals, however, are not the same as reasoned explanations for why an action is chosen or how the chosen action will effectuate the stated goals.”); accord NSF, 2025 WL 1725857, at *17 (“These statements express goals and conclusions, not reasoning.”). Defendants also fail to explain in the Policy why universities are singled out under the Rate

Cap Policy. While there very well could be valid policy reasons to treat universities differently than other grant recipients, the Rate Cap Policy makes no attempt to provide even a single reason to do so. See generally Hegseth Memo, Michael Memo. This failure to articulate any reason, let alone a “satisfactory explanation for [the] action,” renders the Rate Cap Policy unlawful. Penobscot, 164 F.3d at 719; see also Encino Motorcars, LLC v. Navarro, 579 U.S. 211, 224 (2016) (“Whatever potential reasons the Department might have given, the agency in fact gave almost no reasons at all. In light of the serious reliance interests at stake, the Department’s conclusory statements do not suffice to explain its decision.”).65 Moreover, the underlying idea for the Policy—that indirect costs are “waste” and “bureaucratic fat,” Hegseth Memo at 2, that are less worthy of funding than direct costs—is at least conceptually irrational and ignores the realities of research, as demonstrated by the record

evidence. The record is clear that indirect costs support critical resources and infrastructure, without which the research cannot proceed.66 See, e.g., BU Decl. ¶ 5 (explaining that the Rate Cap Policy would “end or seriously jeopardize” BU’s research for DOD); Dkt. 13-6 (“Brown Decl.”) ¶ 9 (“The newly announced DOD 15% rate cap will make many, if not all, of Brown’s proposed research projects infeasible relative to the entire scope of each project at the budget requested.”); Dkt. 13-7 (“Caltech Decl.”) ¶ 6 (“The 15% rate cap will make many, if not all, of Caltech’s proposed research projects infeasible.”); Dkt. 13-8 (“UC Decl.”) ¶ 7 (“DOD’s proposal to cut F&A cost rates to 15% for new awards would end or seriously jeopardize research projects.”);

65At the hearing, Defendants argued that there was no need for additional explanation, given that universities are singled out for different treatment in other portions of the regulations. Tr. at 38:9–13 (“[THE COURT:] So if the court could conclude that a class is created in an arbitrary and capricious way absent some justification, what is the justification for creating the class of IHEs? [DEFENDANTS’ COUNSEL]: Well, here, your Honor, again, the IHEs are treated as a class in the regulations themselves.”). First, this fails to provide “a ‘rational connection between the facts found and the choice made,’” as the fact that in some situations universities are treated differently than other grant recipients does not explain why that distinction makes sense with regard to indirect cost reimbursement. See Penobscot, 164 F.3d at 719 (quoting Motor Vehicle, 463 U.S. at 43). Indeed, many of the relevant regulations single out not only IHEs but non-profit organizations generally. See, e.g., 2 C.F.R. § 200.414. Second, this explanation is not included in the Rate Cap Policy, nor do Defendants point to any contemporaneous explanation with this reasoning, and thus it is not an appropriate basis on which Defendants may rely to defend the Policy. See Melone, 100 F.4th at 31 (explaining that the Court “may consider only the agency’s explanation given at the time the relevant decision was made”). 66At oral argument, Defendants pointed to libraries as an example of a perhaps-less-justifiable indirect cost. Tr. at 55:15–25. But libraries are an excellent example. Their budgets cover journal subscriptions that give grant recipients access to essential primary research. While any researcher would consider these subscriptions vital, their costs cannot easily be assigned to a specific project. As a reminder, any indirect cost charged back to a federal research grant must have already been deemed allowable. 2 C.F.R. § 200.402. Federal grants do not reimburse generally for “libraries”—rather, library costs are broken up and scrutinized individually, just as every other research cost, direct or indirect. Dkt.13-17 (“UMCP Decl.”) ¶ 9 (same); Dkt. 13-19(“OSU Decl.”) ¶ 8(same); see also NIHI, 770 F. Supp. 3d at 306–07 (“In cutting indirect costs without identifying a countervailing funding stream for such costs of research, the only reasonable outcome will be the discontinuing of research supported by the slashed F&A rates.”). Congress itself has called attention to these consequences, observing that a rate cap like the one here would “radically change the nature of the

Federal Government’s relationship with the research community” and “throw[] research programs across the country into disarray.” S. Rep. 115-150, at 109; accord H.R. Rep. 115-244, at 50. Embarrassingly, it is not obvious that Defendants are even fully aware of what constitutes an indirect cost—Defendants cite “‘tenured faculty salaries’” as an example of “expenses that can and should be borne by universities.” Opp.at 28 (quotingDkt. 13-10 (“CSU Decl.”)¶ 20).67 But faculty (i.e., researcher) salaries are generally treated as direct costs because a researcher’s efforts can be allocated directly to their specific project. See 2 C.F.R. § 200.413(b).68 Indirect costs reflect merely an accounting method in recognition of the fact that some allowable costs cannot easily be attributed to a particular project and that there are benefits to resources that can support

multiple projects, rather than requiring institutions to separately purchase the resources for each project or forgo them entirely. See, e.g., Dkt. 13-9 (“CU Boulder Decl.”) ¶¶ 9–10 (detailing 67 The referenced declaration mentions “tenured faculty salaries” as an example of one of the university’s “long-term obligations” for which it must budget prospectively. CSU Decl. ¶ 20. The previous sentence, however, made clear that the declarant was talking about both “both direct and indirect” costs. Id. 68 Faculty salaries, like all costs under the Uniform Guidance, are limited by what the Government deems reasonable under the circumstances. 2 C.F.R. § 200.430(b). Moreover, “[c]harges to Federal awards for [faculty] salaries and wages must be based on records that accurately reflect the work performed.” Id. § 200.430(g)(1). The Uniform Guidance sets out particular requirements for how those records must be kept. Id. §§ 200.430(g), (i). To comply with those requirements, IHEs develop comprehensive internal policies and systems for faculty members and other employees to track what percent of their overall efforts go to specific, directly funded objectives (i.e., research projects), indirectly funded objectives (i.e., grant administration), and non-funded objectives (e.g., teaching). See, e.g., Effort Reporting Policy, BROWN UNIVERSITY, (June 2020), https://division-research.brown.edu/research- cycle/manage-award/manage-expenses/effort-reporting-guidance/effort-reporting-policy [https://perma.cc/PA9Y- WV3F] (“Individual effort reports are required for all faculty, graduate students, administrative and professional employees, undergraduate students, nonexempt and limited-duration employees with effort allocated to sponsored projects.”). resources and facilities shared among various DOD-funded research grants); Dkt. 13-18 (“MIT Decl.”) ¶ 23 (same). Critically, for an indirect cost to be reimbursable, it must still be allowable. 200 C.F.R. § 200.402. Defendants also cite to “indirect cost rates allowed by several non-profit foundations that provide financial assistance to IHEs.” Michael Memo at 2. Yet Defendants make no effort to

distinguish a recent decision from this District that rejected this same comparison when made without further explanation as to how that comparison makes sense and supports capping indirect costs. See NIH I, 770 F. Supp. 3d at 305 (rejecting same comparison as arbitrary and capricious where the policy failed to explain the rationale and failed to address the significant differences between DOD grant funding and private foundation funding approaches or the reliance interests in federal grant funding). While Defendants critique Plaintiffs’ argument as a “mere disagreement with DoD’s decision,” Opp. at 35, even in their brief, Defendants provide no explanation or reasoning for why or how comparing IHEs’ grants from DOD to those from private foundations supports the Rate Cap Policy, see NIH I, 770 F. Supp. 3d at 305 (finding that such “conclusory

statements hardly rise to the level of ‘reasoned decisionmaking’ required by the APA”). Defendants’ position also ignores key differences between funding from private foundations and the Government, including that many foundations define and calculate indirect costs differently from the federal government; DOD imposes various requirements that increase indirect costs that are not imposed by private foundations; and universities accept lower indirect cost rates from foundations in reliance on the grants they receive from the Government.69 See NIHI, 770 F. Supp. 3d at 305 (noting that “private organizations calculate direct and indirect costs differently than the

69 See Jocelyn Kaiser, NIH Plan to Reduce Overhead Payments Draws Fire, SCIENCE (June 2, 2017), https://www.science.org/content/article/nih-plan-reduce-overhead-payments-draws-fire [https://perma.cc/F4JG- ZRFN]. federal government (including many indirect costs in their direct costs calculation) and that private foundations are more likely to fund different types of research with lower F&A costs than the government”). Defendants “fail[] to provide an intelligible explanation,” which “amounts to a failure to engage in reasoned decisionmaking.” Constellation Mystic Power, LLC v. Fed. Energy Regul. Comm’n, 45 F.4th 1028, 1057 (D.C. Cir. 2022) (quoting FPL Energy Marcus Hook, L.P.

v. Fed. Energy Regul. Comm’n, 430 F.3d 441, 448 (D.C. Cir. 2005)). Defendants also wholly fail to address multiple judicial decisions that recognized valid reliance interests in the negotiated rates for indirect costs. While Defendants contend that Plaintiffs have no reliance interest in future grant awards, “[n]on-governmental organization[s] have relied on the Federal provision of indirect costs for decades.” NIH I, 770 F. Supp. 3d at 308; see also id. at 309–10 (explaining reliance interests at stake even in future grants and noting that “[a]t the very least, these reliance interests exist for the length of the NICRA, which institutions negotiated with the cognizant agency for the purpose of consistency until the conclusion of the agreement”). Plaintiffs and declarants have detailed how they rely on the negotiated rates to structure budgets

and planning—including what buildings to construct and which and how many employees to hire—and how they have operated this way for decades. See, e.g., Dkt. 13-21 (“Pitt Decl.”) ¶ 17 (explaining how Pitt relies on its negotiated rate to estimate funding as part of its multi-year budgeting process); OSU Decl. ¶ 20 (same); Dkt. 13-22 (“USC Decl.”) ¶ 20 (same); Dkt. 13-23 (“UToledo Decl.”) ¶ 15 (same); Dkt. 13-24 (“UW Decl.”) ¶ 19 (same). Further, Plaintiffs and declarants have already submitted hundreds of millions of dollars’ worth of grant proposals to DOD based on the reasonable expectation that, if selected, they will be able to seek reimbursement according to their negotiated rates.70 Defendants have not accounted for the burden and expense universities have undertaken in applying for grants that Defendants now seek to issue, under the Rate Cap Policy, on materially different terms than advertised. See NSF, 2025 WL 1725857, at *8 (recognizing reliance interests stemming from use of negotiated indirect cost rates in pending grant proposals). The Michael Memo itself recognizes that DOD will continue to award grants to

universities. See Michael Memo at 3 (“The academic community’s work with [DOD] remains essential for the development of transformative research concepts that drive our future capabilities.”).71 It is “arbitrary or capricious to ignore such matters.” Fox Television Stations, 556 U.S. at 515. Finally, Defendants argue that the Court may not “second guess” the agency’s policy choice, but a policy choice still must be supported by a sufficient explanation for the change to survive under the APA. Seeid. at 513, 515 (explaining that an agency “need not demonstrate . . . that the reasons for the new policy are betterthan the reasons for the old one,” but an agency must “examine the relevant data and articulate a satisfactory explanation for its action” (emphasis in

original) (quoting State Farm, 463 U.S. at 43)). Defendants contend that DOD determined that

70See, e.g., ASU Decl. ¶ 9 (123 pending proposals worth approximately $275 million); Brown Decl. ¶ 6 (37 pending proposals worth over $22 million); Caltech Decl. ¶ 5 (18 pending proposals worth $29 million); CU Boulder Decl. ¶ 5 (48 pending proposals worth $58 million); CSU Decl. ¶ 5 (73 pending proposals worth over $69 million); Dkt. 13-12 (“Duke Decl.”) ¶ 6 (91 pending proposals worth $92.9 million); Dkt. 13-13 (“UIUC Decl.”) ¶ 6 (161 pending proposals worth over $192,500); Dkt. 13-15 (“KUL Decl.”) ¶ 5 (27 pending proposals worth over $27 million); Dkt. 13-16 (“KUMC Decl.”) ¶ 5 (42 pending proposals worth $47.5 million); UMCP Decl. ¶ 6 (117 pending proposals worth over $170 million); MIT Decl. ¶ 11 (hundreds of DOD proposals submitted every year, with many currently under review); Dkt. 13-20 (“Penn Decl.”) ¶ 5 (77 pending proposals worth $101.2 million); Pitt Decl. ¶ 5 (102 pending proposals worth $92 million); USC Decl. ¶ 5 (270 pending proposals worth over $384 million); Dkt. 13-25 (“UW-Madison Decl.”) ¶ 5 (81 pending proposals worth over $70 million). 71 Despite this statement, the Michael Memo fails to acknowledge, let alone account for, how decreased funding will impact universities’ ability to conduct this research. This failure to explain the rationale behind the Rate Cap Policy provides further support for the conclusion that the Policy is arbitrary and capricious. See Encino Motorcars, 579 U.S. at 224 (“Whatever potential reasons the Department might have given, the agency in fact gave almost no reasons at all. In light of the serious reliance interests at stake, the Department’s conclusory statements do not suffice to explain its decision.”). the Rate Cap Policy “will foster the ‘development of transformative research concepts that drive [the Department’s] future capabilities’ and ‘strengthen[] the United States military,’” Opp. at 37 (citing Michael Memo at 3; Hegseth Memo at 2), and that “Plaintiffs are not in a better position than DoD to gauge DoD’s research objectives and Plaintiffs’ argument is an improper attempt to substitute their judgment for that of the agency,” id. But the Rate CapPolicy neither outlines these

supposed research objectives nor explains how an indirect cost rates limit would help accomplish those goals. DOD’s unexplained preference for funding direct costs rather than indirect costs does not on its own establish the necessary foundation to justify the Rate Cap Policy, especially where the Policy so dramatically upends the prior status quo of IHE research, and where there is abundant evidence of the impossibility of DOD’s plan. See DOE, 2025 WL 1414135, at *13 (“The Rate Cap Policy wholly disregards the prior mutually understood reality that these indirect cost payments, while perhaps not sufficient for direct research to occur, are necessary to that research.”). Importantly, in reviewing the Rate Cap Policy, the Court does not review any DOD

decision to fund one type of research over another, or to issue grants to one institution versus another. The only policy preference identified contemporaneously with the Rate Cap Policy was to save money and to promote research by divertingfunds to direct, rather than indirect,costs. See Hegseth Memo at 2 (explaining that the objective is “not only to save money, but to repurpose those funds — toward applied innovation, operational capability, and strategic deterrence”); Michael Memo at 2 (explaining that the Rate Cap Policy will “lead to cost savings”). But the implied relationship between “applied innovation, operational capability, and strategic deterrence” and direct costs is far from self-evident. For example, DOD recognizes cybersecurity as a major threat to national security.72 Relevant computer science research will naturally utilize expensive equipment and electricity, and it will almost certainly take place in buildings. Those are all likely to be charged as indirect costs.73 The Rate Cap Policy assumes a one-to-one relationship between direct costs and actual research that is just fundamentally wrong. In the absence of any contrary explanation, the Court cannot conclude that the Policy has a rational basis.

DOD had a duty under the APA to “examine[] the pertinent evidence, consider[] the relevant factors, and articulate[] a satisfactory explanation for its action including a rational connection between the facts found and the choice made.” DOE, 2025 WL 1414135, at *11 (quoting Penobscot, 164 F.3d at 719) (internal quotation marks omitted). Plaintiffs are likely to succeed in establishing that DOD failed to meet this duty in promulgating the Rate Cap Policy, and thus the Rate Cap Policy is arbitrary and capricious. 2. Irreparable Harm Plaintiffs contend that Defendants, in adopting and enforcing the Rate Cap Policy, have irreparably harmed Plaintiffs’ institutions, and will continue to do so absent immediate injunctive relief. Mem. at 48–54. For their part, Defendants contend that any alleged harm is merely

speculative, self-inflicted, or redressable after a final judgment through an award of monetary damages. Opp. at 44–50. “‘Irreparable injury’ in thepreliminary injunction context means an injury that cannot adequately be compensated for either by a later-issued permanent injunction, after a full

72 See Army Maj. Wes Shinego, DOD Leaders Urge Congress to Bolster Cyberdefenses, DOD (May 16, 2025), https://www.defense.gov/News/News-Stories/Article/Article/4188996/dod-leaders-urge-congress- to-bolster-cyberdefenses/ [https://perma.cc/BGU8-5E4Z]. 73 See 2 C.F.R. § 200.439 (explaining that the use of buildings and equipment is reimbursed through depreciation, consistent with indirect cost principles); id. § 200.413(b) (listing utilities among costs that would ordinarily be treated as indirect costs except under special circumstances). adjudication on the merits, or by a later-issued damages remedy.” Rio Grande Cmty. Health Ctr., Inc. v. Rullan, 397 F.3d 56, 76 (1st Cir. 2005). “A finding of irreparable harm must be grounded on something more than conjecture, surmise, or a party’s unsubstantiated fears of what the future may have in store.” Baptiste v. Kennealy, 490 F. Supp. 3d 353, 381 (D. Mass. 2020) (quoting Charlesbank Equity Fund II v. Blinds To Go, Inc., 370 F.3d 151, 162 (1st Cir. 2004)). “District

courts have broad discretion to evaluate the irreparability of alleged harm and to make determinations regarding the propriety of injunctive relief.” K-Mart Corp. v. Oriental Plaza, Inc., 875 F.2d 907, 915 (1st Cir. 1989) (quoting Wagner v. Taylor, 836 F.2d 566, 575–76 (D.C. Cir. 1987)). This Court is persuaded that Plaintiffs have demonstrated that they would face irreparable harm should the requested injunction not issue. Plaintiffs and declarants have credibly asserted in sworn declarations that the 15% indirect cost rate is financially and operationally unsustainable for institutionsin the immediate term.74 See, e.g., Caltech Decl. ¶ 6; Dkt. 13-11 (“Cornell Decl.”) ¶ 5; Dkt. 13-12 (“Duke Decl.”) ¶ 7; UC Decl. ¶ 7; UMCP Decl. ¶ 7; Dkt. 13-20 (“Penn Decl.”)

74 Defendants also argue that “each Plaintiff—and each institution represented by an [Organizational] Plaintiff—must, on its own, make a clear showing of irreparable harm.” Opp. at 44 (citing Adams v. Freedom Forge Corp., 204 F.3d 475, 485–86 (3d Cir. 2000)). As to each University Plaintiff, the Court expressly so finds. As to the Organizational Plaintiffs and their members, courts in this Circuit have “no difficulty” finding a likelihood of irreparable injury based on “illustrative” examples combined with facts showing that the risk of irreparable harm faced by some members is representative of the risk faced by all members. See Orr v. Trump, — F. Supp. 3d. —, 2025 WL 1695941, at *12 (D. Mass. June 17, 2025) (quoting United Steelworkers of Am., AFL–CIO v. Textron, Inc., 836 F.2d 6, 8 (1st Cir. 1987) (affirming preliminary injunction applicable to non-party union members)). “[W]hen a court infers a risk of harm to all individuals although only a few testify, it is reasoning inductively. . . . [This] is the daily work of fact-finders.” Adams, 204 F.3d at 487. All IHEs, for decades, have had to follow the Uniform Guidance, its cost principles, and its accounting requirements. See notes 10–18, supra, and accompanying text. It follows that all Organizational Plaintiffs’ member institutions with DOD funding would experience the same kind of upheaval and disruption as a result of that system’s being suddenly and drastically replaced by one that explicitly seeks to shift the cost of DOD research to those institutions. In considering and rejecting a very similar rate cap limit, Congress said that the harm would be felt “nationwide,” S. Rep. No. 115-150, at 109, and “across the country.” H.R. Rep. No. 115-244, at 50. Accordingly, the Court finds that there is more than “an adequate foundation from which one could draw [the] inference[],” Adams, 204 F.3d at 487, that all Organizational Plaintiffs’ member institutions subject to the Rate Cap Policy are likely to suffer the irreparable harm detailed by those that have testified. See also, e.g., Dkt. 13-2 (“ACE Decl.”) ¶ 15 (“The declarations submitted by ACE members in this litigation will vividly illustrate the types of harms felt by all ACE members with DOD funding.”). ¶6; USC Decl.¶ 6. In support of that position, Plaintiffs detail the disruption of ongoing research, loss of essential and specialized staff, and disruption to hiring, budgeting, and planning that will occur imminently as a result of the Rate Cap Policy, all of which constitutes irreparable harm.75 E.g., CSU Decl. ¶ 18 (“[A] reduction in the indirect cost rate will necessarily and immediately result in staffing reductions.” (emphasis added)); Duke Decl. ¶ 16 (“This means deciding now to

cease projects, reduce head count, and stop operating facilities.” (emphasis added)); Caltech Decl. ¶ 17 (“We would immediately begin a process of reducing the number of graduate student researchers at Caltech.” (emphasis added)); see also New York v. Trump, 133 F.4th 51, 73 (1st Cir. 2025) (denying stay of preliminary injunction where district court found plaintiffs “would irreparably suffer . . . impediments to planning, hiring, and operations[,] and disruptions to research projects by state universities”); NSF, 2025 WL 1725857, at *21 (characterizing “disruption of ongoing research, loss of essential and specialized staff, and disruption to hiring, budgeting and planning” as irreparable harm); DOE, 2025 WL 1414135, at *19 (finding plaintiffs would suffer irreparable harm in the form of inordinately delayed or abandoned research, loss of talent, and risk

of degradation of physical infrastructure); NIH I, 770 F. Supp. 3d at 320 (finding, inter alia, “the suspension of ongoing clinical trials” and “the potential loss of human capital and talent to virtually every Plaintiff” sufficient to demonstrate irreparable harm); cf. Bos. Celtics Ltd. P’ship v. Shaw, 908 F.2d 1041, 1048 (1st Cir. 1990) (concluding losing services of athlete with “unique skill and ability” would constitute irreparable harm). Additionally, Plaintiffs face imminent, irreparable harm by way of their pending research proposals. Plaintiffs and declarants have detailed hundreds of millions of dollars’ worth of

75 Defendants object that the declarations also mention “long-term” consequences of the Rate Cap Policy. That Plaintiffs also identify eventual harms does not undermine the imminent harms that Plaintiffs have identified. proposals pending before DOD prepared in reliance on their individual NICRAs, which they contend could not proceed at the 15% indirect cost rate, with more proposals submitted “every day.” Mem. at 48–49.76 Plaintiffs have described how they have been told they must either resubmit proposals that had used the institutions’ negotiated rates, or not be considered at all, see JHU Decl. ¶ 27 & Ex. 1, and how their proposals are being automatically rejected if they use

negotiated rates, see BU Decl. ¶ 18. Defendants contend that these harms cannot constitute irreparable harm because they can be addressed monetarily after final judgment. Opp. at 45. This argument has been rejected by other courts. See, e.g., DOE, 2025 WL 1414135, at *18 (“In other words, the harm is the massive disruption and its collateral effects, rather than issues around reimbursement.”); see also NIH I, 770 F. Supp. 3d at 320 (finding that the “suspension of ongoing research” would constitute “irreparable harm”); New York, 133 F.4th at 73 (same with respect to “disruptions to research projects by state universities”). Further, even if funding were later restoredto the universities, that would not resolve the harm created by inordinately delayed or abandoned research, nor would it

fix the loss of talent. See, e.g., Brown Decl. ¶ 21 (describing difficulties replacing talent lost due to layoffs even if funding is later restored); JHU Decl. ¶ 8(f) (“A halt in the trial due to JHU’s inability to recover its costs would compromise the health and quality of life of current and future patients and make it difficult, if not impossible, to restart the trial. And even if it were possible to

76See, e.g., ASU Decl. ¶ 9 (123 pending proposals worth approximately $275 million); Brown Decl. ¶ 6 (37 pending proposals worth over $22 million); Caltech Decl. ¶ 5 (18 pending proposals worth $29 million); CU Boulder Decl. ¶ 5 (48 pending proposals worth $58 million); CSU Decl. ¶ 5 (73 pending proposals worth over $69 million); Duke Decl. ¶ 6 (91 pending proposals worth $92.9 million); UIUC Decl. ¶ 6 (161 pending proposals worth over $192,500); KUL Decl. ¶ 5 (27 pending proposals worth over $27 million); KUMC Decl. ¶ 5 (42 pending proposals worth $47.5 million); UMCP Decl. ¶ 6 (117 pending proposals worth over $170 million); MIT Decl. ¶ 11 (hundreds of DOD proposals submitted every year, with many currently under review); Penn Decl. ¶ 5 (77 pending proposals worth $101.2 million); Pitt Decl. ¶ 5 (102 pending proposals worth $92 million); USC Decl. ¶ 5 (270 pending proposals worth over $384 million); UW-Madison Decl. ¶ 5 (81 pending proposals worth over $70 million). resume the trial if funding was restored, the lack of continuity would compromise the scientific results of the trial.”). Plaintiffs have detailed the difficulties in recruitingstaff with the knowledge, experience, and tools to work on highly specialized projects, including security clearances and other approvals needed for sensitive DOD research. See, e.g., UC Decl. ¶ 15 (“[R]ecruiting staff who have the requisite knowledge, experience, and security clearances to work on such projects is

exceedingly difficult. Even if funding were later restored, it would be difficult to either re-recruit or find qualified individuals to fill these positions.”); UMCP Decl. ¶ 22 (same); see Nat’l Council of Nonprofits v. Off. of Mgmt. & Budget, 763 F. Supp. 3d 36, 56–57 (D.D.C. 2025) (inability to recruit replacements following layoffs is irreparable harm).77 Defendants also characterize these harms as “speculative downstream effects from receiving less money from DoD” and argue that these harms would be “self-inflicted” because the universities could reallocate funds, Opp. at 45, but that argument ignores the reality of the situation and mischaracterizes what it means for an injury to be irreparable. Defendants fault Plaintiffs for failing to “explor[e] ways to mitigate any potential budget shortfall,” id. at 45–46, but the

significant weight of the evidence demonstrates that such reallocation of funds as proposed by Defendants is not possible,78 see, e.g., BU Decl. ¶ 16 (detailing financial difficulties and legal restrictions to using additional endowment funds or reallocating endowment funds to cover indirect costs); Brown Decl. ¶¶ 26–28 (same); CSU Decl. ¶ 24 (same); Duke Decl. ¶12 (same);

77 Defendants argue that DOD funding is not a “unique product,” Opp. at 45, but having to forgo a DOD grant means more than a loss of funds. Federally funded research projects are often seeded with proprietary data—in some instances, this data is sensitive or classified, see UMCP Decl. ¶ 11, meaning that it is essentially non-fungible. Intellectual property that results from federally funded research is typically owned by the grantee, subject to a government license. See, e.g., 32 C.F.R. § 34.25(b)(1). Researchers may develop specialized knowledge and connections that lead to further development. See, e.g., UMCP Decl. ¶ 25 (describing a faculty-founded company that subsequently received a $402 million contract with the U.S. Army). In short, the opportunities attending these research grants clearly go beyond their bottom line. 78Defendants provide no support for their contention that any of the universities could use their endowments to address the loss caused by the Rate Cap Policy. See generally Opp. Dkt.13-13(“UIUC Decl.”) ¶ 24 (same); see also NSF, 2025 WL 1725857, at *21 (“[C]ontrary to Defendants’ suggestion, the significant weight of the declarations demonstrates that, because of donor restrictions, competing demands, lack of surplus, and other limitations, endowments and other funding sources are insufficient to compensate for reduced indirect cost funding from NSF such that Universities cannot avoid the irreparable injuries discussed above.”); DOE, 2025 WL

1414135, at *18 (“Faced with the 15 percent rate cap, Plaintiffs will immediately have to make the difficult choices, described in their affidavits, including whether and how to curtail research, freeze hiring, and lay off staff.”). Even just the requirement that the universities reallocate existing resources—which would merely shift the harm by removing funds from other areas of the universities’ budget—is itself an irreparable injury.79 See NSF, 2025 WL 1725857, at *21 (“And even where institutions can replace lost NSF funding from other sources, the reallocation of existing resources disrupts University budgeting processes, which is an irreparable injury itself.”); see also Metro. Transport. Auth. v. Duffy, — F. Supp. 3d —, 2025 WL 1513369, at *45 (S.D.N.Y. May 28, 2025) (finding irreparable harm where “in reliance on the federal government’s approval

of a program,” a plaintiff makes “‘substantial resource investments in planning for the implementation of that program,’ only to have that approval rescinded,” stymying expectation interestsrelated to those investments(quoting Texas v. Brooks-LaSure, 2021 WL 5154219, at *12 (E.D. Tex. Aug. 20, 2021))).

79 IHE negotiated rates typically span multiple years. See 2 C.F.R. pt. 200, App. III(C)(4) (recommending renegotiation every 2–4 years); see also, e.g., BU Decl. ¶ 4 (explaining that BU’s negotiated rate is in effect through June 30, 2028). IHEs fairly rely on those rates when budgeting. Pitt Decl. ¶ 17 (explaining how Pitt relies on its negotiated rate to estimate funding as part of its multi-year budgeting process); OSU Decl. ¶ 20 (same); USC Decl. ¶ 20 (same); UW Decl. ¶ 19 (same). It is therefore no wonder that suddenly upending years of indirect cost coverage would, as Congress predicted, “throw[] research programs across the country into disarray.” S. Rep. No. 115-150, at 109 (2017). While this may very well be “a simple and clear reduction of overhead reimbursement,” Opp. at 47, such a sudden reduction comes at an immense burden given the extent of future budget planning that universities undertake. Defendants further argue that “[f]uture, unissued grants cannot be the basis of irreparable harm” because “Plaintiffs have no entitlement to DoD funding merely because DoD accepts proposals and has previously issued a certain number of awards.” Opp. at 45–46. As discussed in more detail above, supra Section II(A), “the underlying premise that IHEs have nothing but a speculative interest in future [DOD] grants that would incorporate indirect costs is divorced from

reality.” DOE, 2025 WL 1414135, at *19. The Hegseth Memo and Michael Memo both make clear that DOD seeks to continue funding IHEs—which includes Plaintiffs—and there is no basis to believe that DOD will change anything about its grant process other than the Rate Cap Policy at issue. See Hegseth Memo at 2 (describing plans to repurpose the funds saved from decreasing indirect costs for “applied innovation, operational capability, and strategic deterrence”); Michael Memo at 3 (“The academic community’s work with [DOD] remains essential for the development of transformative research concepts that drive our future capabilities.”);see also Opp. at 35 (“The Hegseth Memo does not suggest DoD will abandon research, but simply that the money will shift ‘from overhead to outcomes.’”). As another session of the court explained with regards to DOE:

The DOE and IHEs co-exist in a symbiotic, mutually beneficial relationship, fostered over decades, which is built on the DOE’s need for cutting edge research and the IHEs’ ability to provide it. As a result of this relationship, each IHE has a reasonable, non-speculative assumption that it will obtain future grants because, through the past grants and its own investment, it has built the highly technical, often one-of-a-kind infrastructure needed to conduct the research with the assistance of those future grants, as well as assembled a team of specially trained, uniquely qualified researchers and support staff needed to operate and maintain that equipment. These grants are forthcoming to IHEs because, unless the DOE intends to stop funding this type of research, they must be. The Rate Cap Policy itself manifests an intent on the DOE’s part to continue funding IHEs; all that has changed is that the DOE now wants IHEs to foot more of the bill, without regard to the individual circumstances and overhead costs of each IHE, which Plaintiffs have adequately established they cannot do.

DOE, 2025 WL 1414135, at *19. Given the weight of the evidence that the immense harms to Plaintiffs’ programs cannot be remedied by a later monetary awardand Defendants’ utter lack of evidence to the contrary, the Court is convinced that Plaintiffs have demonstrated irreparable harm. 3. Balance of the Equities and the Public Interest Finally, the Court must weigh the balance of the equities and determine whether a

preliminary injunction would be in the public interest. “These two inquiries merge in a case like this one, where the Government is the party opposing the preliminary injunction.” Devitri v. Cronen, 289 F. Supp. 3d 287, 297 (D. Mass. 2018) (citing Nken v. Holder, 556 U.S. 418, 435 (2009)). Defendants argue that an injunction would harm the public interest by “prevent[ing] DOD from effectuating a policy that would further the democratically-accountable Executive Branch’s assessment of the public interest,” especially in the area of national security, and because it would force the Government “to pay out funds that it may not be able to recover.” Opp. at 50. “To begin with, the Plaintiffs’ likelihood of success on the merits lightens [Defendants’] stated interests.” Huisha-Huisha v. Mayorkas, 27 F.4th 718, 734 (D.C. Cir. 2022). The Supreme Court has confirmed that “our system does not permit agencies to act unlawfully even in pursuit

of desirable ends.” Alabama Ass’n of Realtors v. Dep’t Health & Hum. Servs., 594 U.S. 758, 766 (2021); see also Nat’l Fed’n Indep. Bus. v. Dep’t of Labor, 595 U.S. 109, 120–21 (2022) (staying an illegal vaccine mandate even though the Government said the mandate would save more than 6,500 lives); Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 582 (1952) (affirming district court’s preliminary injunction of an illegal executive order even though a wartime president said his order was “necessary to avert a national catastrophe”). Defendants cite Department of Education v. California, which concerned a temporary restraining order enjoining the Department of Education from terminating grants and requiring it to pay out past-due grant obligations. 145 S.Ct. 996, 2025 WL 1008354, at *1 (2025) (staying the order). However, while California concerned the Government’s alleged breach of existing grant agreements, this motion concerns Defendants’ Rate Cap Policy and DOD’s authority to review grant proposals and issue grants inconsistent with the law.80 “There is generally no public interest in the perpetuation of unlawful agency action.” League of Women Voters of U.S. v. Newby, 838 F.3d 1, 12 (D.C. Cir. 2016). Thus, the Government “cannot suffer harm from an injunction that

merely ends an unlawful practice.” Rodriguez v. Robbins, 715 F.3d 1127, 1145 (9th Cir. 2013). Nonetheless, even if promulgation of the Rate Cap Policy is ultimately vindicated, the temporary halting of the rate change, at most, causes Defendants to suffer the minimal harm of continuing their approach to research cost reimbursement that has endured for decades. Regardless of what approach it uses, DOD is required to disburse the same funds in support of research, just in different ways. This can hardly be said to impose such a burden to conclude that an injunction would go against the public interest, especially where Plaintiffshave demonstrated a likelihood of success on the merits and irreparable harm. Despite Defendants’ claims to the contrary, this minimal burden does not frustrate DOD’s goals to maximize support of “applied innovation,

operational capability, and strategic deterrence,” Hegseth Memoat 2, as the evidence demonstrates

80 Additionally, after raising its concern regarding “the Government’s representation that it is unlikely to recover the grant funds once they are disbursed,” the Supreme Court questioned whether the plaintiffs could establish irreparable harm such as to justify the temporary restraining order because the plaintiff states had “represented in th[e] litigation that they have the financial wherewithal to keep their programs running.” California, 2025 WL 1008354, at *1. Here, Plaintiffs have not made any similar representation. Instead, as discussed above, Plaintiffs have demonstrated that the Rate Cap Policy will force Plaintiffs to terminate ongoing research efforts, terminate staff, and hinder future research irreparably. See, e.g., BU Decl. ¶ 5 (explaining that the Rate Cap Policy would “end or seriously jeopardize” BU’s research for DOD); Caltech Decl. ¶ 6 (“The 15% rate cap will make many, if not all, of Caltech’s proposed research projects infeasible.”); CSU Decl. ¶ 18 (“[A] reduction in the indirect cost rate will necessarily and immediately result in staffing reductions.”); Duke Decl. ¶ 16 (“This means deciding now to cease projects, reduce head count, and stop operating facilities.”); Caltech Decl. ¶ 17 (“We would immediately begin a process of reducing the number of graduate student researchers at Caltech.”); JHU Decl. ¶ 8(f) (“A halt in the trial due to JHU’s inability to recover its costs would compromise the health and quality of life of current and future patients and make it difficult, if not impossible, to restart the trial. And even if it were possible to resume the trial if funding was restored, the lack of continuity would compromise the scientific results of the trial.”). that the Rate Cap Policy will “diminish institutions’ ability to support both existing and potential research,” NIH I, 770 F. Supp. 3d at 327. Thus, the balance of equities and public interest factors weigh in favor of a preliminary injunction. IV. Remedy

“[I]n drafting equitable relief, courts must consider ‘what is necessary, what is fair, and what is workable.’” Id. at 328 (quoting North Carolina v. Covington, 581 U.S. 486, 488 (2017)). Shortly after the parties completed briefing on this motion, the Supreme Court clarified in a separate case the scope of relief available from federal courts, holding that injunctive relief must be limited to “administer[ing] complete relief between the parties.” Trump v. CASA, Inc., 606 U.S. —, 2025 WL 1773631, at *11 (June 27, 2025). As such, this Court directed the parties to submit additional briefing on the remedy sought here and the impact of CASA. See Dkt. 63 (clerk’s notes from the hearing). Plaintiffs seek either a stay of the Rate Cap Policy under section 705 of the APA,81 or a preliminary injunction enjoining the Policy with respect to University Plaintiffs and Organizational Plaintiffs’ member universities. See generally Dkt. 67 (“Pls.’ Suppl. Br.”).

Defendants contend that a stay under “section 705 is bounded by traditional equitable principles, just like the Judiciary Act of 1789” and thus subject to the same prohibition against universal relief under CASA, Dkt. 66 (“Defs.’ Suppl. Br.”) at 4–6, but otherwise largely agree with Plaintiffs regarding the scope of a preliminary injunction, should one issue, id. at 3–4.82

81Section 705 of the APA provides that, at a preliminary stage in the proceedings and “[o]n such conditions as may be required and to the extent necessary to prevent irreparable injury,” courts may “issue all necessary and appropriate process to postpone the effective date of an agency action or to preserve status or rights.” 5 U.S.C. § 705. 82The specific disagreements are discussed below. As an initial matter, the Court does not necessarily agree that a stay under the APA is subject to the same limitations espoused in CASA. In CASA, which concerned the scope of the Judiciary Act of 1789, the Supreme Court explicitly left open “whether the [APA] authorizes federal courts to vacate federal agency action.” CASA, 2025 WL 1773631, at *8 & n.10 (citing 5 U.S.C. § 706(2) (authorizing courts to “hold unlawful and set aside agency action”)). TheSupreme

Court itself has previously upheld the stay of agency action while there was a pending legal challenge, without questioning the propriety of either the interim or final remedy. See, e.g., West Virginia v. Env’t Prot. Agency, 577 U.S. 1126 (2016) (staying an Environmental Protection Agency rule pending the outcome of the case); see also CASA, 2025 WL 1773631, at *21 (Kavanaugh, J., concurring) (explaining that courts may, after CASA, still “grant or deny the functional equivalent of a universal injunction” by “preliminarily setting aside or declining to set aside an agency rule under the APA”). “Moreover, courts ‘routinely stay already-effective agency action under Section 705.’” Haitian Evangelical Clergy Ass’n v. Trump, — F. Supp. 3d —, 2025 WL 1808743, at *10 (E.D.N.Y. July 1, 2025) (quoting Texas v. Biden, 646 F. Supp. 3d 753, 770

(N.D. Tex. 2022) (collecting cases)). That said, the Court finds that a preliminary injunction is the more appropriate remedy at this time, given the parties’ agreement that an injunction limited in scope to University Plaintiffs and Organizational Plaintiffs’ member universities83 would address Plaintiffs’ most immediate concerns.84 Defs.’ Suppl. Br. at 3; Pls.’ Suppl. Br. at 5. The parties also agree that such an

83 The Supreme Court left untouched district courts’ “authority to grant relief to the . . . members” of organizational plaintiffs. CASA, 2025 WL 1773631, at *4 n.2. 84While the Supreme Court has previously referred to vacatur as a “less drastic remedy” than an injunction, see Monsanto Co. v. Geertson Seed Farms, 561 U.S. 139, 165–66 (2010), in this instance, a stay would have a greater practical impact than the injunction issued here, given a stay’s more expansive scope, both in terms of the universities it would cover and the portions of the Policy it would address. injunction must prohibit the Government from, inter alia, “rejecting or treating adversely proposals for DOD funding submitted at universities’ negotiated rates rather than the 15% rate.” Defs.’ Suppl. Br. at 3 (quoting Dkt. 13 at 2). In the face of this agreement, the Court proceeds to consider the relatively narrow disagreements between the parties with regards to the scope of the preliminary injunction.

First, Defendants argue that relief should cover only “the members of associational plaintiffs that have established standing and demonstrated irreparable harm.” Id. at 4. As discussed above, Organizational Plaintiffs have established standing to sue on behalf of all of their member universities, see supra Section II(A), and Defendants misstate the requirements for an organization to demonstrate its members’ irreparable harm, see supra SectionIII(B)(2). TheCourt concludes that no additional showing is necessary to afford relief to all university members of Organizational Plaintiffs. Second, the parties dispute whether the injunction should cover “awards or proposals that involve Plaintiffs or their members in any manner, including but not limited to as sub-recipients

or collaborators.” Pls.’ Suppl. Br. at 6. Defendants contend that this would be vague and overbroad. Dkt. 68 (“Defs.’ Resp.”) at 7. The Court tends to agree that including the language “in any manner” is vague and overbroad for a preliminary injunction. However, the Court finds that the injunction should prevent Defendants from applying the Rate Cap Policy against University Plaintiffs and Organizational Plaintiffs’ member universities with respect to any grant in which they participate, regardless of whether that is as primary grantee, sub-grantee, or collaborator. Not only does the Court find no meaningful distinction in the preliminary injunction analysis with regards to primary grants as compared to sub-grants or collaborations, but the Court also finds that Defendants’ proposed limitation would provide them with an unjustified end run around the terms of the injunction. Finally, Plaintiffs ask the Court to require Defendants “to provide all funding recipients affected by the . . . Policy written notice” of any Court order granting Plaintiffs preliminary relief and to “notify the Court when they have done so.” Pls.’ Suppl. Br. at 6. Defendants note that

Plaintiffs failed to provide any authority that a district court may require such notice to “non-parties or others to whom the preliminary relief does not extend.” Defs.’ Suppl. Br. at 7–8. In the absence of any authority or meaningful explanation for such notice, the Court declines to require it. V. Bond Federal Rule of Civil Procedure 65(c) provides that a “court may issue a preliminary injunction or a temporary restraining order only if the movant gives security in an amount that the court considers proper to pay the costs and damages sustained by any party found to have been wrongfully enjoined or restrained.” That said, there is “ample authority for the proposition that the provisions of Rule65(c) are not mandatory and that a district court retains substantial discretion to dictate the terms of an injunction bond.” Int’l Assoc. of Machinists & Aerospace Workers v.

E. Airlines, 925 F.2d 6, 9 (1st Cir. 1991); see also Pineda v. Skinner Servs., Inc., 22 F.4th 47, 57 (1st Cir. 2021) (concluding that district court did not abuse its discretion when it did not require low-wage laborers to post a bond). “Further, a number of cases have found that a bond is ‘is not necessary where requiring [one] would have the effect of denying the plaintiffs their right to judicial review of administrative action.’” DOE, 2025 WL 1414135, at *21 (quoting Nat. Res. Def. Council, Inc. v. Morton, 337 F. Supp. 167, 168 (D.D.C. 1971)(collecting cases)); see also Doe v. Noem, — F. Supp. 3d —, 2025 WL 1099602, at *20 n.33 (D. Mass. Apr. 14, 2025) (rejecting Government’s argument to require a bond because of “a grave risk” it would “deter individuals from enforcing their procedural rights to challenge agency action”), appeal docketed, No. 25-1384 (1st. Cir. Apr. 18, 2025). For these reasons the Court finds that no bond is warranted. VI. Conclusion For the foregoing reasons, Plaintiffs’ motion for preliminary relief is GRANTED. (i) The Court, therefore, pursuant to Rule 65(a) of the Federal Rules of Civil Procedure,

PRELIMINARILY ENJOINS Defendants, their agents, and anyone acting in concert or participation with Defendants from implementing, instituting, maintaining, or giving effect to the immediately effective portions of the Rate Cap Policy, i.e., those portions implementing a 15% cap for all awards issued on or after June 12, 2025, including but not limited to rejecting or treating adversely proposals for DOD funding submitted at universities’ negotiated rates rather than the 15% rate, in any form with respect to University Plaintiffs and Organizational Plaintiffs’ member universities until further order issued by this Court. This includes where a University Plaintiff or Organizational Plaintiffs’ member university is a sub-grantee or collaborator. (ii) The Court also ORDERS Plaintiffs to provide Defendants with a list of all Organizational Plaintiffs’ member universities by July 21, 2025. Defendants must submit by

July 25, 2025, under the pains and penalties of perjury, a declaration from an attorney of record in this case that certifies that notice of the preliminary injunction and list of covered universities has been provided to all individuals potentially involved in DOD’s grant-approval process. So Ordered.

/s/ Brian E. Murphy Brian E. Murphy Dated: July 18, 2025 Judge, United States District Court