CLOSING THE TOUHY GAP: THE APA, THE FRCP, AND NONPARTY DISCOVERY AGAINST FEDERAL ADMINISTRATIVE AGENCIES

CLOSING THE TOUHY GAP: THE APA, THE FRCP, AND NONPARTY DISCOVERY AGAINST FEDERAL ADMINISTRATIVE AGENCIES

In the 1951 case United States ex rel. Touhy v. Ragen, the Supreme Court determined that courts can’t hold federal agency officials in contempt for refusing to comply with nonparty subpoenas if they do so pursuant to valid agency regulations. Though the Court suggested that litigants could still challenge these noncompliance decisions, it didn’t flesh out what that process would look like. Following Touhy, federal courts have split. When it comes to civil, federal court litigation, a plurality of circuits evaluate agencies’ noncompliance decisions under the Administrative Procedure Act (APA), while a minority of circuits do so under the Federal Rules of Civil Procedure (FRCP).

This Note serves two primary purposes. First, it estimates the effect of the APA–FRCP split on nonparty discovery outcomes. Using a logistic regression analysis, it finds that a litigant proceeding under the FRCP can expect about a twenty-six percentage-point greater chance of obtaining discovery compared to a similarly situated litigant proceeding under the APA. Second, it proposes ways to mitigate the breadth and potency of the split. Courts can limit the number of contexts where the circuit split comes into play by applying traditional tools of interpretation to the statute giving agencies authority over their employees’ subpoena responses. And plurality-approach courts can close the discovery-outcome gap (where the split remains) by ensuring their analyses import into the Touhy context the APA’s administrative law safeguards, not just its deferential arbitrary and capricious standard.

The full text of this Note can be found by clicking the PDF link to the left.

Introduction

An estimated 2.68 million civilians work for federal administrative agencies. 1 Jennifer L. Selin & David E. Lewis, Admin. Conf. of the U.S., Sourcebook of United States Executive Agencies 10 (2d ed. 2018), https://www.acus.gov/sites/default/files/documents/ACUS%20Sourcebook%20of%20Executive%20Agenices%202d%20ed.%20508%20Compliant.pdf [https://perma.cc/R2Z7-7PF5]. Their work spreads across more than one hundred agencies, each with its own area of expertise. 2 Id. at 12. This estimate uses the categorization of the Freedom of Information Act’s (FOIA) website. See FOIA, https://www.foia.gov/#agency-search [https://perma.cc/8J6Z-GVKB] (last visited Oct. 7, 2019). The statutory definition of “agency” in the Administrative Procedure Act (APA), however, is ambiguous. See 5 U.S.C. § 551(1) (2018); Selin & Lewis, supra note 1, at 12. Therefore, estimates of the number of federal agencies vary from one source to another. See Selin & Lewis, supra note 1, at 12. As then-Professor Felix Frankfurter remarked in 1927 and is even more so the case today, the subject-matter expertise of administrative agencies runs “the whole gamut of human affairs.” 3 Felix Frankfurter, The Task of Administrative Law, 75 U. Pa. L. Rev. 614, 614 (1927). It’s no wonder then that parties involved in lawsuits often turn to agencies for information to help build a case or mount a defense, 4 See, e.g., CF Indus., Inc. v. Dep’t of Just. Bureau of Alcohol, Tobacco, Firearms & Explosives, 692 F. App’x 177, 178–79 (5th Cir. 2017) (describing a defendant-corporation’s attempt to retrieve DOJ records to show that a criminal actor—and not the corporation—caused a factory explosion that killed fifteen people and was the subject of a wrongful death suit); Plaintiffs’ Original Complaint for Judicial Review Pursuant to the Administrative Procedure Act at 5, 7–9, Hasie v. Off. of the Comptroller of the Currency, No. 5:07-cv-208-C (ECF), 2008 WL 4549881 (N.D. Tex. May 9, 2008), 2007 WL 3311850, at *2–3 (seeking suspicious-activity reports from the Office of the Comptroller of the Currency to prove lack of probable cause in a malicious-prosecution suit). even when the federal government isn’t a party to the underlying litigation. 5 See, e.g., Exxon Shipping Co. v. U.S. Dep’t of Interior, 34 F.3d 774, 779 (9th Cir. 1994) (“[T]he National Weather Service alone receives hundreds of requests a year from private litigants seeking to introduce evidence about weather patterns . . . .”); Alex v. Jasper Wyman & Son, 115 F.R.D. 156, 157 n.3 (D. Me. 1986) (noting that over 1,500 subpoenas are served annually on DOL employees); Joshua Jay Kanassatega, The Discovery Immunity Exception in Indian Country—Promoting American Indian Sovereignty by Fostering the Rule of Law, 31 Whittier L. Rev. 199, 228 (2009) (“Perhaps the largest non-party source of facts and information is the United States government.”).

Take, for example, a recent lawsuit against the sheriff’s office of a small county in North Carolina. 6 Lamb v. Wallace, No. 7:16-CV-44-FL, 2018 WL 847242 (E.D.N.C. Feb. 13, 2018). Larry Lamb, having spent two decades in prison for a crime he did not commit, alleged that the Duplin County Sheriff’s Office deprived him of his constitutional rights by coaching a witness to fabricate her testimony and failing to disclose that the witness had a long and less-than-truthful history as a criminal informant for both state and federal law enforcement. 7 Id. at *4–6. To prove his claim, Lamb served a subpoena on the FBI requesting documents related to the Bureau’s work with the witness. 8 Id. at *4. The FBI, however, flatly rejected Lamb’s subpoena, citing its own regulations for the authority to do so. 9 Id.

Subpoenas—like the one Lamb served on the FBI—seeking deposition testimony, trial testimony, the production of nontestimonial evidence, or some combination of the three  from  parties  not  involved  in  an  underlying  action  are  called  nonparty subpoenas. 10 See, e.g., Natasha Breaux, Comment, Analysis of the Proposed Amendments to Federal Rule of Civil Procedure 45 Pertaining to Nonparty Subpoenas for Documents, 50 Hous. L. Rev. 191, 192–94 (2012). Generally, the relevant rules of procedure govern whether a litigant can expect a response to a nonparty subpoena, with nonparty discovery tending to be a relatively straightforward and requester-friendly process. 11 See, e.g., Fed. R. Civ. P. 45(d)(3)(A) (permitting federal courts to quash subpoenas only if they are procedurally defective, exceed a court’s territorial jurisdiction, would create an “undue burden,” or would require the disclosure of privileged material); Cal. Civ. Proc. Code § 1987.1(a) (2020) (permitting California state courts to quash or modify subpoenas if they make “unreasonable or oppressive demands”). But when the subpoena recipient is an agency (e.g., the DOJ) or a subcomponent thereof (e.g., the FBI), the process gets more complicated.

Before an agency official can comply with a nonparty subpoena, an agency head—or an official with delegated authority—must determine that the litigant satisfied the department’s Touhy regulations. 12 These regulations are named after the Supreme Court case, United States ex rel. Touhy v. Ragen, 340 U.S. 462 (1951), which section I.A.3 details below. These regulations, which are named for a midcentury Supreme Court case and differ slightly from agency to agency, govern whether an agency employee is authorized to submit to judicial process. 13 See COMSAT Corp. v. Nat’l Sci. Found., 190 F.3d 269, 272 n.3 (4th Cir. 1999).

Agencies get the authority to promulgate Touhy regulations from the Federal Housekeeping Statute, which reads: “The head of an Executive department or military department may prescribe regulations for . . . the custody, use, and preservation of its records, papers, and property.” 14 5 U.S.C. § 301 (2018). The statute’s next sentence lays out what, at first blush, might look like a broad caveat but, in reality, is a narrow clarification of agencies’ authority: “This section does not authorize withholding information from the public or limiting the availability of records to the public.” 15 Id.; see also infra section I.A.4 (discussing the addition of this sentence to the Housekeeping Statute). Under this statute, nearly every administrative agency has adopted Touhy regulations restricting to some degree its employees’ ability to comply with work-related subpoenas. 16 See, e.g., 6 C.F.R. §§ 5.41–5.49 (2020) (DHS); 10 C.F.R. §§ 202.21–202.26 (2020) (Department of Energy); 15 C.F.R. §§ 15.11–15.18 (2020) (Department of Commerce); 21 C.F.R. §§ 20.1–20.3 (2020) (FDA); 28 C.F.R. §§ 16.21–16.29 (2019) (DOJ); 32 C.F.R. §§ 97.1–97.6 (2019) (DOD); 38 C.F.R. §§ 14.800–14.810 (2019) (VA); 40 C.F.R. §§ 2.401–2.406 (2019) (EPA).

Touhy regulations fall into two broad categories. Regulations in the first category are procedural. For example, a party may need to submit, in addition to a subpoena, a letter providing a “summary of the information sought and its relevance to the proceeding.” 17 28 C.F.R. § 16.22(d). For the most part these procedural regulations are easy enough to comply with, 18 See Juan G. Villaseñor, How to Properly Seek Testimony or Documents from a Federal Agency, Colo. Law., Aug. 2016, at 37, 40–41 (“The potential roadblocks that a party may encounter by subpoenaing a federal agency or employee without complying with the agency’s Touhy regulations are avoidable.”); see also Gregory C. Sisk, A Primer on Civil Discovery Against the Federal Government, Fed. Law., June 2005, at 28, 33 (arguing that “being forewarned is to be forearmed” in the Touhy context and litigants should “be able to respond appropriately” to agencies’ procedural requirements). though they do pose obstacles for unwary litigants, especially those proceeding pro se. 19 Santini v. Herman, 456 F. Supp. 2d 69, 72 (D.D.C. 2006) (“Although [Plaintiff] is representing herself and is entitled to some leeway as a pro se litigant . . . she must follow the procedures of the DOJ’s Touhy regulations . . . .”); Meisel v. Fed. Bureau of Investigation, 204 F. Supp. 2d 684, 689–91 (S.D.N.Y. 2002) (dismissing a case for lack of jurisdiction because the litigant did not attach a satisfactory “Statement of Scope and Relevance” to his subpoena as required by the FBI’s Touhy regulations).

Regulations in the second category are substantive. These regulations, which are the focus of this Note, tend to be extremely difficult to satisfy. 20 See infra section II.B (describing the unlikelihood of obtaining discovery under a standard employed by a plurality of circuits that gives great weight to agencies’ Touhy regulations). For example, a party may need to show that an agency’s compliance with a subpoena serves the “public interest” even when considered against the government’s need “to avoid spending . . . time and money . . . for private purposes” and the risk that compliance would undermine the agency’s “performance . . . of its mission and duties.” 21 6 C.F.R. § 5.48(a)(3), (5), (7). No easy task.

If an agency determines a litigant failed to satisfy its Touhy regulations, the litigant can challenge the agency’s decision in court. But what happens next is the subject of a circuit split that’s now over twenty-five-years old. 22 See infra section I.B; see also Exxon Shipping Co. v. U.S. Dep’t of Interior, 34 F.3d 774, 778–80 (9th Cir. 1994) (creating the circuit split by holding that discovery disputes against nonparty agencies should be evaluated under the FRCP). A plurality of circuits require a litigant to challenge the agency’s subpoena noncompliance under the Administrative Procedure Act (APA). The litigant bears the burden of showing, with reference to the agency’s own regulations, that the agency’s action was arbitrary and capricious. 23 See infra section I.B.1. A minority of circuits, however, permit a litigant to proceed as if the government was any run-of-the-mill nonparty. The Federal Rules of Civil Procedure (FRCP) govern, and the agency bears the burden of showing that compliance would be unduly burdensome or require the disclosure of privileged material. 24 See infra section I.B.2.

This circuit split has produced a large body of literature, with the bulk of it advocating for a particular resolution of the split. 25 See infra note 126 and accompanying text. Two untested assumptions characterize much of the subject’s literature. The first is that litigants proceeding under the FRCP prevail at a significantly greater rate than litigants proceeding under the APA. 26 See infra note 127 and accompanying text. The second is that the circuit split will be readily resolved by either an outside actor (i.e., Congress or the Supreme Court) stepping in or a lower-court-generated consensus trending toward application of the FRCP. 27 See infra notes 128–129 and accompanying text.

This Note tests these two assumptions. While the first withstands scrutiny, the second does not. Based on a logistic regression analysis, a litigant proceeding under the FRCP can expect a roughly twenty-six percentage-point greater chance of obtaining discovery against a nonparty federal agency compared to a similarly situated litigant proceeding under the APA. 28 See infra section II.B. But neither Congress nor the Supreme Court has shown much interest in resolving the APA–FRCP split; and the suggested trend toward employing the FRCP enjoys little support in reality. 29 See infra section II.C.

Given these two findings, it’s time to face an odd state of affairs as the circuit split inches toward three decades of existence: The circuit split significantly undermines federal court uniformity, but its resolution doesn’t appear to be on the horizon. As a result, courts employing the plurality, APA-based approach need to take care that litigants in their jurisdictions are not disadvantaged compared to litigants in courts employing the minority, FRCP-based approach. Fortunately, this Note argues, plurality-approach courts can do just that while avoiding major departures from their precedent. Plurality-approach courts can mitigate the unfairness otherwise created by the Touhy-derived circuit split by vigilantly applying traditional tools of statutory interpretation to the Federal Housekeeping Statute and ensuring that their approach imports into the Touhy context the APA’s administrative law safeguards, not just its deferential arbitrary and capricious standard.

This Note proceeds in three Parts. Part I provides an overview of the Touhy doctrine, describing its historical development, its foundational cases, and the circuit split over its reach in the federal-civil context. Part II serves two purposes. First, it tests the two above-described assumptions (differing success rates and imminent resolution), finding support for the former but not the latter. Second, it explains how these two findings place federal courts in the middle of two key commitments: federal court uniformity and adherence to precedent. Part III then offers ways of mitigating this tension by limiting the scope of intercircuit disagreement and making APA review of Touhy decisions rigorous enough to narrow the APA–FRCP gap in discovery success rates.