Welcome back to Ninth Circuit Review-Reviewed, your monthly recap of administrative law before arguably “the second most important court in the land.” Let’s get straight to January’s cases.

In Gill v USDOJ, the Ninth Circuit made hash of an important administrative law doctrine. To be precise, the panel seems to have either ignored or implicitly overruled its own precedent regarding an exemption from APA rulemaking requirements for “general statements of policy.” 5 U.S.C. § 553(b).

At issue is the famous distinction between nonlegislative rules and legislative rules: The former are exempt from APA § 553 notice-and-comment procedures, while the latter are not. Of course, the APA identifies two types of nonlegislative rules: “general statements of policy” and “interpretative rules.” Under black-letter law, the two types of nonlegislative rules are distinct. Interpretative rules are “rules or statements issued by an agency to advise the public of the agency’s construction of the statutes and rules which it administers,” while general statements of policy are “statements issued by an agency to advise the public prospectively of the manner in which the agency proposes to exercise a discretionary power.” Attorney General’s Manual on the Administrative Procedure Act 30 FN 3 (1947). For this post, my focus is on the Ninth Circuit’s doctrine for “general statements of policy.”

In an insightful article, Prof. Ronald Levin observes that federal courts generally have accepted the “binding norm” test to discern “general statements of policy” from “legislative rules.” Simply put, the crucial question is whether the document has “binding effect” or “binding practical effect.” If it does, then it’s a legislative rule. Notwithstanding the wide acceptance of the “binding norms” test, Prof. Levin identifies one doctrinal question where courts diverge: Whom do the rules bind?

In its original form, the “binding norms” test pertained to the extent to which the agency’s issuance was binding on regulated parties. See Pacific Gas & Electric Co. v. FPC, 506 F.2d 33, 38 (D.C. Cir. 1974). In CNI v. Young, however, the D.C. Circuit held that a policy statement is a legislative rule if it binds the agency instead of the public. See 818 F.2d 943, 948 (D.C. Cir. 1987); see also Texas v. United States, 86 F. Supp. 3d 591, 610 (S.D. Tex.), aff’d, 809 F.3d 134 (5th Cir. 2015) (employing this reasoning to find unlawful the Obama-era deferred action immigration policy).

Prof. Levin notes, however, that “not all courts accept the idea that a guidance document becomes legislative if it binds agency personnel.” As an example, he discusses a 2004 Ninth Circuit case:

In Erringer v. Thompson, Medicare claimants argued that a “local coverage determination” was unlawful under § 553 because Medicare contractors (insurance companies) were required to follow [it] (although ALJs to whom the contractors’ decisions were appealable were not). The Ninth Circuit disagreed by saying that the correct inquiry was whether the decision was binding on persons outside the agency, and for this purpose the contractors should be considered “inside” the agency.

To recap, there is a dispute among courts whether the “binding norms” test—which is employed to distinguish the scope of the APA exception for “general statements of policy”—applies to the government. In its 2004 Erringer decision, the Ninth Circuit sided with those who argue that the test applies only to the public.

Notwithstanding Erringer, in Gill v. US DOJ, the panel applied the “binding norms” framework to the government rather than the public. This is the opposite position that the court took in Erringer. Yet the court didn’t mention its precedent.

At issue in Gill was an avowed “general statement of policy” to implement the National Strategy for Information Sharing, a post-9/11 initiative that established “fusion centers” to ensure Suspicious Activity Reports (SARs) are disseminated to appropriate government authorities. To govern this nationwide effort to standardize information sharing, the Justice Department issued a document known as the Functional Standard. The court describes the Functional Standard as follows:

The Functional Standard defines suspicious activity as “[o]bserved behavior reasonably indicative of preoperational planning associated with terrorism or other criminal activity.” After receiving a report of suspicious activity, an officer creates a SAR. The SAR then undergoes a two-part evaluation process. An analyst determines whether the SAR meets certain behavioral criteria and has a potential nexus to terrorism. If the analyst concludes that it does, the SAR becomes an ISE-SAR, and is uploaded to the eGuardian repository, where it is available to all [program] participants.

Plaintiffs-appellants are U.S. citizens who are the subjects of the program, none of whom has been charged with a crime. Inter alia, they alleged that the Functional Standard is a legislative rule that unlawfully dodged rulemaking procedures. For its part, the government argued that the Functional Standard is exempt from APA § 553 requirements because it is a general statement of policy.

Below, I’ve excerpted the Gill court’s operative reasoning:

While the Functional Standard employs a combination of mandatory and discretionary language, it does not compel analysts or agencies to disseminate SAR information, nor does it require analysts to create an ISE-SAR when the information reflects a certain number of behavioral categories. Rather, if the SAR contains at least one of the categorized activities, the analyst must still ascertain whether it has a potential nexus to terrorism. No single category of behavior or aggregation of categories is determinative.

…

This significant discretion retained by agencies and their analysts in determining whether to disseminate information compels our decision that the Functional Standard is not a legislative rule. Therefore, the Functional Standard was exempt from the notice and comment requirement.

Despite its prior holding in Erringer, the court’s analysis in Gill focuses solely on whether the “general statement of policy” is binding on the government. Indeed, the court nowhere mentions its precedent to the contrary in Erringer. The upshot is that, after last month’s decision in Gill, the Ninth Circuit now has taken both sides in the ongoing debate among courts about how to employ the binding norms test for discerning nonlegislative rules from legislative rules.

Jury Instructions as Article III Regulatory Process

Congress does not always delegate regulatory power to administrative agencies. Lawmakers also provide courts the initiative to effectuate regulatory policy.

While agency administrative processes are well-studied, scholars have paid far less attention to judicial mechanisms of regulating. And because this area of administrative law is arguably underserved by the scholarship, administrative law nerds (like me) may be interested in the Ninth Circuit’s decision in Frost v. BNSF, which pertains to one ad hoc process by which courts implement regulatory policy—jury instructions.

With passage of the Federal Railroad Safety Act, Congress provided safeguards for railroad workers from employer retaliation for statutorily protected activity, including the filing of injury reports and work safety complaints. Congress could have delegated responsibility for adjudicating these statutory rights to an administrative tribunal. But, instead, the Act established a cause of action in federal district courts, and thereby rendered employers liable for damages for statutory violations. See 49 § U.S.C. 20109(d)(3).

Frost v. BSNF involved such a cause of action. In the facts underlying the case, Michael Frost narrowly missed being hit by a train on the job, and afterwards he filed an injury report and a work safety complaint. Subsequently, Mr. Frost was dismissed from his job, and he filed suit seeking damages in federal trial court, alleging that he was fired for activity protected by statute. For its part, the company claims he was dismissed for a related safety violation.

To establish a rebuttable claim of unlawful discrimination under the Act, the plaintiff must prove by a preponderance of the evidence that his or her protected conduct was a “contributing factor” in the unfavorable personnel action alleged in the complaint. 49 U.S.C. § 42121(b)(2)(B)(iii)). At trial, BSNF argued that this federal statutory provision requires that the employee demonstrate that the employer acted with discriminatory intent. That is, the company argued that statute required retaliatory scienter. Persuaded by the company’s argument, the district court judge instructed the jury that BNSF could not be liable if it terminated Frost due to an “honest belief” that he violated the company’s safety rules. The jury returned a verdict for BNSF.

The Ninth Circuit panel overturned the district court. According to the panel, “[w]hat BNSF misses is that the only proof of discriminatory intent that a plaintiff is required to show is that his or her protected activity was a ‘contributing factor’ in the resulting adverse employment action … [T]here is no requirement that FRSA plaintiffs separately prove discriminatory intent.” As a result, the court reversed the trial court and remanded for a new trial.

I think this case is neat because of the regulatory role played by the jury instructions. The law’s scope hinged on those instructions—it’s the difference between having to prove intent or not, which obviously narrows the Act’s remedial regime.

To be sure, the district court was overturned based on its jury instructions. But there’s no guarantee that a different circuit (or even a different panel) wouldn’t have decided the case differently. For that matter, it’s easy to imagine the Ninth Circuit not getting the chance to review the lower court—Mr. Frost easily could have decided against the cost of further litigation, or his lawyer could have failed to preserve an appeal. The point is that district court Judge Donald Molloy wielded significant regulatory authority when he issued his instructions to the jury.

To my eyes, the jury instructions in this case are akin to many of the procedures regularly employed by regulatory agencies. Jury instructions are a discrete document. They carry an immediate legal effect on the regulatory program. Moreover, jury instructions can influence future cases, as they are mined by trial lawyers for applicable language.

Other Ninth Circuit News

I don’t have enough time or space to write about all the neat developments in administrative law before the U.S. Court of Appeal for the Ninth Circuit. Therefore, I list below the best recent commentary on the circuit.

• Anne Roller, National Law Review, “Due Process and Primary Jurisdiction Defense to Website Accessibility Claims Fall Like Dominoes in the Ninth Circuit” (regarding Robles v. Dominoes Pizza, filed 1/15/19)
• Devin Watkins, Open Market, “Ninth Circuit Strikes Down Soda Labeling Ban for Wrong Reasons” (regarding American Beverage Assn. v. City & County of San Francisco, filed 1/31/19)
• Niels Lesniewski, Roll Call, “Debate over ninth Circuit Seats Is Latest Battle in Trump-Senate Judicial Wars” (for an interesting take on “blue slips” in the judicial advise and consent process, see this post from Above the Law’s David Lat, “Bye-Bye Blue Slips”)
• Steve Brachmann, IP Watchdog, “No License No Cry: Ninth Circuit Nixes Jammin Java Appeal in Bob Marley Trademark Case” (regarding Hope Road Merchandising LLC v. Jammin Java Corp., filed 1/9/19)

William Yeatman is a research fellow at the Cato Institute’s Robert A. Levy Center for Constitutional Studies.