On Monday, we saw the contours of state-action immunity from federal antitrust law, and on Tuesday, we saw the basic facts of the North Carolina Board of Dental Examiners v. FTC case, recently decided by the Fourth Circuit and now up at the Supreme Court on a cert petition. Today, we’ll talk about the disarray in the federal appellate courts as to how to decide whether an ostensibly “state” agency is in fact private for purposes of the state-action doctrine. Because of this disarray, it’s a good idea for the Supreme Court to take the case.

Let’s recall the law on state-action immunity from federal antitrust law, first discussed in Monday’s post. Initially, Parker v. Brown holds that the acts of the state legislature itself are absolutely immune from antitrust law, as a matter of statutory interpretation (but informed by federalism concerns). Next, Midcal gives us a two-part test for lower-level actors: (1) the anticompetitive policy must be clearly articulated by the state and (2) the state must actively supervise the implementation of the anticompetitive policy. But in Town of Hallie, the Supreme Court said that prong (2) isn’t necessary for municipalities, and suggested that “it is likely that active state supervision would also not be required” for state agencies.

That suggestion was dictum, but the influential Areeda-Hovenkamp antitrust treatise agrees with it: “Dispensing with any supervision requirement for municipalities implies, a fortiori, the same for the ‘public’ departments and agencies of the state itself.” The treatise adds that “[t]oday the courts uniformly agree with that conclusion,” which gives us a nice three-part doctrine where legislatures themselves get blanket immunity, public state agencies and municipalities are subject to the first Midcal prong, and private parties are subject to both Midcal prongs.

But the apparent uniformity might be misleading, since—as the treatise says—“determining whether an actor is sufficiently ‘public’ so as not to require supervision has often proven difficult.”

As a result, we have at least three separate approaches in the federal appellate courts. One, a “cursory approach”, prevails in the Second, Fifth, and Tenth Circuits. The second, an “intermediate approach”, prevails in the First, Seventh, Ninth, and Eleventh Circuits. The third, the “categorical approach”, reflects the FTC’s view, prevailed here in the FTC proceeding, and was essentially adopted by the Fourth Circuit.

*     *     *

The cursory view.

In Earles v. State Board of CPAs of Louisiana (5th Cir. 1998), the Fifth Circuit considered the status of Louisiana’s state board of CPAs, which had made rules prohibiting CPAs from “engaging in the practice of . . . ‘incompatible professions’” like selling securities. The Fifth Circuit held that this board was public enough to be exempt from the active supervision requirement. Even though it was “composed entirely of CPAs who compete in the profession they regulate,” the “public nature of [its] actions mean[t] that there [was] little danger of a cozy arrangement to restrict competition.” The Board was thus “functionally similar to a municipality.” The analysis here was unfortunately fairly cursory.

In Porter Testing Laboratory v. Board of Regents for Oklahoma Agricultural & Mechanical Colleges (10th Cir. 1993), The Tenth Circuit did no better. This case arose in the context of a public university that had allegedly conspired to “monopolize certain agricultural testing services” in the state. The Tenth Circuit held that the active supervision requirement was unnecessary “[g]iven the nature of these defendants, a constitutionally created state board, its executive secretary, and a state created and funded university.”

The Second Circuit’s analysis of why an urban development corporation was exempt from the active supervision requirement seems to have the same flavor. In Cine 42nd St. Theater Corp. v. Nederlander Organization, Inc. (2d Cir. 1986), it held that the development corporation at issue could be presumed to be public-interested because it was “by statute a political subdivision of the state.” This one-factor test—essentially relying on how the state labels the entity—seems certainly incorrect in light of the Supreme Court’s holding in Goldfarb v. Virginia State Bar (1975) (as reinterpreted in Town of Hallie), that the Virginia State Bar, though a state administrative agency, was a “private part[y]” subject to the active supervision requirement.

*     *     *

The intermediate view.

Other circuits don’t take the publicness of a state agency for granted or rely on formal labeling. The Oregon State Bar adopted a rule making itself the sole legal provider of malpractice insurance for the state’s lawyers. In Hass v. Oregon State Bar (9th Cir. 1989), the Ninth Circuit ended up exempting it from the active supervision requirement, but only after analyzing a number of factors. It considered not just the Bar’s formal classification as “a public corporation and an instrumentality of the State of Oregon,” but also how many of its members “must be nonlawyer members of the public,” the requirement that its records be “open for public inspection” and its accounts “subject to periodic audits by the State Auditor,” its open meeting requirements, and the fact that its members “are public officials who must comply with the Code of Ethics.” “These requirements leave no doubt that the Bar is a public body, akin to a municipality for the purposes of the state action exemption.”

The First Circuit seems to also follow this sort of nuanced approach. In FTC v. Monahan (1st Cir. 1987), then-Judge Stephen Breyer discussed whether the Massachusetts Board of Registration in Pharmacy was acting anticompetitively in limiting pharmacist advertising, mail-order pharmacies, and “branch offices” or “pick-up stations” (“where patients could drop off, and pick up, prescriptions that the ‘main office’ (in the interim) would fill in batches”). According to Judge Breyer, whether the pharmacy board was essentially private for the purposes of the active supervision requirement would depend “upon how the Board functions in practice, and perhaps upon the role played by its members who are private pharmacists.”

Just half a year earlier, in Interface Group, Inc. v. Massachusetts Port Authority (1st Cir. 1987), Judge Breyer had used a similarly pragmatic approach, holding that the Massachusetts Port Authority was similar to a municipality because it possessed “such typical governmental attributes as the power of eminent domain, rulemaking authority, bonding authority, and tax exempt status.”

In Bankers Insurance Co. v. Florida Residential Property & Casualty Joint Underwriting Ass’n (11th Cir. 1998), the Eleventh Circuit summarized its own (and other circuits’) cases as focusing on how “public the entity looks” through an analysis of “the government-like attributes of the defendant entity”:

Factors favoring political-subdivision treatment include open records, tax exemption, exercise of governmental functions, lack of possibility of private profit, and the composition of the entity’s decisionmaking structure. The presence or absence of attributes such as these tells us whether the nexus between the State and the entity is sufficiently strong that there is little real danger that the entity is involved in a private anticompetitive arrangement.

A similar multi-factor analysis—where the entity’s nonprofit form also plays a role—can be found in the Seventh Circuit; see Fuchs v. Rural Electric Convenience Cooperative, Inc. (7th Cir. 1988).

*     *     *

The FTC/Areeda-Hovenkamp view.

The intermediate multi-factor analysis of the First, Seventh, Ninth, and Eleventh Circuits, while more nuanced than the simple “public is public” view of the Second, Fifth, and Tenth Circuits, still isn’t good enough for the FTC. Rather than adopting a “laundry list of attributes” approach, the FTC prefers to focus on one particular aspect of the challenged bodies: the extent to which they’re driven by private self-interest.

In yesterday’s post, we saw the FTC’s position that the North Carolina Board of Dental Examiners was private, and therefore required active supervision to benefit from the state-action exemption, because “the state agency’s decisions are not sufficiently independent from the entities that the agency regulates”—in this case, because the agency has a “financial interest in the restraint that [it] seeks to enforce” and is “controlled by private market participants” “who [stand] to benefit from the regulatory action.” The FTC also noted that the Board was even more problematic because its dentist members (and dental hygienist member) were elected only by other practicing dentists (or dental hygienists), though this political accountability element doesn’t seem to have been part of the FTC’s formal test.

The Fourth Circuit adopted the FTC’s holding, at least to the extent of requiring active supervision when both of the FTC’s factors were present—domination by and accountability to market participants.

The Areeda-Hovenkamp treatise seems to take a position closer to the FTC’s, not requiring the additional element of accountability to market participants:

We would presumably classify as “private” any organization in which a decisive coalition (usually a majority) is made up of participants in the regulated market. This presumption would be rather weak . . . where the competitive relationship between the decision maker . . . and the plaintiff is weak and the potential for anticompetitive effects not particularly strong. It would be weaker still where the decision maker responds to the court, governor, or legislature directly and on an ongoing basis. But the presumption should become virtually conclusive where the organization’s members making the challenged decision are in direct competition with the plaintiff and stand to gain from the plaintiff’s discipline or exclusion.

The Areeda-Hovenkamp treatise thus takes issue with the Ninth Circuit’s approach in the Oregon State Bar case, agreeing with the dissent that the self-interest of the lawyers composing the Bar should make the Bar private for state action immunity purposes. All the more does the treatise disagree with approaches like that of the Second Circuit: “state legislative declarations that the body is a ‘public’ corporation,” or “state mandates that the organization serve the ‘public interest,’” should count for little. Nor should an entity’s nonprofit status count for much: “the typical trade or professional organization is itself a nonprofit organization dedicated to improving the welfare of its members. The key is not the profit or nonprofit status of the organization, but the identity of its decision-making personnel.”

*     *     *

Is this circuit split illusory?

The FTC and Solicitor General have filed an opposition to the Board’s cert petition. In the first place, it argues that the Fourth Circuit’s decision is correct. That’s all well and good (the merits are something I’ll address in another post), but if there’s a true circuit split, it’s still worth granting cert, even if just to affirm the Fourth Circuit and impose a single approach on the circuits that go the other way.

Therefore, a big chunk of the FTC’s brief is devoted to rebutting the claim that there’s a circuit split. Though the FTC grants that there’s “some variation” between this case and those in other circuits, it goes on to say that there are significant differences between this case and those others:

Although the board in Hass was dominated by bar members, the challenged rule—which required Oregon attorneys to participate in the state bar’s own professional liability fund—did not implicate competition among bar members or between bar members and non-attorneys, because board members were not themselves engaged in the business of liability insurance. Similarly, although the board in Earles was “composed entirely of [accountants] who compete in the profession they regulate,” the court concluded that the challenged rule—which forbade accountants from simultaneously practicing “incompatible occupations” (such as selling securities)—was not a “cozy arrangement to restrict competition.” Although the board in Hass was dominated by bar members, the challenged rule—which required Oregon attorneys to participate in the state bar’s own professional liability fund—did not implicate competition among bar members or between bar members and non-attorneys, because board members were not themselves engaged in the business of liability insurance. Similarly, although the board in Earles was “composed entirely of [accountants] who compete in the profession they regulate,” the court concluded that the challenged rule—which forbade accountants from simultaneously practicing “incompatible occupations” (such as selling securities)—was not a “cozy arrangement to restrict competition.” (Citations omitted, and spelling corrected from Haas to Hass.)

Therefore, the FTC says, the Fifth Circuit (in Earles) and Ninth Circuit (in Hass) weren’t faced with precisely similar facts: if those circuits were faced with a fact pattern like that of the North Carolina Board of Dental Examiners, nothing would stop them from finding that that Board was private.

Moreover, the board in Earles was appointed by the governor (from a slate of candidates chosen by the profession) and confirmed by the senate; this distinguishes it from this case, where the Fourth Circuit put some weight on the fact that the board members were directly elected by the profession. And Hass, says the FTC, may not even be good law in the Ninth Circuit anymore, given some intervening Supreme Court precedent (Patrick v. Burget (1988)) and a remand in light of that precedent (Washington State Electrical Contracts Ass’n v. Forrest (9th Cir. 1991)).

I don’t find all this super-convincing. Of course Earles and Hass both implicated competition. In Earles, the board of CPAs was regulating other CPAs and preventing them from engaging in other lines of work. Some CPAs presumably were involved in selling securities and others weren’t, and those that did were placed in a worse competitive position relative to those that didn’t.

And in Hass, the state bar made itself the sole provider of malpractice insurance, which means lawyers couldn’t choose from a variety of providers. Reading Hass on the malpractice crisis makes one think of adverse selection and Obamacare: 35% of private attorneys in Oregon had previous been without malpractice insurance. It’s not hard to see why: it’s hard for insurance companies to gauge an attorney’s true malpractice risks (an attorney’s past record is useful but not perfect), so any pool of insureds will have some high-risk and low-risk people that the insurance company can’t tell apart. The high-risk people will drive premiums up, which might make the insurance too expensive for the low-risk people. You end up with high-priced insurance for the high-risk people, and widespread non-participation by the lower-risk people. The state bar’s becoming the sole provider—in a program with mandatory participation (sound familiar?)—inevitably means that low-risk people were going to cross-subsidize high-risk people. So all this does implicate competition among members of the profession.

Moreover, on what basis can we conclude that these circuits would come to a different conclusion if they were faced with these facts? The Fifth Circuit said in Earles that the “public nature of [the CPA board’s] actions mean[t] that there [was] little danger of a cozy arrangement to restrict competition.” Would it give the board’s actions less of a “public nature” if the board members were elected by members of the profession, rather than appointed by the governor based on a slate submitted by the profession? Sounds iffy to me, but in any event, Earles doesn’t hint at it.

And what about Hass and recent Ninth Circuit precedent? I’ve checked out Washington State Electrical Contracts Ass’n v. Forrest, and it doesn’t cast any doubt on Hass (which in any event remains a Ninth Circuit precedent). Here’s a quote from Forrest:

The terms of our remand . . . asked the [district] court to determine the extent of actual state control over the wage-fixing activity of the Apprenticeship Council. It is now apparent, after further briefing and argument, that the Apprenticeship Council may not qualify as a state agency. The council has both public and private members, and the private members have their own agenda which may or may not be responsive to state labor policy. The council is more nearly like the Oregon Board of Medical Examiners which failed in Patrick to satisfy the Supreme Court that it was a governmental agency sufficiently qualified to engage in anticompetitive activity on behalf of the state to avoid scrutiny of the degree of state control over its activities.

Note, first, that this isn’t a holding by the Ninth Circuit, but just a discussion of issues related to a remand. The Apprenticeship Council may not qualify as a state agency; this was all something to be hashed out by the district court in the first instance. But in any event, the idea of asking the district court “to determine the extent of actual state control” is consistent with the Ninth Circuit’s multi-factor test. It’s not enough that the council has some private members; the Ninth Circuit suggested that the degree to which the private members’ agenda “may or may not be responsive to state labor policy” is relevant. So characterizing Forrest as calling Hass into question is way overreading the case.

But let’s also go back to first principles. Ultimately, it’s clear that the different approaches I’ve sketched above are in fact different approaches. They use entirely different factors: the cursory view just looks at the “public nature” of the acts or the formal labeling of the entity by state law; the intermediate view looks at a laundry list of factors; the Fourth Circuit view looks at just private membership and political accountability. In terms of black-letter law, these tests are different, regardless of whether one can show that they would lead to different results in one particular case. That’s what would make Supreme Court review useful here. Different tests are different because they lead to different results in a ton of cases, many of which never reach the appellate courts, and have an ex ante effect on how states structure their regulation. Given that the issue is important enough to make a difference in this case, and given that different circuits have tests that are clearly phrased in terms of different factors, a Supreme Court resolution would be highly useful in clarifying when these boards are potentially liable and when they’re not.