If all raisin producers got together in a room and privately agreed to limit the quantity they produced in order to raise price, this cartel would be a per se violation of the Sherman Antitrust Act. The agreement not only would be unenforceable but also would carry significant civil and criminal penalties.
But what if the state of California created a Raisin Control Board charged with imposing the same regime by statute—perhaps after lobbying by these same raisin producers? This might be worse than the private agreement. Private agreements often break down (to the benefit of consumers) because some of the producers cheat on the deal by producing more than the agreed-on quantity or undercutting the agreed-on price, or because new producers enter the market to take advantage of the high prices. But the statutory regime would be able to control these “problems” with the force of law, and would continue until repealed by the legislature—i.e., possibly forever.
Nonetheless, in Parker v. Brown (1937), the Supreme Court held that the Sherman Act had nothing to say on the subject. The legislature that passed the Act in 1890 surely didn’t mean to control states’ sovereign activity, even if it was anticompetitive; moreover, a due respect for federalism counseled against preempting state policy in this way. This was the genesis of “state action” immunity to federal antitrust law.
Since 1937, the Supreme Court has explained in greater detail what it takes for state governments’ actions to be immune from antitrust law. What if the state authorizes private individuals to engage in price fixing? Surely this isn’t immune: states can’t invalidate federal law. For private action to be immune, the Court held in California Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc. (1980), it has to be both (1) clearly authorized by state law and (2) actively supervised by the state. This two-prong Midcal test remains the law today. Moreover, the Supreme Court has stated on numerous occasions that state-action immunity—this judge-made exception to the Sherman Act—should be applied narrowly.
But what if the state authorizes a municipality to engage in anticompetitive activity? There, the Court held in Town of Hallie v. City of Eau Claire (1985), the first prong of Midcal still applies: the municipality isn’t immune from federal antitrust law unless its acts were clearly authorized by state law. But the second prong isn’t necessary, because when a political entity like a municipality is involved, we aren’t that concerned about purely private profit-seeking. Moreover, the Court suggested, the rule is likely the same for state agencies.
But what if the state agency is composed of active market participants? The North Carolina Board of Dental Examiners was a state regulatory agency, but was also mostly composed of practicing dentists. Is this more like a private party (which doesn’t get antitrust immunity unless it can satisfy both prongs of Midcal) or more like a traditional (i.e., public) state agency (which can get immunity after satisfying only the first Midcal prong)?
In N.C. State Board of Dental Examiners v. FTC (2015), the Supreme Court held that when a board is dominated by active market participants, the harms of anticompetitive self-dealing targeted by the Sherman Act are present in full force, so the full two-prong Midcal test must apply before the Board can claim immunity. (This case, before it got to the Supreme Court, was discussed in this blog post from July 2013, as well as this policy brief from May 2014.) In particular, such market-participant-dominated boards must be actively supervised by the state if they want to not have to worry about being sued for violations of federal antitrust law.
* * *
The N.C. Dental decision was a blow in favor of competition. Whatever one thinks of antitrust law generally, government-sponsored anticompetitive activity should be at least as tightly controlled as private anticompetitive activity. (Admittedly, some might be uncomfortable, on federalism grounds, with federal interference with how a state chooses to regulate.) But it raised as many questions as it answered. What does it take for a market-participant-dominated agency to be “actively supervised”?
A case quickly arose to test the contours of active supervision. Teladoc, Inc. is a provider of “telehealth services”; it provides health-care services by phone or video, instead of by traditional in-person consultation. Typically, employers contract with Teladoc for a subscription fee; its employees create personal accounts and upload their medical records. These individuals can then contact Teladoc at any time and be connected with a doctor trained in phone or video treatment and diagnosis. (This doctor is of course licensed in Texas and has to abide by the same ethical rules as other Texas doctors.) The doctor talks with the patient and consults the patient’s records; the doctor’s medical advice can include prescribing medications or referring the patient to a physical doctor or an emergency room. Teladoc’s services cost a fraction of traditional doctors’ services and are especially useful for people who work in remote areas where doctors are in short supply.
In Texas, the practice of medicine is regulated by the Texas Medical Board, which is dominated by active market participants—doctors of one kind or other. In 2010, the Board amended its regulations to provide that doctors cannot prescribe medications to patients without a face-to-face visit. The result was, in effect, to force Teladoc to shut down its operations in Texas.
Teladoc sued in federal district court, alleging (among other things) that the Board had violated the Sherman Act by excluding Teladoc from the market. First, the parties agreed to ignore the question of immunity and just ask the court to decide, on the merits, whether the Board’s actions violated the Sherman Act. In May 2015, the court determined that there was indeed a Sherman Act violation. It was clear that the Board’s regulations had the effect of limiting supply and increasing price. Of course, in such cases, the Board is allowed to offer a pro-competitive justification for these restrictions: the concern that medical practice without face-to-face consultations would decrease the quality of medical care. But the district court concluded that the evidence for this proposition was scant and anecdotal, and was rebutted by contrary evidence produced by Teladoc, including a study showing positive health outcomes from the use of Teladoc by a large employer in California.
Thus, everything rested on the question of immunity. If the Board was immune, it could safely ignore the court’s ruling that its actions violated the Sherman Act. In a second decision, issued in December 2015, the court reached this question, and analyzed whether the Board was immune. Given N.C. Dental, the main question was whether the Board was “actively supervised” by the state.
Usually, when the Supreme Court imagines active supervision, it’s talking about supervision by some official in the executive branch of state government. If a disinterested executive official signs off on every Board decision and endorses its merits, then one can comfortably say that the Board’s decision is that of the state, and then immunity properly applies. This is, roughly, the way that Georgia has reacted to the N.C. Dental decision; now the Governor must sign off on any contested decision by a market-participant-dominated board.
But such review doesn’t exist in Texas, so the Board’s most substantial argument focused instead on state-court judicial review. Whenever any agency adopts a new rule or interpretation of law, any affected party can sue in state court, arguing that the agency is violating the law. The Board argued that this state-court judicial review was sufficient to constitute active supervision. The district court disagreed, therefore holding that the Board wasn’t immune. The Board then appealed to the Fifth Circuit (the regional federal appellate court that covers Texas).
* * *
At the Fifth Circuit, a group of 55 antitrust and competition scholars—mostly law or economics professors—filed an amicus brief agreeing with the district court’s (and Teladoc’s) analysis and disagreeing with the Board’s arguments (more specifically, disagreeing with a better version of the Board’s arguments, as reformulated on appeal by the Texas Solicitor General). These scholars included Rebecca Haw Allensworth of Vanderbilt, Aaron Edlin of UC Berkeley, and Einer Elhauge of Harvard, the authors of an antitrust professors’ Supreme Court amicus brief in N.C. Dental (which I joined), as well as other noted scholars like antitrust treatise author Herbert Hovenkamp. I was the main author of that amicus brief.
First, we argued that state-court judicial review can’t confer immunity if it occurs only after costly litigation. State courts won’t review a rule that nobody challenges; and affected firms might decide that the cost of a legal challenge is just too great. Even worse, a Board rule might be a disguised form of cartel enforcement (as in the raisin case in the introduction), so perhaps no firm is even interested in challenging the rule. Dispersed consumers may likewise find that challenging the rule isn’t cost-effective for them. This amounts to “the mere potential for state supervision,” which the Supreme Court has held isn’t good enough. It doesn’t help that judicial review often doesn’t even occur until after injury is suffered—which additionally discourages firms from challenging these rules in court.
Even worse, we argued, state-court judicial review is deferential. This is an issue familiar to anyone who studies judicial review of federal agencies, which is broadly similar to its state counterpart. Say the FCC writes a regulation to implement the Communications Act—suppose it interprets the federal prohibition against broadcasting indecency to include the broadcast even of “fleeting expletives,” as when Bono said at the MTV Music Awards that “this is f***ing awesome.” The FCC uses this new legal interpretation to impose liability on a TV station, and the TV station sues, arguing that this is the incorrect interpretation of “indecency.”
Under federal administrative law, it’s not enough to convince the federal court that this interpretation is incorrect. The federal court has to be convinced not only that the agency was incorrect, but that it was unreasonably incorrect—in other words, that the agency’s decision was “arbitrary and capricious” or an “abuse of discretion.” When broad discretion can be exercised (or an ambiguous statute can be interpreted) in several ways that are not unreasonable, the Supreme Court has held that a reviewing court’s job is to uphold any of these reasonable choices, and not to substitute its own policy judgment for that of the agency. In federal administrative law, this is called the Chevron and State Farm doctrines, and the Texas Supreme Court has similar doctrines for Texas agencies.
Is this sort of judicial review “active supervision”? The point of active supervision, as elaborated by N.C. Dental, is to determine whether a challenged action by an agency “accords with state policy” as determined by disinterested state officials. But the statutes involved here involve huge amounts of discretion; one of the statutes, for instance, requires that doctors “practice medicine in an acceptable professional manner consistent with public health and welfare.” Does this statute require examinations at an “established medical site”? Does it authorize disciplinary action against those who prescribe drugs based on an “online or telephonic evaluation by questionnaire”? There are many possible implementations of this statute that are not unreasonable. But judicial deference means that, in the presence of this broad grant of discretion, judges do not ask whether the agency’s choice reflects state policy; instead, judges allow the (self-interested) agency to set its own policy. This is actually the antithesis of active supervision.
The Board also argued generally that Texas law contains many features that, in its view, reduced its own possibilities for self-dealing and enhanced its political accountability. These include appointment and removal of Board members by the governor, the fact that the Board members are specialists from different fields, good-government laws, reporting requirements, and legislative oversight. None of these features actually constitute active supervision: The active supervision requirement demands not just generalized accountability, but actual oversight of the specific action being challenged. Rather, the Board argued that these features “reinforce[]” or “buttress[]” active supervision; because all these features together minimize the risk of self-dealing and maximize accountability, the court need not enforce the supervision requirement very stringently.
We argued that several of these details, like legislative oversight, were highly speculative, weak forms of oversight. We also argued that this sort of “sliding scale” of active-supervision scrutiny, where a court calibrated the stringency of the active-supervision requirement to how strong it thought the risk of self-dealing was in any particular case, would be judicially unadministrable.
But more importantly, we argued that these details were irrelevant to whether there is sufficient active supervision. In N.C. Dental, the Supreme Court did consider the extent of self-dealing and accountability to be important. The Court considered these factors important to the threshold question of whether to require active supervision in the first place—not to the subsequent question of whether there was active supervision. “Self-interest,” we wrote, “determines whether a Board needs supervision, not whether it is supervised.” The N.C. Dental Court painted with a broad brush, explaining that the risk of self-dealing was the problem that infected all market-participated-dominated agencies—which is why it required that the second prong of Midcal be satisfied. Because the Board had already conceded that (as an obvious consequence of N.C. Dental) it needed supervision, there was no further cause to consider these institutional details.
* * *
It would have been nice to have had our arguments endorsed in a precedential opinion by the Fifth Circuit. But Teladoc instead gained a victory of a different sort. In October 2016, perhaps moved by the arguments in our amicus brief—as well as those in other amicus briefs (the Cato Institute also filed a brief, which discussed other issues such as the right to earn a living)—the Board withdrew its appeal. This leaves standing the district court opinion in favor of Teladoc and against the Board’s antitrust immunity. That opinion has no precedential value, but it does allow Teladoc to continue its operations in Texas, and it may have some persuasive value if the same issue arises in other states. The trend is thus in favor of keeping a tight lid on market-participant-dominated agencies, and requiring that, if a state wants these boards to be immune from antitrust liability, it had better subject them to real, disinterested oversight.
Alexander “Sasha” Volokh is an associate professor of law at Emory Law School. An archive of his previous Reason.org articles is available here.