On Friday, HHS released a proposed rule that would make a number of adjustments to the rules governing insurance exchanges for 2019. The rule is long and detailed; there’s a lot to digest. Among the most noteworthy changes, however, are those relating to the essential health benefits. They’re significant, and I’m not convinced they’re legal.

By way of background, the ACA requires all health plans in the individual and small-group markets to cover a baseline roster of services, including services falling into ten broad categories (e.g., maternity care, prescription drugs, mental health services). Taken as a whole, the essential health benefits must be “equal to the scope of benefits provided under a typical employer plan, as determined by the Secretary.”

The ACA’s drafters anticipated that HHS would establish a national, uniform slate of essential health benefits. Instead, the Obama administration opted to allow the states to select a “benchmark plan” from among existing plans in the small group market (or from plans for state employees). The benefits covered under the benchmark were then considered “essential” within the state.

At the time, Helen Levy and I concluded that HHS’s approach brushed up against the limits of what the law allowed. We noted, among other things, that the ACA tells HHS to establish the essential health benefits—not the states. And it’s black-letter administrative law that an agency can’t subdelegate its powers to outside entities, states included.

At the end of the day, however, Helen and I concluded that the Obama-era regulation passed muster. Our rationale bears repeating:

Although a federal agency cannot delegate its powers to the states, it “may turn to an outside entity for advice and policy recommendations, provided the agency makes the final decisions itself.” Here, the secretary gave the states a constrained set of options (e.g., choose a benchmark plan from among the three largest small-group plans in the state) and retained the authority to select a benchmark for any state that either does not pick a benchmark or chooses an inappropriate one. As such, the secretary remains firmly in control. Nothing in the ACA prevents her from deferring to states that select benchmark plans from among the few options she has provided. That choice to defer is itself an exercise of her delegated powers.

The Trump administration’s proposed rule would vastly enlarge this Obama-era subdelegation. For starters, the rule would allow a state to adopt another state’s benchmark, or part of a state’s benchmark, as its own. Michigan, for example, could borrow Alabama’s benchmark plan wholesale, or it could incorporate Alabama’s benchmark for mental health and substance use disorder treatment. More significantly, the rule would allow a state to “selec[t] a set of benefits that would become the State’s EHB-benchmark plan.”

You read that right: if the rule is adopted, each state can pick whatever essential health benefits it likes. No longer will it be choosing from a preselected menu; it’ll be picking the essential benefits out of a hat. In so doing, the proposed rule looks like it would unlawfully cede to the states the power to establish the essential benefits.

This extraordinary subdelegation of regulatory authority is subject only to the loosest of constraints: benefits can’t be “unduly weighted” toward any one benefit category or another, and the benchmark must “[p]rovide benefits for diverse segments of the population, including women, children, persons with disabilities, and other groups.” The selected benefits also can’t be more generous than the state’s 2017 benchmark (or any of the plans the state could have selected as its benchmark), but that’s a ceiling, not a floor, so states have lots of room to pare back.

The only meaningful constraint is that the benefits covered by the state’s benchmark must be “equal to the scope of benefits provided under a typical employer plan.” But another portion of the proposed rule would hollow out that requirement:

[W]e propose to define a typical employer plan as an employer plan within a product (as these terms are defined in §144.103 of this subchapter) with substantial enrollment in the product of at least 5,000 enrollees sold in the small group or large group market, in one or more States, or a self-insured group health plan with substantial enrollment of at least 5,000 enrollees in one or more States.

In other words, HHS is saying it will treat as “typical” any employer plan, in any state, that covers more than 5,000 people.

This looks like an innocuous change. It’s not. If the rule is adopted, it means that a single outlier plan can now count as typical, even if it’s way stingier than any other plan in the market. It also makes me wonder if HHS already has in mind some large employer with an unusually narrow health plan—maybe some hospital-based “administrative services only” plan, as Dave Anderson speculates. If so, voilá, the states can all ratchet down their essential benefits to that plan’s level.

I don’t think that’s legal. To know if a slate of health benefits is typical, you have to know something about how many health plans cover those benefits and how many don’t. The proposed rule eschews that comparative inquiry, and instead defines typicality with reference to the number of people who are covered by a single plan. Some random self-insured plan that excludes appendectomies could be treated as typical, even if it’s the only plan in the nation that does so.

In other words, HHS wants to define a “typical employer plan” to include atypical plans—which the agency emphatically cannot do. Yes, plans that enroll 5,000+ people are less likely to be outliers than smaller ones. But in a country as big and complicated as ours, there are bound to be some idiosyncratic quirks even in large plans. Those quirks would all be considered typical under HHS’s rule.

This definitional change, combined with the choose-your-own-adventure option to devise a benchmark, means that states will have wide authority to water down the essential health benefits requirement. Whether that’s good or bad is hard to say. Requiring plans to cover lots of services assures comprehensive coverage, but it also raises the cost of insurance. Because there’s no single “best” way to strike the balance, I think there’s a lot to be said for giving states the freedom to choose for themselves.

Wise or not, however, I’m skeptical that the Trump administration’s effort to hollow out the rule governing essential health benefits is legal. If HHS presses ahead with the rule, it could face tough sledding in the courts.

@nicholas_bagley