Arkansas Proposal Still Fiddly in Practice

by Adrianna McIntyre & Karan Chhabra

Still following Arkansas Medicaid expansion developments? Us too!

On Thursday, Sara Rosenbaum, Chair of the Department of Health Policy at George Washington University School of Public Health, published a technical and authoritative summary on the legal grounds for Arkansas’ expansion. It mostly gels with our take last weekend. We gather two clarifications:

  • Rosenbaum treats the proposed HHS regulation (rule 435.1015) as uncontroversial—it doesn’t conflict with the Medicaid statute. We defer to her judgment here.
  • Though 435.1015 might be valid, she also finds it vague. As we excerpted in the original post, to use Medicaid funds for private insurance, the cost of the new plans “must be comparable to the cost of providing direct coverage under [Medicaid].” Rosenbaum takes issue with comparable”—what does that mean, empirically? We suggested “at or below the cost of Medicaid,” but perhaps HHS intended something different. For clarity, we use that term instead of “cost-effective” now.

If HHS creates a narrow operational definition of “comparable” in rule 435.1015 and the Arkansas proposal is passed without meeting that criteria on paper, HHS is contradicting it’s own regulation. This would be—legally—a problem [1].

So, the Arkansas deal is on sound footing as long as plans offered on the exchanges are of “comparable” cost, however HHS defines that. We never disputed this—but we have trouble understanding how legislators think they’ll meet that threshold. There is good reason to think that plans on the exchanges will cost substantially more than a Medicaid expansion would. Massachusetts is useful for lessons like this; it’s had an insurance exchange in place since 2006. The average cost of an individual exchange plan was $5,143 in 2009. That same year, their average Medicaid expenditure per nonelderly/nondisabled adult was $2,965. See a difference? For a coherent breakdown of why private costs aren’t likely to come in line with Medicaid, see Austin Frakt’s recent analysis over at The Incidental Economist.

We still need to know how Arkansas legislators intend to see plans cut without violating essential health benefits or any other Medicaid/exchange requirements. Some have cited savings from churn reduction, but I haven’t seen evidence that suggests savings would scale in a way that makes sense. David Ramsey usefully points out that total administrative costs for a conventional Medicaid expansion—churn would only account for a fraction of this—would cost Arkansas $14 million annually; that’s small potatoes compared to projections for the total overall cost of the proposal. Moreover, a “conventional” expansion already has some buffer for those who cycle in and out of the 100-138% FPL range [2].

There’s also a “risk pool” question. As a general trend, people of higher incomes tend to enjoy better health status—which tends to make them cheaper to insure. The Arkansas Surgeon General believes that including the newly-eligible in the insurance exchanges increases the overall risk of those pools—that’s not to say that they’re very sick, but they might, in the aggregate, have poorer health than their 138-400%FPL exchange counterparts. Insurance companies usually hedge against higher risk by raising premiums. Except, now, we’re ostensibly increasing overall risk while asking these same insurance companies to decrease their costs—to levels approaching Medicaid. That’s going to be tricky.

From a practical (if not legal) perspective, we also have to consider how those costs—comparable or not—are covered. Recently publicized emails among AR state leaders reveal that, assuming this deal is more expensive, they could make up state budget shortfalls through cuts to the existing Medicaid program’s benefits—by adding copays, shifting enrollees to other programs (like the exchanges), cutting payments to providers, and eliminating funding for uncompensated care and Community Health Centers. Before we raise a ruckus, let’s be clear that emails are not the same thing as policy. Still, it leaves us concerned: in essence, the effect is to take from the most poor (currently in Medicaid) to benefit the less poor (those in the Medicaid/insurance expansion). Again, this is far from actual policy, and details are sparse, but nonetheless it’s troubling. (More information in this superb, though gated report.) (Addendum: See this new piece from David Ramsey of the Arkansas Times for more nuanced detail about these cuts; notably, they are “transfers and savings that would have happened under the original Medicaid expansion plan.”)

There is, of course, an alternative scenario: HHS could define “comparable” generously enough to make this feasible. That would mean this gets expensive for the nation—fast. According to emails obtained under FOIA by Charlie Frago of the Arkansas Democrat-Gazette (paywall), a top CMS official told Arkansas and Ohio “not to lose sleep over cost comparability. They [CMS] apparently are comfortable in their interpretation of the rules on this.” (Addendum for clarity: these are Arkansas state officials characterizing discussions with CMS, not emails from CMS officials.)  New York and Texas are also toying with this idea, and Rosenbaum expects that it’d “spread like wildfire.” We’ll spare you a rehash of our analysis; we covered the implications pretty thoroughly last week.

Curious creatures that we are, we have some outstanding questions.

  • If Arkansas’s proposal is “comparable” on paper, but not in practice, who’s on the hook for the difference? Arkansas or the federal government? The statutory language doesn’t create a “hard cap,” and the regulations don’t seem to clarify .
  • Do all plans on the exchange have to meet the comparability criteria, since newly eligible would be empowered to “shop around” like any other consumer?
  • If so, won’t that discourage variety among different exchange plans, since it  creates a price ceiling? Might it discourage insurers from participating on the exchange at all?

Footnotes (Yes, more footnotes!)

1. This actually creates a second tricky question: if HHS violates its own regulation, who has standing to sue HHS? This is something I touched on briefly in my correspondence with Timothy Jost. I can’t, as a taxpayer, point to the agency and cry foul (well I can, but it wouldn’t have any legal force). Someone could theoretically sue because they were forced to seek private insurance over Medicaid, but they might have a hard time demonstrating “harm,” if that’s necessary. Perhaps there’s a violation of contract interpretation that insurance companies could probe? I’m not sure.

2. People who fall between 100-138% FPL qualify for Medicaid or for subsidies on the exchanges, so there’s room to maintain coverage if one’s income fluctuates into this range. That means churn will only occur where someone with >138% FPL falls below 100% FPL and vice versa. (Addendum: There may be more complexity to buffer offered (or not) by this overlap. Refer to the comments below for more information.)


Click here to subscribe, for the latest news in and around health policy. 

Adrianna works in clinical research and is a graduate student in public policy & public health at the University of Michigan. Follow her on Twitter @onceuponA.

Karan is a first-year student at Robert Wood Johnson Medical School and Duke graduate who previously worked in strategic research for hospital executives. Follow him on Twitter @KRChhabra.


12 thoughts on “Arkansas Proposal Still Fiddly in Practice

  1. […] to standard Medicaid. But what does that mean? I’m not going to get into it because Adrianna McIntyre and Karan Chhabra already have in a new post. Just read […]

  2. Sarah says:

    Regarding footnote #2, I believe that in general a person is not eligible for tax subsidies on the exchange if they have access to affordable coverage elsewhere either through Medicaid or an employer. E.g. if your employer offers individual coverage that costs less than 9.5% of your income then you can’t get a tax subsidy. The same for Medicaid. This is the source of the recent worry about family coverage because employers could offer an individual policy that meets this affordability test but family coverage for the rest of the household could be much more expensive but they would not be eligible for tax credits on the exchange. A strict reading might say as soon as you are Medicaid-eligible you are no longer tax credit-eligible.

    The proposal I’ve seen from MACPAC for reducing churn is to structure Medicaid with 1 year presumptive eligibility for everyone. That’s currently done in some states for children and for CHIP but usually not for adults who have to recertify more often. They often face coverage gaps now, not because they make too much money but just through administrative slipups.

    I’ve less sure how eligiblity would work the other direction, e.g. buying an exchange plan but then losing a job and becoming eligible for Medicaid. You can keep the exchange plan but you might automatically loose the tax credits that make it affordable. And Medicaid is cheaper so you would have incentive to switch.

  3. Ethan says:

    On FN 1, perhaps someone could sue, for example, for not being availed of the due process protections afforded by Medicaid because presumably beneficiaries who purchase through the exchange do not have the same protections as do traditional Medicaid beneficiaries for appeals, equal access, etc. Also,

  4. Ethan says:

    Sarah, true on FN 2, which raises for me a question…That means that under the law Arkansas would have to force newly eligibles to purchase through the Exchange, i.e., they couldn’t have the option of allowing them to enroll in a traditional Medicaid program because if they are eligible then under the law they wouldn’t be eligible for a subsidy no matter who pays for it.

    • While I don’t dispute the claim about coverage offered through employers, all of the experts I’ve conversed with — people in a position to know — have indicated that the 100-138%FPL overlap exists with the intention of buffering churn. If there wasn’t a purpose to the overlap, it wouldn’t exist (either Medicaid would only extend to 100%FPL, or exchange subsidies would start at 138%FPL).

      I’m open to being wrong, but can someone point me to documentation suggesting that people who fall between 100-138% (without employer sponsored coverage) must default to Medicaid if it’s cheaper?

      • Ethan says:

        The 100-138% eligiblity for tax credits is likely there for several reasons: (a) it is a relic of a previous version of the bill that would have allowed for choice between the exchange and Medicaid (I believe Baucus was on this); (b) it has the appearance of mitigating churn; and (c) many states are at 100% or above for current Medicaid eligibility (though not AR of course). The problem with the churn idea is that the law prohibits those eligible for public programs–Medicaid, CHIP, etc.–from receiving tax credits. Don’t have a citation to the provision but Kaiser notes it here:

        I don’t think, though, that it’s a question of whether it’s cheaper; all I’m saying is that it appears that AR would be unable to offer an option if this is the case. That is, they would have to force newly eligibles into the Exchange because it’s the only option.

      • Eileen Ellis, Michigan says:

        This is a “back door” inference. In the December 10, 2012 Q & A document from CMS that has the discussion of premium subsidy options, there is another question that relates to your topic. The response to question 31 indicates that for a state that does not do Medicaid expansion, individuals with incomes above 100 percent of the FPL who ARE NOT eligible for Medicaid, CHIP, etc will be eligible fore premium tax credits and cost sharing reductions. The implication is that somewhere else CMS has decreed that those eligible for Medicaid must take Medicaid. However, Section 1401 of ACA has no Medicaid exception clause. I believe those betweeen 100% and 133% of FPL are legally entitled to a choice. However I suspect that the “rules engines” for Exchanges will funnel them straight to Medicaid.

  5. Eileen Ellis, Michigan says:

    Ethan’s right about the Kaiser citation, but I don’t see it in the law.

    • Thanks for your clarifications! I missed these developments. I’ll update the footnote with an addendum to refer to these comments for more detail.

    • Mark Regan says:

      The provision excluding people who get Medicaid from getting premium tax credits is section 1401(c)(2)(B), which excludes from the definition of “coverage month” any month during which someone is eligible for minimum essential coverage other than individual insurance. Medicaid is one of the forms of minimum essential coverage other than individual insurance.

  6. […] Subsidies reduce the risk and degree of adverse selection, but nobody knows by how much. It probably varies by state, commensurate with variation in the culture of acceptance of the ACA and its mandate. Of course, the higher the subsidy, the more likely an individual is to enroll when healthy. So, another way to reduce adverse selection and its cost is to allow exchange access to people who would be subsidized at an even higher rate. The Medicaid expansion population does the trick. So, there’s a selection-based argument for the Arkansas style “private option.” […]

  7. […] didn’t get the approval that Arkansas did to allow federal money to be used to buy private plans for Medicaid-eligible people on the […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: