The Strange Making of the “Marketplace Stabilization Rule”

On April 13 CMS published the agency’s final “market stabilization” rule.  The proposed rule was summarized by THCB’s editors on February 15, the day it was published, and on March 22 THCB published my essay in which I noted CMS provided no evidence any of the proposed reforms would actually stabilize the state marketplaces.  The final rule, ostensibly a carbon copy of the proposed, finalizes the six proposed changes without, again, providing any evidence these changes will stabilize the markets by increasing enrollment and issuer participation.

Briefly, the final rule will reduce the 2018 enrollment window from three months or to six weeks, or from November 1 to December 15.  The rule narrows the definition of guaranteed availability by allowing issuers to apply re-enrollment payments to outstanding debt.  The rule will require 100 percent verification for enrollees’ attempting to acquire insurance during a Special Enrollment Period (SEP) and places other payment, eligibility and exceptional circumstances restrictions on SEP enrollment.  The rule finalizes an increase in de minimus variation from +/- 2 percent to -4/+2 percent except for bronze plans which increases to -4/+5 percent.  The rule will allow states to determine plan  network adequacy or make a determination using an issuer’s accreditation status.  The rule finalizes a reduction from 30 to 20 percent of plan providers being defined as an Essential Community Provider (ECP).  For plans that cannot meet the 20 percent determination, CMS will allow for a narrative explanation.

In the final rule’s “impact analysis” CMS is again unable to provide any evidence these changes will produce the intended effect.  In the proposed rule CMS stated “on the one hand” premiums could fall but “on the other hand” premiums could increase.  In the final rule, CMS stated, “the net effect of these provisions on enrollment, premiums and total premium tax credit payments are uncertain.”  “Premiums will tend to fall if more young and health individuals obtain coverage . . . .”  “However,” CMS followed, “if changes such as shortened open enrollment period, pre-enrollment verification for special enrollment periods, reduced AV of plans, or less expansive provider networks result in lower enrollment, especially for younger, healthier adults, it will tend to increase premiums.”

What we do know or still know is these changes will likely reduce enrollment or destabilize the markets.  As noted in my March 22 post, those aged 18 to 24 are far less likely to complete an enrollment verification process due to “hassle costs” than those aged 55 to 64.  Expanding actuarial values would, per the Center for Budget and Policy Priorities estimates, increase out of pocket costs for families.  A shorter open enrollment period will likely produce the same result.  In a recent Health Affairs blog post, using Kentucky marketplace data, Paul Shafer and Stacie Dusetzina showed that 25 percent of new enrollments took place in the final week, 60 percent of those eligible for financial subsidies enrolled during the latter half of the enrollment period and those who changed their plans also did so in the latter half of the enrollment period. 1

CMS proposed and went final with this rule because the agency believes it must act.  In the final rule CMS argued “we considered maintaining the status quo,” however, “we determined that the changes are urgently needed to stabilize markets, to incentivize issuers to enter or remain in the market and to ensure premium stability and consumer choice.”

From a certain perspective it’s unsurprising CMS went forward with this rule.  The agency was responding to industry pressure that absent changes, they would withdraw from participating.  The day the final rule was released,  AHIP, not surprisingly, stated, “This final rule adopts some important changes that have been needed for some time in order to improve the functioning of the individual market, and we appreciate those changes. Those improvements include tightening up rules for special enrollment periods, greater flexibility in product and benefit design, and simplified administrative processes.”

What is surprising is the fairly obvious confirmation bias exhibited here.  The Trump administration has stated over and again it believes the ACA is a failure.  The marketplaces are in a death spiral.  If you are biased toward a particular belief you are then prone to interpret information in such a way as to confirm your pre-existing beliefs.  Even if CMS truly believed the agency had to act, what is also surprising is the agency, again, promulgated and finalized a rule lacking any evidence.  Instead, the final rule is based simply on the agency’s beliefs.  CMS’ explanation for these policy changes in sum amounts to “we believe” – a phrase the agency uses over 50 times in the final rule.  In his recent work, The End of Expertise, The Campaign Against Established Knowledge and Why It Matters,” Tom Nichols cautioned when everything becomes a matter of opinion whether policy is made based on weak or no factual basis becomes irrelevant.  Imagine if the FDA approved a drug that the FDA admitted would have an “uncertain” effect.

1. Paul Shafer and Stacie Dusetzina, “Looking Ahead to 2018: Will a Shorter Open Enrollment Period Reduce Adverse Selection in Exchange Plans?” Health Affairs Blog (April 14, 2017), at: http://healthaffairs.org/blog/2017/04/14/looking-ahead-to-2018-will-a-shorter-open-enrollment-period-reduce-adverse-selection-in-exchange-plans/.

Categories: Uncategorized

1 reply »

  1. David, to some extent the CMS vagueness and hedging should be commended as a form of humility in the face of uncertainty. It would be worse if they convinced themselves they had hit on the right path, and used that strong confirmation bias to resist future modifications if things aren’t working out. There is evidence that certain types of changes, in isolation, have certain types of effects in health insurance purchasing, but these changes don’t all move in the same direction and the net effect after interactions is unknown.

    If you increase the premium price when purchasing insurance is optional, those who are healthy will be relatively less likely to purchase insurance. If you require proof that a person is eligible to get insurance outside the open enrollment period, rather than merely an attestation of eligibility, people will be less able to game the system to buy insurance after they have a problem, and the risk pool will be protected (other things equal, in normal circumstances). A health plan with higher cost sharing (lower actuarial value) is likely to be more attractive to a healthy person than to a person with one or more chronic conditions that need regular attention. Smaller networks are less of an issue for healthy people than sick ones, and less an issue for young people than old, because older and sicker people tend to have deeper relationships with specific providers and/or beliefs about the importance of being able to freely choose whoever you want to see by whatever quality standard you want to apply.

    Now let’s say premiums go up by 3% due to medical inflation, plus an additional 4% because of the need to make up for the last year’s risk pool being worse than expected, but reduced by 2% due to lower actuarial values, and by 3% due to smaller, more tightly contracted networks, and by 1% because fewer people can enroll in special enrollment periods. Is that a net increase of 1%? Even assuming you could accurately attribute factors as I have suggested, there are no controlled studies showing the net real impact on enrollment and medical costs (and thus profitability and next year’s premiums) when you modify all these factors at the same time. Oh, and meanwhile the word gets out that there will be no enforcement of an insurance mandate, and the only coverage requirement relates to a short-term penalty for buying insurance after a gap in coverage. What then?

    Some guesses on the net impact on enrollment and the risk pool are better than others, but it is hubris to believe you are doing more than guessing. We are changing too many variables at once in a dynamic policy environment. I don’t think you disagree, I’m just saying: what more could you expect of CMS?

    On the other hand, if something like the AHCA passes and instead of these small percentage point differences we see a doubling or tripling of the premium cost, or out of pocket expense for people between 100% and 400% of poverty, then we can be quite confident of what will happen. The market for individual health insurance will shrink dramatically, and if guaranteed issue is retained, a death spiral will truly begin.