My co-author, Dan Ludwinski, and I have a new research letter in Health Affairs Scholar on Section 1332 reinsurance waivers for the ACA individual health insurance markets. We wanted to know what states thought they were buying with the waiver. So we read all of the applications and used long division. There is a lot of variation in the cost of state funds per newly projected covered life.
Section 1332 waivers are a voluntary program in the ACA that have been adapted by all sorts of states. The dominant waiver is reinsurance. In these waivers, which I’ve written about with Coleman Drake and Ezra Golberstein, states add some state funds to the pool of funds that are available to pay some high cost claims. This lowers premiums as premiums are no longer 100% responsible to cover all the claims. Lower premiums reduce federal subsidy costs. Those federal savings get passed back to the state for the reinsurance pool.
States have to file waiver applications that demonstrate with actuarial analysis that the waivers meet the four criteria for approval.
- Coverage at least as comprehensive as the no waiver coverage
- Coverage at least as affordable as the no waiver coverage
- Coverage for at least as many people as the no waiver enrollment [my emphasis]
- Does not increase the federal deficit
We don’t look at the comprehensive, affordable or federal deficit criteria in this paper (My Georgia paper looks at affordability and enrollment.) In previous work with Drake and Golberstein, we learned that every state projected that the only enrollment impact of Section 1332 waivers would occur among individuals who were unlikely to receive premium subsidies as their incomes were over 400% Federal Poverty Level. Dan and I look at what states think they the cost of buying an additional enrollment in terms of only state funds.
We are not analyzing what actually happened. We just look at the waiver applications and figured out the average cost of a projected new enrollee.
And it is all over the place. Some states think it is pretty cheap to cover a new person. Some states think it is extremely expensive to cover a new person. If we bring in Goldin et al as a rule of thumb/back of the envelope, the low cost states likely see reinsurance as a cost-effective public health intervention while the most expensive states we would be wary about cost-effectiveness of the intervention with the proviso that what the states project actually occurred.
There is a growing body of evidence that reinsurance in the ACA context is mostly a transfer payment with minimal market changing impacts and potentially negative subsidized enrollment effects. States have tremendous variation in the cost of what they think they are buying and we need to understand what is driving this variation to figure out how much and if these programs are actually worthwhile.
Ohio Mom
Aren’t we afraid this question will soon be moot?
I can’t help but think the attack on Medicaid is mostly an attack on Obamacare, for obvious reasons.
David Anderson
@Ohio Mom: No, most of the attacks on the ACA are cover for attacking Medicaid. Repeal and Replace in 2017 was mostly a CUT MEDICAID bill.
Ohio Mom
@David Anderson: I am in over my head. I thought the funding for the ACA expansion — the federal money given to states to allow their residents to join the Marketplace — was a form of Medicaid.
the pollyanna from hell
You “used long division.” As a qualified numerical analyst I resent you cheating the guild of our customary bribe!
Ohio Mom
@Ohio Mom: Can states continue to afford having Marketplaces without federal subsidies?
They certainly won’t be able to serve the populations currently supported by other aspects of Medicaid at their current levels.
The cuts in funding would make the Waiver program for the DD population unrecognizable (translation: Ohio Son would most likely be booted).
I don’t know enough about Medicaid support for nursing homes, mothers and children, and whatever else to begin to imagine how they would change.
David Anderson
@Ohio Mom: Nope, I can see where you’re coming from, but the states per se don’t get the ACA individual market money. That is directly transferred from the feds to insurance companies once they validate a monthly enrollment.
EB
Were states only required to propose a single point “estimate for first year state costs and incremental enrollment gains”? I would have expected that they’d be required to report a range or distribution. That would make a between-state comparison more reasonable.
Also, I didn’t quite understand why one of MN’s applications was excluded. It wasn’t excluded based on the “coverage for at least as many people as the no waiver enrollment” requirement you were look at, was it?
David Anderson
@EB: We excluded Minnesota’s 2nd application as there was no central statewide estimate of impact; everything was at the rating area level.
There is a single estimate with no CI required. It is a CHOICE.