Monday, July 06, 2015

Saved from the Medicaid Gap

Prompted in part by my observation last week that at least a third of Florida's 1.4 million private plan enrollees on healthcare.gov would have been Medicaid-eligible if the state had accepted the ACA Medicaid expansion, Richard Mayhew poses a question (or rather, elaborates on one posed in a comment on my post):
Liberal technocrats have been assuming that the states which refuse to expand are giving up massive amounts of money and thus economic growth by refusing to expand Medicaid will eventually expand.  However, are we accounting for the additional cash flow coming in as premium and cost sharing subsidies for people making between 100% and 138% Federal Poverty Line.
Leaving aside the financial question, we can make some reasonable estimates as to what percentage of those who would have been eligible for Medicaid had their states not refused the expansion are now in subsidized private health plans purchased on healthcare.gov. (All but one of the states that refused the expansion use the federal exchange).

On one side of the equation, we have Kaiser's state-by-state estimates of how many people fall in the Medicaid gap -- that is, have household incomes under 100% of the Federal Poverty Level (FPL) but are shut out of Medicaid in their state. On the other side is the number of subsidized private plan buyers in non-expansion states who have incomes in the 100-138% FPL range -- those who would have been eligible for Medicaid if their states had accepted the expansion.
We don't have exact tallies for those with incomes between 100% and 138% FPL. But HHS did report that 47% of private plan buyers in nonexpansion states had incomes in the 100-150% FPL range, versus 22% in states that implemented the Medicaid expansion.  In expansion states, "100-150% FPL" effectively means "138-150 FPL," since those under 138% FPL are Medicaid-eligible (and ineligible for private plan subsidies). By comparing the income breakouts for expansion vs. nonexpansion states, I concluded that about 67.8% of those in the 100-150% FPL range in nonexpansion states would have been eligible for Medicaid had their states expanded. At present that comes out to 1.6 million private plan holders in the 21 non-expansion states using healthcare.gov.

Kaiser estimates (as of April 17, 2015) that 3.7 million people remain stuck in the Medicaid gap. We might estimate then that about 30% of those who would have been Medicaid-eligible had their states accepted the expansion (1.6m out of 5.3m) are currently in subsidized private plans. Probably 90% of those are in silver plans enhanced by Cost Sharing Reduction, which raises the plan's actuarial value to 94% at that income level. [UPDATE, 7/7: It occurs to me that this estimate excludes those in the 100-138% FPL range in nonexpansion states who declined to buy subsidized QHPs and remain uninsured, This spring, Avalere Health found that 76% of those in the 100-150% FPL range who were eligible for subsidized private plans did in fact buy them. If that's accurate, the percentage is probably somewhat higher below 138% FPL, as silver premiums are capped at 2% of income below that threshold.]

HHS enrollment data released last week makes it possible to estimate state-by-state the percentage of private plan enrollees who would have been Medicaid-eligible had their states accepted the expansion. That estimate can be set next to Kaiser's estimate for those who fall in the Medicaid gap in that state. These estimates come with three caveats: 1) we know the percentage of enrollees in each state with incomes in the 100-150% FPL range, but not the percentage at 100-138% FPL, i.e., Medicaid-eligible. For the latter, I use my estimate of 67.8% of those in the wider band, which is probably conservative.  2) The new enrollment breakouts are based on enrollment as of mid-February. In June, HHS released updated state-by-state enrollment numbers (as of March 31) that showed significant attrition, dropping overall enrollment from 11.7 million to 10.2 million. We don't know whether the attrition was proportionate across income groups. 3) Kaiser's Medicaid gap estimates are also only just that -- estimates.  Some states such as Michigan that have expanded Medicaid belatedly have exceeded enrollment targets.

All that said, here are snapshots based on my recent look at the new enrollment stats for Florida, Alabama and Mississippi. Links are to the recent posts.

Florida: As of mid-February, 53.6% of Florida enrollees had incomes in the 100-150% FPL range. That comes to about 758,440 of the states's March 31 enrollment total. Of those, by my formula (67.8%), about  514,200 were likely in the Medicaid eligibility range. Kaiser estimates 669,000 Floridians in the Medicaid gap. That would place a whopping 43.4% of Florida's potential Medicaid-eligibles in private plans at present. Given Florida's nation-leading private plan enrollment surge in 2014, that rather outsized proportion is perhaps not implausible.

Alabama: As in Florida, 53% of the state's private plan enrollees had incomes in the 100-150% FPL range. As of March 31, Alabama had 145,763 total enrollments through hc.gov. Assuming that 77,254 of them were in the 100-150% FPL range, I'd estimate 52,378 Medicaid-eligible enrollees. Kaiser estimates 176,000 Alabamans in the Medicaid gap. We can therefore posit that about 23% of those who should have been eligible for Medicaid are now in private plans.

Mississippi: A whopping 58.6% of the state's private plan enrollments as of February were in the 100-150% FPL income range. The state suffered much larger-than-average attrition: enrollment was down from 104,538 in mid-February to 80,011 by March 31. If that large disenrollment was proportionate across income groups, there would now be about 46,900 enrollees from 100-150% FPL and, according to our formula, about 32,210 who would be Medicaid-eligible had the state expanded. Kaiser estimates 107,000 in the Mississippi Medicaid gap. That suggests that 23% of those who should be eligible for Medicaid are in private plans.

I will continue these state estimates in future posts.

No comments:

Post a Comment