Is there any remaining "upper coverage gap" in nonexpansion states?

 Note: All xpostfactoid subscriptions are now through Substack alone (still free), though I will continue to cross-post on this site. If you're not subscribed, please visit xpostfactoid on Substack and sign up.

What's left of the upper coverage gap?

For some years, the Kaiser Family Foundation (KFF) has published estimates of people in the “coverage gap” in states that have refused to enact the ACA Medicaid expansion — that is, uninsured people who would be eligible for Medicaid if the state enacted the expansion. (As of now, 10 “nonexpansion states” remain.) In some briefs, KFF has divided the estimate into two groups: uninsured below 100% FPL, who in most cases are ineligible for government-supported insurance in nonexpansion states, and uninsured in the “upper coverage gap” — those with income in the 100-138% FPL range, who are eligible for subsidized marketplace coverage.

KFF estimates of the uninsured are based on the Census Bureau’s American Community Survey, which generally lags ACA enrollment data by two years. In April 2021, I noted that marketplace enrollment at 100-138% FPL in nonexpansion states in 2020, laid beside KFF’s estimates of uninsured in that cohort as of 2019, indicated that a bit more than half of those eligible for marketplace coverage in this income bracket had enrolled in plans. In a followup post, I noted that enrollment gains in 2021 at 100-138% FPL (a 17% increase in the 12 nonexpansion states then remaining) should be making inroads on the uninsured population in the upper coverage gap.

At the time of that writing, the American Rescue Plan Act (ARPA) had just rendered benchmark silver coverage with Cost Sharing Reduction free at incomes up to 150% FPL, and an emergency Special Enrollment Period, which extended through August 15, 2021, was just gathering steam. Since then, enrollment at 100-138% FPL (85% of which is in the remaining nonexpansion states) has more than doubled, from 3.3 million as of the end of the Open Enrollment Period (OEP) for 2021 to 6.9 million in OEP 2024.

Just last month, KFF updated its estimates of the uninsured in the coverage gap, including the upper coverage gap (100-138% FPL). The estimate, again, is based on the American Community Survey and only goes through 2022. Since that point, enrollment gains the 100-138% FPL income bracket exceed — in fact, more than double — KFF’s 2022 estimate of uninsured in the bracket. Enrollment figures below are from the Marketplace Open Enrollment Period Public Use Files, 2022 and 2024.

Updating survey results with enrollment figures is a somewhat questionable exercise. Survey results and hard enrollment figures are different kinds of data. The total population in this income bracket in these states has doubtless grown since 2022. Incomes also have grown. The federal poverty level has been sharply adjusted for inflation. The ACS uses a different definition of “household” to calculate FPL (though KFF takes steps to reconcile that difference). Enrollment totals are as of the end of the Open Enrollment Period (OEP), whereas the ACS surveys people throughout the year, asking if they are insured at the time of the interview. In OEP 2024, the Medicaid unwinding — the resumption of Medicaid disenrollments after a three-year pandemic-induced moratorium — pushed about 3 million Medicaid disenrollees into the ACA marketplace, and in nonexpansion states, many of them may have crossed into the 100-138% FPL bracket.

All that said, the marketplace enrollment gains in these states at this income bracket are so large that I think laying those gains beside the uninsured estimate is instructive. In every one of the states tallied above except Wyoming, enrollment at 100-138% FPL in 2024 exceeds the total of enrollment in 2022 plus KFF’s estimate of the uninsured in the income bracket.

One possible factor in this enrollment growth may be that in the course of the pandemic, significant numbers of people with income hovering near or below 100% FPL, the threshold for ACA subsidy eligibility, learned to estimate incomes over the 100% FPL threshold on the marketplace application. Knowledge is precious on this front: most applicants have no idea that there’s a minimum income requirement for subsidy eligibility (which defies reason and decency) — and future income estimates are inherently uncertain and therefore massageable. A rule change in May 2021, mandated by a court order, ensured that HealthCare.gov would not disallow subsidies if the “trusted data sources” it taps indicate an income lower than the estimate (if data sources indicate an income significantly higher than the estimate, the marketplace may require verification and reduce or end subsidies if the estimate can’t be verified).

I have written several times (e.g., here) about the possibility of estimating income optimistically at income levels where income is often fluctuating and difficult to predict. The influx of brokers into the marketplace in recent years, and the Biden administration’s investment in enrollment assistance in HealthCare.gov states (increased tenfold over Trump administration funding, and about 50% over Obama-era levels), may be increasing “threshold awareness” and ways to cope with the cruel and arbitrary effective ban on government aid for health insurance for most adults with income under 100% FPL in nonexpansion states.

The main factor in enrollment growth at low incomes, however, is doubtless the ARPA boosts to marketplace premium subsidies, rending high-AV coverage free to those in the upper coverage gap and beyond.

Update, 11:30 a.m.: I should have added that the Medicaid unwinding may have swelled the ranks of the uninsured as well as marketplace totals. As of December, total Medicaid/CHIP enrollment nationally was down 9 million net from March 2023. While most people in the 100-138% FPL income bracket in nonexpansion states would not be Medicaid eligible, income is volatile, particularly at low incomes.

Also, it appears my brains are turning to fish juice, in that I updated estimates of takeup in the upper coverage gap in April 2022 and April 2023. The 2023 post is based on a KFF update using 2021 ACS data.