About a month ago, I once again reiterated that the official year over year ACA Open Enrollment Period plan selection drop from OEP 2025 to OEP 2026, which currently stands at around 1.26 million people (23.06M in 2026 vs.
Your post actually got me thinking about "effectuated enrollment", and to double check that that doesn't mess up the thing that I reported on here (and I know you are aware of):
where my conclusion: "ACA Exchanges: I get 3.5 million More (4.7 million) non-returned in 2026 (vs. 1.2 million in 2025)" is still a correct conclusion, in particular the 3.5 million increase, and shows how many people who had exchange coverage last year looked at the prices on the ACA exchanges for 2026, and said "too expensive now--better drop". (I.e., the amount more in the number of people making a decision not to continue coverage for 2025 is 3.5 million.)
(I checked the definitions of the KFF vs CMS numbers in the I was using, and each of the 4.7 million and 1.2 million non-returning numbers I calculated are off a smallish bit due to non-effectuation and people enrolling or unenrolling as each year goes on, but the numbers are off only by the same small, roughly identical, amount, due to identical causal factors. So my "3.5 million more" is fine!
Obsessively, I also tossed on a note on the issue being a non-issue on my post!)
--
-->However, Charles, if you see this comment, you mentioned you have non-effectuation numbers handy since you just used them in an analysis.
If you can plop down a typical percent non-effectuation, say 2024 or 2025 when there is no issue of changing subsidy schedule making the year non-typical, that might help my understanding of the whole business.
For you and any other wonky people, the basis of what I did is I looked for proportion returning for coverage, to compare 2025 to 2026. (Aware that non-effectuation will not be available for 2026 at this early date. So, just comparing signed up without effectuation should give a lower bound, at least, on how many people were turned away by the new higher prices.)
Now, with the CMS data sources for status as of early 2025 and early 2026
It has "number of returning consumers". But, it doesn't have number "eligible to return": which the footnote says is those who have exchange coverage through an exchange through END of the prior year.
If it it did have number "eligible to return", then we just subtract returning-(eligible to return) for each of 2025 and 2026, and we're done.
However, since I didn't have number of "those who have exchange coverage through an exchange through END of the prior year", I went with the KFF-plotted number enrolled for the prior year as "eligible to return", which is slightly different. But, the causality pattern on the difference between the number wanted and the number used is the same for the two years involved, which is 2024 and 2025, which are both years of the expanded subsidy.
That is, the 1.2 million for 2025 non-returning is a little off, and so is the 4.7 million non returning for 2026, but they're off by roughly the same amount. And so the difference, 3.5 million more 2026, is very close to correct. (On not-considering-effectuation open-enrollment declining-re-enrollment people.)
Though, I think that this year, as it catches those chased away by the higher prices, and separates ("unconfounds") those people from those other people deciding to enroll for the first time despite the higher prices, has a special usefulness that I don't know of any other way to capture.
Thanks, Charles.
Your post actually got me thinking about "effectuated enrollment", and to double check that that doesn't mess up the thing that I reported on here (and I know you are aware of):
https://normspier828307.substack.com/p/aca-exchanges-i-get-35-million-more
where my conclusion: "ACA Exchanges: I get 3.5 million More (4.7 million) non-returned in 2026 (vs. 1.2 million in 2025)" is still a correct conclusion, in particular the 3.5 million increase, and shows how many people who had exchange coverage last year looked at the prices on the ACA exchanges for 2026, and said "too expensive now--better drop". (I.e., the amount more in the number of people making a decision not to continue coverage for 2025 is 3.5 million.)
(I checked the definitions of the KFF vs CMS numbers in the I was using, and each of the 4.7 million and 1.2 million non-returning numbers I calculated are off a smallish bit due to non-effectuation and people enrolling or unenrolling as each year goes on, but the numbers are off only by the same small, roughly identical, amount, due to identical causal factors. So my "3.5 million more" is fine!
Obsessively, I also tossed on a note on the issue being a non-issue on my post!)
--
-->However, Charles, if you see this comment, you mentioned you have non-effectuation numbers handy since you just used them in an analysis.
If you can plop down a typical percent non-effectuation, say 2024 or 2025 when there is no issue of changing subsidy schedule making the year non-typical, that might help my understanding of the whole business.
For you and any other wonky people, the basis of what I did is I looked for proportion returning for coverage, to compare 2025 to 2026. (Aware that non-effectuation will not be available for 2026 at this early date. So, just comparing signed up without effectuation should give a lower bound, at least, on how many people were turned away by the new higher prices.)
Now, with the CMS data sources for status as of early 2025 and early 2026
https://www.cms.gov/newsroom/fact-sheets/marketplace-2025-open-enrollment-period-report-national-snapshot-2
https://www.cms.gov/newsroom/fact-sheets/marketplace-2026-open-enrollment-period-report-national-snapshot-2
It has "number of returning consumers". But, it doesn't have number "eligible to return": which the footnote says is those who have exchange coverage through an exchange through END of the prior year.
If it it did have number "eligible to return", then we just subtract returning-(eligible to return) for each of 2025 and 2026, and we're done.
However, since I didn't have number of "those who have exchange coverage through an exchange through END of the prior year", I went with the KFF-plotted number enrolled for the prior year as "eligible to return", which is slightly different. But, the causality pattern on the difference between the number wanted and the number used is the same for the two years involved, which is 2024 and 2025, which are both years of the expanded subsidy.
That is, the 1.2 million for 2025 non-returning is a little off, and so is the 4.7 million non returning for 2026, but they're off by roughly the same amount. And so the difference, 3.5 million more 2026, is very close to correct. (On not-considering-effectuation open-enrollment declining-re-enrollment people.)
Oh, I see! From a little poking around, I see that, if I wait a few weeks or month for the next-year version of this
https://www.cms.gov/files/document/effectuated-enrollment-early-snapshot-2025-and-full-year-2024-average.pdf ,
then I will have much closer to the 2025 and 206 "eligible to return"s.
(I guess I'll have to do a revision of my 3.5 million when it comes out.)
On search, I also caught that xpostfactoid routinely publishes what will be my revised number, such as: https://xpostfactoid.substack.com/p/aca-marketplace-renewals-reflections .
Though, I think that this year, as it catches those chased away by the higher prices, and separates ("unconfounds") those people from those other people deciding to enroll for the first time despite the higher prices, has a special usefulness that I don't know of any other way to capture.