OK, HERE WE GO: The First Monthly HHS Enrollment Report, broken down (UPDATED)
As I noted on Saturday, I was out of town for a few days (at a water park resort in Ohio, if you really want to know) with my family. We had a good time at the actual water park, but I strongly recommend avoiding the TGI Friday's attached to it. HHS Sec. Burwell noted last week that the first Big, Official, Comprehensive Monthly ASPE enrollment report (broken out by state and other demographics) would be released sometime this week. I knew it would come out either today or tomorrow; I was hoping they'd hold off for one more day, but I guess they wanted to put it out there so the HHS staff could go home for New Year's, which is reasonable.
As a result, I was driving back from Ohio (well, my wife was anyway...I drove down, she drove us back) when the news hit earlier today, and wasn't really in a position to do a full analysis from a moving car, though I did tweet back & forth about it quite a bit.
In addition, a ton of other state-exchange data updates were released, including several I wasn't expecting yesterday or today (Rhode Island, DC) and others which I was (Minnesota and Massachusetts). Ironically, the one Big Update I've been expecting for a week now--Washington State's "deadline day" update--still hasn't come out yet, which is surprising.
So, I'm back now and have spent the past couple of hours furiously updating no fewer than 7 states: Michigan, DC, Maryland, Massachusetts, Minnesota, Rhode Island and Oregon. And none of this has anything to do with today's ASPE report, which only runs through--this is important--December 15th. A lot has happened since then, and I'm not just talking about the HC.gov autorenewals.
Anyway, that brings me to the report. Remember what I said on Saturday:
This is the same type of report which HHS released once a month last year, and it's the official record for purposes of tracking enrollment. It should include all 50 states plus DC. However, I have no idea whether it will run:
- November 15th through November 30th (ie, a partial calendar month, which wouldn't be particularly useful); if so, I'd expect around 930K QHPs via HC.gov and 1.24M total.
- November 15th through December 15th (everything up through the enrollment deadline for January coverage in most states; this would be useful, but wouldn't include any autorenewals); if so, I'd expect around 3.52M QHPs via HC.gov and 4.70M QHPs total.
- November 15th through December 19th (actually 5 full calendar weeks, which is what I'm guessing it'll include and which would be the most useful, since autorenewals for most states would be included). If so, HHS has already given the total as roughly 6.39M via HC.gov, and I'm estimating roughly 8.52M total (remember, my 8.65M estimate includes 4 extra days, through 12/23).
As it happens, the report ended up taking the 2nd tack: It runs from the first day of 2015 Open Enrollment (11/15) through the official deadline for January coverage (12/15), 31 days total, which is fine.
So, to the report itself: Skipping ahead to Pages 16 & 17, when you add up all 50 states +DC, it looks like the grand total as of 12/15/14 was 4,048,861, as opposed to the 4.7 million I was expecting as of 12/15. Oops...that means I overshot by nearly 14%, right??
WRONG.
There are several MAJOR things to remember about this report, and in all the furious Twittering and "breaking news!" stories I saw on my way up from Ohio, I've already seen at least a half-dozen reporters/pundits forget about some or all of them.
The first, most obvious one is this: It doesn't include active/manual renewals for California or New York.
* Indicates that the data reported for the current month only include new consumers, and exclude reenrollees.
47 states, plus DC, include both manual renewals (that is, current 2014 enrollees actively logging into their existing accounts and either renewing their current plan or switching to a different one) as well as brand-new enrollments. Some states may not have these numbers separated out, but they're all there...at least as of 12/15, anyway.
HOWEVER, the two largest state-based exchanges...for the largest state in the country (California) and the 3rd (or 4th?) largest (New York), still haven't provided their active renewal/re-enrollment data, even to HHS. This surprised me, frankly; when CoveredCA stated that they wouldn't be releasing their renewal numbers until January, I assumed that just meant they wouldn't be issuing a press release about it. I kind of figured that they'd be required to provide it to the HHS Dept, at least...but apparently not.
So, how many renewals could there have been by 12/15 in those states? Well, CA was reporting about 1.12 million current 2014 enrollees as of mid-October, and New York had 379,000 as of mid-September, so there's potentially up to 1.5 million. Assuming that 10% of those didn't renew (either actively or automatically), that leaves around 1.3 million. Further assuming that 40% of them did so "actively" (that's how the 6.4M via HC.gov played out), that should mean around 520K had done so as of 12/15.
If that's accurate, that SHOULD mean that the actual total QHP tally as of December 15th should be right around 4.57 million...or only around 2.8% less than my 4.7M estimate.
However, it could be higher or it could be lower than that. Unfortunately, untill CoveredCA or NY State of Health issues their data, we won't know that particular data point.
OK, with that out of the way, what is included?
Well, for HealthCare.Gov itself, I was estimating around 3.52 million QHP selections as of 12/15. The actual number ended up being 3,416,023: Off by 3%.
There's a ton of other interesting tidbits, and other reporters/pundits are poring over all of it right now. Some of it is way outside of my area of expertise, but here's some items which jump out at me:
- OREGON: Holy Crap. Just an hour ago I was noting that OR's QHP enrollment jumped up from just 40.6K on 12/14 up to 67.5K on 12/21, but I had no idea how many of those were squeezed in under the 12/15 wire. I figured it would be around 8,000 or so. Instead, check it out:
WOW. If 40,581 had enrolled as of 12/14 and 73,152 were enrolled by the following day, that means over 32,000 people enrolled in a single day in Oregon. Whew. That's mainly noteworthy because it means that a good 95% of 2014 enrollees managed to avoid a coverage gap in January in Oregon, if I'm reading this correctly (ok, perhaps 90%...a few thousand are presumably new additions).
However, a word of caution: According to the actual Oregon State Government website, they only had 67,467 exchange-based QHP enrollees as of 12/21...nearly 5,700 fewer people, 6 days later. Neither the 73K nor the 67K figures account for payments, so I really don't know what to make of this discrepancy. Strange.
Either way, Oregon is now at least 75% of the way towards HHS's 2015 enrollment target (89K), or at least 54% of the way towards mine (125K).
- NEVADA: Speaking of which, Nevada has an even better siutation: They only had 32,460 people still enrolled in 2014 policies as of mid-October, but managed to enroll 40,285 by the 12/15 deadline. Considering that NV only managed to enroll 45,390 people during the entire 2014 open enrolment period, that means they've already hit 89% of that number in the first 1/3 of the 2015 period. NV is now at 68% of HHS's 2015 target (59K) or 55% of the way towards mine (73K).
- Also worth noting: Last year, the monthly reports pretty much broke the states into two categories: Federally Facilitated (HC.gov) and State-Based (15 others). They changed the shading for Idaho and New Mexico to recognize that both of them were expected to move off of HC.gov this year, but kept them lumped in otherwise. This year, partly due to the musical chairs (Idaho moving off of HC.gov, NV & OR moving onto it and NM staying put), but also likely due to the upcoming King v. Burwell SCOTUS case, they've broken the listings out into 3 categories: Federal (HC.gov), State-Based and "State-Based Using the HealCare.Gov Platform", which includes NV, OR and New Mexico.
OK, what else? Well, that brings me to the second major missing data point: It only runs through December 15th and does not include autorenewals for any (most) states.
- UPDATE: Ok, strike that: It turns out that a few states (CT, ID, KY, VT & WA) do include autorenewals in this report, which just confuses things further.
In some cases this isn't relevant because they have no autorenewals anyway (MA, MD, RI) or because all renewals are categorized as "new" anyway (OR & NV).
However, those missing 4 days from 12/15 to 12/19 make a huge difference in the total numbers; HC.gov alone jumped from 3.42 million to 6.39 million, mostly due to nearly 3 million autorenewals being added to the total.
In that sense, this report is equally interesting and frustrating, because it's difficult to get a true grasp on some of the figures flying around.
The third major thing to remember is that while 12/15 was the official deadline for January coverage in most states, it wasn't the deadline for quite a few of them, including:
- California: Final Deadline was 12/21
- Connecticut: Final Deadline was 12/19
- Hawaii: Final Deadline isn't until 12/31
- Idaho: Final Deadline was 12/20
- Maryland: Final Deadline was 12/18
- Massachusetts: Final Deadline was 12/23
- Minnesota: Final Deadline isn't until 12/31
- New York: Final Deadline was 12/20
- Rhode Island: Final Deadline isn't until 12/31
- Vermont: Final Deadline isn't until 12/31
- Washington State: Deadline was 12/23
Add all of those up and that's still potentially a ton of renewals/re-enrollments/new additions, not to mention more autorenewals in most of those states.
Finally, HHS also released the 6th week "snapshot" report which runs the HC.gov total out through 12/26; the grand total there is 6.49 million, about 96,000 higher than the 12/19 total.
Add all of that up--including the half-dozen or so state-level updates which I posted earlier today--and you have a grand total of 7,403,558 confirmed QHP plan selections via the various ACA exchanges.
I'll have much more to say about today's report tomorrow, but I think that's plenty to digest for today. For now, I'm just going to plug the official numbers into the Spreadsheet (except for states where I have more recent official reports from their exchanges) to see where things stand.
Much more tomorrow...
UPDATE: ARRRGH!! OK, Here's another tidbit to look out for: It turns out that the QHP totals from the state-based exchanges--all 14 of them--only run through December 13th, not December 15th (except for California, which runs through December 14th).
In other words, the two busiest days of the 2015 enrollment period to date are not included in this report for 13 states (and the busiest day isn't included for the single largest state exchange, CoveredCA).
Are those extra 1-2 days across 14 states enough to make up the remaining 130,000 difference from my 4.7M estimate? Well, 25,408 of it is specifically noted in California alone. That drops the likely difference down to around 105K. Colorado's press release specifically stated 12,600 enrollments on the 15th, bringing the likely difference down to 93,000. Massachusetts reported 8,902 QHP determinations on the 14th & 15th, and has consistently shown roughly 50% of those actually enrolling on the same day; that's around 4,400 more, leaving around 89,000 for the remaining 11 states.
I don't have specific "last-day only" or "deadline weekend only" enrollments for those states, but it's safe to assume that between them they chalked up easily 89K or more.
So, to summarize, while the HHS Report does include a ton of useful info...
- It doesn't include any renewal data from the two largest state exchanges
- It doesn't include any data from the two busiest days of the period for 13 states or the busiest day for the largest state
- It doesn't include any deadline date data for 11 states (not blaming them for this one)
- It doesn't include any autorenewal data for either 45 or 46 states (again, not blaming them for this one)
...all of which makes this an incredibly messy, confusing and supremely un-helpful report for my purposes, in spite of the gobs of data that it does include.