Even if QHPs reach 13 Million, they won't reach 13 Million (and that's perfectly fine)

(sigh) I might as well get this out of the way now.

I've been promising for some time to write a post explaining the difference between the CBO's "13 Million QHPs" projection, the HHS Dept's "9.1 million" projection and my own "12.5 Million" projection. I've alluded to this here and there, but a lot of people are still confused, so it behooves me to dedicate a full post to the topic (which I can then add to the menu for quick access).

I was planning on waiting another week or so before writing this, but with the CBO issuing their latest 10-year Budget Outlook, which quietly reduces their "13 million" figure down to 12 million, it seems like now is the time. I'm already anticipating people thinking that this means that my own projection is now higher than the CBO's...which it isn't.

In a nutshell, here's what the different numbers mean:

  • 9.1 Million: HHS's Paid QHP Enrollee Projection
    This is how many people the HHS Dept. believes will select a private policy and have paid their first premium in time for them to be fully enrolled for either January 1st, February 1st or March 1st coverage. In other words, let's say someone picks a plan by the deadline for March 1st coverage (2/15). They may not actually pay their premium until early to mid-March, but they'll have coverage via the 2015 Open Enrollment Period. The 9.1 million figure is on the right-hand side of The Graph, in red (9.1M Paid)
  • 10.4 Million: HHS's Total QHP Selection Projection
    This is how many people the HHS Dept. believes will select a private policy as of 2/15/15, whether they end up paying their first premium or not. Assume 88% of these folks pay up and you have the 9.1 million above actually paid/enrolled by mid-March. This is also on the right-hand side of The Graph in red. NOTE: Technically speaking, HHS just gave a range of "10.3M - 11.2M" total plan selections during open enrollment, not 10.4M specifically. However, this corresponds to the range of 9.0 - 9.9M paid enrollments...which basically maeans 88%. Since HHS Sec. Burwell did give 9.1M as a "hard" number, the 88% corresponding "selection" number to that would be 10.4M even.
  • 11.0 Million: ACA SIGNUPS Paid QHP Enrollee Projection
    This is how many people I think will select and pay for/be enrolled in their policy by March (my equivalent of HHS's 9.1M above). This is the tan/beige section of The Graph, and the 11.0M figure is on the right-hand side in blue.
  • 12.5 Million: ACA SIGNUPS Total QHP Selection Projection
    This is how many people I think will select a private policy via the exchanges by 2/15/15, whether they end up paying or not. This is The Big Number...the one that the media, pundits, and healthcare junkies are all watching most closely. This is the main number being reported by the various exchanges, and is the highest, dark blue section on The Graph.

And finally, we come to the Congressional Budget Office.

Way up at the top-right corner of The Graph is the 13 Million figure.

Most pundits, followers, ACA supporters and detractors alike seem to think that if the total QHP selections as of February 15, 2015 reach 13 million (or don't reach it), the exchanges will have achieved (or not achieved) the CBO's target, just like reaching 7 million (or not) was considered make-or-break last year.

The thing is, neither of these were, or are, the case, either last year or this year. What do I mean by this?

Well, on a big-picture level, neither the 7 million number nor the 13 million number ever were "make or break". The earth wasn't going to open up and swallow the HHS Dept. or the country if we came up a bit short. Last year's enrollment total was either 8 million, 7.1 million or 6.7 million, depending on whether you're counting the total plans selected as of last April, the total number of that 8 million who actually paid their first premium (7.1 million) or the number who were still enrolled and paying by the end of the year (6.7 million).

Breaking--or not breaking--that 7 million number didn't in and of itself magically mean the ACA was a success or failure, but it certainly provided a lot of PR/boasting/attack point fodder for the media and politicians.

However, on a pure numbers basis, even if the grand total number of QHP selections by 2/15 ends up being 13 million after all (surpassing both HHS and my own projections), that still does not mean that the CBO's original projection has been achieved.

Yes, I'm the one saying this. Better for ACA supporters to hear it from me first than from someone opposed to the law.

Here's why: When the CBO said that they expect 13 million people to enroll in private healthcare policies via the ACA exchanges in 2015, there's two important points to remember:

  • First, they're talking about over the course of the entire calendar year, not just the open enrollment period. Remember, last year around 270K people per month were still enrolling from May - October (in November, of course, this was mostly replaced by people enrolling/renewing for 2015).

On the surface, this makes it sound like even if the 2/15 total is 12.5 million, it should be easy to break 13 million, right? 270K x 8 months (March - October) = 2.16 million more! 14.66 million! Heck, even if you only count the 88% who pay up, that's still 12.9 million, close enough, right? Woo-hoo!

However, there will also likely be a slightly higher number of people dropping their coverage every month as well due to similar life changes (getting married, divorced, having a child/losing a child, moving to a new state, getting a job with benefits, aging into Medicare eligibility, falling on hard times and moving to Medicaid, dying...). In fact, if last year provides any guidance, we'll probably end the year with more like 10.5 million people who are still paying & enrolled by the end of 2015 (about 84% of the starting total).

OK, you're saying: If we're only counting those who are still enrolled by the end of the year, how many would it take to achieve the CBO's 13 million?

Well, if you have 15.5 million people actually select plans by 2/15, then via a combination of non-payments, off-season additions & off-season subtractions we should end the year with around 84% of that still enrolled...or 13.0 million even.

OK, so 15.5 million it is, then. Pretty tall order, but...


Because here's the other really important thing:

  • When the CBO says "13 million enrolled in 2015", they're talking about the average number enrolled each and every month.

I've talked about this several times before, starting as early as way back on April 4th of last year, when this point was brought to my attention by the Washington Post's Glenn Kessler (their "fact check" guy). I again discussed it at length in response to Chris Conover, a conservative Forbes writer with whom I have a somewhat colorful history, back on June 29th.

The reason why the CBO needs to know how many people are enrolled in exchange-based QHPs on average per month is because they need to have some sort of idea how much federal money has to be sent out in the form of tax credits. Their chief concern isn't so much how many people are covered, but how much the treasury will be impacted by the ACA subsidies.

Therefore, "13 million" doesn't mean "13 million signed up" or even "13 million paying the 1st month". It doesn't even mean "13 million still enrolled by the end of the year".

What "13 million" means to the CBO is an average of 13 million enrolled each and every month. Since the HHS Dept. knows what the average tax credit per enrollee is, that means the CBO just multiplies that dollar number x the average enrolled per month in order to know how much the credits will cost.

There are only 2 ways to achieve 13 million average enrollees per month: You can have 13 million pay up in time for January coverage and then continue to stay enrolled throughout the year (or at least be replaced 1:1 by new people joining as they drop off), or you can have a high enough number enroll that it offsets the number who either didn't start in January or who drop their coverage before December.

So, how likely is this to happen in 2015? Well, by my estimates, here's how I expect things to play out (click below for full-size version):

As you can see, of the 12.5 million QHP selections I'm expecting, only around 7.61 million of them were actually paid and enrolled for January coverage.

Another 790K or so should be enrolled starting on February 1st, and assuming my prediction of a huge mid-February spike holds true, an additional 2.6 million should be enrolled as of March 1st (remember, this includes most people who have been enrolling for the past 10 days as well).

Then, assume 270K/month during the off season (tapering off sharply in November and December for obvious reasons), and you get a total of around 14.5 million people selecting a plan throughout the year, of which around 12.8 million pay for their first month's premium...but only 10.5 million of whom are still enrolled and paying by the end of the year, assuming a 2% monthly attrition rate above and beyond the non-payers.

(Note that last year, around 112K people were dropped involuntarily in September due to their legal residency/citizenship/immigration status not being rectified; I'm assuming that this issue should be mostly resolved this year, so perhaps just 50K will be dropped this time around).

Now, take a look at the number in yellow at the bottom of the chart: 10.27 million. That's what the average monthly enrollment would be if this scenario plays out.

Here's the thing, though: NONE OF THIS IS CAUSE FOR ALARM, or for ACA supporters to worry about. That's still over 10 million people each and every month who are enrolled in fully ACA-compliant policies, about 85% of whom are receiving tax credits to help them afford the cost.

For all the fuss being made by supporters and opponents alike (including this very site), the reality is that there are only two reasons why the exchanges "need" to enroll a certain number of people:

  • First, because of the risk pool. Here, the mix of enrollees, healthy to unhealthy, young to old, etc. is more important than the actual number. Furthermore, it's my understanding that most (not all) of the off-exchange QHPs are part of the same risk pool anyway, which nearly doubles the total number of people in the pool.
  • Second, the exchanges (especially the state-based ones) have to have a certain minimum number of enrollees in order to have enough revenue to support themselves. This, admittedly, is a bigger issue; several states are struggling to ramp up enrollment enough to be self-sustaining. However, assuming they're able to pull this off, any additional enrollments beyond that number are pretty much a bonus.

Remember, the primary goals of the ACA are not specifically to enroll a certain number of people on the exchanges, but to enroll as many people as possible in quality, affordable healthcare plans. The exchanges help make achieving this easier, but aren't the end-all be-all goal. If corporations go on a hiring binge, snap up 30 million uninsured people and give them all ESI this year, I don't think anyone at the HHS Dept. is going to be complaining that they aren't "leaving enough people" for the exchanges.

Heck, from the CBO's perspective, this is a good thing, since it means that about 20% of the tax subsidy budget they thought the IRS would have to shell out can stay in the treasury dept's coffers...or something like $8 billion for the year.

(As an aside, for all the fuss made last year about QHP enrollments only "really" turning out to be 6.7 million after the infamous "dentalgate" brouhaha--which, I might add, I predicted about 2 weeks earlier, though I didn't know the cause of the discrepancy--it was kind of irrelevant, because going by the CBO's standards, the monthly 2014 average enrollment was just 5.5 million for the year anyway).

No, in order to achieve a 13 million average per month for all of 2015, the rest of the year would have to look something like this:

Yup. Assuming an 88% average payment rate, a 2% monthly attrition rate (after subtracting non-payments) and the same 270K/month during the off-season, a whopping 7 million people would have to enroll in the final month of the current open enrollment period, for a total of around 16.5 million by 2/15.

Simply put, this is not going to happen.

The payment rate may be higher than 88%; the attrition rate may be lower than 2%/month; the off-season additions may be higher than 270K/month...but there's simply no reasonable expectation that nearly 7 million people are going to show up over the next 20 days.

I'm optimistic in general, but I'm also realistic. If a last-minute surge of that magnitude does happen over the next 3 weeks, I'll be ecstatic...but I just don't see any way of that happening.

How about the just-lowered CBO projection of 12 million for 2015? All other numbers being equal, it would take 5.5 million in the 3rd month of open enrollment to end up with a monthly average of 12 million even. That would mean 15 million would have to select plans by 2/15, which is also unlikely to happen; I still expect just shy of 3 million between 1/16 - 2/15.

In other words, the CBO is still more optimistic than I am about one or more of the following metrics:

  • They expect more than 3 million people will select a plan by 2/15.
  • They expect more than 88% of those who select a plan will end up paying their first premium.
  • They expect more than 9,000/day (270K/month) additional people will enroll during the off-season.
  • They expect less than 2%, on average, of those who pay the first month will drop off in subsequent months.
  • Some combination of any or all 4 of the above.

So, there you have it: 13 Million Enrollments doesn't necessarily mean 13 Million Enrollments in the eyes of the Congressional Budget Office...and that's just fine.