How many people are enrolling each day?

Update (11:50 AM) : With the new number just released for enrollments through 12/24, we can now estimate that ~160,000 individuals enrolled on 12/23, nearly twice the rate of the preceding days. That’s because ~200,000 enrolled on 12/23 and 12/24 via, bringing up the total enrollment through the federal site to ~1.1 million and the nationwide enrollment to ~2 million. Expect to see updated graphs soon!

Short Answer: In the four days (12/19-12/22) leading to the big enrollment deadline (i.e., the deadline to sign-up for health insurance in time for coverage on January 1st), the enrollment rate was ~85,000/day. To put that in perspective, in October a paltry 4,200/day were enrolling. November fared only a little bit better at ~10,000/day.

Long Answer: We all know October was terrible for Very few Americans successfully enrolled in insurance plans due to lots of glitches. November was essentially sacrificed by the White House to fix the website rather than sign folks up.  Now, it’s December. By all accounts, both the federally-facilitated (FFMs) and state-based marketplaces (SBMs) are significantly improved (for the most part) – seeing heavy loads and boasting significantly higher enrollment numbers than between October and November.  But, has it been enough to reach the oh-so magical “7 million enrollments” target for success? Last week we documented the December surge, now it’s time to delve a little deeper into the numbers. 

Given the history of the roll out and the reported surge of enrollments, we were interested in determining the approximate number of  people enrolling each day. This has been a bit of an arduous task because of the haphazard nature of how the enrollment data has been publicized. For that reason, I want to emphasize that the graphs below are estimates and are limited by the quality, quantity, and accuracy of the enrollment data that has been publicized. In this regard, I give a ton of credit and appreciation to the work by Brainwrap and colleagues at who have meticulously documented the enrollment for the 50 states and DC from the beginning, even as many other groups have failed to keep-up with the deluge. Thanks to their work, we have been able to compile enrollment numbers in parallel.

Since we could compile an enrollment history for each state (and via updates from the Administration, the federal exchange as a whole), we’ve been able to make an estimate (guestimate?) of the number of people who enroll daily. We calculated this number the simplest way possible – a linear interpolation between enrollment reports. For example, if 7,000 individuals enrolled between 10/1-10/7 and 14,000 enrolled between 10/8-10/14, then we would input it as 1,000 enrollments/day and 2,000 enrollments/day respectively. Of course, this is not a true measure of the enrollments rate. In our hypothetical scenario, there might have been a server outage on one of the days resulting in significantly fewer enrollments on that day (and correspondingly more on other days). Also, this model disregards  “ramping up” trends and introduces discontinuous jumps in the enrollment rate (e.g., from 1,000  [10/7] to 2,000 [10/8]). A linear interpolation assumes the least, but will result in flat curves between enrollment reports and discontinuous jumps corresponding to new enrollment data. A more “accurate” model, such as a quadratic interpolation, would help solve the latter problems, but not the former. However, our purpose is not to figure out exactly how many people enrolled on a given day, but to observe general trends.

Alright. Caveats aside, let’s see those enrollment numbers!

Click to Enlarge!

In the four days leading up to the Dec. 23rd deadline, the enrollment rate was staggeringly more than in the past. This is true for both the state and federal marketplaces. A combined 85,000 individuals were enrolling  per day, split evenly between FFMs and SBMs. Perhaps surprisingly, the enrollment in the state-based exchanges ramped up significantly in recent days, even though these marketplaces were working far better in October and November than the federal exchanges.

This gives some credence to the theories that:

(1) the state-based exchanges, while generally better than the federal exchange, have also improved;

(2) the  bad publicity and woes for the federal marketplace significantly dampened the enrollment on the state-based exchanges;

(3) the ramp up of advertisements from private health insurers and state/ federal officials in December influenced enrollment;

(4) we are a nation of procrastinators.

The comparatively lower enrollment of SBMs in early December does suggest FFMs observed some “pent-up demand” – people who would have enrolled in October and November if was working properly.

So how do these number compare to earlier trends? Well, for the federal exchange, the current rate of enrollment is 50 times that in October (!!), 11 times that in November, and 2.5 times that in the first week of December. For the state exchanges, the current rate of enrollment is 14 times that of October, 7 times that of November, and 3 times that of the first week of December (see the table for more details).

Date FFM SBM Total
October                837           3,365           4,202
November            3,736           6,320         10,056
December 1-7          16,000         14,334         30,334
December 8-18          42,883         24,179         67,062
December 19-22          42,000         43,049         85,049

If you were one of those kids who was super awesome at calculus, you may have noticed this surprising fact:  sometime around 12/15, the federal marketplace had enrolled more total individuals (~590,000) than the state-based marketplaces (~570,000). As of 12/22, the federal marketplace has enrolled 890,000 and the state-based marketplaces have enrolled 830,000. And for those of you who can’t tabulate the area under the curve by sight, fear not! Below, we did the calculus for you!

Screen Shot 2013-12-28 at 4.35.59 PM

This figure is fairly similar to the one done by, but a little smoother because of our linear interpolation method. The enrollment, thus far, has been exponential.  As of 12/23,  ~1.8 million Americans had enrolled (the figures are only through 12/22 because there has been no update from the federal marketplace about the number enrolled). Suffice it to say, by the time all the data comes it is likely close to 2 million Americans will have enrolled for coverage that begins January 1st. Of course, this is just for private health insurance selected through the marketplaces. It is unclear how many people purchased their plans directly from the insurer (people who do this are not eligible for the subsidies). Also not included are the approximately 4 million individuals who will gain coverage through Medicaid/SCHIP. A significant number are also likely to enroll through the marketplace’s small business exchanges (SHOP). 

So…is Obamacare on track to meet it’s enrollment target in March?  

Unfortunately, the media is less interested in the number of actual people helped or the nuances of the law than they are about this mythical 7 million enrollment number (which, in case you were wondering, was an estimate made by the CBO for the number of people enrolling in private health insurance through the marketplaces by the end of 2014).

Fairly or unfairly, the law will be judged by whether the Administration hits the 7 million mark. With that in mind, is there any hope to reach the mark given the glitches and errors the marketplaces had and still have? It’s hard to tell because enrollment is not linear. It will surge before major enrollment deadlines (like the recent cut-off for coverage by January 1st), and be relatively calm otherwise (like October and November would have been even if was fully functional, as we can glean from the state-based exchanges). That being said, the enrollment rate in December suggests that meeting the 7 million mark is not out of the realm of question. 

If there was a linear rate of enrollment, in order to reach the 7 million target, about 38,500 individuals would have enroll each day during the open enrollment period (red line). With the slow October and November enrollment, the necessary enrollment rate for the remainder of the enrollment period (blue line) crept up to 55,500 a day at one point. Since early December, this number has declined as the actual enrollment rate has been higher the required rate (now at ~53,000). Of course, we should expect the enrollment rate to decline post-deadline, as the December enrollment (to get coverage for January) and the March enrollment (to get coverage before the end of the open enrollment period) are likely to be biggest enrollment months. But if we continue to see enrollment rates that persist between 30,000-50,000 during the “lull” periods (e.g., December 25-January 7 before the January 15th deadline to enroll), then hitting 7 million enrollments is not outside the realm of possibilities – a testament to how far the exchanges have come since October 1st.

Here’s the link to the data used to derive those graphs.

Also, if all this “exchange” gibberish has you confused, check out our “Obamacare 101” page for a brief primer.


One response to “How many people are enrolling each day?

  1. Pingback: Enrollments Hit 3 Million | obamacarenerds·

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s