No, Obamacare Enrollment Is Not Strong, Not By A Longshot

By Chris Conover. Excerpts:

"The reason it is not is buried at the tail-end of the story where the reporter notes "the enrollment period ends Dec. 15, which is about half as much time as people had to sign up last year."

Yipes! If enrollees have only half the time to sign up, then by pure arithmetic, the daily enrollment pace needs to be double last year's in order for total enrollment at the end of the enrollment period to match the level reach at the end of last year's enrollment period: 12,216,003.

But if current enrollment is 128% of last year's when it needs to be 200%, a more accurate way to frame this year's performance would be to say that Obamacare is on track to sign up 36% fewer enrollees than last year (i.e., 128/200=64% which would meaning signing up 36% fewer). That's a pretty bad news story rather far removed from the rosy picture painted by The Hill's headline."

"Clearly, it's possible that there will be a massive surge in enrollments prior to December 15 that will allow this looming shortfall to be averted. But how probable is that prospect?  This year's total enrollment period (Nov. 1-Dec. 15) is only 45 days long [1].  If 2.3 million signed up in the first 18 days, that means that 9.9 million must sign up in the final 27 days to reach last year's enrollment total. That implies the daily rate of sign-ups has to be 2.9 times as fast as the rate we've observed to date (and remember the rate we've observed to date already is 28% higher than last year's pace!).

Even the guru of ACA enrollment figures--Charles Gaba, who runs ACASignups.net [2]--concedes that total Exchange enrollments this year likely will top out at only 10 million. Yet even this prediction (made just before the open enrollment began) now already seems optimistic. To hit that target, the pace of enrollments for the balance of the open enrollment period would have to be 2.2 times as high as we've observed in the first 19 days! Not impossible, to be sure, but neither is this highly probable.

As an example, last year the pace of enrollments was 42% higher during the final 71 days of open enrollment compared to the first 21 days. This same thing could happen this year, but remember that we only have 24 days of open enrollment left. So even if average daily enrollments climb from their current level of 128,000 daily to 182,000, that will add only 4.9 million more enrollees by December 15. If this happened, total enrollment for 2017 would end up at 7.2 million (which would be lower than the 8 million who signed up in Obamacare's first year)."

"Actual vs. Projected Enrollment. The first is to compare actual enrollment against projections made by the Congressional Budget Office (CBO) in July 2012. I use these projections because they take into account the Supreme Court ruling on Obamacare that was handed down in June of that year. That ruling--by making Medicaid expansion discretionary rather than mandatory across all states--meant that CBO analysts expected additional individuals to end up purchasing coverage through the Exchanges rather than being forced onto Medicaid.

Produced by Christopher J. Conover, Duke University, based on CBO projections from July 2012, CMS data on ACA enrollments, and ACASignups data on paid (effectuated) enrollments and projected enrollment at end of 2017-18 open enrollment period.

Fig. 1

CBO made its projections based on paid enrollments, which are lower than the gross enrollment figures I've cited earlier. From the standpoint of accurately measuring gains in coverage, it is paid enrollments that matter since those who fail to pay their premiums can be legally tossed off their policy, thereby becoming uninsured.

According to figures from Charles Gaba, average monthly paid enrollments in 2014 (the first year Obamacare Exchanges were fully operational) were 5.46 million, i.e., 32% lower than the gross enrollment figure of 8.0 million mentioned earlier.  Things improved in year 2 insofar as Mr. Gaba's estimate of average monthly paid enrollment (9.35 million) was only 20% lower than the gross enrollment figure of 11.7 million.

Mr. Gaba has never produced similar figures for subsequent years, so for purposes of analysis, I have generously assumed that paid enrollments equal 80% of gross enrollments rather than only 68%. Yet even using figures stacked in favor of Obamacare, Fig. 1 shows that estimated actual average paid monthly enrollments have fallen far short of CBO projections every single year; moreover, this shortfall generally increased with each passing year.

In year 1, Exchange enrollments fell 39% below CBO projections, improving slightly to a shortfall of 33% in year 2. But in every subsequent year, the shortfall has grown steadily, from 56% in Year 3, and 61% in year 4. We do not yet know the actual result for year 5, but if Mr. Gaba is correct that total enrollments will reach 10 million, that translates into average monthly paid enrollments  of only 8 million, or 69% below where CBO once thought we would be in 2018. If instead 2018 final enrollment numbers follow the pattern of last year's open enrollment period, the shortfall would reach 78%."

"Year-Over-Year Changes in Enrollment. An even clearer picture of just how persistently and swiftly the air has gone out of Obamacare's tires in shown in Fig. 2.

Produced by Christopher J. Conover, Duke University, based on CBO projections from July 2012, CMS data on ACA enrollments, and ACASignups data on paid (effectuated) enrollments and projected enrollment at end of 2017-18 open enrollment period.

Fig. 2

Each year, the gains in enrollments in the Obamacare Exchanges have gotten smaller and smaller, actually turning negative by year 4. So this year's expected decline in coverage will simply continue that pattern except that the absolute loss in enrollments will be anywhere from 4.5 times last year's loss (if Mr. Gaba is correct) to 10 times as large (if last year's enrollment pattern is repeated this year).

This bleak trend was not foreordained. The CBO did expect the year 2 gains in enrollment to be somewhat smaller than those in year 1. However, in year 3, the gain in enrollment was expected to match those in year 1 and in no year did CBO expect Exchange enrollment to decline (and certainly not be 4 million people in a single year)."