Avalere agrees w/my 2015 QHP projection...if you read their press release carefully

Last Friday I bumped up my own 2015 QHP projection from around 12.0 million as of 2/15/15 up to 12.5 million.

As you can see on The Graph, I'm also assuming roughly an 88% payment rate (ie, how many of those 12.5M I expect to pay for their first month's premium), which should be roughly 11.0 million even.

However, there's one number missing, which I left off in the interest of avoiding confusion: How many I expect to still be enrolled by the end of 2015. This is pretty simple, though: Around 10.5 million. How do I arrive at that number? Well, this year the net attrition rate resulted in around 6.7 million people still being enrolled nationally as of mid-October. That's around 84% of the 8.0 million "total" enrollment figure as of last April. In other words, out of 8 million people, about 12% never paid and another 4% gradually dropped off the exchanges by October.

Assuming a similar pattern next year, 84% of 12.5 million is 10.5 million who should still be enrolled via ACA exchanges as of next October.

That brings me to Avalere Health, which just released their own projection/analysis this morning:

A new analysis by Avalere Health estimates that exchange enrollment will total 10.5 million people by the end of 2015, though enrollment could fluctuate up or down by approximately 1 million individuals.

Note that they hedge their bets by including a range of 9.5 - 11.5 million, which is fine given how much uncertainty there is...and this doesn't even include the GOP Elephant in the Room, the SCOTUS/King/Halbig case decision expected next June, which could pretty much nullify both Avalere and my own projections completely.

As of late 2014, 6.7 million individuals enrolled in coverage through state and federal exchanges*. In the first three weeks of open enrollment for 2015, 664,000 new consumers applied for coverage and selected a plan through healthcare.gov.** Since the federally facilitated exchange has generally represented 68 percent *** of total exchange enrollment, Avalere estimates that roughly 977,000 individuals signed up for coverage across all exchanges between November 15 and December 5.

Avalere makes an error here: Yes, HC.gov's 5.45 million was indeed 68% of the 8.02 million total last April, but that was the cumulative average--that is, it included October and November, when HC.gov was a disaster and only held 25% & 43% of the total respectively. By the time all the worst kinks were worked out, HC.gov was actually enrolling around 75% of the total nationally, which is what I've been using for my model this year (it's actually probably even higher by now, perhaps 77-78%, but I'm using 75% to err on the side of caution).

Still, it's interesting that they're approaching this from a different angle than I did. They're starting with how many new enrollees they anticipate, using the first 3 weeks of data, then working their way backwards towards the renewed estimate. I did it the other way around.

Based on the initial pace of sign-ups in the first three weeks of open enrollment, Avalere estimates that new enrollment by February 15 is likely to be between 4.5 and 6 million individuals.

I had originally estimated 12 million total, split into roughly 6.1M renewals + 5.9M new, which is at the high end of their estimate. When I increased my total projection to 12.5M, I assumed roughly 3.0M manual renewals + 2.9M autorenewals + another 300K or so renewals over the next week from states with later deadlines, or around 6.2 million renewals total, plus another 6.3 million new enrollees, for 12.5 million total.

However, based on the 2014 exchange experience, about 16 percent of these individuals may drop out of coverage by the end of the year—either by failing to pay premiums or electing to disenroll. Furthermore, it is unclear what portion of 2014 enrollees will reenroll in plans for 2015.

This is the 12% nonpayment + 4% gradual attrition factor I noted above.

“Altogether, total 2015 enrollment is tracking between 9.5 and 11.5 million, including people renewing their plans,” said Caroline Pearson, vice president at Avalere. “While 2015 enrollment may fall short of CBO estimates, it is still likely to exceed the Administration’s projections.”

Here's where Avalere's press release is confusingly worded if you aren't paying attention. To me, "total 2015 enrollment" means "everyone who goes through the enrollment process throughout the course of the year, whether they paid or not, and regardless of whether they dropped their policies at some point later that year". In other words, I initially read their "10.5 million total" as being equivalent to my own 12.5 million total figure. The fact that they then make mention of this "exceeding the Administration's projections" but "falling short of CBO estimates" adds to this interpretation.

HOWEVER, Ms. Pearson clarified this morning that this isn't what that number refers to at all: It does indeed refer to their estimate of how many people they still expect to be enrolled via the exchanges at the end of 2015:

.@charles_gaba Always good to have some consensus. You are right that we are 10.5 year end net of attrition.

— Caroline Pearson (@CPearsonAvalere) December 16, 2014

In other words, it seems to me that Avalere is agreeing with me that the total plan selections as of 2/15/15 should be roughly...12.5 million...which is exactly what I've been saying since Friday :)

As Ms. Pearson notes, the fact that both they and I arrived at the same conclusion using different methodology is a promising sign.