UPDATE x2: Two right-wing outlets use my work. One gets it right. The other...

I've given a lot of interviews to various media outlets over the past 2 years. Most of them are fairly neutral ideologically; some are obviously left-of-center; and once in a blue moon, I'll even provide input to a right-of-center outlet which seems to be intellectually honest about how they present themselves.

Until today, the only anti-ACA media outlet I've ever trusted to be honest about how they present my work is, ironically, one which has also not only mangled my data in the past but which once actually posted a hit job on me, insinuating that I was "cooking my books" in some sort of collaboration with the HHS Dept. to try and "hide" the infamous DentalGate debacle a year ago. The irony, of course, is that not only was this utter bullshit, I was actually the one who discovered and reported the "missing" exchange enrollment problem in the first place. I reported that exchange QHPs had likely dropped to 6.8 million or lower in October before Alex Wayne broke the "DentalGate" story a week or so later. (Wayne's story was still a genuine scoop, however; I knew the numbers didn't match up, but I had no idea why until he wrote his story about CMS inadvertantly double-counting several hundred thousand standalone dental plans).

The thing is, while Investor's Business Daily is mostly full of shit, they do have one honest reporter by the name of Jed Graham, who I actually have a lot of respect for. For instance, just yesterday he wrote a story about the same topic (how much have effectuated exchange QHPs dropped since June) in which he cites my work properly:

5 Big Insurers Lost 8% Of ObamaCare Enrollees In Q3

Cumulatively, the five insurers lost 230,800 exchange customers during Q3 out of a total 2,791,000 at the end of June, reporting by ACASignups.net found. If national enrollment declined at that rate, it would have sunk to a bit over 9.1 million at the end of September, notes Charles Gaba, who runs the enrollment-tracking site.

...National enrollment may not have fallen quite so sharply. ACASignups.net noted that a small sample of states shows roughly flat enrollment and an increase in Massachusetts, where there is no cut-off date for signing up.

The headline, while not telling the whole story (those 5 insurers only make up 28% of all enrollees, which means people might have just moved to other carriers), is factually accurate: Those 5 insurers are indeed "big" and did indeed lose over 8% of their enrollees from June through September...and it's possible that no, those 8% didn't just shift to other exchange carriers but did in fact drop off the exchange completely.

In addition, both the "Enrollment is DOWN!" and the "Enrollment is UP/FLAT!" parts from this article of mine from a few days ago are included.

You'll note that while Graham's article is more negative on ACA enrollment than mine, he's presenting my data honestly, including both the negative and positive findings.

I've learned to trust Jed Graham enough over the past year or so that when Richard Pollack of the Daily Caller (a very much openly right-wing outlet which I've criticized before at this site) contacted me a few days ago to discuss my "2016 Average Rate Hike" project, stating that he was genuinely interested in presenting the full picture, I decided to take a gamble and trust him.

With that in mind, let's take a look at the Daily Caller's piece:

Obamacare Premiums To Soar 3 Times Faster Than Feds Claim

Obamacare premium costs will soar 20.3 percent on average in 2016 instead of the 7.5 percent increase claimed by federal officials, according to an analysis by The Daily Caller News Foundation.

Let's stop it there: 20.3% is not "3 times" 7.5%; it's 2.7x higher. Whether that's a quibble or not is subjective, I suppose.

The discrepancy is because the government excluded price data for three of the four Obamacare health insurance plans when the officials issued their recent forecast claiming enrollees would face only a 7.5 percent average rate increase in 2016.

When data for all four plans are included, premium costs will actually rise on average 20.3 percent next year. The 2015 Obamacare price hike was 2.0 percent.

The Obamacare program’s federal exchange operates in 37 states where officials declined to set up state-run exchanges. Officials in the U.S. Department of Health and Human Services Center for Medicare Services, which manages Obamacare, only calculated price changes for the health insurance program’s Silver plan, thus ignoring data for the Bronze, Gold and Platinum plans.

This is the gist of Pollack's article. Obviously, as this is an anti-ACA outlet, he's taking an accusatory stance, which I was expecting. As such, I'm not going to criticize any subjective opinions by others quoted in the article; instead, I'll focus on the part which quotes and cites me:

Charles Gaba, a data analyst who tracks Obamacare trends and is an Obamacare supporter, reported earlier this year that Obamacare consumers in all 50 states will experience an average 14.4 percent increase. His analysis can be found on his web site, acasignups.net.

“I was hoping they would include all of the rates,” Gaba told The DCNF. “I would love it ideally if they had all the medal [sic] levels.”

Gaba called the CMS price analysis, “fairly representative, but there’s the Gold, the Platinum, the Bronze, the catastrophic plan even, and there’s also a variety of Silver plans. So there are a bunch of different ones in addition to the benchmarks which they did not include.”

OK, so what's the problem here? The article he links to does include a table in which I show a 14.4% weighted average increase vs. the CMS Dept's. 7.5% (actually, I get 7.8% but I'm assuming there's some rounding issues going on there). Well..

  • As the article clearly states and I explained to Mr. Pollack, this version of the rate hike table only includes the 37 states run through HealthCare.Gov, not "all 50 states". This was done deliberately to try and match the CMS's analysis. Many of the missing states (California, New York, etc.) have much lower average rate increases.
  • As the article clearly states and I explained, this version of the rate hike table includes requested rate hikes for many states, not final ones, whereas the CMS analysis includes final, approved rate hikes only. While some state regulators have actually increased average rate hikes and most keep them the same, many reduce them before approval.
  • As the article clearly states and I explained, this version includes off-exchange policy enrollees, which the CMS table doesn't. I have no idea whether this has a net effect of increasing or decreasing the overall ACA-compliant rates, to be honest...but it should've been mentioned.
  • I also sent Mr. Pollack the other link to my main "Average Rate Hike" analysis, which does include all 50 states + DC. While the other caveats still apply, my overall findings there are a 12-13% average increase, not 14.4%.

So...even ignoring the other "apples to oranges" caveats above, I found the average rate hikes to be about 67% higher than CMS's "benchmark" report (12.5% vs. 7.5%), not "3 times" as much.

So...is Mr. Pollack presenting my work and our discussion honestly in his piece? I'll leave that to you.

I've contacted Mr. Pollack and requested that he make the above corrections. We'll see if he does so or not.

UPDATE: Well, it's been a couple of days and Pollack/the Daily Caller still hasn't corrected even their most basic factual error (the "50 states" vs. "37 states" error).

I should also note, by the way, that I have no idea where they got their "20.3%" average. It's possible that this is correct since they only used exchange-based policies vs. off-exchange, and they only used approved rates vs. requested...but the approved rates should be slightly lower than the requested ones for the most part.

UPDATE x2: I just wanted to give a quick shout-out to Peter Suderman of Reason.com (another fairly right-wing outlet), who also cited my work but did so properly:

But that’s only one way to look at premiums, and it doesn’t provide a complete picture. Silver plans only account for about two thirds of sign-ups, and averages that attempt to capture rate changes across the full spectrum of  plans have found bigger raises. Charles Gaba, an Obamacare supporter who runs the remarkably useful ACA Sign-Ups site, estimated a weighted average increase of 14.4 percent across Obamacare-compliant plans (on and off exchange) in federal exchange states. (A separate estimate which included states running their own exchanges found an average increase of 12-13 percent.) As Gaba cautions, it’s not a strict apples-to-apples comparison, but it suggests that the story on premiums is more complicated than the headline HHS estimate.

There, was that so hard, Daily Caller? The larger point you're trying to make is the same (ie, HHS's "7.5%" figure is kind of misleading), but you can do so without being misleading and exaggerating in doing so.