CMS: Indy Market costs per enrollee stayed flat year over year

With all the concern about the ACA exchange risk pool being sicker than expected as well as plenty of grumbling about how the ACA's Risk Adjustment program is working out in practice, earlier today Kevin Counihan, the CEO of the Health Insurance Marketplace (i.e., the guy in charge of HealthCare.Gov) posted a blog entry laying out some changes that CMS has in mind for the RA program going forward. It's interesting stuff for health insurance wonks, but to be honest, I was more interested in a different document also released today (referenced in the RA blog):

Changes in ACA Individual Market Costs from 2014-2015: Near-Zero Growth Suggests an Improving Risk Pool

• Per-enrollee costs in the ACA individual market were essentially unchanged between 2014 and 2015. Specifically, after making comparability adjustments described below, per-member-permonth (PMPM) paid claims in the ACA individual market fell by 0.1 percent from 2014 to 2015. For comparison, per-enrollee costs in the broader health insurance market grew by at least 3 percent

• Available evidence indicates that the slow ACA individual market cost growth resulted at least in part from a broader, healthier risk pool. In particular, states that saw stronger-than-average enrollment growth in 2015 saw greater-than-average reductions in PMPM costs. For example, in the 10 states with the highest 2015 growth in ACA individual market member months, PMPM claims costs fell by an average of 5 percent.

• Nearly all states saw continued growth in Marketplace enrollment in 2016, suggesting continued risk pool improvement. Moreover, the 2015 claims data also predate important steps CMS has taken over the six months to further strengthen the Marketplace risk pool. These steps include implementing new processes to prevent misuse of Special Enrollment Periods, reducing the number of consumers losing coverage or financial assistance due to data-matching issues, helping consumers who turn 65 move from the Marketplace onto Medicare, and proposing to curb abuses of short-term plans.

The study goes on to detail the methodology used and conclusions. There's one key point which needs to be stressed, however:

On June 30th, the Centers for Medicare & Medicaid Services (CMS) released data on reinsurance payments for 2015.1 Reinsurance payments are based on issuers’ claims paid amounts for the full individual market, excluding grandfathered and transitional plans; the data include all plans sold on the Health Insurance Marketplace, including the federal HealthCare.gov Marketplace and the individual State-based Marketplaces, as well as off-Marketplace plans that are subject to the same pricing and coverage rules. Therefore, these data provide the first snapshot of how costs in the ACA individual market evolved from 2014 to 2015.

The highlighted point is vitally important. In 2014, there were somewhere around 5.5 million people actually paid and enrolled in ACA on-exchange policies on average per month...but there was another large chunk of people who enrolled in off-exchange ACA-compliant policies (roughly 5 million or so, as far as I can tell). On top of this, there were perhaps another 5 million people still enrolled in transitional or grandfathered policies (which are completely separate from the ACA-compliant risk pool altogether, either on or off-exchange). Add all three groups up and the Kaiser Family Foundation estimated the total individual market to be roughly 15.6 million people for 2014, of which around 11 million were part of the ACA-compliant risk pool (half on exchange, half off).

In 2015, thanks both to a 46% increase in exchange-based QHP selections during open enrollment (11.7 million vs. 8.0 million) as well as most of those enrollees getting an earlier start (ie, 74% of 2015 enrollees started coverage on January 1st and 100% started coverage no later than March 1st, vs. only 28% of 2014 enrollees starting in January and over half of them not starting coverage until April or May), there were an average of around 9.4 million people enrolled in exchange policies each month. This point is confirmed later in the memo:

Overall, total ACA individual market member months increased 66 percent in 2015, reflecting higher Marketplace enrollment, increased enrollment duration, and shifts from grandfathered and transitional plans into the ACA individual market.

Well, OK...9.4 / 5.5 = 71%, not 66%, so it sounds like I either slightly underestimated 2014 or slightly overestimated 2015, but it's pretty much in the ballpark.

Meanwhile, the off-exchange enrollment numbers shrank somewhat, mostly due to people dropping transitional/grandfathered policies and moving either onto exchange policies or other types of coverage. As far as I can tell, the 2015 individual market was roughly 9.4 million exchange-based, 5 million (?) off-exchange but ACA-compliant, and around 3 million or so grandfathered/transitional. That would mean a risk pool of around 14.4 million ACA-compliant enrollees.

So, in 2014, around 50% of the ACA-compliant individual market was exchange-based, while in 2015 it increased to around 65%. For 2016, as far as I can figure, it's likely to be somewhere around the same (perhaps 10 million/month on average on exchange and another 6 million/month off exchange), or slightly higher. We won't know the final net attrition numbers for this year until next spring, however.

Anyway, the CMS memo goes on to say:

In the broader health insurance market, such as employer coverage and Medicare, per-enrollee costs grew 3 to 6 percent from 2014 to 2015. For example, the CMS Office of the Actuary estimates that per enrollee growth in employer sponsored insurance (ESI) grew 3 percent ; the Kaiser Family Foundation’s annual survey and the Medical Expenditure Panel Survey both estimate that average premiums for employer-based family coverage grew 4 percent; and insurers’ projections of medical cost trend for 2015 averaged 6 percent.

In contrast, in the ACA individual market, per-enrollee costs were essentially unchanged from 2014 to 2015. Specifically, after making comparability adjustments described below, per-member-per-month (PMPM) paid claims in the ACA individual market fell by 0.1 percent from 2014 to 2015. Moreover, this estimate likely overstates the true growth in per-enrollee costs, since it does not account for improvements in data reporting which likely increased measured PMPM costs. Available evidence implies that the slow ACA individual market cost growth results at least in part from a broader, healthier risk pool. Supporting that interpretation, states that saw stronger enrollment growth in 2015 saw larger reductions in costs. For example, in the 10 states with the highest 2015 growth in ACA individual market member months, PMPM claims costs fell by an average of 5 percent. Likewise, states with higher enrollment growth saw larger improvements in risk adjustment program risk scores. These data are encouraging for the future, since the Marketplace and the broader ACA individual market continue to grow.

As you'd probably expect, the insurance industry sees things a bit differently...

AHIP's Marilyn Tavenner (former CMS head) blasts today's federal report about #ACA risk pool as "overly optimistic." pic.twitter.com/DcesT1G7pf

— Bob Herman (@MHbherman) August 11, 2016

On the one hand, it's easy to dismiss Tavenner's gripes as simply the head of a major for-profit trade organization spinning a sob story. And to be honest, my opinion of Tavenner has been pretty sour ever since she (while still head of CMS herself) claimed that 400,000 more people were enrolled in ACA exchange plans than actually were...because someone at CMS screwed up and double-counted standalone dental plans as Qualified Health Plans...and then left just a couple of months later, only to pop back up again as the head of AHIP just a few months after that.

On the other hand...what's actually going on here? If CMS is correct that the average cost per member per month stayed essentially identical year over year, while the average premium prices paid increased around 8%, why are the carriers still claiming to be losing big bucks on the indy market?

Well, the answer seems to be that, as noted in Counihan's blog entry (and by myself a few days ago)...

Meanwhile, independent researchers recently estimated that 2016 Marketplace premiums are between 12 percent and 20 percent below what the Congressional Budget Office (CBO) initially predicted.

Let's say that the CBO assumed that ACA-compliant enrollees would cost $800 per member per month in 2014, $900 PMPM in 2015 and $1,000 PMPM in 2016...but the carriers only priced their 2014 policies to account for $600 PMPM, then raised rates by 7% in 2015 (to $642 PMPM) and another 8% last year (to $693 PMPM). Even if the actual 2015 cost turned out to still only be at $800 PMPM, that would still leave the premiums a good 13-14% below where they "should" be. Plus, of course, the carriers also want to bake in elbow room for their up-to-20% in overhead expenses/profits/etc. In other words, this "same PMPM cost year over year" news is a very good thing...but it still leaves a gap on the other side of the equation.

As for the "improvements in data reporting", SEP crackdown and attempts to quash short-term plans, there's no way of knowing what impact those measures will have until next year, since none were implemented until 2016 and some only kicked in over the past few months.

If the PMPM cost for 2016 also holds steady, then hopefully the ugly rate hikes being asked for 2017 will be the last time they hit double digits, as the rates finally catch up with the costs. If so, then we'll hopefully see things settle down into the low-single digit range (3-6%, perhaps) as things stabilize.

If not, however...