Last week, the Centers for Medicare and Medicaid Services released the 2016 premium data for the "benchmark" plans in the states using federal exchanges. Those are the second-lowest-cost Silver plans in each area, a benchmark chosen by health-care experts using arcane methods involving chicken entrails and a pound of dry rhinoceros horn.
(Just kidding! They don't use rhinoceros horn, which would be illegal. They use the ground-up bones of an ox slaughtered at midnight on the summer solstice.) This data, which showed premiums rising an average of 7.5 percent, is useful. But it is limited. We'd like to think that this tells us "how much premiums went up," but it's not that simple.
First of all, while the subsidies are pegged to the benchmark plan, not everyone buys that plan. Some people buy the cheapest Silver plan. Some people buy a more expensive one. Some people buy an even cheaper Bronze plan, and some people spring for Gold or Platinum coverage. Knowing how the price of the benchmark has changed tells us a great deal about how the subsidies will be calculated in 2016, but not nearly as much as we'd like to know about what people are experiencing in the marketplace.CartoonDavies' latest cartoon: Transition of powerCommentSubmit your letterReader essaysGet published in Newsday
Moreover, the change in the benchmark rate doesn't even tell us about the cost of any particular plan. Last year's second-cheapest policy could be this year's most expensive, if the insurer decides that it got the pricing wrong, while some other insurer slides into the benchmark spot.
Does that matter, you may ask? This tells us what price people can buy insurance at.
But insurance is not an undifferentiated good, like pork bellies or concentrated frozen orange juice. People don't want to "be insured" so much as they want to "get health care" -- from providers they like, with the drugs and services they want in the benefit list. If the same package you got last year cost more, you probably aren't comforted to know that a different package you didn't want is still very affordable.
Besides, most people don't actively shop for health insurance every year; well under half of people who bought a policy in 2014 returned to the exchange to look for a new one in 2015. If their policy went up by more than the benchmark rate, then they will be paying the difference out of pocket.
And that matters a lot, because subsidies are probably the biggest thing keeping the markets from sliding into the dreaded adverse-selection "death spiral," in which sick people are more likely to buy insurance, so the pool is sicker and more costly, so prices go up, so the healthiest folks decide the insurance isn't worth it, so the pool gets sicker and more costly..
Contrary to expectations, the mandate really doesn't seem to be doing much to get people to buy insurance, at least yet (the penalty is set to go up again this year, and that may get people to pay attention). The subsidies, on the other hand, clearly have a large effect, which is why the customer base for the exchanges is so disproportionately composed of folks who are getting large amounts of taxpayer assistance to buy insurance. Anything that increases the gap between the cost of the insurance and the subsidy they are getting is therefore worrisome, if you want the exchanges to get and stay healthy.
It is possible to find out what different rates are being charged by insurers in many states; I have a giant spreadsheet of the 2016 rate increases for those states that make them available, and boy does that make me popular at cocktail parties.
Unfortunately, we still don't know what rate increases people are facing, because we don't know what individuals had in 2015, or what they'll buy in 2016.
This is frustrating. But a consulting company, Avalere Health, has provided at least slightly more data than we had before, supplementing the administration's release of the information on benchmark plans by looking at the cost of the cheapest Bronze and Silver policies. It's still far more limited than one would like, but looking at those rates does give us additional information.
The biggest thing they tell us is that, as I suspected when I wrote about the CMS release, the whole bottom of the market is undergoing a fairly massive repricing. In most states, the cost of the cheapest Silver plan, relative to the cheapest one last year, rose even more than the benchmark rate. And in most states, the cost of the cheapest Bronze plan went up by more than the cost of the cheapest Silver plan. (The average increase was 13 percent, but it looks as if it's unweighted, while the government used a weighted average. As I noted last week, the weighted average is usually the most meaningful national metric, but the unweighted can be revealing as well.) That still doesn't tell us what happened at the top of the market, with the carriage trade splurging on policies that cover nearly everything. But I think there's one thing we can say for certain: Insurers significantly underestimated how much people would spend on health care, even if they chose stripped-down plans that have narrow provider networks and require consumers to cover a significant portion of their own costs.
Some of this may reflect pent-up demand, if folks who had been putting off treatment for years went out and got all the services they'd been delaying. But the consensus seems to be that the pool is sicker than expected(and that includes unsubsidized folks who are buying insurance off of the exchanges, but form a single actuarial risk pool for the purposes of pricing the insurance). It's also older; the administration was looking for about 40 percent of exchange enrollment to come from folks 18 to 35, and at the end of the last enrollment period, that number was stubbornly stuck around 28 percent.
This is not surprising, exactly, but it is perhaps somewhat disappointing. Subsidies or no, people who don't get a lot of value out of insurance don't seem to be buying it. As a result, the cost of the cheaper policies -- the ones you'd expect the healthier folks to be buying, since they don't expect to use it much -- is going to have to go up, because the average cost to cover the people who remain is higher than initially hoped.
Does this mean that Obamacare is in the early stages of a death spiral? We're certainly not in death spiral territory now, and it's too early to say whether we're headed there, because the program is still evolving. As I say, the mandate has so far proven surprisingly irrelevant. But this year it goes up to a pretty hefty sum -- the higher of $695 per adult and $347.50 per child, or 2.5 percent of your annual income -- and perhaps people will look at the higher fees, sigh, and finally decide that they might as well pay a little bit extra and get some insurance. More realistically, perhaps they will freak out in spring of 2017, when they see what the mandate penalty does to their 2016 tax refund, and head over to the exchanges when the next open enrollment period finally rolls around.
To be sure, 2016 is also the year that the risk adjustment programs end, which will tend to push prices up. This law has a lot of moving parts, and it's hard to assess it while they're in motion. All we can do is wait and see where they finally end up.