7 is the line between long and short

Rate filings are starting to come in for the 2020 ACA plan year. Insurers will file their initial rates that are based on all of 2017 and 2018 experience and a little bit of 2019 claims. These rates will change. Right now, it is an opening bid in a complex blended game of euchre and Uno with varying information being shared and hidden between competitors (this is a paper I want to write some day).

I’ve been fascinated by Medical Loss Ratio rebates for a while now. I long thought that 2018 was going to be overpriced and it was. We will see big rebate checks issued this fall for the policy years 2016-2018. The initial rate filings will tell us if 2019 will be a “normal” year or a slightly overpriced year which will determine the size and spread of 2020 MLR rebate checks that will be handed out five weeks before the election:

The MLR calculation is based on a three year rolling average. The 2018 rebates are based on a high MLR 2016, a normal MLR 2017 and a wicked low MLR 2018. Many insurers will see 2016 cancel out 2018 for this current round of rebates. However in 2020, the 2019 rebates will be based on a normal 2017, a wicked low 2018 and either a low or normal 2019

I think 7% average increase (plus or minus a little) will be the dividing line between 2019 being priced roughly right and being overpriced.

Insurers are not allowed to set future rates to make up for a bad past year, but they can use the information of a bad year to adjust their estimates on morbidity, enrollment, duration and other relevant factors. If we assume that insurers fundamentally got 2019 right in their projections of morbidity and expenses, we should expect to see a trend increase of five points or so and then another couple of points for declining morbidity due to more underwritten exit points being available.

Insurers that got 2019 wrong and thought that 2019 was going to be way healthier and cheaper to cover will probably ask for and get rate increases significantly above seven percent. Insurers that got 2019 wrong in the other direction because they thought the 2019 cohort would be way more expensive will probably ask for low single digit increases or small decreases.

There will be significant variation between initial rates and final rates. Final rates will be based on a different competitive market as new insurers enter and old insurers exit particular regions. They are also based on another several months of claims run-out so what actuaries are inkling in May can be more confidently projected in July and August. Furthermore, several states are still looking into filing new reinsurance waivers while California is looking to get creative with a non-waiver affordability package.

But 7% is a decent rule of thumb. If insurers are asking for 7% or more in a state, we can probably say that their 2019 experience will be a “normal” mid-80s MLR for 2019. If insurers are asking for flat rates or decreases, then I’m fairly confident in saying that 2019 will have a sub-80% MLR which will lead to bigger rebate checks in 2020.

5 replies
  1. 1
    wvng says:

    Just want to say that even if I don’t comment on your posts, they are always fascinating dives into the health insurance maze in our country. I always learn from your writings. Thank you.
    Since this post is on the MLR, do you have a sense for how much insurance company gaming goes on of this key cost reduction element of the ACA?

  2. 2
    Butch says:

    If our current rate goes up 7 percent I will have to go without health insurance.

  3. 3
    narya says:

    First, it’s “euchre.” Second, one of the interesting thing about watching people play euchre–particularly people (**cough**from Wisconsin**cough) who grew up playing it–is that they often do not play out a hand. They often know, based on the first couple of tricks, what the outcome will be. That is, there is actually quite a bit of information around the table. I’m wondering how much of the health insurance industry functions similarly. Your posts and analyses show that, for people such as yourself, there is a good bit of information available–and I suspect that all insurers have their own internal algorithms. THAT is the part I would like to see–what are the pieces of information they have, perhaps from their claims, that they do not share and that affect how they play the yearly rate-setting “hand.”

  4. 4
    Kraux Pas says:

    That’s a pretty provocative title you got there.

  5. 5

    @Kraux Pas:
    Get your mind out of the gutter

Comments are closed.