Medical Loss Ratios: An actuarial wrinkle

NAIC appears ready has asked for another month to recommend rules for medical loss ratios, and including loss-control costs in the definition of losses appears to be getting serious consideration.

Recall that Obamacare requires insurers to post minimum medical loss ratios of 80% to 85%, and NAIC gets to recommend by June July 1 how medical loss ratios are defined. Lower loss ratios trigger a rebate to customers. So far most of the talk has centered on whether loss-control costs should be included as medical losses.

On its web site, NAIC’s Accident and Health Working Group had a meeting scheduled for today (May 24) to discuss a draft rebate calculation regulation, which it helpfully posted here. The MLR definition includes claims plus “any expenses to improve quality.” Loss ratios would be calculated by state by licensed legal entity and split by individual, small group and large group plan.

Thanks to the American Academy of Actuaries, there are two more adjustments, one a smoothing for large losses and one to reduce the chance of a false positive (that a plan was properly priced but booked a loss ratio below the minimum).

AAA, in a May 12 letter to NAIC, was concerned about credibility. Suppose a bloc of business is really tiny. Its MLR might be under 80% because of good fortune, not because the business was overpriced. Should the insurer have to issue rebates? AAA says no and offered three alternatives:

  • Roll up results from multiple blocs of business and use that number.
  • Smooth the loss ratio for large losses. AAA is talking about the sort of smoothing you’ll often see in casualty ratemaking, removing excess losses and replacing them with a factor that represents the average amount of excess losses. (ISO’s non-cat excess wind procedure has sort of the same idea.) That way the presence or absence of a large loss doesn’t drive the result.
  • Add an adjustment to the MLR that covers the amount of variability a bloc of business would expect. For example, a bloc of 50,000 customers might have an adjustment of 3.2 percentage points. If an insurer needed with that size of a book records a 78% MLR, it would record an 81.2% loss ratio for purposes of calculating the rebate. AAA notes that Medicare Supplement policies (Medigap) allow for such an adjustment. AAA proffers a table of adjustments but says it’s for illustration only.

The NAIC appears to have rejected the first idea but embraced the latter two. But its this last one that will probably generate the most heat, because it allows insurers to book a low loss ratio but not pay a rebate. And I don’t have a lot of faith in mainstream media to explain the math well.

But AAA argues that you need to make some adjustment because the rebates are one-sided, meaning the insurer may have to rebate if results are good, but would not be able to surcharge if rates tanked. So business priced to an 80% loss ratio would actually run slightly higher than 80%.

Example: Suppose you will either run a 40% loss ratio on $100 premium or you will run a 120%. Both are equally likely. If you book the 40% loss ratio, you’ll have to rebate $50. Under one scenario your loss ratio is 80% and under the other, it’s 120%. Your expected loss ratio is 100%.

The smaller the program the worse the problem, AAA notes in another letter. Even after the adjustment they recommend, plans would still run slightly above 80%, by a tenth or two.

Of course, a good actuary would want to price this stuff like any loss-sensitive policy or dividend plan and increase the premium need so that the expected loss ratio was 80%. In my example, they’d charge $150. With $40 of losses, they’d rebate $75 $100 and book an 80% MLR. With $120 of losses, they’d keep all $150 and book, well, 80%. (In real life, health business isn’t as variable as my example, so the rebates would be much smaller.)

Rebates would be due Aug. 31 the year after the plan year ends, which sounds to me like an incentive to have plan years end Jan. 1 instead of the standard Dec. 31.

Advertisements
Tagged

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: