CSR assumptions and impacts

Yesterday’s letter from Blue Cross and Blue Shield of Tennessee illustrates the strategic dilemma insurers face as they have to set prices for next year. What should they assume on Trump administration monkey wrenching?

Given the potential negative effects of federal legislative and/or regulatory changes, we believe it will be necessary to price-in those downside risks, even at the prospect of a higher-than-average margin for the short term, or until stability can be achieved. These risks include but are not limited to the elimination of Cost Sharing Reduction subsidies (CSRs), the removal of the individual mandate…

Now it is time for some game theory.

Let’s assume that executives at insurers are risk loss averse. The pain of losing $100 million dollars outweighs the benefit of making $100 million dollars as the executives may get fired for losing money. There are two cases we need to think through. The first is the simpler case when a single insurer is the only insurer in a region. The second case is when there are multiple insurers with two subcases.

Let’s start simple.

Here the insurer faces two choices. Assume that CSR will be paid or not. The government can make two choices: pay CSR or not pay CSR.

If the insurer and government agree (Yes, Yes) or (No, No) then the product is priced on the underlying assumptions and contours of the market. It is business as usual. The interesting cases are where there is disagreement.

If the insurer thinks the government will fund CSR and thus price their products at medical growth rate and then the government does not pay CSR, the insurer is in severe trouble. A very large, very well capitalized insurer can take the hit of a loss but smaller insurers will go under in weeks as the state regulators have a legitimate reason to worry about solvency. If the insurer thinks the government will not fund CSR and thus prices very high but the government pays CSR, the insurer makes extremely outsized profits.

The Medical Loss Ratio (MLR) requirements will come into play. However MLR is a three year rolling average of claims over premiums. If an insurer had an MLR above 80% in 2016 and/or 2017, they can carry forward the credits to smooth out some of the excess gains so there would be less pay back to members. In a skewed way, this outcome cell in this scenario is an extremely convoluted and inefficient risk corridor.

Now what happens if there are multiple insurers in the market? We have two sub-cases. The first case is when all insurers make the same assumption. The second case has insurer decision making variance.

Same decision making is a bit more interesting. (Yes, Yes) and (No,No) has all of the insurers operating in a normal market. Insurers that offer the best products with the best value proposition regarding relative post-subsidy price should pick up most of the membership. This is effectively boring.

But things get a bit odder when all carriers agree but get the government action wrong. If the carriers think the government will pay but the government does not, the outcomes are unequal. The carrier with the absolute worst value proposition is not hurt that much. No one wants to buy that product anyway. It is a minor hit for an ugly plan.

Now if a insurer offers a very attractively priced plan with a good network and good customer service, they attract all of the membership. That usually is a good thing! Not in this case. Offering the most attractive plan means that plan takes all of the losses. And if that plan goes under, the special enrollment period refugees will choose the next most attractive plan, sending it under. This is an odd winner’s curse scenario.

If there are multiple insurer and they disagree the dynamics are interesting.  Insurers that assume they won’t be paid CSR always survive.  They won’t get much membership as they are massively overpriced for the market.  They will keep some portion of their sickest membership so they will be net risk adjustment recipients and their administrative costs relative to premiums will be very high.  But they survive.

In this split decision making scenario, the carriers that are optimistic that CSR will be paid are in an all or nothing scenario.  If they are right and the government pays CSR, they get almost all of the membership in the market.   Assuming they priced appropriately, they should make good money for the year.  If they guess wrong and CSR is not paid, they get all of the membership and the state regulators shut them down due to either a premium deficiency reserve (PDR) event as they have to come up with an extra 20 points of actuarial value that is not being compensated by the government, or there is a risk based capital problem as they have too many members to be safely covered by their current reserves.

But the dynamic changes once all of the optimistic insurers fold, liquidate or withdraw.  The surviving insurer(s) will have priced appropriately to handle the special enrollment period refugees from the optimistic insurers.  The pessimists will do fine in this scenario.

I am having a hard time seeing (with a strong risk aversion assumption) why a carrier would price on the assumption that CSR will be paid. There is little pay-off and great risk.

 

/game theory