Working through CSR actuarial value games 2014-2017

WARNING This is esoteric even for me.

I’m trying to wrap my head around incentive structures for insurers to design silver plans in a Silverload world. I need to write out the old system first.

Cost Sharing Reduction (CSR) subsidies in the 2014-2017 period offered insurers some interesting incentives on how they could line up their actuarial value of the base silver plan and the CSR variants. CSR was a dollar for dollar reconciliation of the difference between what the insurer would have paid on any given claim for the baseline silver plan and what it paid out on a CSR variant plan. Insurers (in most states) have the option of designing their own silver plan benefit structures. Games can be played once the mechanics are understood.

CSR is only available to Silver buyers. It is structured as an add-on instead of a separate plan even though it is experienced as a distinct plan from the buyer’s point of view. Insurers build a base silver plan that is used to generate the premium level and thus the advanced premium subsides. The CSR benefit is a variant of the base plan. The insurer pays a claim and then reconciles the difference between what the base plan would have attributed to beneficiary cost sharing and what the CSR variant attributes to beneficiary cost sharing. That incremental difference is paid on a claim by claim basis by the federal government. The Feds will front an estimated CSR payment and reconcile on the back-end. The payments are very timely and CSR keeps the insurers’ doctors and hospitals happy as they don’t have to chase anywhere near as much bad debt.

From 2014-2017, silver plans had a target actuarial value of 70%. This meant for a standardized statistical pool at a typical contract rate, the insurer would pay 70% of the claims incurred for essential health benefits and the people in the pool would pay 30%. That 30% is made up of some combination of deductible, co-insurance and co-pays. Those combinations can vary wildly between insurers and within insurers.

The 70% target had wiggle room. The “de minimas” variation was +/- 2 points. A silver plan could be anywhere from 68% to 72% actuarial value. Higher actuarial value means, all else being equal, lower cost sharing and higher monthly premiums. Lower actuarial value means higher cost sharing and again, all else being equal, lower premiums.

CSR variants (73% AV, 87% AV, 94% AV) have slightly different allowable error bands. They are restricted to +/- 1 point from the target. A 94% AV CSR silver plan could actually be anywhere from 93% AV to 95% AV.

The base plan actuarial value is completely independent of the CSR variant’s actuarial value. An insurer could choose to have a 72% AV silver base plan and a 95% CSR variant. That same insurer could choose to offer a 68% AV silver base plan with the same 95% CSR variant or it could pair that 68% AV silver with a 93% CSR variant. These choices are not regulated. An insurer has a twenty seven point spread to play with the 94% AV CSR variant. An insurer can offer whatever combination it wants.

Let’s assume that all else being equal lower AV silver plans have lower premiums than higher AV silver plans. We need to remember that the benchmark for advanced premium tax credits (APTC) is the second least expensive silver plan in counties with multiple silver plans or the only silver plan in a single silver county.

I am curious about the following game —
In counties where there is a reasonable amount of competition, we should expect aggressively smart insurers to capture the benchmark point and the least expensive silver point. This can be done by either offering a low actuarial value silver or having such a pricing advancing on provider contracts and a very narrow network compared to other insurers that a few AV points don’t matter. And from there, a smart insurer with a low AV silver as the cheapest plan should be offering a very high AV CSR variant. We should expect a lot of 68-95 combinations and very few 72-93 combinations for the insurer that captures the cheapest benchmark position.

Doing that would have allowed the insurer to offer a great price proposition reasonably healthy buyers while not foregoing any actuarial value advantage to the most price sensitive buyers.

I also think that this matters far less in monopoly markets. Altruistic insurers and insurers that own or are owned by dominant hospital groups (like my former employer UPMC Health Plan) may be motivated to offer high AV CSR variants as well.

I’m not sure how this changes in a Silver Loading universe. I think loading the CSR into the silver premium rejiggers the incentives but I am not sure which way those will move yet.