Improving Meaningful difference

Jed Graham raises a good point at IBD on the problems of medicaid managed care providers spamming the exchanges in Mississippi:

A closer look at Centene’s much-heralded success reveals that competition between insurers on the ObamaCare exchanges has become toxic. Rather than being a boon to consumers looking for a good deal, competition is hurting the very people it is supposed to help…

only 40% of the subsidy-eligible exchange population is enrolled in Mississippi, putting it dead last among states which didn’t expand Medicaid and, therefore, cater to a more heavily subsidized group.

Yet the fallout was entirely predictable. Although subsidies are pegged to the premium of the second-lowest-cost silver plan in each market, Centene has been scrunching down its silver premiums by making the plans more bronzelike…. That in turns shrinks the government subsidies that are supposed to help make coverage affordable. If customers can’t afford, or don’t like, the cheap Centene silver plans on offer, they’ll have less of a subsidy to buy a bronze plan or a silver plan from a competitor.

In the largest HealthCare.gov market in 37 states, the second cheapest silver plan costs, on average, 27% more than the cheapest bronze plan this year. But in Mississippi, where Centene’s Ambetter brand is the low-cost provider, the gap between bronze and silver is only 12.6% in 2016, down from 34.8% in 2015.

Ambetter/Centene has adapted a Silver spam strategy that has been a hobby horse of mine for a while.  The goal is to proliferate a bunch of plans tightly clustered around the subsidy benchmark point.  For markets where all insurers are paying near commercial rates, this is not a big deal.  In markets where one insurer is paying Medicaid plus something or Medicare like rates and everyone else is paying commercial rates to providers, this is a big deal. It forces heavily subsidized buyers to choose between half a dozen affordable plans that are functionally similar from the same insurer or not really be able to afford a choice.

This is due to the lack of good meaningful difference regulation.  Currently Meaingful Difference regulation is a farce.

One of six characteristics has to change for CMS to consider a cloned plan to be meaningfully different (from the 2015 regulations where we first saw this strategy explode).

The key characteristics were proposed in paragraphs (b)(1)–(b)(7), and include (1) cost sharing; (2) provider networks; (3) covered benefits (including prescription drugs); (4) plan type (for example, HMO or PPO); (5) premiums; (6) health savings account eligibility; and (7) selfonly, non-self-only, or child-only coverage offerings. We proposed that, at a minimum, a reasonable consumer would have to be able to identify two or more of the characteristics proposed at § 156.298(b) as different in order for the plan to pass the meaningful difference test…

To address concerns with the proposed meaningful difference standard, we have modified § 156.298(b) to have the standard set at one material difference rather than two, and have removed premiums as one of the characteristics among which plans must be different.

There are two ways to avoid the Silver Spam problem. The first is to have an active buyer exchange. Covered California allows insurers to offer a single product cluster to be introduced into a market. A single insurer can offer multiple products (see yesterday’s post on San Francisco for an example) if there is significant space between the two products in design and pricing).

That solution can work in states that have their own Exchanges and more importantly the political will to fight insurance companies.

A more straightforward solution is available. Fix the meaningful difference regulation as well as build better decision support tools into Healthcare.gov. CMS can do this in their normal rule making operations. I think it was not done as a means of getting good headlines on how low cost the exchange plans have been.

There are two types of fixes to the regulation. The first could be a set of check boxes. — a plan is meaningfully different if it has three elements different from its peers or four elements or two elements or whatever number of elements. One or two elements still allows for significant gaming. Three or four elements would get rid of most of the clones. I would also be inclined to mandate a standard benefit package (ie a Silver has a $X deductible, Y% co-insurance and A/B/C co-pays for drugs etc) as that is the cheapest area for an insurer to tweak things to beat a regulation.

The second is a bit more open ended. For a reference individual (I like using the 40 year old non-smoker with no dependents), a plan offered by the same insurer is meaningfully different if the premiums vary by Z percent. That Z could be 3% or 5% or 7%. After that, CMS would not care how the difference is achieved? Is it a PPO instead of an EPO? Who cares? Is it a narrow network versus a broad network? Who cares? Is it HSA eligible or not HSA eligible? Who cares?

The insurance companies could then design products where there actually is a useful difference in the products instead of attempts to game/hack the subsidy formula for business strategy reasons.

The decision support tools would require a standard format to allow a buyer to compare the (far fewer) plans offered by a single insurer in a single band. Mayhew Silver Value HMO is different than Mayhew Silver SuperDuper PPO in these 3 key elements of network, out of network benefits and PCP gatekeeping. Buying Mayhew Silver SuperDuper PPO will cost you 25% more.

This reduces information complexity for the buyer as it gets rid of a number of isomorphic clones while increasing the probability that there is a meaningful and realistic set of practical choices for subsidized buyers. The downside is that this will increase the cost to the federal government as advanced premium tax credit payments will increase. I am fine with that.