Selling across state lines (again)

Earlier this week, the Center for Medicare and Medicaid Services (CMS) sent out a request for information on how it should think about implementing Section 1333 regulations of the Affordable Care Act (ACA). Section 1333 allows for voluntary interstate compacts where a plan approved in one state would be approved for all member states of that compact. This was supposed to be launch in 2016 but the regulations have never been written.

There are three different proposed ways a single insurer can sell across state lines.

  • Interstate compact where participation is an opt-in
  • Open the gate and invite anyone to sell
  • Race to the bottom credit card regulation style

Maine and Georgia, among other states, have elected to open up their gates and invite any approved insurer into their state.  As of this week, no out of state insurer has elected to sell a non-locally approved product in the individual market.  The Race to the Bottom was the AHCA.  Interstate compacts are Congressionally authorized and approved agreement among several states.  There are different flavors of compacts and a health insurance compact may not necessarily be a 1333 agreement.

I am not too worried about Section 1333 from a race to the bottom perspective as they are voluntary and limited.  Massachusetts could theoretically approach Rhode Island for a combined market and Rhode Island could theoretically agree.  They would have to hammer out regulatory differences but there would be a shared understanding that they don’t want Texas or Oklahoma standards in their market.  Oklahoma and Texas could also talk and form a compact based on an understanding that they don’t want Massachusetts standards in their market.  It is based on a commonality of interest and approach.

I think that if two or more states were to enter a 1333 compact, the states would be fairly similar in approach and expectations so there would be minimum bottom dropping concerns.

I don’t think that selling across state lines will do much if we assume that there is not a race to the bottom in standards.  It does have some value.  Many state based individual market risk pools are small.  I have 2018 enrollment data open for another piece right now; 16 states had less than 100,000 initial effectuated buyers on-Exchange, 9 had less than 50,000 effectuated on-Exchange buyers, and 4 had less than 25,000 effectuated on-Exchange buyers.  Small risk pools means high idiosyncratic risk which has to be covered by either reinsurance policies (which can be expensive if bought through a third party) or higher premiums.

There is value in merging small risk pools.  I could see the logic of North Dakota, South Dakota and Wyoming (three of the four sub 25,000 states) cooperating in order to minimize catastrophic claim risk.  That is valuable and it would reduce premiums by a couple of bucks per member per month.  It won’t solve the underlying structural cost drivers of care, but it would shave off a point or two in costs.

I can also see the value of an interstate compact between West Virginia, Ohio and Pennsylvania that applies only to the Northern Panhandle.  Insurers in Ohio and Pennsylvania already have the hospitals in Wheeling and Weirton (near Balloon-Juice world headquarters) in network because that stretch of the state is a long jog wide at most.  A Pennsylvania insurer (Highmark) and an Ohio insurer (Care Source) already sell in West Virginia so a compact could reduce administrative costs by a few dimes per member per month.  Other Pennsylvania and Ohio insurers could theoretically be interested in selling in the West Virginia Northern Panhandle if there were lower barriers to entry regarding licensing.

As long as the interstate compacts are voluntary, opt-in agreements, this is an edge smoother.  It is not a game changer.



MLR is coming

We should expect significant Medical Loss Ratio (MLR) checks to be written this summer for the ACA individual market. This is independent of whatever the courts decide on Cost Sharing Reduction (CSR) litigation. If the current decisions that there has been a broken contract with no mitigation through Silverloading stands, MLR checks will be bigger and could come in two waves.

This is not surprising. In June 2018, it was evident that insurers were overpricing 2018.

Over the past couple of years, there have not been large checks cut as insurers weren’t making large profits anywhere in the individual market. That is changing. 2017 looks to have been a very profitable year for insurers. 2018 looks to be even more profitable. There is a good chance that the 2016-2017-2018 time period will produce several states with an average MLR well below 80% as the first quarter results plus initial 2019 rate filings strongly suggest that insurers in many states overpriced their premiums for 2018.

States that have mostly monopolistic insurers and large Silver loads are more likely to have significant MLR checks.

The actuaries are starting to confirm my inkling.

2018 was massively overpriced. It is some combination of monopoly markets, uncertainty over the behavior of CSR eligible individuals shifting to other plans, Silverloading bringing in healthier than expected individuals and continued policy/political uncertainty. Any insurer that is competently run should be Scrooge McDucking it. Milliman shows this.

MLR rebates are normally calculated in mid-summer with the checks being sent out in early fall.

It looks like the three year rolling average will have much smaller MLR checks landing in voters’ mailboxes a few weeks before the presidential election than this year. I had a moment of severe cynicism in 2017 when I figured out how to game the system so that insurance commissioners could make themselves look real good right before an election:

in the fall of 2020, ambitious state insurance commissioners will be handing out rebate checks in late September as they are running for Governor or the Senate. Or if they are a bit less ambitious, they are supporting the incumbent party by handing out checks and injecting new federal money into the state and making the fundamental background economic picture a bit better than it otherwise would have been.

I might be getting too cynical today.

There are games that can and will be played with MLR rebates, but this nakedly cynical one looks to be either isolated or non-existent.

Paying enough for prevention

Most vaccines create a cost in the current time period for a benefit at some distant point in the future of an avoided disease. This creates a “have to” versus “want to” situation for insurers.

Insurers, for most preventive care services, are in the same mindset I was in as a teenager when it came to cleaning my room. Good enough was a low bar to clear. Clean claims will be paid on time and in full but without extra motivation, the business case to promote most prevention services is weak for an insurer that figures they’ll not cover that person in a few years. Motivation is needed.

Insurers will pay claims if they come in as they have to. They may not want to encourage more people to get the preventive services who otherwise would not have.

My Duke Margolis colleague, Dr. Peter Ubel, raises this point from the provider perspective on HPV vaccination rates and reimbursement. HPV vaccines are administered to young teens. The pay-off is a decade or more later in avoided cervical, throat and lip cancers.

something else that’s preventing kids from getting vaccinated ….

The HPV vaccine is expensive. According to the CDC, the 3 doses needed for complete vaccination cost almost $500….

According to a study out of the CDC, there is sizable geographic variation in how well providers are reimbursed for vaccine administration. The most generous state is Pennsylvania, where private insurers pay an average of $194 to physician practices for administering the vaccine, and thus almost $600 for all three. The next most generous state is Nebraska: go cornhuskers! But in last place stands the terrapin state, Maryland, where providers can expect to receive an average of only $150, with neighboring Washington, D.C. not far behind (ahead?) at a rate of only $154.

What is the result of this stinginess? Areas with lower vaccine reimbursement rates also have lower vaccination rates. According to the CDC team, a $1 decrease in reimbursement for the vaccine is associated with 25,000 fewer adolescents getting at least 2 doses of the vaccine.

This is an area where the business case for vaccination falls apart from the point of view of the current insurer. They spend money now and there is no way in hell they will ever see a benefit in averted costs. Therefore, they’ll pay the HPV claim if it is presented to them but they will work hard to not get the services sent to them.

The business case solution is to make offering HPV a break even or a profitable service for both the clinicians and the insurer. That means raising the payment rate insurers send to clinicians so that the docs will shoot their patients up with the vaccine. It also means adding HPV and other potentially expensive vaccines with long/slow payoffs to risk adjustment systems. Doing that will create a strong business reason for insurers to both want to and have to cover HPV vaccinations.

Right now, the solution of mandating HPV vaccine claims to be paid is not working as the insurers have to but don’t want to.

Silver or Gold benchmarks

Aang asked a great question in comments yesterday:

How difficult would it be to make the benchmark plan the second cheapest gold plan? Would it be worthwile? Seems it would give bigger subsidies and less cost sharing for those who don’t qualify for CSR.

This is an important question because there is a wide variety of ACA 3.1 plans floating around out there. Most attempt to use an appropriation of CSR to pay for other things. Those plans attempt to hold harmless beneficiaries of Silverloading by either expanding CSR eligibility up the income scale or resetting the benchmark to a Gold plan instead of a Silver plan.

Charles Gaba has a great chart on the options.

There are trade-offs to beefing up a Silver benchmark with more CSR versus less CSR but resetting the benchmark to Gold.

Keeping the benchmark at Silver lowers average actuarial value for the lightly subsidized folks. There is a chance that these folks will see a below benchmark premium for a single Silver plan and an array of Bronze plans. There is little chance (using 2017 data to back out Silverloading) that someone making 400% FPL will see a Bronze plan for under $100 a month. And the people who will see that low cost Bronze plan will be married late 50s and early 60s in very high cost regions; they will not improve the risk pool.

Resetting the benchmark to Gold will make CSR less relevant as the benchmark is doing a lot of the work of upping actuarial value to middle income individuals and families. The market will be more sorted by health status as healthier buyers may choose to move from an 85% AV Gold CSR plan to a 70% Silver plan to save on premiums while individuals and families that anticipate high costs will stay in the CSR plans. This is a risk adjustment challenge but it is surmountable.

More importantly, for the families that are not CSR eligible, they will be exposed to far more zero dollar plans in Bronze. Zero premiums may matter for incremental enrollees who are flipping a coin between buying or not buying insurance. Dr. Coleman Drake of Pitt, and I have some preliminary work that we’ll be presenting at Academy Health’s Annual Research Meeting in June on this subject.

From a mechanical standpoint, insurers will want to know what the rules are a good nine to twelve months before open enrollment. Once they know what the rules are and are confident that they are stable for the upcoming policy year, the implementation of a Silver or a Gold benchmark should not matter too much to the insurers. At the policy level, it is a harder decision as Gold benchmarks, all else being equal, will lead to higher enrollment of the marginal buyer which means incrementally healthier risk pools.

Industry’s proposal to use CSR

Late last week, the Blue Cross and Blue Shield Association released a three page proposal to for an ACA 2.0 bill. This proposal identifies the biggest issues as not enough young people in the pool, idiosyncratic risk and premiums being too high for folks who currently are not eligible for premium subsidies. The pay-for is appropriating Cost Sharing Reduction (CSR) subsidies.

  • Younger adults pay a lower percentage of their income (at a given level) for the benchmark plan
  • Older adults are held harmless
  • All individuals, regardless of income, are eligible for subsidy assistance
  • CSRs appropriated
  • CSRs expanded
  • Full advertising and outreach funded
  • Health insurance premium tax suspended

This is an industry proposal.  It is a first cousin of the Democrats’ proposed ACA 2.0.  The Democrats in HR 5155 were indifferent to the health insurance premium tax but they sought to expand premium tax credit eligibility and upped CSR payments to higher income groups.

These are discussions that are going in the same general direction.

The ACA was the result of several years of stakeholder consensus building. Advocates were involved. Wonks were involved. Hospitals were involved. Physician groups were involved. Insurers were involved. Pharma was involved.

It looks like the insurers are trying to lay markers for where they want to see things in 2021 or 2022.  They are looking at a fix and expansion of the current paradigm instead of a complete replacement of the system.

We should look to see what the other major interest groups that can elect to drop $100 million in support or in opposition to major health policy bills.  There policy papers will shape the contours of possibility and the trade-offs of ambition.