Price levels by payer

Health Affairs has a new article out that looks at relative prices paid by a variety of insurers for common primary care office visits.

Third party means the insurer and out of pocket is the combined co-insurance and co-pay that the patient pays.  There are a few big take-aways.

The three private programs (Employer, Marketplace/Exchange and other (underwritten) individual market) look a lot alike in what they pay the docs.  Government programs pay docs less.    Cost sharing paid by the patient is a lot less as well for the government programs.   Medicare pays docs more than Medicaid.  Medicare has higher cost sharing than Medicaid.

This paper is not a revelation.  It is confirmation of plenty of previous work on relative pricing of different payment mechanisms.  The useful addition is the information that the Exchange payment levels looks a lot like other private insurance options and not like Medicare or Medicaid.  From here, this leads to the obvious insight that insurers that can offer networks priced like Medicare or Medicaid will have significant pricing advantages over insurers that pay their docs like they are part of a group network.

Last July, I wrote that many proposals are attempts to get a larger wedge of services paid at near Medicare rates:

Most liberal health policy goals have a very simple summary: get more people on insurance that pays providers rates that are closer to Medicare rates than commercial large group rates. Large group rates pay providers between 40% and 100% more than Medicare for physical health service. Moving the entire employer sponsored coverage universe to paying Medicare like rates would knock 30% off of the current bill.

We see this in Exchange. There is significant configuration convergence caused by both the subsidy formula and the risk adjustment formula. Plans that are profitable tend to be paying providers Medicare plus a little bit while offering narrow networks. We see this in the proposal to move the Medicare buy-in age to 55. We see this in the proposal to have a public option. The 2009 House public option was pricing out at Medicare plus a bit. All of these efforts are just different ways to achieve an underlying goal of reducing provider compensation by lowering the average payment per service by having more people move from high payment to provider coverage to Medicare based pricing.

This figure shows why that is the case.


1332 as a state block grant

The Section 1332 waiver in the Affordable Care Act (ACA) is a state innovation waiver.  It allows states to meet or beat the standard coverage of the ACA through whatever means possible. The state can use the funds that the federal government would have used for Premium Tax Credits (PTC) and Cost Sharing Reduction (CSR) subsidies to fund the state plan. There are four major guardrails that measure whether or not a state’s proposal qualifies.  A 1332 must:

  • Cover at least as many people
  • Cover those people at least as well
  • Cover those people with the same or less cost sharing
  • Cost the federal government no more than the federal government would otherwise spend.

The current guidance tightly defines these criteria.  The Alexander-Murray Senate bill proposed looser criteria.

There have been a good amount of interest in states using 1332 waivers.

Some of the waivers are minor technical changes; Hawaii is using an approved 1332 to close the state SHOP exchange and use that funding for another pre-ACA state program that performed a similar function.

Other changes are significant.  Alaska, Oregon and Minnesota have approved re-insurance waivers that use some of the APTC/CSR money to fund reinsurance in order to bring down non-subsidized premiums.  Iowa and Oklahoma have very ambitious, conservative leaning proposals that they pulled in the Fall of 2017 because the Center for Medicare and Medicaid Services was not going to rush their analysis.   Idaho has an extremely creative waiver it released that integrates Medicaid and the ACA in a creative fashion with creative rules lawyering.

One of the interesting things, to me at least, is that the waiver process is not being framed as a state level block grant.  The states are given quite a bit of freedom to choose a “better” method that fits their local needs.  Once the waiver is approved, the federal government writes a big check and then performs light monitoring of the state’s customized program.  That to me sounds like a block grant.



A short review of 2017

Now that I’m back from vacation, I should stop procrastinating and do a quick year in review.

My biggest miss

The first week after the 2016 election, I was convinced that the entirety of the ACA minus the Medicare Advantage cuts was dead.  Medicaid expansion gone. Pre-existing conditions gone. Essential Health Benefits gone. Subsidies gone.

I was wrong.  The individual mandate is gone and Cost Sharing Reduction (CSR)subsidies are a mangled mess of ineffective sabotage.

My most important post

CSR and the limited time fuse

I argued that the threat to terminate payments for CSR subsidies was limited in scope and duration.

The CSR threat loses its ability to blow up the market by sometime in the fall.

This post led me to believe that CSR payments required Republican concessions and not the conventional wisdom of Democratic concessions. From there, it led me to believe and argue that there will never by an appropriation for CSR again.

Proudest moment

Every time every one of you picked up the phone and called.  Every time that we stood for our values.  Every time that we looked at our world and tried to figure out how to make it better and not worse.

End Notes

2017 was a huge transition year for me.

Read more

Associations and Word Salad

President Trump spoke at length with the New York Times last week. In Adam’s post he finished up with this task and charge:

There’s a lot more at the link, including some stuff about health care associations that doesn’t make a lot of sense. Perhaps Dave will eventually try to make sense of it for us.

So let’s find this particular piece of word salad:

TRUMP: Wait, wait, let me just tell you. … Also, beyond the individual mandate, but also [inaudible] associations. You understand what the associations are. …

[Cross talk.]

TRUMP: So now I have associations, I have private insurance companies coming and will sell private health care plans to people through associations. That’s gonna be millions and millions of people. People have no idea how big that is. And by the way, and for that, we’ve ended across state lines. So we have competition. You know for that I’m allowed to [inaudible] state lines. So that’s all done.

Now I’ve ended the individual mandate. And the other thing I wish you’d tell people. So when I do this, and we’ve got health care…….

I’ll tell you something [inaudible]. … Put me on the defense, I was a great student and all this stuff. Oh, he doesn’t know the details, these are sick people.

There is a charitable read and there is the read one has after going through It’s A Small World four times in a forty eight hour period.

The charitable read is the president is overclaiming credit for actions that have not yet occurred but are in the pipeline to occur.

Read more

Premium Spreads for Subsidized Buyers in a no Mandate World

The individual mandate is gone. And this helps subsidized buyers.

The Congressional Budget Office thinks that the lack of a mandate will raise individual market premiums by 10% because healthier people who either receive low subsidies or no subsidies will leave the market. A sicker risk pool on average leads to higher premiums. The pain overwhelmingly falls on the non-subsidized as the subsidy eligible portion of the ACA market, those families who make between 100% and 400% ($12,020-$48,080 for a single adult) Federal Poverty Line could see even better subsidies.

I’ve been fascinated for years now that the relevant number for subsidized buyers is not the absolute premium but the spread between any plan and the Silver Benchmark. For plans that cost less then the Benchmark Silver, the subsidized buyer gets a better deal. The bigger the gap, the lower the monthly premium. The lower the monthly premium, the more likely a low utilizing individual will buy a plan. The lower the monthly premium, the more likely a low utilizer will stick with a plan for the entire year.

The graphic below is just a toy model with 2017, 2018 and 2019 pricing for a single county with a single insurer offering a single plan on a single network in Bronze, Silver and Gold. I am making up the baseline pricing for illustration and intuition.

Let’s work through this underneath the fold:
Read more

Enrollment changes by CSR coping mechanism

We looked at a quick and dirty change in enrollment on yesterday. Overall enrollment on was down by about 4%.There was wide variance with Kentucky gaining more than 10% of its 2017 enrollment and Louisiana dropping over 20% of its enrollment. The ACA faced several massive shocks this year. The termination of Cost Sharing Reduction (CSR) was one of them. Most states had a plan to cope with CSR but that plan varied.

Below is a very quick state level chart of how the four different strategies played out. Silver Load meant that a state had every insurer put all of the CSR costs only onto the Silver plans. Broad Load had a state put the cost of CSR into all plans. No CSR Strategy was the North Dakota response of nothing while Mixed responses had insurers in a given state do many different things.

I am not controlling for anything. I think that there is a strong case to control for price spreads, network quality, insurer longevity and probably political orientation. An analysis that goes into a peer reviewed journal probably needs to be focused at the county level with a boat load of controls. I would be fascinated to see what has happened in the Mixed Strategy states as there are a couple of different flavors of mixed strategies.

This primarily is for my intuition. And it looks like nothing is obviously statistically significant. The outliers that I am very curious about are Louisiana and Arizona on the downside while Georgia and Kentucky seem to be upside surprises. I think there is preliminary evidence that Silver Loading helped enrollment compared to either Broad Loading or mixed responses but this will be dependent on better data and better methods than I have.

My data is here:

2018 Enrollment data is at CMS.

2017 state level data is from CMS PUF.

Exchange plan quality

DaddyJ asks a good question:

we here in NW suburban Chicago are seeing a palpable degradation in the quality of care being offered by BCBS narrow networks. My family (non-subsidized) felt driven onto a far more expensive Silver plan rather than stick with last year’s narrow-network Bronze, as we have seen with our own eyes in the past 6 months what looks like hospitals/docs in the narrow network having the life squeezed out of them by the dominant insurer….

This is a really good question.  Let’s start with some data and then theorizing.

The University of Pennsylvania LDI team tracks network size on the Exchanges.  Their most report showed that the super narrow networks are becoming less common:

We find that the overall rate of narrow networks is 21%, which is a decline since 2014 (31%) and 2016 (25%). Narrow networks are concentrated in plans sold on state-based marketplaces, at 42%, compared to 10% of plans on federally-facilitated marketplaces. Issuers that have traditionally offered Medicaid coverage have the highest prevalence of narrow network plans at 36%,

I think the major thing we need to think through is the incentives of pricing structures. Narrow networks, all else being equal, will have two sources of pricing advantages over broad networks. First, the narrower the network, the less likely it will include all the doctors and facilities that a person who knows that they need a lot of care will want to use in the network. This aspect of a narrow network drives some people who are disproportionally expensive to choose the broader network. Secondly, the narrow network is a subset of all available providers in a regional broad network. The narrowness gives the insurer a stronger ability to say no which means the average cost per unit of service can decrease.

The ACA subsidy formula gives a push for relatively healthy, subsidized individuals to buy the least expensive plans possible. This probably means either the least expensive or Benchmark Silver plans for people who qualify for Cost Sharing Reduction (CSR) help or Bronze plans for people who make more than 200% (~$24,040) and less then 400% (~$48,080) FPL.

We looked at Silver Spamming in October 2015:

We know that most subsidy receiving individuals who are buying on the Exchange are post-subsidy price sensitive. Owning the #2 Silver and then seeing a large gap between #2 Silver and the first competing Silver means that almost every price sensitive shopper who is healthy (as they don’t need a broad network) will buy Celtic/Ambetter. Individuals with known medical conditions are less likely to go to a narrow network plan because they already have relationships with providers that work for them. It is an attempt to buy healthy membership and dump sicker people to other insurers.

In these markets where there is a Mediciad-esque narrow network insurer, there is a strong incentive for every insurer to make their networks smaller and cheaper so as to claw back some of the reasonably healthy membership that otherwise would have gone to the hyper narrow network insurer.

This is the incentive to continue to go narrow and cheap. Most people don’t use many services so they are buying on premium more than anything else so steps that lower premium are valued by the insurer more than steps that make the experience pleasant.