CBO Report: The Fine Print of Creating 23 Million Uninsured

If you want to know the basics of what’s in it, there are plenty of headlines today. If you want more in depth analysis, there is also no shortage out there. As always, I’m left with the luxury of writing about only what I want to. Today, I’m going to leisure cherry-pick sections of the report I find interesting and/or underreported. Now – we know about cognitive biases here, and I don’t deny that I’m more likely to choose the sections that re-affirm my priors, proving that I was right all along. I don’t claim to be impartial about a bill that I think is utter garbage, written by a group of people who I think have no little interest in improving the market for health insurance. After reading this report, it seems pretty clear that the CBO agrees with me, but hey, maybe we’re both wrong, along with basically every health advocacy group, policy shop and think tank.

All of the quotes you see here are direct quotes from the Report linked above, with page numbers. It’s going to get technical a few times, my apologies – but asking questions is what Facebook comment section is for!

“…The agencies expect that some people would use the tax credits authorized by the act to purchase policies that would not cover major medical risks and that are not counted as insurance in this cost estimate.” (Page 1)

“CBO and JCT broadly define private health insurance coverage as consisting of a comprehensive major medical policy that, at a minimum, covers high-cost medical events and various services, including those provided by physicians and hospitals.” (Page 2)

“In this estimate, people who have only such policies are described as uninsured because they do not have financial protection from major medical risks.” (Page 2)

We already knew that the CBO considers a person holding only non-comprehensive insurance as being uninsured. They re-affirm this at the beginning; the Report includes some more details about what it means to be comprehensive. The CBO also recognizes that the Government will be spending money on subsidies that go to people who do then not become insured (by the CBO definition). More on this later.

“Under the legislation, a few million of those people would use tax credits to purchase policies that would not cover major medical risks.” (Page 4)

I have to disagree with the CBO here; why would an uninsured person not take advantage of free money to buy whatever coverage they can? States would have to permit it – but they would also be incentivized to let their citizens get their hands on this sweet, sweet Federal cash. The CBO agrees that insurance companies will be incentivized to offer these kinds of plans.

I don’t see where the CBO includes the cost of subsidizing these plans, so I did a back of the envelope calculation. I use the CBO’s estimate of the uninsured and their breakdown by age and Trumpcare’s subsidy levels. I assume that 20% of the uninsured will claim their tax credits in 2018, going up to 50% in 2027; I think this is conservative, as 25 million-odd people would be missing out on free money. The cost for the Government to provide subsidies to people who are still uninsured? Around $750,000,000,000 for the 10-year period. That’s billion. More subsidies would be going to uninsured people than insured people – a truly impressive feat. The CBO assumes “a few million” but I don’t see why this assumption makes any sense. In their analysis, 40-50 million people would give away free money.

“[In the current ACA-based market] Several factors could lead insurers to withdraw from the market—including lack of profitability and substantial uncertainty about enforcement of the individual mandate and about future payments of the cost-sharing subsidies to reduce out-of-pocket payments for people who enroll in nongroup coverage through the marketplaces established by the ACA." (Page 4)

Here the CBO says it explicitly: the work of House Republicans and the Trump White House to un-guarantee the CSR payments and question the Individual Mandate is increasing premiums and causing insurers to leave.

“However, the agencies estimate that about one-sixth of the population resides in areas in which the nongroup market would start to become unstable beginning in 2020.” (Page 5)

The CBO says that states that take the MacArthur waivers will destabilize their health insurance markets. This is the “choice” and “returning decisions to the states” that the GOP always talks about. It goes in to describe that for states that waive the ACA’s health insurance regulations, medical underwriting will return and people with medical conditions will not be able to buy insurance.

Referring to states that eliminate only some of the ACA’s requirements:

“Average premiums in the nongroup market would be roughly 20 percent lower in 2026 than under current law, primarily because, on average, insurance policies would provide fewer benefits.” (Page 6)

“Out-of-pocket payments for people who have relatively high health care spending would increase most in the states that obtained waivers.” (Page 7)

Basically, the reduction of premiums in these states results from the product set being worse. Now, Paul Ryan would tell you that this is a choice: people are buying plans that cover what they need and not spending money on things they don’t. This is NOT what the CBO says – let’s remember the Fundamental Theorem of Health Insurance:

If it's not required for everybody, it will not be available for everybody. If everybody has the choice, then nobody will have a choice.

Going back to the CBO:

“Insurers generally would not want to sell policies that included benefits that were not required by state law. Plans with additional benefits that were not mandated would tend to attract enrollees who would use them and thus increase insurers’ costs.” (Page 25)

“For example, maternity benefits could be sold as a rider by some insurers in states that did not include maternity care as an EHB. Insurers would expect most purchasers to use the benefits and would therefore price that rider at close to the average cost of maternity coverage, which could be more than $1,000 per month. (The average cost of pregnancy care and delivery is about $17,000 for women with private insurance coverage.) Alternatively, insurers could offer a lower-cost rider providing less-than-comprehensive coverage—with, for example, a $2,000 limit. Either type of rider would result in substantially higher out-of-pocket health care costs for pregnant women who purchased insurance in the nongroup market.” (Page 26)

The CBO has just confirmed our Theorem! If there are benefits that are optional, then they will generally be priced at a level that makes no sense for people to buy them. It won’t make sense for anybody to buy maternity coverage, or drug coverage, or mental-health coverage. That’s not a choice.

“A reduction of $834 billion in federal outlays for Medicaid;”

“Savings of $665 billion stemming mainly from eliminating, in 2020, the ACA’s subsidies for nongroup health insurance—which include refundable tax credits for premium assistance and subsidies to reduce cost-sharing payments…”

“A cost of $375 billion for the new tax credit for nongroup insurance established by the legislation in 2020;”

“A reduction in revenues of $210 billion from eliminating the penalties paid by uninsured people ($38 billion) and employers ($171 billion);”

“An increase in spending of $117 billion for the Patient and State Stability Fund grant program.” (Page 13)

These are the costs/savings from what I consider coverage provisions. They total to a savings of 797 billion dollars over the period in question. Against this, there are around 204 million “uninsured-person-years” created (i.e. an average of 20 million additional uninsured per year for 10 years). Back of the envelope – it costs the Government $4,000 per year to subsidize a person to buy insurance. Obviously, a big difference depending on the person and the program, but this seems about right.

“Also, more employers would offer coverage to their employees because the available nongroup coverage would tend to have higher out-of-pocket premiums for people currently eligible for subsidies and because the plans would tend to provide fewer benefits.” (Page 17)

Several times in the report, the CBO says that having crappier insurance (sorry, that’s another technical term) available will lead to more Employer Sponsored Insurance (ESI). I don’t know why they think this – history doesn’t bear it out. Remember all the way back to Volume 2, when we looked at how the percent of the labor force with ESI changed leading up to the passage of the ACA:

This was a period where the insurance available on the non-group market was getting markedly worse and more expensive – just as the CBO says would happen if the AHCA were passed. Employers did not respond by offering more or better ESI. Why would this time be different?

“To evaluate the potential effect of the $8 billion fund [for high-risk pools], looking back at the high-risk pool program funded by the ACA prior to 2014 is useful. Within two years, the combined enrollment of about 100,000 enrollees in that program resulted in federal spending of close to $2.5 billion.” (Page 29)

The $8 billion set aside for high-risk pools are woefully underfunded, as we already knew. The pool sets aside $1.6 billion per year for these pool; this amount was previous enough to cover a bit more than 100,000 people. My estimate is that there will be 15 million people in the non-group market with pre-existing conditions; it could be more depending on how the MacArthur waivers are taken-up. So, there will be enough money to fund 1 out of every 150 people who need to use the pool. Oh, and after five years there is no money at all!

“In addition, the marketplace reports the amount of the credit paid on an enrollee’s behalf throughout the plan year to the Internal Revenue Service (IRS), which uses the information to verify what appears on the person’s tax return.” (Page 29)

“It is unclear how quickly implementation could take place or what type of information would be provided to the IRS to enable it to match payment records against information reported on tax returns, particularly in states that would obtain waivers for market regulations.” (Page 30)

Remember all the complaints from the GOP about the IRS being involved in your healthcare decisions? Like every other problem the GOP claims about Obamacare, the AHCA does nothing to solve that. It’s also interesting to note that at several points the CBO talks about how difficult the AHCA would be to implement, especially as states gain different MacArthur waivers over time.

I could go on – but that’s a good summary of a few things that didn’t make the headlines. This is a devastating paper, far worse than the previous one. I keep going to back to my old description of Trumpcare: no matter what you think the problem is in our health care system, this doesn’t solve it.