Avik Roy, to his credit, had some appropriately harsh words for at least parts of the first version of the American Health Care Act, the House bill to replace the Affordable Care Act. He was particularly critical of those parts that would make coverage much more expensive for low-income and older persons. And yet, given that AHCA v. 2.0 is considerably worse than the original, I found a lot wrong with Roy’s claim in an NYT oped today that the House bill is “fixable.”

Roy claims that the CBOs score of the new version—the one that showed it to reduce coverage by 23 million—is “overly pessimistic” and that the bill can be fixed by raising premium subsidies to provide more help to those who can’t afford coverage (Roy wrongly dings the CBO, which has actually gotten much of this right).

Put aside for a moment that he is (reasonably) recommending an approach to premium subsidies that tracks existing ACA policy much more so than it does the R’s replacement. I wanted to highlight three misleading aspects of his critique, ones which underscore the extent to which this bill is far from fixable.

First, and most egregiously, Roy neglects to tell readers what the AHCA does to Medicaid, a shortcoming addressed by my CBPP colleagues in this new analysis:

“The [AHCA] would radically restructure Medicaid’s federal financing and effectively end the ACA’s Medicaid expansion, reducing enrollment by 14 million people by 2026 and cutting federal spending by $834 billion over ten years. The bill’s other Medicaid changes would cut another $19 billion over ten years. All told, the AHCA would have a devastating impact on health care for over 70 million people who rely on Medicaid, including over 30 million children and millions of seniors, people with disabilities, pregnant women, and low-income adults.”

Roy has consistently been highly critical of Medicaid and may view these cuts as more of a feature than a bug of the AHCA. But he’s wrong; Medicaid is an efficient program and provides access to care that is comparable to private insurance. It is more well-liked than private insurance for good reason (see second figure here; in fact, look at all of those figures; then look at this). As health analyst Tim Jost has written, studies upon which Roy bases his anti-Medicaid claims “do not actually show what Roy claims they do.” Ignoring the Medicaid cuts at the heart of the AHCA – cuts that are responsible for most of CBO’s projected coverage loss – may be a marginal improvement over making incorrect claims about the program, but it is hardly any more credible.

Second, Roy neglects to mention what makes AHCA v.2 even worse than v.1. Because the first version of the bill wasn’t harsh enough for hardliners in the House, the new one gives states the opportunity to seek waivers that would allow insurers to once again base premium charges on health status (i.e., it would usher in the return of “medical underwriting”). States can also waive out of covering essential health services required under the ACA.

CBPP:

“Before the ACA, plans frequently excluded coverage for services like maternity care, mental health treatment, or prescription drugs. CBO estimates that, under the bill, half the population would live in states where plans would no longer cover such services.  That means that even if people with pre-existing conditions could afford their premiums, their health insurance might exclude the treatment they need.

• Plans could again put annual and lifetime limits on coverage — including employer-sponsored coverage. Before the ACA, 105 million people, most with coverage through their employers, had plans with lifetime limits on benefits, meaning their insurance coverage could end exactly when they needed it most. Waiving the ACA’s standards for the services that plans must cover in effect waives its banson annual and lifetime limits as well.
• People with pre-existing conditions could face unaffordable premiums. CBO estimates that one-sixth of Americans would live in states that would let insurers set premiums based on health status in a substantial part of the non-group market. In these states, CBO concludes, “less healthy individuals (including those with preexisting or newly acquired medical conditions) would be unable to purchase comprehensive coverage with premiums close to those under current law and might not be able to purchase coverage at all.”

Third, Roy’s premium cost comparisons are misleading, as he’s comparing (rotten) apples and oranges. That is, he compares premium costs in the pre-ACA market, where protections like those listed above didn’t exist and where less healthy people were excluded from the market entirely, to premiums in the post-ACA market (the study he cites uses the same wrong methodology). Analysis that tries to account for such differences finds that premium costs for the comparable benchmark ACA plan are growing more slowly post-ACA.

Moreover, as I stressed here, premium subsidies, which reach 85 percent of those in the exchanges, help to offset rising premium costs, though Roy and I might agree that they should be more generous.

To the extent the AHCA is “fixable,” and I don’t believe it is, it would mean keeping, if not raising, the subsidies in the ACA, as Roy acknowledges. But it would also mean getting rid of the very coverage changes hard right conservatives insisted upon, rejecting the proposal to convert the Medicaid program to a per capita cap or block grant, and restoring the ACA’s Medicaid expansion. In other words, fixing the AHCA means losing the “H” and actually building on the ACA’s progress towards affordable, quality coverage for more Americans.