What CBO Analysis of Partial ACA Repeal Really Means

Policy makers should strive to get a clearer handle on important factors such as the Medicaid expansion population, the demonstrated efficacy of the coverage mandate, any recent corrections of prior CBO projection errors, and the critical distinctions between partial ACA repeal, full ACA repeal, and any subsequent replacement for the law.

The just-released Congressional Budget Office (CBO) analysis of partial Obamacare repeal will undoubtedly precipitate some sensational headlines, broadcasting dire effects on health insurance coverage and premiums if lawmakers proceed with plans to repeal the law.  For those interested in the policy issues in play behind the hype, the following may be of use.

Key Fact #1:  CBO Analyzed Partial ACA Repeal, Not Total Repeal, Nor Repeal and Replace.  CBO’s title judiciously refers to “portions of the Affordable Care Act (ACA).”  This is not an evaluation of ACA repeal.  Rather, it’s an evaluation of repeal of certain select parts of the ACA – specifically, the parts moving through the budget reconciliation process.

Per Congress’s rules, only legislation with certain budgetary effects can be moved via budget reconciliation.  This requirement exists to ensure that lawmakers don’t simply label any and all legislation as being budget-related, exploiting that designation to move it via privileged processes.  The ACA provisions moving through the budget process reflect a particular view of which provisions would pass muster as germane to budget reconciliation.  CBO has concluded that repealing only those specific parts of the ACA will have certain problematic effects.

One can quibble with the numbers, but qualitatively it should be uncontroversial that partial ACA repeal is problematic in the ways CBO has identified.  Consider CBO’s description of the various insurance regulations that would remain on the books if ACA repeal were limited to those provisions moving through reconciliation:  

“Importantly, H.R. 3762 would leave in place a number of market reforms—rules established by the ACA that govern certain health insurance markets. Insurers who sell plans either through the marketplaces or directly to consumers are required to:

  • Provide specific benefits and amounts of coverage;
  • Not deny coverage or vary premiums because of an enrollee’s health status or limit coverage because of preexisting medical conditions; and
  • Vary premiums only on the basis of age, tobacco use, and geographic location.”

One needn’t malign the value judgments underlying the ACA to recognize that these provisions drive up health insurance costs for most consumers.  Because the ACA prohibits insurers from charging different premium amounts based on the insured’s health status, most participants must face higher premiums to offset the cost of covering those with expensive health conditions.  Other specific ACA requirements also increase many participants’ costs, such as its limitations on premium variance by age: under the ACA, young adults’ premiums must be kept at least one-third as high as those charged to older Americans, even though seniors’ health costs are several times higher.  

In sum, CBO has found that if a law such as the ACA increases health insurance costs for most people, then unless federal law requires people to carry it many healthy people will drop coverage, reducing total coverage numbers and leaving premium costs higher for the remaining insured (sicker) population.  This dynamic would further destabilize insurance plan markets, causing more insurers to drop out, reducing competition and driving consumer costs even higher.  

Notably, CBO found that if the ACA’s insurance regulations were also repealed, the insured population would remain 6-9 million larger than under partial repeal.   Average premiums would also likely be substantially lower under full repeal.  All of this would have substantial effects on the federal budget and could well strengthen a tactical argument that repealing these regulatory components of the ACA should also be considered germane for budget reconciliation purposes.

Key Fact #2:  CBO Finds Most People Will Consider ACA Insurance Unattractive Unless They’re Penalized for Not Carrying It.  CBO projects that if the ACA is partially repealed, insurance coverage nationwide will decline sharply.  CBO bases this projection almost entirely on repeal of the ACA’s coverage mandate penalties, allowing individuals to drop coverage and precipitating an exodus of insurers from the market:

“Most of those reductions in coverage would stem from repealing the penalties associated with the individual mandate. However, CBO and JCT also expect that insurers in some areas would leave the nongroup market in the first new plan year following enactment. They would be leaving in anticipation of further reductions in enrollment and higher average health care costs among enrollees who remained after the subsidies for insurance purchased through the marketplaces were eliminated. As a consequence, roughly 10 percent of the population would be living in an area that had no insurer participating in the nongroup market.”

It’s remarkable that CBO bases its latest projection largely on the power of the individual mandate penalties to incent health insurance coverage.  This is effectively a finding that most people would not find it in their interest to carry ACA insurance unless compelled to do so.  

CBO is undoubtedly correct that many individuals will, absent a penalty, conclude that ACA insurance is not worth the cost.  Indeed, many have declined coverage despite the penalty, defying previous projections and leading to substantial insurer losses and instability in the ACA’s exchanges, long before the recent election placed “repeal and replace” on the legislative agenda.  My Mercatus colleague Brian Blase has found that “CBO’s model has consistently and significantly overestimated the effect of the individual mandate in inducing people to enroll in the exchanges.”  It’s important to know whether these sources of previous projection error have been addressed, because otherwise they will carry over into estimates of the effects of repeal.

Key Fact #3: Repeal Wouldn’t Return Medicaid Enrollment to Pre-ACA Levels.  CBO projects that partial repeal would result in 19 million fewer Medicaid enrollees by 2026 as a result of repealing the ACA’s Medicaid expansion.  It is worth knowing more about this projection, specifically how much of recent Medicaid enrollment increases consist of people made newly eligible by the ACA, vs those who were already eligible pre-ACA.

The ACA significantly expanded Medicaid eligibility and thereby brought many new participants onto the rolls. Simplifying broadly, the ACA (after a subsequent Supreme Court decision) gave states the option, with the federal government picking up nearly all of the tab, of expanding coverage to those with incomes up to 138 percent of the poverty line.  This group is needier than the general population but less needy than the historical Medicaid enrollment population, which consisted of people of generally lower incomes and heightened medical needs along with pregnant women and children.  The ACA also established procedures for enrolling those already eligible for Medicaid who had not yet done so.

It appears that a significant chunk of the post-ACA Medicaid coverage expansion has consisted of so-called “woodwork” enrollees – people who were already eligible prior to the ACA, who signed up under its outreach processes, and would keep their coverage even after repeal.  Brian Blase estimated on the basis of work by Jonathan Gruber that “about 40% of new Medicaid enrollees added by 2014 were made eligible by the ACA and about 60% were woodwork enrollees.”  If the ACA is repealed, these people would remain covered. It’s important to know how CBO’s estimates of woodwork enrollment compare with the Blase/Gruber estimates.

Another key point: although ACA substantially increased Medicaid eligibility and federal funding, it did not appreciably change the supply of health care services available through Medicaid.  Accordingly, the primary effect of the ACA’s Medicaid coverage expansion was to require the most sympathetic and vulnerable Medicaid population (lowest-income enrollees, pregnant women, children, etc.) to face more competition for health services from a marginally less vulnerable population (childless adults of somewhat higher income).  While repeal would likely adversely affect the expansion population, there would be offsetting health access gains for the most vulnerable individuals.  Again, it would be good to know more from CBO about how access to care has been redistributed between these populations due to the ACA’s enactment.

There is much more to know about the effects of ACA repeal—and certainly of the intended repeal and replacement—than has been discussed here. But for starters, policy makers should strive to get a clearer handle on important factors such as the Medicaid expansion population, the demonstrated efficacy of the coverage mandate, any recent corrections of prior CBO projection errors, and the critical distinctions between partial ACA repeal, full ACA repeal, and any subsequent replacement for the law.