Health Care Reform Is a “Three-Legged Stool”

The Costs of Partially Repealing the Affordable Care Act

Jonathan Gruber examines the costs of partially repealing the Affordable Care Act in certain states.

Missouri voters for and against Proposition C—a ballot measure in Missouri to block the federal health insurance law—debate during a rally in Kansas City, Missouri. (AP/Charlie Riedel)
Missouri voters for and against Proposition C—a ballot measure in Missouri to block the federal health insurance law—debate during a rally in Kansas City, Missouri. (AP/Charlie Riedel)

Download this memo (pdf)

The recent ballot measure in Missouri, along with litigation in the federal courts, challenges the Affordable Care Act’s requirement that all individuals hold health insurance. Simultaneously, some members of Congress advocate repealing this requirement and other elements of the Affordable Care Act, claiming that some parts of the transformative legislation will work even if other parts are removed. This paper shows that these claims are false by analyzing what would happen to the Affordable Care Act’s coverage and affordability effects if some parts of the legislation were repealed. It focuses on these effects in 2019—the end of the budget window that the Congressional Budget Office uses.

The paper’s analysis shows that:

  • Repeal of the requirement to buy insurance would mean more people would wait until they get sick to buy insurance in the new nongroup exchanges, which would increase the average premium by 27 percent in 2019.
  • Retaining the law’s insurance reforms, but repealing the subsidies as well as the requirement to purchase insurance, would further discourage people from buying insurance when they’re healthy. Premiums in 2019 would cost twice as much as projected under the law as a result.
  • Retaining the law but repealing the mandate would newly cover fewer than 7 million people in 2019 rather than the 32 million projected to be newly covered by the law. Federal spending, however, would decline by only about a quarter under this scenario since the sickest and most costly uninsured are the ones most likely to gain coverage.
  • Retaining only the insurance reforms in the law—repealing both the mandate and the subsidies—would not increase the number of people with insurance, leaving 55 million people uninsured in 2019.

The “three-legged stool” of health reform

The Affordable Care Act represents the most significant transformation of our health insurance market in more than 40 years. One of the law’s key goals is to fix the broken small group and nongroup insurance markets—where small businesses and people not covered through their jobs get their health insurance. Insurance prices are very high and variable in these markets today, and sick individuals who most need coverage are not able to get it.

At the health law’s core is a “three-legged stool” approach to reforming these markets: new rules that prevent insurers from denying coverage or raising premiums based on preexisting conditions, requirements that everyone buy insurance, and subsidies to make that insurance affordable. But some confusion exists about how the stool’s three parts fit together—confusion that’s compounded by claims that some parts will work without others and by efforts to repeal key elements of the new law.

The truth is that all three legs of the stool are necessary to assure affordable coverage. The first “leg” is regulations that require insurance companies to offer insurance to any applicant with premiums based on age (and tobacco use) and not on underlying health status. Insurance companies are also prohibited from excluding coverage due to preexisting illnesses.

This is a highly popular reform, but it doesn’t work in a vacuum. If insurance companies must charge the same price to people whether they’re sick or healthy many healthy people will view this as a “bad deal” and not buy insurance. This results in higher prices that chase even more people out of the market. The result is a “death spiral” that leads only the sick to purchase insurance at very high prices. Several states tried such community rating reforms—offering health insurance policies within a given territory at the same price to all persons without medical underwriting—in their nongroup markets over the past two decades, and sharp rises in insurance prices ensued along with rapidly shrinking market size.

This fact motivated Massachusetts in 2006 to add a second “leg” to the stool: a requirement that all residents purchase insurance. In this way the state could ensure a broad distribution of health risks in the market and fair “communityrated” pricing to all.

The problem with this solution in a vacuum, however, is that many families cannot afford health insurance at those community-rated prices. Massachusetts therefore added a third “leg” in the form of subsidies that make health insurance affordable for those below three times the poverty line (as well as some targeted exemptions from the mandate for those who were above the subsidized level but could not afford coverage). This reform has shown very encouraging results, with the number of uninsured in the state falling by 60 percent and nongroup premiums falling by 40 percent.

The Affordable Care Act is similarly designed as a three-legged stool. A recent ballot measure in Missouri and litigation in federal courts would repeal the law’s coverage requirement and leave other elements unchanged. At the same time, legislation has been introduced in Congress to repeal some parts of the health law while keeping others—most notably the insurance market reforms. Critics who propose to “repeal and replace” the Affordable Care Act don’t seem to understand that all three legs of the stool are critical for reform. Pulling out any of the legs while leaving one or two intact will critically undercut gains from reform.

The following table illustrates this fact by estimating the impacts of removing various aspects of the law in 2019, the last year of the projected budget window. The Gruber Microsimulation Model, or GMSIM, was used to develop the estimates. It models the reform’s effects in the same manner as the Congressional Budget Office, and therefore reproduces fairly closely the CBO estimates of the law as passed. It’s used here to compare three scenarios:

  • The law as passed
  • The law as passed minus the individual mandate requirement to purchase insurance
  • The law as passed minus the mandate, tax subsidies for individuals, and the Medicaid expansions—and retaining the small business tax credit, the insurance market reforms, and insurance exchanges

Table on the cost of partial repeal

The table shows the effect of these three scenarios on 1) singles and family premiums in the new exchange for a plan of a fixed generosity (an “actuarial value” of 0.7, corresponding to the “silver” level in the new exchanges); 2) changes in the number of uninsured covered from what was projected prior to enactment; and 3) federal spending. The partial repeal’s impact can be seen by comparing the second column (no mandate) and the third column (no mandate and no subsidies) to the first.

Why repealing certain portions of the law won’t work

Both the mandate and subsidies are crucial to keeping exchange premiums low: The simple logic imbedded in the law is that it is potentially destructive to reform insurance markets without mandating purchase because only the sick buy insurance and prices remain high. We have seen examples of this in states such as New York and Massachusetts (before its most recent reform), which both imposed modified community rating without a mandate and saw prices skyrocket in their nongroup markets. When Massachusetts implemented its comprehensive reform in 2006 it saw a striking decline in nongroup premiums of 40 percent.

Comparing premiums for the silver plan in the exchange under the law (column 1) with premiums for the same plan under the repeal scenarios reveals the mandate’s importance for nongroup premiums in the exchange. For singles, removing the mandate (as shown in column 2) raises premiums by 27 percent—in other words, individuals purchasing insurance in the exchange would pay 27 percent more for their coverage without a mandate. Insurance reforms without a mandate and without subsidies (column 3) would have an even more dramatic impact and would double the single premium in the exchange to almost $16,000 per year. (The impact on family premiums is more modest, as the selection effects are much stronger for young healthy singles).

The individual mandate is critical for increasing insurance coverage: Removing the individual mandate cuts the reduction in uninsured by more than three-quarters. Rather than covering almost 60 percent of the 55 million uninsured in 2019, the bill without the mandate would cover only about 12 percent of the uninsured. If the subsidies are removed—as in the last column—the coverage effects fall further so that there is essentially no increase in insurance coverage from simply setting up the exchange with small business credits and insurance market reforms.

Repealing the mandate greatly erodes coverage by employers: The Affordable Care Act leads to a modest erosion of employer coverage of 4.1 million persons, or about 2.5 percent of projected coverage. But repealing the mandate would reduce employer coverage by 13.5 million persons, or over 8 percent of baseline projections. This is because repealing the mandate would eliminate the enrollment that will come from people meeting the requirement to purchase insurance from employers offering insurance to employees who need to meet that requirement.

The mandate means much more “bang for the buck”: While removing the mandate cuts the legislation’s coverage gains by more than 75 percent, it only reduces the spending under the legislation by less than one quarter. This is because without the mandate the uninsured gaining coverage are the sickest ones taking advantage of the market reforms and subsidies, while the healthy uninsured remain out of the system. Repealing the mandate further increases federal spending by creating a large movement out of employer coverage and into public insurance and the subsidized exchange.

The mandate and the subsidies are critical to building an insurance market that includes the healthy and the sick: The exchange insures far more people under the Affordable Care Act than under either of the other scenarios—26.8 million people compared to 17.2 million without the mandate and 11.4 million people without the mandate and the subsidies. The reason is that the larger exchange under the law includes healthy and sick people. Partial repeal—new rules for insurance but no mandates and no subsidies—means people are far more likely to participate only when they need health care, producing the substantial increase in average premiums, and, ultimately, the “death spiral” in which only the sick purchase insurance at very high prices.


Removing the Affordable Care Act’s mandate would eviscerate the law’s coverage gains and greatly raise premiums. And going further by only keeping the market reforms and the small business tax credit would virtually wipe out those coverage gains and cause an enormous premium spike. Without all three legs, the stool— and effective health reform—will not stand.

Download this memo (pdf)

Jonathan Gruber is a professor of economics at the Massachusetts Institute of Technology.

For more information, see:

The positions of American Progress, and our policy experts, are independent, and the findings and conclusions presented are those of American Progress alone. A full list of supporters is available here. American Progress would like to acknowledge the many generous supporters who make our work possible.