Seven Promises Made for the Affordable Care Act, Tested Against the Record

A promise is a claim about the future, and the future has a habit of filing a reply. When Congress sold the Affordable Care Act to the country, its sponsors made a set of specific promises. Specific promises can be checked. That is the whole point of making them.

So let us check. The method is simple: state each promise as its defenders would state it, look at the record, and give a verdict. I will say where the evidence is mixed and where the other side has a fair point, because an honest audit has to keep both columns. Where a number moves from year to year, I will point you to the official source and not quote a figure I cannot stand behind.

The ground rules

Three definitions keep this clean.

A promise is a statement made by the law's sponsors or its principal advocate in order to win support. Not every feature of the law was a promise. I am looking at the ones that were.

A test is something a reader can verify from public records: statutory text, a court opinion, a budget office estimate, an agency report.

A verdict is my judgment, with the reasons given. I am a partisan, and you should know it. But I would rather be a partisan who gives you the evidence than one who gives you the adjectives.

Seven claims follow.

Promise one: "If you like your plan, you can keep it"

The claim. President Obama repeated, during the 2009 and 2010 debate, that if you liked your health plan you could keep it. The sentence was repeated so often that critics have quoted it back ever since.

The test. The law set minimum standards for plans, and plans that did not meet them could not be sold to new customers. Plans that existed on March 23, 2010 could be "grandfathered," but a grandfathered plan lost that status if the insurer cut benefits or raised costs beyond limits the regulations set. In the autumn of 2013, as the standards took effect, insurers sent cancellation notices to many people in the individual market whose plans no longer qualified. Those notices were the reality that the sentence had denied. In November 2013, the administration allowed insurers to continue some of those plans for a limited time, which is an admission that the promise as stated could not be kept.

The fair point from the other side. The cancelled plans were, in many cases, thin plans with narrow benefits, and the replacement plans covered more. A person who had a bare-bones plan did get a more complete one.

The answer. That may be true, and it is beside the point. The promise was not "you will get a better plan whether you like it or not." The promise was that you could keep the one you liked, and a person who liked a cheaper plan with fewer benefits had exactly the preference the sentence promised to respect.

A plan you were forced to replace with one you did not choose is not a plan you kept. It is a plan you were allowed to remember.

Verdict: broken. The record in 2013 shows it plainly.

Promise two: family premiums would fall by $2,500

The claim. During the 2008 campaign, candidate Obama promised that his health plan would lower the typical family's premium by up to $2,500 a year. The figure became a reference point for the whole debate.

The test. The Department of Health and Human Services and the Centers for Medicare and Medicaid Services publish premium data, and so does the Kaiser Family Foundation in its annual employer survey. Check them. What you will find, as a matter of mechanism, is that the law's new requirements, which included essential benefits, a ban on pricing by health status and limits on pricing by age, raised the price of individual-market coverage for many buyers, while subsidies shielded those who qualified. Employer premiums kept rising: the Kaiser Family Foundation's survey put the average annual family premium for employer coverage at $12,680 in 2008 and $23,968 in 2023. I tell you where to look because the figures vary by year and by state, and I do not want you to rely on my memory when the sources are public.

The fair point. Defenders say premium growth slowed after 2010, and that without the law the increases would have been worse. That is a counterfactual, and no one can observe the world that did not happen. The recession and other forces also affected costs.

The answer. The promise was not that increases would be a little smaller than a model predicted. It was that families would see their premiums fall. A promise to lower a number is tested by whether the number went down.

Verdict: broken. The sticker price did not fall for most families.

Promise three: the law would bend the cost curve

The claim. The law's advocates said it would "bend the cost curve," which means slow the rate at which national health spending grows, through payment reforms, new boards, accountable care organizations and the like.

The test. National health spending growth did slow in the years after 2010. But the slowdown began around the time of the Great Recession, and economists disagree about how much of it was due to the law, how much to the recession and how much to other trends. Spending as a share of the economy did not shrink, and the law's own major cost-control tools were small compared with the new spending it authorized, the subsidies and the Medicaid expansion. One of the cost-control bodies, the Independent Payment Advisory Board, was never staffed and was repealed in 2018.

The fair point. Some of the Medicare payment experiments produced savings in some programs, and some hospital readmission penalties changed behavior. Those are real.

The answer. A few successful pilots are not a bent curve. The promise was a general slowdown of cost growth, and the evidence for the claim is contested at best.

Verdict: unproven. The honest finding is that the law added coverage and added cost, and that no one has shown it changed the underlying growth rate on its own.

Promise four: the law would pay for itself

The claim. The sponsors said the law would not add a dime to the deficit. The Congressional Budget Office, scoring the final law in March 2010, estimated that it would reduce deficits over the following decade.

The test. Read how that score was built. The CBO scores what the statute says, and the statute assumed several things that critics said were fictions. It counted hundreds of billions of dollars in Medicare reductions that Congress had a long record of reversing. It counted premiums collected by a new long-term care program, called the CLASS Act, in the first years, before the benefits would be paid, a trick I return to in promise five. And it left out the cost of the annual "doc fix," the congressional patch that blocks scheduled cuts to physician payments, by handling it in a separate bill. A budget that counts the taxes and cuts, and omits the cost of the promises, will balance.

The fair point. The CBO is a nonpartisan office that did the scoring it was required to do, and its estimate was not corrupt. A law that raises taxes and trims Medicare payments really does bring in money.

The answer. The question is not whether the CBO did arithmetic correctly with the assumptions it was given. The question is whether the law was self-funding in practice. Medicare payment cuts were repeatedly delayed or reversed in later legislation, the CLASS program collapsed, and the subsidies and the Medicaid expansion have created a spending commitment that grows with premiums. The federal debt has passed $40 trillion, as Treasury's figures showed in August 2026, and it did not get there because of one law, but nobody can call this law a contribution to thrift.

Verdict: false as stated. It paid for itself on paper, in the way that a house paid for with borrowed money pays for itself.

Promise five: the CLASS Act would provide long-term care

The claim. The Community Living Assistance Services and Supports Act, the CLASS Act, was a section of the law that created a voluntary, government-run insurance program for long-term care, financed by premiums that workers would pay in.

The test. The law required the program to be actuarially sound, meaning its premiums had to cover its costs without help from the Treasury. In October 2011, the Secretary of Health and Human Services, Kathleen Sebelius, announced that the department could not find a way to implement the program consistent with that requirement. The reason was adverse selection: the people most likely to sign up for a voluntary long-term care plan are the people most likely to need it, and the premiums needed to cover them would price out everyone else. Congress repealed the program in the American Taxpayer Relief Act at the start of 2013.

The fair point. Americans do worry about the cost of long-term care, and the problem the CLASS Act aimed at is real. Its defenders point out that the failure was a failure of one program's design, not of the idea of covering long-term care.

The answer. I agree that the problem is real. But note what the episode proves. The law's score counted the early premiums from the CLASS Act as money available to help balance the books, and the program's own administrators concluded it could not work. Washington's budget office has seen many creative accounting schemes, but a pension that spends the contributions before anyone retires has a long and unhappy history. A program that was designed to be counted before it was designed to be paid is a gimmick, and the collapse of the gimmick is a fair test of the whole claim to fiscal honesty.

Verdict: failed. It was dropped in 2011 and repealed in 2013.

Promise six: private exchanges would deliver the coverage gain

The claim. The law was sold as a way of building a regulated marketplace where Americans would shop for private plans with the help of subsidies. Its architects spoke of choice and competition.

The test. The law did increase the number of Americans with health coverage. Now ask where the increase came from. The Congressional Budget Office and the Census Bureau publish the numbers, and I encourage you to read them. A large share of the gain came through the expansion of Medicaid, and I would have you check the split in the budget office's baseline tables for yourself. Medicaid is a government program that pays doctors less than private plans do and gives its enrollees no choice among competing insurers in the usual sense.

The fair point. Coverage is coverage, and a person with Medicaid has access to care he lacked. A conservative who denies the gain loses the argument.

The answer. I do not deny it, and I do not need to. The claim being tested is not that coverage rose. It is that the competitive marketplace was the engine of the gain, and the record gives the marketplace a smaller role than the sales pitch did. The expansion went mainly to able-bodied adults, which created a financing anomaly, explained in The Medicaid Match That Pays States More for Adults Than for Disabled Kids: Washington pays a larger share of the cost of an expansion adult than of a disabled child. Whether work expectations belong in the program is a separate argument, addressed in Asking Able-Bodied Adults to Work for Medicaid Is Not Cruelty, It Is Dignity.

Verdict: mostly not so. The gains were real, but the market was a junior partner.

Promise seven: the mandate is not a tax

The claim. The law required most Americans to carry insurance or pay a penalty, and its sponsors insisted, including the President in a 2009 television interview, that this penalty was not a tax. The statute itself called it a "penalty."

The test. When the law was challenged, the government's lawyers defended the mandate chiefly under the Commerce Clause and, in the alternative, under Congress's power to tax. In National Federation of Independent Business v. Sebelius in 2012, the Court rejected the Commerce Clause argument and upheld the mandate only as an exercise of the taxing power. Chief Justice Roberts wrote that the penalty could be read as a tax, because it was paid to the Treasury with the tax return and operated like a tax in practice. The case is analyzed in John Roberts Called the Mandate a Tax, and Conservatives Still Argue About Why.

The fair point. Nothing in the Constitution stops Congress from taxing, and a tax is a legitimate tool. And the decision settled the matter as a legal question.

The answer. Granted on both counts. But the legal victory contains the political defeat. The sponsors told the voters one thing and told the Court another. A law that survives only because its central provision is a tax, after its sponsors denied that it was one, did not keep its promise. It rested on a technicality that its own advocates had disowned.

Verdict: broken. The Supreme Court called it a tax.

The scorecard

Put the seven side by side.

I expect a reader who supports the law to say that this scorecard leaves out its genuine accomplishments, and they would be right. The ban on turning people away for preexisting conditions was a promise, and it was kept. Young adults staying on a parent's plan until 26 was a promise, and it was kept. Both are popular for a reason. But notice what that proves. Features that help people are not an argument for a mandate, a trillion dollars of subsidies, or a structure built on a promise that turned out to be false. The question for conservatives is how to keep the protections people value without the machinery they do not. The answer in High-Risk Pools Covered the Sick Before 2014, and They Could Do It Again is one starting point.

The Founding question

Why does it matter whether the promises were kept? Because consent is the heart of the American idea. The Declaration says that governments derive "their just powers from the consent of the governed." Consent obtained by a promise the government knew, or should have known, it could not keep is thinner than it looks. A citizen who supported the bill because he was told he could keep his plan was consenting to something other than what he got.

That is no small thing. A republic depends on the ability of voters to hold officials to what they said. If the pledge and the performance part ways, the voter cannot tell what he voted for, and the next promise sounds hollow. That, more than any single clause, is the damage of the law's sales campaign.

The case in brief

1. The promise to keep the plan you liked was broken by the cancellation notices of 2013.
2. The promise of lower family premiums and a bent cost curve was not delivered, and the claim of self-funding rested on counting premiums and cuts that did not hold.
3. The CLASS Act, counted as money in the bank, was abandoned in 2011 and repealed in 2013.
4. A large share of the coverage gain came through Medicaid, not through the competitive marketplace, and the mandate was upheld only as a tax after its sponsors said it was not.
5. Real protections, such as coverage for preexisting conditions, can be kept without the structure that came with them, and conservatives should say how.

What you can do: pick one promise from this list and look up the official source, the Congressional Budget Office or the Centers for Medicare and Medicaid Services, then bring the page to your next conversation about health policy.