However one rates the past performance of the U.S. health care system, by 2009 it was clearly in trouble, on two fronts.

First, a growing number of working-age Americans were uninsured. Coverage of children improved between 1999 and 2009, largely thanks to an expansion of a government program associated with Medicaid, the Children’s Health Insurance Program, and seniors were covered by Medicare. But as Figure 18-7 shows, the percentage of working age Americans (considered to be those between the ages of 18 and 64) without insurance was on a clear upward trend even before the severe economic downturn of 2007–2009.

Lying behind the growing number of uninsured, in turn, were sharply rising premiums for health insurance, reflecting rapid growth in overall health care costs. Figure 18-8 shows overall U.S. spending on health care as a percentage of GDP, a measure of the nation’s total income, since the 1960s. As you can see, health spending has tripled as a share of income since 1965; this increase in spending explains why health insurance has become more expensive. Similar trends can be observed in other countries.

Why was health spending rising? The consensus of health experts is that it’s a result of medical progress. As medical science progresses, conditions that could not be treated in the past become treatable—but often only at great expense. Both private insurers and government programs feel compelled to cover the new procedures—but this means higher costs, which either have to be passed on in the form of higher insurance premiums or require larger commitments of taxpayer funds.

The combination of a rising number of uninsured and rising costs led to many calls for health care reform in the United States. And so in 2010 Congress passed the Affordable Care Act (ACA) which took full effect in 2014. It was the largest expansion of the American welfare state since the creation of Medicare and Medicaid in 1965. It had two major objectives: covering the uninsured and cost control. Let’s look at each in turn.

Covering the Uninsured On the coverage side, the ACA closely follows a model that has been successfully used in Massachusetts since 2006 (introduced under Republican governor at the time, Mitt Romney). To understand the logic of both the Massachusetts plan and the ACA, consider the problem facing one major category of uninsured Americans: the many people who seek coverage in the individual insurance market but are turned down because they have preexisting medical conditions, which insurance companies fear could lead to large future expenses. (Insurance companies were known to deny coverage for even minor ailments, like allergies or a rash you had in college.) How could insurance be made available to such people?

One answer would be regulations requiring that insurance companies offer the same policies to everyone, regardless of medical history—a rule known as “community rating.” In fact, a number of states have such a rule. But community rating tends to lead to an adverse selection death spiral: Healthy individuals don’t buy insurance until or unless they get sick, and because only people with health problems are getting coverage, insurance becomes very expensive.

To make community rating work, it’s necessary to supplement it with other policies. Both the Massachusetts reform and the ACA add two key features. First is the requirement that everyone purchase health insurance—known as the individual mandate. This prevents an adverse selection death spiral. Second, government subsidies make the required insurance affordable for lower- and lower-middle income families.

It’s important to realize that this system is like a three-legged stool: all three components must be present in order for it to work. Take away community rating, and those with preexisting conditions won’t get coverage. Take away the individual mandate, and community rating will produce an adverse selection death spiral. And you can’t require that people buy insurance without providing subsidies to those with lower incomes.

Will this arrangement, once fully implemented, succeed in covering more or less everyone? The Massachusetts precedent is encouraging on that front: by 2012, more than 96% of the state’s residents had health insurance and virtually all children were covered. Since the ACA is very similar in structure, it ought to produce similar results.

Cost Control But will the ACA control costs? In itself, the expansion of coverage will raise health care spending, although not by as much as you might think. The uninsured are by and large relatively young, and the young have relatively low health care costs. (The elderly are already covered by Medicare.) The question is whether the reform can succeed in “bending the curve”—reducing the rate of growth of health costs over time.

The ACA’s promise to control costs starts from the premise that the U.S. medical system, as currently constituted, has skewed incentives that waste resources. Because most care is paid for by insurance, neither doctors nor patients have an incentive to worry about costs. In fact, because health care providers are generally paid for each procedure they perform, there’s a financial incentive to provide additional care—do more tests and, in some cases, perform more operations—even when there are little or no medical benefits.

The bill attempts to correct these skewed incentives in a variety of ways, from stricter oversight of reimbursements, to linking payments to a procedure’s medical value, to paying health care providers for improved health outcomes rather than the number of procedures, and by limiting the tax deductibility of employment-based plans. Even supporters of the reform admit that nobody knows how well any one of these measures will work, but they point out that ACA incorporates virtually every idea for cost control that has been proposed by health care economists and that some of these ideas are likely to be highly successful.

The ACA’s launch As you might gather, the ACA sets up an insurance system considerably more complex than other government programs like Medicare and Medicaid, which simply provide insurance directly. The ACA sets up special online marketplaces, the “exchanges,” in which insurance companies offer policies from which individuals must choose. To figure out how much these policies really cost, it’s necessary to calculate the subsidy you receive, which depends on your income as well as the premium. Making all of that work is a difficult programming problem. And sure enough, when healthcare.gov, the federally run online exchange, started up in October 2013, the system crashed repeatedly. For the next two months few people managed to sign up.

This was, however, a problem with the software, not the fundamental structure of the law, and by the beginning of 2014 it was largely solved. There was a huge surge of people signing up as the deadline for 2014 coverage approached, and by the time that deadline arrived more than 8 million people had signed up via the exchanges, while millions more had either purchased directly from insurers or been covered by an expansion of Medicaid.

How many of those signing up were newly insured? At first there were warnings that many of the policies being bought through the exchanges might simply be replacing existing coverage. By the summer of 2014, however, multiple independent surveys showed a sharp drop in the percentage of respondents without insurance. At this point it seems likely that the ACA will, indeed, manage to cover many though not all uninsured Americans. Whether it will also succeed in reducing the rate of cost growth remains to be seen, although by 2014 there were early signs that the rate of growth of health care costs was indeed falling.