The Washington PostDemocracy Dies in Darkness

The Affordable Care Act is covering people, holding down costs and not killing jobs

(Joe Raedle/Getty Images)

I’m on a panel this morning talking about the impact of the Affordable Care Act on the broader economy.

My rap starts from the perspective that the ACA, while not perfect, is working remarkably well. The ACA debate, on the other hand, has become inseparable from intense partisan politics and this has led to incessant hand-waving and smoke-blowing such that it’s impossible to get the straight dope on its impact.

Supporters aren’t blameless, as we should have expected (and warned people) that a complex policy dealing with a sector that’s 17 percent of GDP is going to get some things wrong out of the gate. To this day, the ACA needs recalibration (e.g., “thin” individual markets with too few providers).

But the idea that it should be repealed flies in the face of these facts:

Fact 1: The ACA has had the truly remarkable — almost unprecedented — impact of lowering the share of Americans without health coverage. The first figure shows that there are two times over the past 60 years or so that the uninsured rate fell sharply: the implementation of Medicaid/Medicare and that of the ACA. Any student of public policy knows that this is much harder than it looks. It’s all too rare for the enactment of a major public policy to quickly have its intended effect the way you see here.

Fact 2: The efficiencies in health-care delivery that the ACA has helped to usher along are one reason both spending and prices in the sector are growing more slowly. Also remarkable is the fact that current projections for future health spending are below those from January 2010, before the ACA was legislated, meaning these earlier forecasts did not include either the Medicaid expansion or the premium subsidies in the exchanges. As budget expert Richard Kogan recently noted, “As a result, overall health spending is not only projected to be lower but also to cover 20 million more people.”

Fact 3: There’s no evidence for the claim that the ACA is a “job killer.” We are in the midst of the longest period of continuous job growth on record and the job market is closing in on full employment enough that the Federal Reserve is talking about raising interest rates to prevent overheating (based on the absence of inflationary pressures, I wouldn’t go there, but that’s a different discussion). But is there any evidence that’s a bit more ACA-specific?

The first two figures below are scatterplots where each dot is a state. As noted, ACA’s Medicaid expansion and premium subsidies have boosted insurance coverage, so one way to test the job-killer hypothesis is to plot changes in employment growth by state against changes in insurance coverage (this figure shows 2013-2015 changes, but the results are the same for one-year changes). If the ACA killed jobs, states with the biggest gains in insured rates should have seen slower job growth.

But the correlation goes the other way. The positive slope in the first figure means that states with bigger health-coverage gains were also the states with faster job growth. Of course, this was a period with pretty strong job growth, and the correlation could be driven by private, employer-sponsored coverage. But the second figure, which takes employer coverage out of the equation, shows a flatter but still positive correlation.

To be clear, while there’s evidence that the ACA has led to increased job growth in the health-care industry, as you’d expect, I don’t think those positive slopes mean the ACA is a job creator. There are many moving parts to that machine, and the scatterplots are picking up only correlation, not causation. But I do think those positive slopes are not negative slopes: I’ve seen no evidence-based case that the ACA is dampening job growth.

What about job growth in the Medicaid expansion vs. the non-expansion states? The bar chart below shows that the rate of job growth was essentially the same in both groups (the small differences between the heights of the bars are statistically insignificant).

Okay, but what about the shift to part-time jobs? The ACA requires employers with at least 50 full-time workers to offer health coverage to employees who work at least 30 hours a week or pay a penalty. So the incentive to shift full-time workers to part-time schedules exists. But let’s again go to the evidence.

If this incentive was in play, we’d expect to see an increase in the share of workers who were involuntary part-timers (IPTs), i.e., those who want to work full time but are stuck working part time. In fact, the share of IPTs (as a share of employment) shot up in the recession, well before the ACA was on the scene. But as the job market has come back, it has fallen sharply, as you see in the figure below.

However, it’s a highly cyclical variable: it always goes up in recessions (shaded area in figure) and down in recoveries. The question then becomes, is it falling more slowly in this recovery because of the ACA? To answer that, I built a simple statistical model to predict the ups and downs of the IPTs, and estimated it up to 2010, pre-ACA. I then forecast it forward. If the ACA was slowing the expected decline in IPTs, the forecast line would fall faster than the actual line. The fact that both lines track each other pretty closely suggests that the IPTs are declining as you’d expect, given the improving job market.

Like I said, the ACA isn’t perfect (I noted the individual market problem above, but remember: a) only 6 percent of those with coverage are in the non-group market, and b) a public option is a simple solution to this problem). But it’s a vast improvement over what came before, it’s providing affordable coverage to millions, it’s holding down costs beyond many of our expectations, and it’s not hurting job growth.

Those who want to repeal it need to convincingly explain how their plan achieves these goals, and trust me, they’ve not even begun to do so. That’s because their true motivation is because it’s “Obamacare.” And that, my friends, is no plan at all.