Health-Care Reform

Defining Success or Failure for Obamacare

Judging the Affordable Care Act requires agreeing on what it's supposed to do.

Eight years later, what does the evidence show?

Photographer: Win McNamee/Getty Images

Is Obamacare imploding?

I ask this question because so many people have tweeted this Vox column at me, headlined “Obamacare Keeps Refusing to Implode.” Having predicted Obamacare’s implosion so many times, I'm apparently supposed to be eating humble pie.

I never, in fact, predicted that Obamacare would inevitably implode. I have suggested that the Obamacare exchanges might go into a death spiral, and then again, might not -- a statement I’d still stand by. The subsidies might be enough to prevent a death spiral; they might simply slow it down. Some markets might death spiral while others stabilize.

It is fair to say that I’m more bearish on Obamacare than people to the left of me. But you can be bearish on Obamacare without thinking that it is going to melt down in a cataclysm so utter that the living will envy the dead (at least until a premium-induced heart attack causes them to join that happy group). You can be bullish on Obamacare without thinking that this program has achieved the platonic ideal of health insurance, leaving our citizenry with little to do except stroll around this new heaven, singing songs of praise for the almighty Cost-Sharing Reduction. Unfortunately, both sides of the debate are too fond of attributing these ultra-strong opinions to their opponents, and then triumphantly proving them untrue.

What might be more helpful is to ask, “Is Obamacare working as intended?” As it happens, the day after Obamacare passed, I listed the benefits that supporters had been predicting would follow from the law: fewer uninsured people, of course, but also lower premiums and slower health-care cost growth; declining mortality, particularly infant mortality, and a dramatic decrease in personal bankruptcies. In my book, meeting those promises would be a good definition of “success” for the program.

So how are we doing so far? In some cases, it’s simply too soon to tell -- infant mortality data, for example, are published slowly, so we don’t have a good picture of what’s happened post-Obamacare. Overall health-care cost growth didn’t noticeably slow, but again, that data is slow to come in, and the numbers tend to jump around in any given year.

In other cases, one side or the other can claim victory, but only by squinting hard at statistics that have been confounded by outside factors. For example, mortality rose after Obamacare took effect, but that probably has more to do with opioid abuse than Obamacare. (Medicaid, by financing opiate access for vulnerable populations, is possibly making this problem worse, but so far, there hasn’t been a good strong study to show the connection.) On the other hand, bankruptcies did fall quite a bit from 2013 -- but given that we were recovering from the worst financial crisis in decades, that’s not really surprising. And it’s hard to see any discernable post-Obamacare downward inflection in the data.

One thing we can say is that the number of uninsured people did fall, from around 15 percent in 2008, to around 11 percent today 1 . That’s probably not as big a decline as most supporters were expecting, but it’s not nothing, either. The open question is, of course, how sustainable that improvement is, and whether it was worth the cost.

Some Republicans seem to expect the imminent heat death of Obamacare, perhaps next month. This is wrong. Most of the coverage expansion came from Medicaid, and Medicaid is operating about as it always does: expensive, maybe, but here to stay unless Republicans muster more votes than I think they can for radically altering the program. The exchanges are certainly covering fewer people and costing more than supporters expected, but outside of some rural counties, if they die, it will be over a period of years, not months.

On the other hand, the law’s supporters have been engaging in some energetic goalpost-relocation as premiums have risen and enrollment leveled off. So the fact that Ohio found insurers willing to sell in most of their counties is presented as good news, because hey, that’s only one county in Ohio where no one can buy insurance! People talk excitedly about how this or that state government has found a way to reduce premiums -- a way that turns out to consist of pouring extra subsidies into the system. And huge rate increases are waved away because at least they’re not as huge as last year’s. A fictional sales patter I once saw conjured: "Sure, 37 mph isn’t very fast for a sports car, but you have to compare that to hopping!”

Large swathes of the country have only one insurer, and a few may have none. (Those places tend to have little in the way of population, to be sure, but who in 2010 would have said that “success” for Obamacare included completely destroying any rural insurance markets?) In many places, recent premium increases have rivaled or surpassed the kinds of hikes that were once presented as the reason we needed to pass Obamacare. And enrollment has stopped growing well short of market saturation, which means that yes, the individual markets are still vulnerable to a death spiral, particularly in the off-exchange, unsubsidized segments. Does that mean that it will literally be impossible to buy insurance anywhere in the United States? Probably not. But just how many places have to face serious problems before we say Obamacare isn’t working?

That’s a serious question, mind you. Liberals are always going to look at Obamacare and see a lovelier visage than conservatives. But we should develop some sort of common definition for “success” or “failure” beyond “Has it completely gone away yet?” Until we do, both sides are going to pick out the features they want to see, and be confused when the other side of the aisle seems to be looking at a completely different picture.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

  1. The Obamacare bulls see a larger decline in this data, from 18 percent in 2013. Why am I not using this number? Two reasons: First of all, because that 18 percent represented a sharp spike right before Obamacare took effect, which seems likely to be either a result of Obamacare (people dropping insurance in anticipation of the new coverage), or more probably, a statistical outlier. And second of all, because of the abovementioned recession. The number of uninsured rises after recessions, and then gradually falls, and that would have happened regardless of Obamacare’s passage. So I’ve chosen the number that is closest to the start of the Great Recession. But to forestall quibbles, I have rounded generously to generate a slightly bigger improvement. Without rounding, the decline from 2008 is not 4 percentage points, but 3.3.

To contact the author of this story:
Megan McArdle at

To contact the editor responsible for this story:
James Gibney at

Before it's here, it's on the Bloomberg Terminal.