It's all in how you slice the data.

Photographer: John Moore/Getty Images

The Myth of the Medical Bankruptcy

Megan McArdle is a Bloomberg View columnist. She wrote for the Daily Beast, Newsweek, the Atlantic and the Economist and founded the blog Asymmetrical Information. She is the author of "“The Up Side of Down: Why Failing Well Is the Key to Success.”
Read More.
a | A

One of the predictions that was made, with great fanfare, when Obamacare passed, was that our nation’s bankruptcy epidemic would finally come to an end. Last week, veteran liberal commentator Norm Ornstein declared that it had already come to pass.

“Before Obamacare,” he tweeted, “more bankruptcies from health bills than anything else. Now, hardly at all. Do we really want to go back to that?”

Did medical bills single-handedly account for more bankruptcies than anything else? No. This is an exaggerated half-remembering of a series of studies, authored by (among others) Elizabeth Warren, that were themselves exorbitant exaggerations.

I went into detail on the problems with the work seven years ago, but the highlight reel is that these authors have an aggressive tendency to employ any technique that ratchets the count of “medical bankruptcy” upward, while not using similar techniques that would tend to ratchet up other categories and diminish the number of bankruptcies counted as medical, and to present their results in misleading ways -- so as to obscure, for example, the fact that by their own accounting, the number of medical bankruptcies actually fell by hundreds of thousands between 2001 and 2007. Which is why their study tended to be presented in the media as “growing problem” rather than “shrinking threat.”

To be clear: I don’t believe that medical bankruptcies fell by hundreds of thousands between 2001 and 2007. I think this conclusion further suggests just how problematic their methodology is. So does the fact that in 2011, two of Warren’s co-authors issued a new study finding that medical bankruptcy hadn’t fallen in Massachusetts after the passage of Romneycare. These two co-authors are the co-founders of Physicians for a National Health Care Program, an advocacy group that supports single-payer in America, and they have a noticeable tendency to find that --  quelle surprise -- America has enormous problems with medical bills that can be solved only by a single-payer health-care system. They have a lot of latitude to get that answer, because it’s surprisingly hard to know exactly which bankruptcies are medical. Someone who bought three new cars, and also had a hernia, is probably going to blame the hernia.

And we also have to look beyond the bills. The Warren studies tended to get reported, or remembered, as “medical bills cause more than half of all bankruptcies.” That’s not quite what they said. Bad health events do more than land you with big medical bills (which bills can often be settled for pennies on the dollar, because the collectors know they get nothing if you file). Getting really sick also cuts your income as you stop working. If you’ve got debt and no savings, that job loss is going to be catastrophic.

Unfortunately, the incentives of both academic journals and the media mean that dubious research often gets more widely known than more carefully done studies, precisely because the shoddy statistics and wild outliers suggest something new and interesting about the world. If I tell you that more than half of all bankruptcies are caused by medical problems, you will be alarmed and wish to know more. If I show you more carefully done research suggesting that it is a real but comparatively modest problem, you will just be wondering what time "Game of Thrones" is on.

So no, it was never reasonable to think that medical bills, all by themselves, could explain more than a modest fraction of bankruptcies. Nor is it reasonable to think that Obamacare single-handedly reduced those bankruptcies to nothing.

For one thing, the methodology of Warren et al showed that having insurance had little effect on the medical bankruptcies; I think that this is yet another reason to distrust them, but if you’re going to cite their results, you have to accept the inevitable corollary that Obamacare wasn’t going to help much.

For another, we have data on what’s been happening with bankruptcies over the last 10 years. The number of bankruptcies right now is slightly higher than it was in 2007. This is emphatically not the result we would expect if medical bills had been causing a large plurality of bankruptcies and Obamacare reduced those bankruptcies to practically nothing.

I do think medical bills contribute to or cause a significant number of bankruptcies. I also think that Obamacare must have prevented some medical bankruptcies, though I couldn’t say how many. Unfortunately, a lot of unrelated trends have whipsawed bankruptcy statistics around over the last decade. First the Bankruptcy Abuse Prevention and Consumer Protection Act of 2005 tightened up bankruptcy filing rules, so that a huge number of people rushed to file right before the rules changed, causing a spike in 2005 and an offsetting trough in 2006. Then, just as bankruptcy filings should have been reaching whatever their “natural” post-reform level was, we had the financial crisis. Then we had Obamacare.  It’s hard to tease out the effects of one change when so much is happening all at once. It seems likely to me -- indeed, almost certain -- that without Obamacare, the number of bankruptcy filings in 2016 would have been higher than it actually was. I find it difficult to say whether the difference is closer to “major public policy problem” or “a series of unfortunate events.”

But it is possible to say that medical bills did not cause a majority, or even a large plurality, of bankruptcies. And that another problem remains substantial: the number of people who fervently believe something for which there is no good evidence.

  1. This includes business filings, but the difference is trivial, because consumer bankruptcies make up the overwhelming majority of total bankruptcies.

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

To contact the author of this story:
Megan McArdle at

To contact the editor responsible for this story:
Philip Gray at