The Medical Bankruptcy Myth

The debate about American healthcare is being influenced by recent controversial research claiming to show that nearly two-thirds of personal bankruptcies in the United States resulted from uninsured medical expenses or loss of income due to illness. An earlier 2005 edition of this research claimed that just over half of personal bankruptcies were due to these 'medical causes.' The authors of these studies, David Himmelstein, Deborah Thorne, Elizabeth Warren, and Steffie Woolhandler, argue that the problem of 'medical bankruptcies' would be solved by the adoption of a government-run health insurance system like Canada's.

The research has been politically persuasive. President Obama himself cited the dubious link between medical expenses and personal bankruptcy as part of his rationale for a massive increase of government involvement in healthcare. "The cost of healthcare now causes a bankruptcy in America every 30 seconds," he declared in March. "By the end of the year, it could cause 1.5 million Americans to lose their homes."

A July 28 hearing of the House Judiciary Committee titled, Is Our Healthcare System Bankrupting Americans? prominently featured the medical bankruptcy study. More recently, in a USA Today column, Speaker of the House Nancy Pelosi and House Majority Leader Steny Hoyer cited medical bankruptcy to justify their healthcare overhaul efforts.

Yet the medical bankruptcy study has been soundly refuted by several researchers. This includes critiques published by David Dranove and Michael Millenson in Health Affairs and a working paper by the American Enterprise Institute's Aparna Mathur. The idea that large numbers of Americans are declaring bankruptcy due to medical expenses is a myth.

Dranove and Millenson critically analyzed the data from the 2005 edition of the medical bankruptcy study. They found that medical spending was a contributing factor in only 17 percent of U.S. bankruptcies. They also reviewed other research, including studies by the Department of Justice, finding that medical debts accounted for only 12 percent to 13 percent of the total debts among American bankruptcy filers who cited medical debt as one of their reasons for bankruptcy.

As for the notion that greater government involvement in health insurance will reduce bankruptcy, it is helpful to compare personal bankruptcy rates in the United States and Canada. Unlike the United States, Canada has a universal, government-run health insurance system. Following the logic of Himmelstein and colleagues, we should therefore expect to observe a lower rate of personal bankruptcy in Canada compared to the United States.

Yet the evidence shows that in the only comparable years, personal bankruptcy rates were actually higher in Canada. Personal bankruptcy filings as a percentage of the population were 0.20 percent in the United States during 2006 and 0.27 percent in 2007. In Canada, the numbers are 0.30 percent in both 2006 and 2007. The data are from government sources and defined in similar ways for both countries and cover the time period after the legal reforms to U.S. bankruptcy laws in 2005 and before the onset of the 2008 economic recession.

This is important, because the 2005 reforms produced U.S. legal standards for bankruptcy filing that are now very similar to Canada's. Before 2005 it was much easier to file for bankruptcy in the United States, making cross-border comparisons prior to the legal changes meaningless. Further, in 2008 the United States was harmed by massive systemic home mortgage defaults that did not occur in Canada because of differences in mortgage lending practices. U.S. mortgage defaults would have been correlated with increased bankruptcy rates. Therefore, Canada-U.S. comparisons in 2008 are not valid because the data is skewed by other policy differences unrelated to health insurance.

Aside from universal single-payer health insurance, there are few other significant health, social, or legal policy differences between the two countries that could be causally linked to bankruptcy rates. Both countries have employment insurance programs that provide income support in the event of job loss. In fact, unemployment occurs with roughly similar frequency among Canadians and Americans. National unemployment rates in 2007 were 5.3 percent in Canada versus 4.6 percent in the United States.

Drug insurance is also structured almost identically, so exposure to drug costs is similar in both countries. While the entire Canadian population is universally eligible for publicly funded insurance for hospital and physician services, only about one-third of the Canadian population is publicly insured for prescription drugs. In Canada, as in the United States, low-income people, disabled populations, and seniors are eligible for separate publicly funded drug programs, while most employed people obtain drug insurance as a benefit of employment, and the rest of the population pays cash.

Access to medical care for people who experience long-term unemployment, disability from illness, and chronic low-income status is also practically the same in both countries, being facilitated by non-profit, publicly funded community health centers and public programs such as Medicaid in the United States and government-run systems in Canada.

The truth is that the majority of debt among bankrupt consumers in both Canada and the United States is comprised of non-medical expenditures and therefore has little to do with health insurance coverage.

On the rare occasion that medical debts do partially contribute to bankruptcy, they likely accumulate from patients' demands for the kinds of expensive, cutting-edge or end-of-life treatments that would never be covered by government insurance anyway. It is a fact that many of these same types of expensive treatments are increasingly not insured by government healthcare in Canada.

Indeed, if we define medical bankruptcies the way Himmelstein and colleagues did for their study in the United States, we find such bankruptcies also occur in Canada. Survey research commissioned by the Canadian government found that despite having a government-run health system, medical reasons (including uninsured expenses), were cited as the primary cause of bankruptcy by approximately 15 percent of bankrupt Canadian seniors (55 years of age and older).

There is no objective evidence to indicate that a government-run health care system in the United States will reduce personal bankruptcies. The U.S.-Canada comparative analysis strongly suggests that bankruptcy statistics are being exaggerated and distorted for political reasons.