Is the American Dream being bought on a shaky pyramid of unsustainable credit? Are ordinary middle-class Americans being crushed under the onslaught of rising credit card debt and sky-rocketing medical expenses? The media tales of families who helplessly confront insurmountable debt burdens would lead you to think so. The same gloomy anecdotes are trotted out by Barack Obama, John Edwards, and others who claim that at least half of all personal bankruptcies are caused by medical bills. But more representative data suggest we should calm down and discount proposed solutions to an over-hyped problem that is, in effect, “too bad to be true.”

For example, the Survey of Consumer Finances (SCF) paints a far less dire picture than some politicians do. The study uses random samples of approximately 4,500 households every three years to assess families’ financial situations. In 2004, 46 percent of all families surveyed by the SCF had credit card balances. Similarly, in 1995, 47 percent of all families carried such debt. In other words, indebtedness captured by this measure has remained arguably stagnant for nearly a decade.

More representative data suggest we should calm down and discount proposed solutions to an over-hyped problem that is, in effect, 'too bad to be true.'

Over the same period, the mean dollar balance (in constant 2004 dollars) on credit cards increased by about $1,400, to $5,100. This increase amounted to roughly 11 percent of average household income in 2004 (estimated at $46,000 by the Census Bureau). Additionally, the balance’s rise only represented a marginal increase of 2 percent compared to the 9 percent share of household income of $3,700 in 1995. Furthermore, 50 percent of all families with credit card debt had debt lower than $2,200. According to this study, families actually commit a lesser amount of their assets to their lines of credit than they did ten years ago, thus placing less of their income in jeopardy.

It seems, however, that media accounts of credit card debt contain more melodrama than authenticity. James Scurlock’s recent book and movie “Maxed Out” grabs attention as it discusses college students who committed suicide because of overwhelming credit card bills while warning the public against the dangers of living in perpetual debt. Alarming anecdotes should certainly make us cautious about rising indebtedness in the U.S. and the toll it can take on low-and middle income consumers. However, the SCF numbers suggest that the average household seems to be managing its debt adequately. These figures invalidate claims that Americans recklessly juggle, on average, 13 payment cards as they ring up more and more unaffordable purchases.

Another study released this March by Demos, a New York-based public policy group, asserts that medical debts are increasingly being rung up on plastic, as families unable to meet out-of-pocket medical expenses resort to loading them up their credit cards. If truly widespread, such an accumulation of medical debt certainly could jeopardize access to necessary health services and lead to financial insecurity. More notably, health care system critics like Harvard Medical School professor David Himmelstein have argued that medical debts may be responsible for nearly 50 percent of bankruptcy filings. While these claims have been disputed repeatedly by several researchers, including ourselves, they still attract credulous media coverage and receive reflexive “Amen’s” from politicized critics of private health markets.

According to this study, families actually commit a lesser amount of their assets to their lines of credit than they did ten years ago, thus placing less of their income in jeopardy.

A major shortcoming with the recent wave of consumer debt burden studies is what economists dub the “sample selection issue.” Himmelstein and his coauthors conducted a survey of 1000 bankruptcy filers from public court records for the year 2001. They concluded that more than 50 percent of these claims had been filed for medical reasons. However, by limiting this sample to those who had already filed for bankruptcy, the study overstated the incidence of medical debt. To account for causation, the study sample should have, at the very least, included a “control” group of medical debtors who did not file for bankruptcy. The study also should have allowed for the possibility that other household characteristics, such as work status of the filer, average incomes, and other kinds of debts could have influenced the filing. Compounding this problem was an overly broad definition of “medical filers”, which included people with any sort of addiction or uncontrolled gambling problems. Hence, the study actually failed to show a direct causation between medical debts and bankruptcy filings.

Moreover, results can vary from sample to sample. For example, the Office for United States Trustees (in the Department of Justice) found that medical debt was not a major factor in the majority of bankruptcy cases filed in 2000. Only 5 percent of the cases stated that medical debt contributed to one-half or more of total unsecured debt. In the Panel Study of Income Dynamics (PSID), another large household survey, only 9 percent claimed medical bills as the primary reason for filing. In AEI research by one of us (Mathur) published last July on the topic, a study of a large random sample from the PSID reveals that when other household characteristics are controlled for, a 10 percent increase in household debt with primarily medical expenses may at most increase bankruptcy filings by 27 percent.  

One more example of thinly researched findings on medical debt and bankruptcy is another paper released by Demos that studied medical expenses as a component of credit card debt. “Borrowing to Stay Healthy” offered the somewhat unsurprising finding: households that experienced a major medical expense in the previous three years are more likely to pay for it using credit cards. It then concluded that the “medically indebted” (those who reported having a major medical expense in the previous three years which contributed to their current level of debt) are likely to have higher levels of credit card debt compared to those without such major medical expenses, and that average credit card debt was higher for those without health insurance.

Anecdotal snapshots based on few restricted and ultimately unconvincing samples deplete our understanding of complex issues...In a word, they are bankrupt.

Is this a huge problem, requiring regulation of the credit card industry or universal coverage as the study’s authors recommend? The findings are based on a sample of about 300 people, selected based on having revolving credit card debt for three or more months. They also focused on people within this group who have had any level of medical debt in their credit history and experienced a major medical emergency in the previous three years. This group then narrowed to 60 people. Apart from the non-random and biased nature of the sample, it is a stretch to apply the results to the wider population, even if proper weighting procedures could be adopted.

Consider instead our review of the SCF, which includes medical debts with other debts incurred for goods and services. Medical debts have barely budged from around 6.1 percent of all debt in 1989 to 6.0 percent in 2004. Similarly, in the PSID dataset over the entire period from 1989 to 1996, only about 6 percent of debtors in each year reported having to take a loan to repay medical debts. Medical debts simply are not a ballooning fraction of all debts as is often claimed. Although one-time spikes in medical costs due to accidents or even long-term illnesses are likely to place a heavy strain on household finances, the average debtor today has not seen a huge rise in medical debts, even compared to the late 1980s.

Rising levels of indebtedness remains an issue that deserves serious, but carefully balanced attention. Anecdotal snapshots based on few restricted and ultimately unconvincing samples deplete our understanding of complex issues by overdrawing their thin reserves of evidence. In a word, they are bankrupt.

Aparna Mathur is a research fellow at the American Enterprise Institute and Tom Miller is a resident fellow at the American Enterprise Institute.

Image credit: Photo by Flickr user lemonjenny