Health inequality in America, at least by some counts, seems to be falling. The United States Centers for Disease Control and Prevention’s (CDC) recently released a report that estimates that today about 9.2% of the population is uninsured, a decrease from 11.5% in 2014. But disparities in health care do not depend solely on whether a person has insurance or not. Just as the standards of the school that a student attends matters so, too, does the quality of a hospital that a patient visits.

Across the United States, hospital quality has been found to be an important source of care inequality. But what exactly determines how well a hospital treats its patients? Previous studies have found that an essential factor is a hospital’s financial health. This makes sense since the state of the finances reflects an institution’s ability to provide resources necessary for high-quality care. The factors that, in turn, affect the financial health of a hospital, however, may be plenty and hard to disentangle. New research has tried to shed light on this issue.

In a recent paper published in Behavioral Science & Policy, Matthew Manary, Richard Staelin, William Boulding, and Seth W. Glickman seek to explain key determinants of hospital financial health. The authors gathered data for a total of 265 acute-care hospitals in California from 2005 to 2011. The objective of their study was to look at the relationship between the characteristics of a hospital’s patient population and its quality of care to see if hospital financial health was a vital mediating factor.

Manary and co-authors looked at a host of variables on patient characteristics, such as ethnicity, age, and insurance type. Quality of care was measured through a composite measure that included data on patient experience, mortality, and readmission rates. Financial health was based on three variables: the ability to meet short-term obligations, gross operating margin (the ability to generate profits), and return on assets.

The results were clear. Financial health had a strong and substantial relationship to quality of care. The authors present some evidence that healthier finances may increase adherence to clinical guidelines and investments in medical equipment—both of which influence how well patients are treated. In addition, they found that the key driver of a hospital’s financial health is the so-called payer mix, which refers to the fraction of patients that are covered by a private health insurance.

This means that their original hypothesis about financial health being a mediating factor seems correct. Patient characteristics are related to financial health, which helps determine the quality of care. And the characteristic that matters most appears to be the share of patients with private coverage. Are there behavioral policy implications based on these findings? It certainly seems so. Perhaps the most relevant lesson concerns the use of economic incentives.

A growing trend among policymakers is to design monetary incentives in order to achieve certain outcomes. In education, teachers are sometimes rewarded for improving student outcomes. Households may be financially incentivized to reduce energy consumption during peak hours. And in Richmond, California, individuals at risk of shooting each other have even been paid to stay out of trouble with the law.

There are many reasons why incentives can be a powerful instrument in the policymaker’s toolbox. One is that our mental bandwidth is limited and incentives will help us prioritize certain behaviors and objectives. Another is the fact that many desirable outcomes, such as avoiding major floods due to climate change or contracting horrible diseases, are distant in time while the costs of reducing one’s energy consumption or vaccinating a screaming kid are immediate. To reduce our tendency to be present-biased—or subject to so-called hyperbolic discounting—a financial reward can be a suitable policy.

But economic incentives are not uncomplicated. Sometimes they work in the short term, but not in the long term. There is a risk that incentives reduce a person’s or an organization’s intrinsic motivation to achieve a particular goal. If a school pays students to read books, they may start doing so to receive a check rather than because of intellectual curiosity. Financial rewards can also crowd out other forms of desirable behaviors that are not being incentivized. When teachers receive lower salaries if their students do poorly on tests, they may simply start to “teach the test” rather than making sure that kids learn things that do not show up on exams.

Manary and co-authors point to an additional problem of monetary incentives: if the reason that a desirable outcome is not met is due to lack of pecuniary resources then the policy may actually backfire. In particular, they argue that when governments penalize hospitals if they fail to meet quality-of-care targets, this could hurt a hospital’s financial health, causing a reduction in adherence to clinical guidelines and investments in medical equipment, which may further lower quality of care and thus exacerbate rather than curb health inequality.

As an example, the researchers point to the Hospital Value-Based Purchasing Program of the Centers for Medicare & Medicaid Services (CMS). The program makes Medicare payments to hospitals conditional on how well the institutions perform on quality measures, health outcomes, and patient experience. Manary and colleagues suspect that because this incentive-based system singles out winners and losers, the hospitals that do poorly by the metrics and thus receive less funding may end up performing even worse as their financial health deteriorates. So well-intentioned—and to some extent well-founded—as monetary rewards for good behavior may be, they could ultimately backfire if the context is unfitting.

The broader lesson that their study presents is that when policymakers are keen on using economic incentives to promote desirable outcomes, they are well-advised to first investigate the primary factors that are causing the underperformance. If the root of the problem is pecuniary scarcity, alternative policies may be better suited.


Simon Hedlin is a researcher, a contributor to The Economist, and a graduate of the Harvard Kennedy School. His latest paper, co-authored with Cass Sunstein, studies choice architecture and pro-social behavior, and is available at:

Twitter: @simonhedlin