Blog Home


Inequality Is At The Core Of High Health Care Spending: A View From The OECD

October 9th, 2013

It is commonly said that the US spends more than twice as much on health care as other developed countries, yet its outcomes are worse. The inference is that too much care is provided, to no good end.

Such international comparisons are drawn from the Organization of Economic Cooperation and Development (OECD), a group of 34 developed countries. Analyzing these data is a multi-step process, like peeling an onion, and the truth resides deep within its core.

The process starts by adjusting health care spending for “purchasing power parity” (PPP) and expressing it in US dollars. By that measure, per capita spending in the US is 160 percent more than the OECD mean (Panel A, left bracket), and this is the basis for the notion that the US spends more than twice as much. But it is only the first layer.

HA Blog A

The second layer is the economy. The US spends more principally because it is wealthier (Panel A, right bracket), but even in proportion to its gross domestic product (GDP), the US spends more, about 60 percent more. But that is only the second layer.

The third layer is price. Health care prices are inordinately high in the US and inordinately low in many other countries, particularly those that exercise price controls. Therefore, to understand how much care is given, comparisons of health care spending must be adjusted for the purchasing power parity of health care (HC-ppp). When so adjusted, spending in the US is still higher relative to its GDP, but by only 31 percent (Panel B). This represents the core difference in services. Some are administrative, but most are health care services.

HA Blog B

What explains this 31 percent? A large body of evidence suggests that it results from poverty and income inequality, which are more prevalent in the US than in any other OECD country except Chile, Mexico and Turkey. And poverty is associated with substantial increments in spending. For example, the poorest decile of Medicare beneficiaries spends 30-40 percent more than the wealthiest; overall hospital utilization rates in large urban areas are 25-35 percent more than in their wealthiest Zip codes; and hospital readmissions are most prevalent from poor neighborhoods and in safety-net hospitals.

Much of this relates to chronic illness, which is most prevalent among the poor. And chronic illness rates are higher in the US than in most other OECD countries, higher than Canada or England and higher than the average of France, Germany, Italy, Japan, Spain and the UK. Similarly, obesity, which is most common among the poor, is most prevalent in US. And the rates of infant, maternal and preventable mortality, which are often taken as measures of health care effectiveness but are actually markers of poverty and the burden of disease, are all higher in the US than in any other advanced economy.

The Gini coefficient is a measure of income inequality. To estimate the impact of income inequality on health care spending, it was applied to the spending level in each of the OECD countries (Panel C). So doing erased the difference between the price-adjusted level expected from GDP and the actual expenditures in the US. Thus, while the US spends more than twice as much on health care than the mean of other OECD countries, its greater GDP and higher prices explain most of it, and income inequality offers an explanation for the rest.

HA Blog C

But there is more, and it is the core of the core. The OECD measures a host of spending categories in addition to health care. In most, spending in the US is proportional to its GDP. Health care, which exceeds the OECD norm, is one exception. A second is social spending, and it deviates in the opposite direction. Social spending in the US is 33 percent less than predicted from GDP (Panel D). And recent trends are constraining it further by limiting funds for food stamps, housing subsidies, and programs that serve youth, the elderly, and the homeless.

HA Blog D

It is difficult not to connect the dots from inadequate social spending to excess poverty and income inequality to more chronic illness and higher health care spending. These dots reside in the core of the OECD onion, and the failure to cope with them is placing an unsustainable burden on our health care system.

Email This Post Email This Post Print This Post Print This Post

Don't miss the insightful policy recommendations and thought-provoking research findings published in Health Affairs.

1 Trackback for “Inequality Is At The Core Of High Health Care Spending: A View From The OECD”

  1. National Income Inequality and Local Poverty: Correlates of Health Care Spending | PHYSICIANS and HEALTH CARE REFORM Commentaries and Controversies
    October 15th, 2013 at 2:36 pm

7 Responses to “Inequality Is At The Core Of High Health Care Spending: A View From The OECD”

  1. ctilden Says:

    The Affordable Care Act as originally enacted was supposed to expand Medicaid as well as create exchanges. This portion of the ACA was struck down by the Supreme Court, giving states the option to expand Medicaid coverage. As such, we should acknowledge that the original intent was not primarily focused on the middle class, but expanding health care coverage to include nearly all Americans.

  2. Johnathon Ross Says:

    Thank you so much for this succinct graphic summary. I will share it regularly in the future. For many years now I have been educating medical students and residents about the social determinants of health. Every day in our center city clinic and teaching hospital, we encounter the holes and trapdoors that effect mainly the working poor. Our hyper-complex, money focused sickness care non-system leaves 47 million of them with no coverage and tens of millions more with coverage so skimpy that one serious illness can destroy a family economically. Health care bankruptcy simply does not happen in any of the other wealthy OECD countries.

    There is a feedback loop between poverty, mental illness and substance abuse and physical illness. If you develop physical or mental illness you are likely to lose your job or not be able to work which means you lose your health care coverage until you are poor or disabled enough to qualify for Medicaid or Medicare. While I certainly agree that income inequality and a frayed social safety net can overwhelm any healthcare system, let alone one as dysfunctional as ours, a health care system that covers everyone from cradle to grave is part of a that social safety net that saves lives and saves money. That system is not the ACA which sadly adds even more complexity and skimpy insurance. We need a national health insurance program like an improved and expanded Medicare for all and a government that looks for the health all our public policies.

  3. Richard "Buz" Cooper Says:

    In his follow-up comment, Lee Tocchi called attention to social mobility. Yes, Clinton and Obama were successful in climbing the economic ladder. The problem is that fewer move up in the US than in almost any other developed country. For example, when measured as years of schooling achieved as compared with one’s parents, the US ranks 14th out of 16 countries (only Italy and Hungary are worse), and for inter-generational elasticity in earnings, the US ranks 13th of 17 countries (Italy, Chili, Slovenia and the UK are worse).

    The picture gets darker when various areas of the US are examined, as was done by the Equality of Opportunity Project at Harvard. The areas where low-income children are least likely to move up the income scale are in Appalachia, the Deep South and south-west and the rust belt, while success is greatest in the upper-Midwest and plains states. Among large cities, those with the least social mobility include Atlanta, Indianapolis, Detroit, Cleveland, Jacksonville and Milwaukee, each with dense areas of poverty and high health care expenditures, while those with the most social mobility include Salt Lake City, San Jose, San Francisco and Seattle, with some of the lowest health care costs in the nation.

    William Julius Wilson, who applied the term “the truly disadvantaged” to people living in concentrated poverty, attributed the failure of social mobility in such areas to a “spacial mismatch,” which separates poor people in the urban core from jobs in the periphery. This proved to be true in the Harvard study, which found that communities with concentrated poverty had the least social mobility. Other factors correlating with less social mobility included income-inequality, single parent families and low school budgets.

    Yes, Clinton and Obama made it. So did lots of my friends. Some lived in public housing and all had good schools and visits from public health nurses when they were infants. One of my older friends, now a member of the 1%, told me, “when I got out of the service after WWII, I didn’t have two nickels, but I had the GI Bill.” They profited from strong investments in social infrastructure and, as a group, their health care costs are low, less than half the costs in urban ghettos.

    In states that allow the ACA to work, it will contribute to the social infrastructure for the poor and near poor. Sadly the very states with the least social mobility are the ones that reject the ACA. But the ACA is not enough. There are many non-health elements that are important, and they must be strengthened, too. That’s how it’s done in France, Sweden, Japan and other countries, where social equity is high and health care costs are low. The lesson is, pay for a social infrastructure early or pay double for health care later.

  4. leetocchi Says:

    Buzz, I would be a believer in “income inequality” as a large problem if America didn’t have social mobility. The countries you listed have long established class structure. Here people like Bill Clinton and the current president didn’t come from privileged backgrounds and managed to scale the income ladder to the top. We have been in persuit of helping the poor with many of the programs you describe with seemingly no effect for decades. The current progressive thinking is to disparage those with aspirations to slow down. We don’t want to achieve equality by holding back those advancing our society, technology and economy.

    The ACA in particular seems to be off target for political reasons skipped the poor and wet to a middle class benefit.

  5. Richard "Buz" Cooper Says:

    In response to leetocchi: Health care reform has made a big step in the right direction by expanding insurance coverage and decreasing the financial burden of disease for many who are poor or near poor. But what we have failed to do is to follow the lead of countries like Sweden, France and Japan, which have made social equity a national goal. They have invested in material conditions, such as housing, neighborhood safety, job-training, sick-leave, public transportation and access to food, goods and services. Programs have been created that nurture the ability of mothers to nurture their children. And investments have been made in education, from pre-school through high school. As a result, income-inequality has narrowed, health inequality has diminished and health care spending has been controlled.

    In response to jgkahn: It is true that 15% of spending in the US is related to administration, However, most of that spending is already blended into prices. While cross-country comparisons have proven to be difficult, a study from the Congressional Research Service (1) found that, in comparison with Germany, France and Switzerland, which are viewed as valid benchmarks, administrative costs in the US are greater by an amount equal to 2-3% of total spending. The same conclusion was reached by McKinsey Global (2) in a study of a larger group of countries. So it would be valid to shave a few percent off of the 31% of “core services” in panel B, but that would have little effect on its magnitude.

    1. Peterson C L, Burton R. (2007). U.S. health care spending: Comparison with other OECD countries. Washington, DC: Congressional Research Service.
    2. Angrisano C, Farrell D, Kocher B, Laboissiere M, Parker S. Accounting for the Costs of Health Care in the United States. McKinsey Global Institute, Jan 2007

  6. jgkahn Says:

    Terrific blog. I like the stepwise approach, and the consideration of inequality.

    One important elaboration: Mr. Cooper writes, “When so adjusted, spending in the US is still higher relative to its GDP, but by only 31 percent (Panel B). This represents the core difference in services. Some are administrative, but most are health care services.”

    It’s true that most services are clinical. However, the incremental (excess) U.S. expenditure on billing and insurance related administration, at payers and providers, represents 15% of total health spending — nearly half of the observed excess before adjusting for inequality. Thus, with the adjustment for inequality, U.S. spending on real services is actually substantially below that in other OECD countries. That is, it seems we’re getting fewer services. Which is exactly what we know to be true from other data sources, eg on frequency of primary care visits and hospital days.

    Yes we have inequality. Yes it contributes to greater illness and need for health services. But it doesn’t bring us to the service level of other OECD countries, instead *below* them.

  7. leetocchi Says:

    Buz let me get this straight…we spend 60% more on healthcare proportional to GDP than other countries, and at higher prices, We spend more on the lower end than the wealthy. Our social spending excluding healthcare is 30% below proportional to GDP spending in other countries. So the poor don’t get enough and the wealthy already pay too much and we just rearranged the deck chairs to spend more for the care of the middle class… Exchanges covering 133% of poverty to 400%, when the real focus should have been on the poor? Does the logic then say health reform was more about political control and not about health care problems?

Leave a Reply

Comment moderation is in use. Please do not submit your comment twice -- it will appear shortly.

Authors: Click here to submit a post.