Unbiased Analysis of Today's Healthcare Issues

Super-Simple Explanation: Why the U.S. has high health care costs and poor health outcomes

Written By: Jason Shafrin - May• 03•07

The concept of diminishing return is one that pervades much of economics. A Health Affairs article (“U.S Health Care Spending…“) by health economist Uwe Reinhardt and co-authors claims that since the U.S. spends more per-person on health care, the marginal benefit of any treatment is low due to diminishing returns.

This hypothesis got me thinking and I realized while the aggregate diminishing returns argument may explain the high U.S. spending and low marginal returns, it does not explain why Americans’ health levels are not best in the world. According to OECD data, the U.S. ranked 22rd in life expectancy even though it spent 25% more per person on medical care than any other country (this figure would increase to 50% if we exclude Luxemborg).

One explanation for this is the manner in which health care is allocated. Let us assume that each person in every OECD country faces the following health production function.

Health Spending
Health Level
$0 100
$1000 150
$2000 175
$3000 180

We can see that there are diminishing returns to purchases of medical care. Now let us look at a hypothetical scenario which could explain the U.S. high cost of health care as well as its poor outcome measure.

The U.S.

Let us assume that there are 10 people in the U.S. Four of the people are poor and Six of the people are rich. Poor people buy $1000 worth of health care and rich people buy $3000 worth of health care. This means that on average the U.S. will spend $2200 per person on health care. The average health level is 168.


Now let us look at the Canadian situation. Assume Canada has 4 poor people and 6 rich people as well. In Canada, however, everyone is entitled to government provided care which costs $2000 per person. Individuals are prohibited from buying private care, so rich people are not able to purchase more than $2000 worth of care. Total spending is $2,000 per person, which is less than the $2,200 spent in the U.S and Canadian health levels average 175 units. Thus we can see that Canada can spend less and still have better average health simply by targeting health spending towards the poor.


While this model may explain some of the differences in health spending and outcomes across countries, it of course does not explain it all. Below, I list just a few complications.

  • Taxes/Deadweight Loss: One may wonder why Canada only spends $2000 per person rather than $2200 per person if Canada and the U.S. have the same distribution of rich and poor people. The answer is Canada has to raise the money to fund the government run program through taxes and all taxes create some form of deadweight loss. I have assumed that the deadweight loss is $2000 for all of society to make the calculations simple. Even without the deadweight loss argument however, Canadian could simply have a preference for lower average medical spending than Americans do.
  • Lifestyle: It is possible that the U.S. has population which is more obese or has a less healthy life-style than individuals in other countries. This would imply that regardless of spending on medical care, American health outcomes will generally be worse.
  • Patient Preferences: Patients in the U.S. often demand the best care at the lowest price. This means that a person may choose to undergo surgery rather than take pharmaceuticals because there may be a slightly higher expected health level after surgery than using pharmaceuticals. Since Americans with health insurance do not pay for most of the cost of the operation, they are insulated from the true cost. In most European countries, surgeries may be free (or very low cost) as well, but the government rations these services through a much longer waiting period.
  • Wait Times: Continuing with this theme, most cost analysis does not take into account that Canadian and European patients often have to wait a long period of time for their care. The cost incurred by the patients due to the fact that immediate care is not available is almost never included in the cross-country medical cost calculations.
  • Equality: In term of income, the U.S. has a non-egalitarian society compared with those in Europe and Canada. If income is correlated with health and there are diminishing health returns to income, the results found above could arise. The progressivity of the tax and transfer system as well as the spending of the welfare state is a political issue which I will not address here. I will note, however, that generally more progressive systems increase the income of the worst-off at a cost that the high-income individuals have less of an incentive to work due to higher tax rates.
  • Centralized Purchasing: Many single-payer systems receive large discounts on pharmaceuticals and medical services due to their buying power. This lowers the cost of medical care for the individuals living in countries with a single-payer system. It is likely, however, that reducing the returns to medical care will reduce innovation.
  • Patent law: As noted in my post on April 30th, stringent patent laws may be unnecessarily driving up the cost of care.

To summarize, although the distribution of medical resources may be one explanation why the U.S. has higher health spending and worse health outcomes than other OECD countries, it is not the only reason why this phenomenon could be occurring.

You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site.