A recurring trope in American public life is that Americans live shorter lives of lower quality as measured by life expectancy, and life expectancy adjusted for disabilities or health conditions. To some degree this is clearly true, as the Financial Times’s John Burn-Murdoch showed earlier this month. This is not only true in an aggregate sense, its true across the income distribution. It’s not only the poorest 1 percent of Americans who live shorter lives than the poorest 1 percent in other developed countries, the richest 10 percent of Americans also live shorter lives than their counterparts elsewhere.

This fact is often used to argue that this is proof that American health care is failing in providing quality care, and that it is inaccessible. From this, the justification to adopt health care reforms that would make the United States more like Canada is advanced. The life expectancy gap, however, does not speak to the performance of health care providers in the United States as well as many think.

First of all, a large portion of the gap (but not all of it) is due to unique features of American life that are unrelated to health care services. After all, the United States has a higher homicide rate than most other rich nations. Americans also drive far more than people in other rich nations and thus there are more car fatalities in the United States. Finally, the United States has a far higher level of opioid consumption than other rich nations. Deaths from these causes tend to cluster among younger people such that they disproportionately depress statistics regarding life expectancy at birth. Once you account for them, somewhere between 36 percent and 48 percent of the gap in life expectancy for men with other rich nations disappears. For women, these explain between 17 percent and 19 percent of the gap. All these factors are “bads” that we would like to drive toward zero, but they do not speak to health care.

Second, a sizable share of the remaining gap is explained by the exceptionally high levels of obesity amongst Americans. Cardio-metabolic diseases – which are generally results from prolonged episodes of obesity – alone explain 38 percent of the gap for men and 34 percent for women. Obesity is directly tied to health, but not the quality of health care services. After all, there is a large element of personal volition and agency in play to become obese as many economists have pointed out.

Combine the obesity-related diseases with car fatalities, homicides and opioid consumption, and at the lower bound roughly 80 percent of the gap for men and 50 percent of the gap for women is explained by factors beyond the control of the health care system. And these proportions are probably too low as Emily Oster pointed out that the United States has a slightly different (but stricter) reporting of infant deaths that tends to overstate the gap in that age group, which tends to heavily affect life expectancy statistics, which, in turn, overestimates the gap between the USA and other countries. What is left can be reasonably assigned to health care performance.

But even there, we can be skeptical. For example, we can look at cancer-survival rates within a certain number of years (five years is the most commonly-used range).  For breast cancer, the survival rate in the US is the second-highest in the world at 88.6 percent. For prostate cancer, it has the fourth-highest rate at 97.2 percent. It does less well for lung cancer where it stands at the seventh-highest (18.7 percent). And these rates have been improving across all types of cancers since the 1970s due to important development in treatment, biopharmaceuticals, and other aspects of health care delivery. As such, quality does not appear to be a significant issue.

This leaves the question of access to health care. Some could be tempted to blame the fact that the United States does not have a system similar to Canada’s, the United Kingdom’s, and those of other European countries. It could very well be true that universal health care elsewhere explains the remainder of the gap. This assumes, however, that all of the remaining differences go in the same direction.

There are also multiple laws in the United States that restrict the supply of medical services that could be discarded in order to generate improvements in health outcomes without going in the direction of universal health care. One egregious example is that of the Certificate of Needs (CONs). CONs – which are in effect in close to 30 states – are designed to hold down health costs by limiting what some call “an unnecessary proliferation” of medical installations of all types (nursing homes, hospitals, etc.). But in reality, their main purpose is to protect incumbent providers from competition.

These laws are associated with, obviously, fewer health services. In turn, CONs are also associated with greater mortality across multiple types of disease as well as longer emergency wait times. During the COVID pandemic, states with CONs were associated with greater bed shortages and higher mortality rates than those without.

As such, a part of the gap that can be explained by health care quality can be explained by government regulations that essentially restrict the supply of services. This further reduces the weight that can be assigned to the universal health care systems of other countries. Overall, this makes it hard to strongly assert that adopting universal health care would be a remedy to the well-documented gap in life expectancy.