The link between life expectancy and wealth

Last week the Washington Post published a piece “Long life, wealth go together, study says” The Post has an annoying habit by the way of changing the title of an article between the print and online editions of the paper. The online title is “Research ties economic inequality to gap in life expectancy”

But what is the data? They referenced a study at the University of Washington and provide a link to the Institute for Health Metrics and Evaluation (IHME). However the only recent paper at that site is “New data show countries around the world grappling with changing health challenges” which refers to issues of global health. It links wealth and life expectancy at the global level.

The entire Washington Post article relies on the comparison between Putnam and St Johns counties in Florida. There is not doubt that life expectancy and income are higher in St Johns county than they are in Putnam county. But correlation does to prove causation and this is where the post article comes up short. This is especially true when only two cherry picked counties are used to make the point. Much more analysis is need beyond that described in the article.

The agenda behind the discussion appears to be the issue of raising the retirement age for Social Security and Medicare eligibility. This comes out in the citation from Dr. Rockeymoore of Global Policy Solutions who is quoted as saying:

“People who are shorter-lived tend to make less, which means that if you raise the retirement age, low-income populations would be subsidizing the lives of higher-income people,”

The article goes on to cite Monique Morrissey at the Economic Policy Institute who says:

“For many people, raising the retirement age would amount to a significant benefit cut.”

The argument being made is that the poor do not live as long as the wealthy therefore their contributions into the Social Security system subsidize the benefits of the wealthy. The follow-up logic is that this problem will be exacerbated if the retirement age is raised above the current levels. So first they are implying that wealth leads to a longer life. Then they follow up with this then shortchanges the poor in that they collect less from the system and subsidize the wealthy. These are strong claims.

The weakness in the logic of this argument is in the question: Is it wealth that determines life expectancy or is it lifestyle choices? The data cited in the Washington Post article lends credence to the lifestyle choices as a culprit when it points out that in Putnam county 27 percent are smokers and 35 percent are obese, while in St Johns the comparable percentages are 15 percent and 22 percent. What we have is a claimed correlation between life expectancy, wealth, and lifestyle. The article and the references it cite do not address the much more complex problem of the relationship between these variable.

All to easy is it to see a correlation between income and life expectancy and conclude that there is a causal effect. It is much more difficult to prove such a relationship. I won’t say the conclusions are wrong, only that the case remains to be made.

As a reality check I offer two web sites with maps of life expectancy by county. The actual data can be downloaded here. A very interesting exercise would be to download income data from American Factfinder (US Census Bureau) and look at the relationship. Just from the maps alone if income determines life expectancy the Midwest must be the richest part of the country. There are an awful lot of counties there on the higher end of the life expectancy scale. This in turn casts doubt on the claimed correlation between life expectancy and wealth – at least within the United States.


  1. Jeruza says:

    Hello there. I’m a fan of the blog, however I’m somewhat sad to see the ever-loved “correlation is not causation” used here.

    The association between wealth and both length and quality of life has been thoroughly established (I could link ecological studies, population based cohorts, census statistics). Economists have even tried to name it (the “wealth gradient” in mortality), and epidemiologists have learned to adjust all analysis by it.

    It’s true that we are not sure what is the cause for this. Wealth is associated with a number of healthy behaviors (not smoking, exercising, eating well), and adjusting for every one of them might (might!) eliminate the previous correlation.

    However, this is not important. If we are making an argument that raising the retirement age will hurt the poor more, this is true. And yes, they will subsidize the richer, since this group will live longer. It might be because a number of factors, and not because of money, but none of this discussion will make it a better deal for those dying earlier… We don’t need the causation here; because of a number of correlation studies, we simply know how things will work out.

    • Larry says:

      Jeruze, sorry to disappoint you. Perhaps I was not a clear as I could have been. My main point is that the piece in the Washington Post does not demonstrate their claims.

      Sure the poor smoke more and die younger. They die younger because they smoke more, not because they are poor. The cause there is between smoking and living longer. I cannot blame the poverty for that. On the other hand access to health care and the availability of health insurance are frequently, but not always, related to income. There wealth is, I am willing to assume, an important variable. The work needs to be done to reliably quantify the impact of each variable – life style choices and wealth – on life expectancy.

      Raising the retirement age will hurt those who do not live as long, in the sense they they do not get as much out of the system as they put in relative to others who live longer. We live in a society where on average people worked for 30 or 40 years and lived another 10 to 15 years. The retirement lifespan on average has increased yet no one wants to spend more time working. The poor and the wealthy have both gained from the increase lifespan although likely not equally. The consequences remain the same. We either have to save more while we work, or live on less when we are retired. The same amount of savings just will not last longer at the same standard or living. That is something we cannot escape from.

      Is the retirement system “fair” to the poor? No, not if they don’t live as long. It is not “fair” to anyone with a shorter life span. Let me extend that in a way no one will like. The point here is that fair may not be the right criterion when talking about retirement age. If fairness were the criterion men would be retiring somewhere around five younger than the women of this country due to their lower life expectancy. Just as one can claim that the poor are subsiding the retirement of the wealthy it is just as valid to claim that the men are subsidizing the retirement of the woman. I don’t know about you, but I don’t think I want to go down that path.

(Comments are closed)
  • Subscribe to Blog via Email

    Enter your email address to subscribe to this blog and receive notifications of new posts by email.

  • April 2021
    S M T W T F S
    « Jan    
  • Recent Posts