On deaths of despair

Posted on by

Philip N. Cohen writes (pdf download),

This paper uses complete death certificate data from the Mortality Multiple Cause Files with American Community Survey data to examine age-specific mortality rates for married and non-married people from 2007 to 2017. The overall rise in White mortality is limited almost exclusively to those who are not married, for men and women. . . .Analysis by education level shows death rates have risen most for Whites with the lowest education, but have also increased for those with high school or some college.

This is an important finding. I was sent an advance galley of Deaths of Despair, by Anne Case and Angus Deaton, due out in March. I wonder if they will want to revisit the causal narrative that they tell, which strongly emphasizes economic factors, based on the link between (low) education and high mortality rates. Cohen writes,

If White mortality increases are concentrated among people with low levels of education, for whom marriage has become rarer, it’s possible the increased White mortality among single people could reflect the greater share of that group with low education. However, Figure 3 suggests this is not the whole story. . .it appears the overall White marriage mortality ratio is driven both by increasing death rates for everyone at the lowest levels of education, and by increasing marriage disparities at higher levels of education.

32 thoughts on “On deaths of despair”

  1. Dr Kling,

    Could the correlation between education and marriage indicate that education is used as a proxy for earnings potential in assortive mating choices?

    • I rather think about aspects of personality that makes them successful at education are:
      1) Aspects that make them successful at their careers
      2) Aspects that make them successful at their marriage.
      So this is a huge circular function occurring here of self selection bias!

      While all this is true, the reality is the 1970s divorces were dominated by college educated couples so that is one reason why I believe later marriage (28 – 30) lowers divorce as well. (Note young college educated obviously can work and I believe lots of reasons for this.)

      (Note: This is probably an indication that for most couples having more money does improve marriage.)

  2. Of course my question is what is the plan? We say Marriage louder. Despite all the complaints about pop culture don’t most Rom coms end with the couple getting married or engaged. (Actually Rom coms are rare these days.) End video games and on line porn? The reality is the age of marriage in all societies has moved from early 20s in 1960 to 28 – 30 today so it is not just a US reality. (Check out South Korea some time)

    There is a lot of things national or state governments don’t do well and matchmaker is on top of the list. I don’t doubt conservative are correct on lowering the deaths of despair thinking religion and marriage are solutions but I like to know what are the plans.

    (I am dubious enforced young marriage is the solution here as 1970s divorce rates were higher.)

    • The government can go a long way to help by simply fixing family law to make it equitable. Something not spoken about in polite circles which are hidden in these numbers is how much of this despair is driven by divorce agreements which deny men their children or burden them unrealistically with transfer payments to their ex-wives under the guise of child support. I don’t have the numbers and would love to see them but if you are an American guy you would be stupid to get married for any reason and I think that knowledge is becoming common hence contributing to this problem as every guy knows someone personally that got railroaded. As long as your entire life can be effectively destroyed or your children killed based on the whims of a fanciful ex complete with judicial backing, you would be a fool as a man to marry. Or in economics terms, the cost to exit the market outweighs the perceived opportunity cost given the strong ability by the other party to engage in regulatory capture at that time that will destroy the rest of your life for their benefit.

      If the government wants to encourage marriage, they need to do no more than fix family law. That is a political dead non-starter.

      • That is dark and depressing and I hope it is not true. Regardless, the degree that “marriage exit costs” contribute to despair should be something that can be mined out of the data. This holds true for many/most plausible causative factors so better data collection/analysis might be a more important first step than a focus on interventions.

      • I am not sure how modifying divorce (or family) laws is going have any significant increase in marriage rates or decreasing divorce rates. In all honesty, it is in the government interest to have people married and not getting divorced.

    • Collin, I think a Moulder-of-Men type program can indirectly increase marriage rates. These programs can be both secular and community/charity driven. Rather than lecturing/preaching, these can be about measuring/ranking. Think Fitbit for Marriage with an associated Marriage Potential score based on “character” related measures rather than education/career measures. I don’t know what such a program should look like but I’m guessing that a bunch of experiments could find new innovative solutions.

      • Sure they are good programs but these programs are primarily built locally where these programs are more effective. And probably the ‘best’ national program was the military ‘draft’ of the WW2 that continued through the Vietnam years. The number of young men that joined the military after Korea was still high back then. (And then the draft sent young men to Vietnam which was not effective.)

        But these programs are best administered locally and many local areas depopulation is a reality. And the past 1950s the role of organizing the community were the housewives of the upper classes who now replaced by woman with careers. (I had a Poli Sci Professor in early 1990s state this and had hoped older citizens would fill this role. Unfortunately this has not come true.)

        • Moulder-of-Men was probably the wrong label as it harkens to military or sport programs of yesteryear. I was thinking more along the lines of a brand new cloud app/service or some kind of data driven program like the Weight Watchers point system with social feedback. It could start with the equivalent of the Netflix Prize or Space Prize to encourage ideas. The goal is improve the marriageability of men at the risky end of the socio-economic spectrum. I have no idea what might work best.

          Step 1: ask Robin Hanson how he would design/organize the prize system.

  3. For the record, I think decreasing deaths can be be tackled first and independently of addressing despair. The most important function for the statistical data is to serve as a baseline to judge the efficacy of any future intervention.

      • Isn’t the biggest portion of ‘deaths of despair’ mostly non-college educated people? So changes to higher educations impact would be limited.

        • Dr Kling states that the first paper’s finding that deaths of despair for those with some college is important. I was trying to support that and also to point out that the marriage rate and increase in the average age at marriage are not only related to deaths of despair but also impact the lives of the college educated. College involves trade offs and those pushing even more college for everyone are not perhaps being open about the trade offs.

          • This seems very long logic here and conservatives need to find a replacement for more college education. And in the Post WW2 years our society did have better options for vocational training, on the job training and pathways to middle class without college. (And isn’t a lot of failed college because young person is at least doing something.) And considering the working class labor supply is not increasing (Boomers retiring and lots of young people going college) that might be a significant competitive advantage.

            Honestly, I always thought the Devos hire was a bad Secretary Of Education because although she supported a lot of conservative school goals, I knew she not competent to pursue them. (kinda explains Trump administration IMO)

          • Should be try to decrease the first marriage age in the modern world? I know a lot of conservatives support this goal but I see low age of marriage in early 1960s was followed by the divorce revolution.

            I believe late marriage is one reason divorces are down for a variety of reasons. (I also believe the change of early marriage and kids was the real cause of the post -1990 crime drops.)

    • The paper referenced does not even mention government assistance, much less attribute any correlation. Here is the conclusion:

      Conclusions
      Strong and steady improvements in life expectancy have been observed in industrialized countries for the better part of a century (Oeppen and Vaupel 2002). This salutary pattern has begun to fade in the United States in recent decades (Crimmins et al. 2011; Institute of Medicine and National Research Council 2013). In this article, we identify the geographic areas in the United States where the trends in mortality since 1990, and especially since 2010, are most problematic. We find that improvements in life expectancy have been slowest in nonmetropolitan areas and that actual declines have been observed there in recent years. People living in large central metros, on the other hand, have experienced rapid improvements in life expectancy, producing a striking metro‐nonmetro divergence.

      We have documented this divergence for white men and women by sex, age, cause of death, and geographic region. Our analysis provides some clues about the principal causal processes at work, but it is hardly definitive. Screenable cancers, HIV/AIDS, and influenza/pneumonia helped to widen the gap and suggest the possibility of a role for the quality of medical services. On the other hand, our study shows warning signs about rising mortality resulting from factors less clearly tied to the health care system. In particular, younger adults aged 25–44 have experienced rapid increases in mortality and are now contributing to declines in life expectancy across all metro‐nonmetro categories. This rise in young adult mortality is largely attributable to drug overdose. We have presented suggestive evidence that changes in smoking and obesity are also implicated in the divergence. Metropolitan areas have benefited from a more rapid increase in the population with college degrees, a group with below‐average mortality. Poverty trends are more advantageous in nonmetropolitan areas, but part of the reason is that these areas are receiving larger government benefits associated with labor force inactivity. There are many layers to the mosaic of metropolitan‐nonmetropolitan mortality patterns. Continued efforts to identify the main factors at work are clearly justified by the massive divergence in the length of life that we have documented.

      Does Max Stirner encourage No-Lies-Barred sophistry as well?

  5. Arnold-

    I haven’t read Cohen’s paper yet, but it’s worth pointing out that this increase is likely explained by drug use, especially opioids, and that this problem is clearly not confined to the United States.

    https://www.jec.senate.gov/public/index.cfm/republicans/analysis?ID=B29A7E54-0E13-4C4D-83AA-6A49105F0F43

    Indeed, I recently found some data indicating that opioid prescriptions, illicit drugs, and associated mortality rises with per capita income. This problem is worse in the United States, of course, but it’s likely substantially explained by income levels and associated development (larger health system -> opioid prescriptions -> illicit drugs -> deaths)

    https://randomcriticalanalysis.com/2019/11/07/a-tale-of-two-covariates-why-owid-and-company-are-wrong-about-us-healthcare/#rcatoc-obesity-isnt-the-only-disease-of-affluence

    ~RCA

    P.S., My comment got hung up in moderation. I’m guessing it’s because of the links (href), so I’ll paste them without tags instead here. Feel free to delete the other one. Thanks!

    • this increase is likely explained by drug use, especially opioids

      It is mostly opioid deaths, but opioid deaths of the unmarried.

      • Yes, I’d sort of expect that given age and other factors. I’m just not sure the “despair” framing is the correct way to frame this given the role of prescription opioids, its link to national income levels, and the like, at least not if people mean this in terms of material want….

        • I had always thought that the “despair” in “deaths of despair” was meant to reference, for lack of a better word, spiritual privation.

  6. There is a very practical physical and old way to look at this:
    Ecc 4: 9 Two are better than one,
    because they have a good return for their labor:
    10 If either of them falls down,
    one can help the other up.

    Your spouse can remind you to take your medicine or harass you into not taking opioids drive drunk etc. They can call the ambulance if you have a heart attack.

  7. As a proxy for the problem consider rural American suicides:

    https://www.psychcongress.com/article/rural-america-hit-hard-recent-increase-suicide-rates
    Rural America Hit Hard by Recent Increase in Suicide Rates

    So there is some third variable effecting results, and it looks like sparsity of population. But sparse populations have another third variable of their own, small state divide does not allow economies of scale in federal programs. That is a driver of state population skew, Obamacare drives out young kids from rural Vermont.

    In other words, the symptom is: unmarried people are unhappy. Then there are overlapping, hard third variables which complicate any hypothesis. Don’t plan anything til we get a closer look.

    • The phrase “deaths of despair” is defined to include 1. drug overdose, 2. suicide, and 3. alcohol related liver disease. As commenter Random Critical Analysis states above, you would kind of expect it to include more unmarried people, even for physical reasons like “nobody was around to dial 911” as Floccina rightly points out, but I don’t think anyone predicted it to be so stark; its almost exclusively unmarried people.

      Your get-a-closer-look philosophy is good advice for the original 2015 Case and Deaton “Deaths of Despair” paper and this exclusively-unmarried paper is a great example of why.

    • Yea, US suicide rates and death of despair were a lot higher before WW2 than after where population density really grew. And also the simple reality is a lot of rural areas are losing health clinics that can save a portion of these lives. (What does Obamacare have to with people Vermont?)

Comments are closed.