Lateral thinking

Over at Econlog a few weeks ago I did a post entitled “How I Think.”  I was reminded of that post when reading commenter theories on the poor health outcomes of American whites aged 45-54.  In the post, I said that when evaluating Alan Greenspan’s performance, you don’t want to focus on Greenspan; you want to focus on how other central bankers did during this period (about as well.)  When thinking about China’s growth prospects you don’t want to focus on everything you know about China (too complicated), but rather look at other East Asian countries.  And when deciding whether the gold inflows to Spain (1500-1650) were a resource curse that hurt long-term growth, you want to look at other Mediterranean regions that did not received big gold inflows.

Here’s the graph:

Screen Shot 2015-11-05 at 2.22.21 PMPeople were providing answers that seemed to miss the big picture. The graph shows two surprising results, the poor performance of middle-aged white health after 2000, and the excellent performance of Hispanic health after 2000.  In America, Hispanics tend to be disproportionately low income/working class, the group that has been hit hardest by recent economic trends. They were also especially likely (before ObamaCare) to lack health insurance.  And yet their health is significantly better than the health of French and German citizens, in the same age group.

I have no theories at all—I don’t even know if the data are accurate.  But if I was going to come up with a theory, I sure as hell would make sure it explained the sudden and massive divergence in White/Hispanic health outcomes.  If it didn’t, I’d have zero confidence that my theory was correct.

Paul Krugman has promised us an explanation in a future post.  Let’s see what he comes up with.

PS.  I don’t know about you, but to me that graph undercuts some of the recent anti-immigration hysteria.  If Hispanics are actually so inferior, so likely to degrade our precious “Anglo” civilization, how come they have such superior health outcomes?  In rich countries like America, don’t poor health outcomes often reflect poor lifestyle choices?  Just asking.

Update:  Commenter Mike Scally linked to an Andrew Gelman post that says the US data is biased about 5% upward due to compositional effects (the 45-54 age group is getting slightly older, due to boomers passing through).  So instead of rising slightly, US white (middle aged) death rates at a given age have been essentially flat. Of course other death rates fell about 30%, so there’s still a pretty big mystery.



46 Responses to “Lateral thinking”

  1. Gravatar of Nicholas Weininger Nicholas Weininger
    5. November 2015 at 11:55

    Also, the across-the-board drop in crude death rate for that age cohort over the past 25 years is enormous. Something is going really right for most middle aged people in industrialized countries and for US Hispanics, and any analysis has to start with an understanding of what that is. Less smoking and/or excessive drinking? Cleaner environment? Better treatments for common ailments of that age group? All of the above?

  2. Gravatar of Dylan Dylan
    5. November 2015 at 12:18

    “If Hispanics are actually so inferior, so likely to degrade our precious “Anglo” civilization, how come they have such superior health outcomes? ”

    Health genetics are different from behavioral/IQ genetics, and latin (in the European sense) countries have better health outcomes than you’d expect based on income. Some quick googling suggests Italy and Spain both have life expectancies over 82, while all of the richer northern European nations are 80 or 81.

    The white component of Hispanic ancestry in America has a predilection for better health than the core Irish/English population. I also wonder if the Aztec native population handled a corn heavy diet better than whites do, just like Asians handle rice better.

  3. Gravatar of Steve Steve
    5. November 2015 at 12:20


    Death by prescription drug and heroin has been skyrocketing.

    There’s huge co-morbidity. Over half of heroin addicts start out on prescription drugs, and they also have 3x marijuana abuse and 2x alcohol abuse vs general population.

    Suicide rates are also quite high:

    “White males accounted for 70% of all suicides in 2013.” – A statistic you will never hear on MSM, where white males are claimed to have ‘privilege’ and are also claimed to be undercharged for health insurance.

    Additionally, the co-morbidity is likely quite high between heart disease and alcohol/marijuana/heroin/prescription drug addiction, especially in the 45-54 age group.

    It will be interesting to see what political theory Krugman is able to invent to explain this. His title “HEARTLAND” of darkness is a hint. Given Krugman’s lack of intellectual curiosity, I doubt he will try to break out deaths by underlying cause.

  4. Gravatar of Steve Steve
    5. November 2015 at 12:28

    I suspect that cultural factors are at play, in the sense that whites are expected (by parents, friends, and society) to become homeowners with nuclear families. When that doesn’t work out (bad economy and divorce), the sense of social isolation and failure are crippling.

    In contrast, large extended families are acceptable in Asian and Hispanic communities, which provides both an emotional and a financial safety net. Hispanics also have the benefit of religion.

    I have no idea how to test the empirical validity of this thesis.

  5. Gravatar of ssumner ssumner
    5. November 2015 at 12:49

    Everyone, Interesting theories, I’ll keep an open mind.

  6. Gravatar of Ray Lopez Ray Lopez
    5. November 2015 at 12:59

    C’mon man, don’t tease us like that. You promised a graph on Conquistador Spanish resource curse and gold inflows in Med countries, which would have been very cool. Do you have this data? Please upload it.

  7. Gravatar of E. Harding E. Harding
    5. November 2015 at 13:16

    “and the excellent performance of Hispanic health after 2000.”

    -Shouldn’t that be “unexceptional”?

    “but rather look at other East Asian countries.”

    -I still prefer to keep an eye on both East Asian and post-Communist countries.

    “If Hispanics are actually so inferior, so likely to degrade our precious “Anglo” civilization, how come they have such superior health outcomes?”

    -Exercise? Less smoking? And the inferiority is in culture (out-of-wedlock births), IQ (two-thirds of the way to that of Blacks), and politics (excluding White Hispanics). Think of Greeks, basically, only more so.

  8. Gravatar of Christian List Christian List
    5. November 2015 at 13:45

    Genetics explain the absolute numbers. The only really interesting thing I see is the diverging trend of the white US males from all the others – assuming the data is correct.

  9. Gravatar of Christian List Christian List
    5. November 2015 at 13:51

    Jugman (aka Krugman) will say it’s because of the wrong economic policy by the GOP. He will say that the GOP left white men behind when in fact the Dems left white men behind (if anybody). It’s really not hard to imagine, isn’t it?

  10. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    5. November 2015 at 14:03

    I’m betting on the data being wrong. It’s just too big and too strange an outcome.

    Or, maybe, for the US governments, white lives don’t matter.

  11. Gravatar of marcus nunes marcus nunes
    5. November 2015 at 14:35

    At least the timing and the mostly affected group fits:
    What explains the current low rate of employment in the US? While there has been substantial debate
    over this question in recent years, we believe that considerable added insight can be derived by focusing
    on changes in the labor market at the turn of the century. In particular, we argue that in about the year
    2000, the demand for skill (or, more specifically, for cognitive tasks often associated with high educational
    skill) underwent a reversal. Many researchers have documented a strong, ongoing increase in the demand
    for skills in the decades leading up to 2000. In this paper, we document a decline in that demand in
    the years since 2000, even as the supply of high education workers continues to grow. We go on to
    show that, in response to this demand reversal, high-skilled workers have moved down the occupational
    ladder and have begun to perform jobs traditionally performed by lower-skilled workers. This de-skilling
    process, in turn, results in high-skilled workers pushing low-skilled workers even further down the
    occupational ladder and, to some degree, out of the labor force all together. In order to understand
    these patterns, we offer a simple extension to the standard skill biased technical change model that
    views cognitive tasks as a stock rather than a flow. We show how such a model can explain the trends
    in the data that we present, and offers a novel interpretation of the current employment situation in
    the US.

  12. Gravatar of Christian List Christian List
    5. November 2015 at 14:40

    Don’t forget that two medical journals rejected the study.
    I think for good reasons.

  13. Gravatar of Willy2 Willy2
    5. November 2015 at 14:40

    What I DO know is that the white part of the US population is shrinking & ageing, and therefore is impoverished at a much highr rate.

    But we’ll have to see what Krugman coes up with.

  14. Gravatar of Jose Romeu Robazzi Jose Romeu Robazzi
    5. November 2015 at 15:32

    Do people have an incentive to declare hispanic ethnic group ? Quotas somewhere? Just asking, because of one cannot trust the data…

  15. Gravatar of Steve Steve
    5. November 2015 at 15:35

    The WSJ has a good article with a 2nd graph showing that at least half of the gap is explained by overdoses, suicides, and liver disease. My hunch is that the other half is explained by co-morbid (with pain and depression) illnesses like accidents and heart disease.

    The better question is why. WSJ postulates “Many people are reaching midlife short of a cornerstone of the American dream: making more money than their parents did.”

    I expect Krugman will blame lack of health insurance and income inequality.

    The problem with the first explanation, is that the majority of heroin deaths start out as prescription drug addictions, which implies a doctor was involved at some point. And high-deductible Obamacare plans aren’t going to cover psychotherapy, only the catastrophic injuries that result once health goes down the crapper. Heroin is cheaper anyway.

    The problem with the second explanation, is that these behaviors are rising in all income categories. It’s not clear if people who are suffering are the ones who would benefit from income redistribution, or the ones who would be asked to pay for it. Likely both.

    I suspect the root explanation includes social isolation and family breakdown, often resulting from economic hardship and the weight of high expectations.

  16. Gravatar of Brian Donohue Brian Donohue
    5. November 2015 at 15:36

    The data look suspect. I attended a conference a couple years ago where Jay Olshansky, a well-regarded demographer, was talking about mortality diverging between classes in America, with lower socioeconomic strata seeing worse outcomes.

    Maybe that was wrong, or maybe it can be reconciled with the graph above, but I’m skeptical.

  17. Gravatar of bill bill
    5. November 2015 at 16:20

    I’m curious to see similar data for African Americans too. Why would that be left out?

  18. Gravatar of ssumner ssumner
    5. November 2015 at 19:46

    Christian, I don’t think anyone “left white men behind” the position of Hispanics.

    Steve, You said:

    WSJ postulates “Many people are reaching midlife short of a cornerstone of the American dream: making more money than their parents did.”

    Interesting, Somehow I can’t picture people in a rich country killing themselves because they aren’t quite as rich as their parents. But who knows, I don’t have any other explanation—and it would sort of explain the Hispanic divergence, as even poor Hispanics often do better than their parents did in Latin America.

  19. Gravatar of Steve Steve
    5. November 2015 at 20:31

    “Somehow I can’t picture people in a rich country killing themselves because they aren’t quite as rich as their parents”

    When you phrase it in terms of averages, it sounds really stupid.

    However, when your dad was a union manufacturing worker, and you make tacos at Taco Bell, it is a big gap. Or if your mom and dad are both university professors, and you spend your day as a lab tech putting 100 little cups of urine a day into a medical diagnostic machine.

    It’s also well known that job loss and divorce are often triggering events for depression and substance abuse…if you don’t have a family or community willing to support you, or if your family is small and distant, sometimes you never come back. The causes are multi-factoral: economy, drugs, and family breakdown.

    Remember we are talking about tails, not averages, 4% of white people die between 45-54, vs 3% of Hispanics. Little things shift the tails.

    The person committing suicide is the dad who lost is job and was told he can never see his kids again. That happens in the 45-54 age group a lot. It happens to white men, who don’t have the ‘chauvinistic privilege’ of Hispanic men.

    There are also the prescription drug users who turn to hard drugs; the addicts who are shunned by family; and the unemployed who shack up in unsavory places due to lack of family.

    The pretense of the elites is that the extra 1% caught in the tail can be saved by taxing the shiz out of the 99% and giving all the money to the medical system.

  20. Gravatar of Steve Steve
    5. November 2015 at 20:41

    I should add that I support rehab programs, including the sort where a clinic offers safe doses of narcotic drugs as a substitute.

    My beef is more with the idea that moar government (which crowds out family, civic and religious institutions) while running up the bill is the answer.

    And maybe white culture needs to change, too. Accept your kids, even if they work at taco bell or handle excretions at a hospital. And let fathers see their children.

  21. Gravatar of Don Don
    5. November 2015 at 21:28

    Suicide and overdoses must be from the rising cost of college tuition and declining incomes.

    Study PDF via:

    ps. Didn’t the Oregon study show that insurance access had little effect on health outcome?

  22. Gravatar of Dots Dots
    6. November 2015 at 01:01


    1) many bankruptcies come from medical expenses
    2) many divorces and depressions r caused by bankruptcy
    3) broken homes produce inferior psycho/social outcomes for children

    then Krugman/Europhile conclusions about healthcare policy might not b unreasonable.

    Obamacare would seem to have worsened the situation, tho: before u may not have needed health insurance until an event large enough to bankrupt u or prompt charity, whereas now you’re paying installments that r too large for comfort and too small to merit the courts’ protection. it might also b easier for whites to get access to prescrip drugs for rec use and ODs

    as a neoliberal, I think we (and by we I mean the people smarter than me whose thoughts I shout from the rooftops) the potential for damage to human capital more seriously in choosing/designing claims about policy. I think it would be nice to start by stuffing people through nurse’s aid programs, as oldie care understaffing is sad, employment there seems steady, and promotional opportunities seem obvious

    Oregon health insurance study also claimed that medicaid had good psych effects

    do middle aged whites own more guns than middle aged hispanics?

    if ‘the majority’ looks down on u and u r a unit of ‘the majority’, it’s ur fault. if u’r a minority unit, their gaze seems to u and ur close comrades the same as the gaze directed at them, so u’r not singled out

    we have no romantic struggle against society, we have no streetfighting

    we bald, we r pale, we eat dumb food, our institutions r mocked in pop culture, we follow politics, our language is booby trapped with moral connotations, we bore women, we bowl alone – actually I’m just speaking for myself, here

  23. Gravatar of Engineer Engineer
    6. November 2015 at 03:45

    If you dig a little deeper into the data, you see that it is entirely from Whites with no college education and lack of a decrease in mortality from the larger group is due to a rise in problems due to drug and alcohol from this smaller group.

    This is the time of one’s life where a lifetime of substance abuse catches up with you. As someone who grew up in a redneck area …and saw a large portion of my HS class already abusing drugs and alcohol in high school…I can’t say it really surprises me….

    What I see in white society is a huge cultural split between the educated, wealthier white society and for lack of a better term the redneck society that I never really saw when I was a kid. I think Charles Murray has an entire book on it (have not read it..only reviews…Educated white males no longer marry their secretaries….which also explains a large part of the inequality rise). Unlike minorities who tend to support one another… educated whites are afraid of the lower group…I see that a lot in real estate and school districts..educated white Americans are more afraid to send their kids to redneck areas than minority areas…

  24. Gravatar of Le Comte de Monte Cristo Le Comte de Monte Cristo
    6. November 2015 at 04:11

    The full paper contains answers to some of the questions posted by commenters above. It is an interesting read.

    Black non-Hispanics (BNH) between ages 45-54 have seen a large drop since 1999 in mortality rate (-215) down to 581 deaths/year/100 000 pop.

    Middle age WNH with a college degree or better have seen a drop (-57) down to 178.

    Middle age WNH with high school degree or less have seen an increase (+134) to 736. This is the main reason the overall figure for WNH has increased slightly to 415, bucking the trend send in other countries.

    The paper states that this increase is not seen for older age groups.

    However, the increase deaths related to alcohol and drug poisoning, suicide or liver disease is very similar for all age groups between 30-64 for WNH. About 40 extra deaths/year/100 000 pop.

  25. Gravatar of Morgan Warstler Morgan Warstler
    6. November 2015 at 05:41

    1. I suspect you drag out whites in UK, Sweden, Germany etc, you are going to see such effects…

    2. Force fat assed drinkers to get up and work all day and their BMI goes down, reducing fatty liver disease even if they drink same amount. I can’t say this enough, mother nature didn’t build humans for idleness.

    3. Hispanics which have about same obesity at poverty levels as whites, are far younger, so they aren’t seeing group effects yet.

    Once again #Uber4Welfare – a software driven platform that modifies bad individual behavior – is the proper public health solution.

  26. Gravatar of collin collin
    6. November 2015 at 06:13

    Maybe it is lifetime expectations. In terms of Hispanics and African-American working class, they grew very poor but still felt their lives improved over their parents. A good portion of the data involves people who born before Civil Rights passed. And then it still would take a generation for minorities to feel the full benefit of Civil Rights. No this not some kind of political statement but it takes time for changes.

    In terms of whites, this group was toward end of the Baby Boom and as group they did not surpass their parents livelihood and did achieve the success than their siblings who went to college. I still this group hardest hit by China outsourcing (Rust Belt continued to collapse.) and most cases being Empty-Nesters, it is easier to fall into poor behavior. The authors compared this group to the fall in life expectancy in Russia during the 1990s. It was a group of people who expected better for their lives and in the long run their world changed around them. This was group that the song Born In The USA was about.

  27. Gravatar of Mike Scally Mike Scally
    6. November 2015 at 07:17

    [Apologies if already posted] The following is relevant to discussion.

    Correcting statistical biases in “Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century”: We need to adjust for the increase in average age of people in the 45-54 category

  28. Gravatar of DF DF
    6. November 2015 at 08:01

    Lateral thinking is how we should judge the one-child policy as well.

  29. Gravatar of Capt. J Parker Capt. J Parker
    6. November 2015 at 08:07

    Here’s my 25 cent correlation: First recognize that other write ups of the Deaton study indicate that it is non-college educated middle aged white americans that are most suffering the poorer longevity outcomes.
    Next consider that the health decline begins around 1998. Also beginning in 1998 is a sharp decline in US manufacturing jobs. Manufacturing shed close to 1/3 of its workforce between 1998 and now. Hispanics make up about 16% of the manufactuing workforce and about 17% of the US population so loss of manufacturing jobs doesn’t immediatly explain the divergence in white vs hispanic mortality. However, The median age of US hispanics is only 27, for whites it’s 39 so it’s possible that age distribution differences make the loss of manufacturing jobs more of a problem for middle aged whites of modest education than for hispanics.

  30. Gravatar of Christian List Christian List
    6. November 2015 at 08:22

    “Christian, I don’t think anyone ‘left white men behind’ the [explains?!] the position of Hispanics.”

    You don’t need to explain the Hispanic graph. You focus way too much on that. The Hispanic graph is in trend with all the other graphs. There’s not too much to explain there.

    My military veteran theory still rocks though.

    @Capt. J Parker
    That sounds like a good theory to me.

  31. Gravatar of Christian List Christian List
    6. November 2015 at 08:46

    As I said earlier two pretty big medical journals rejected the paper by Deaton. Now the media makes fun of those journals but I think the medical journals got it exactly right.

    Deaton basically did a simple retrospective cohort study. In medicine those kind of studies have the least value of all studies. You need to be really careful with those kind of papers.

    Think about it this way: Deaton got this huge pile of date. Let’s say the mortality rates of all Americans for example. Now you can subclassify this data even more but you need to be really careful with that. For every subclassification you need a really good reason.

    Let’s say you subdivide the data by gender. This is a justified reason because of the genetic differences between men and women. Deaton also subdivided by race. This might be a justified reason because of genetic and social differences that might occur between races. It’s not as good as gender but it can be done.

    But then I think he made a huge mistake. He subdivided by age. That’s justified when you make rough splittings: Let’s say children vs. elderly people.

    But Deaton handpicked the ages 45-54! Why in the world would you pick 45-54? Why not 40-50? Why not 37 to 47? Why not 35.5 to 55.5? None of those kind of subdivisions would be justified in my opinion. It’s not a good idea. What’s his justification? He got none. He just handpicked the data and found this weird trend that he can not explain. Since when is handpicking in retrospective cohort studies allowed?

    He might have won the Nobel Prize but so did Paul Krugman, didn’t he? Throw this study in the garbage bin. It has not much value. And that’s why the medical journals rejected the paper.

  32. Gravatar of HW HW
    6. November 2015 at 09:43

    I would like to point out that the mortality rate for non-Hispanic whites, according to data gathered from the same increase in mortality rates could be observed for age groups 25-34 and 35-44, between 1999 and 2013. In fact the data for the age group 25-34 was particularly striking, with mortality rates increasing from 91.3 in 1999 to 110.5 in 2013, representing a relative increase of more than 20%. For other groups like Hispanics, non-Hispanic blacks, and Asian or Pacific Islanders, the mortality rates continued to decrease from 1999-2013 for the age groups 25-34 and 35-44.

  33. Gravatar of Floccina Floccina
    6. November 2015 at 09:53

    I would guess that there is some sort of mistake in the USW data.

  34. Gravatar of Christian List Christian List
    6. November 2015 at 11:06

    That would be really interesting.

  35. Gravatar of Christian List Christian List
    6. November 2015 at 11:14

    @Mike Scally
    That might be it.

  36. Gravatar of benjamin cole benjamin cole
    6. November 2015 at 17:39

    I blame tight money policy by the Federal Reserve Board, exactly when men in this ethnic age group are expected to build up financial assets. So they shoot dope instead.

  37. Gravatar of ssumner ssumner
    6. November 2015 at 18:23

    Everyone, Thanks for all the information.

    Steve and Dots, Interesting theories.

    Thanks Mike, I added an update.

    Christian, When the facts clearly refute your military theory, you really need to give it up. These guys didn’t fight in major wars. (With a few exceptions.)

  38. Gravatar of Christian List Christian List
    7. November 2015 at 05:39

    I follow the footsteps of any “good” political person (and economist): I adapt the facts to my theory – not the other way round. I think my idol is Paul K. 😉

  39. Gravatar of ssumner ssumner
    7. November 2015 at 13:18

    Well at least you’re honest!

  40. Gravatar of Engineer Engineer
    8. November 2015 at 07:07

    I suspect that this would be highly correlated with Scots-Irish genetics (that is my genetic background…so I think I know a little bit more that some little Jewish guy from NYC).

    This blog entry pretty much explains the plight….

    In my opinion this group (which also has probably more native American blood than any other group in the US) is genetically highly susceptible to substance abuse.

  41. Gravatar of ssumner ssumner
    8. November 2015 at 18:37

    Engineer, Thanks, that’s interesting.

  42. Gravatar of HW HW
    8. November 2015 at 19:13

    The thing is, Deaton and Case are incorrect; younger age groups have also seen their mortality rates increased (non-Hispanic whites)

  43. Gravatar of Justin Justin
    9. November 2015 at 13:29

    I don’t think anyone serious is using the word “superior”. I think they’re saying that Latin American derived citizens vote 4:1 for the democrats, and have anti market political views. Not to mention showing low trust behaviors.

  44. Gravatar of Scott Sumner Scott Sumner
    10. November 2015 at 16:54

    Justin, Don’t Jews also vote overwhelmingly Democratic? Would more Jewish immigrants to America be a problem? How about immigrants from India, who also vote heavily Democratic?

    As far as low trust, how would you describe the Italians of the 1890-1910 period, mostly from southern Italy?

  45. Gravatar of Justin Justin
    11. November 2015 at 15:11

    Italian immigration was somewhat problematic. See the documentary film “Goodfellas”. New England Yankees weren’t going around robbing trucks and bootleging. It took a few generations of out-breeding, mostly with Irish, to water down their clannishness. Not the end of the world, but it imposed costs on Americans at the time. You could argue that Rhode Island is still dysfunctional because of that migration wave. In any case the numbers were smaller, Italy is culturally closer and isn’t attached to the U.S., plus we had a 40 year moratorium to allow for intermarriage/assimilation.

    I oppose any immigration policy resulting in political changes which threaten my way of life. I think this is reasonable.

  46. Gravatar of Justin Justin
    11. November 2015 at 15:17

    The optimal immigration policy: open borders for females, aged 19-23, from the former Soviet Union. We have an obligation to protect them from Putin’s patriarchy and homophobia &c.

Leave a Reply