A recent Paul Krugman tweet got me thinking about life expectancy in the US:
I’m pretty sure those facts are not accurate, as average life expectancy (in 2016) in Trump states is much higher than 76. The source he cites gives quite different data, if my math is accurate:
Nonetheless, it does seem to be the case that life expectancy is higher in Clinton states, and has been rising more rapidly in Clinton states.
Krugman rightly criticizes conservative theories that try to explain this gap:
Conservative figures like William Barr, the attorney general, look at rising mortality in America and attribute it to the collapse of traditional values — a collapse they attribute, in turn, to the evil machinations of “militant secularists.” The secularist assault on traditional values, Barr claims, lies behind “soaring suicide rates,” rising violence and “a deadly drug epidemic.”
But European nations, which are far more secularist than we are, haven’t seen a comparable rise in deaths of despair and an American-style decline in life expectancy. And even within America these evils are concentrated in states that voted for Trump, and have largely bypassed the more secular blue states.
So something bad is definitely happening to American society. But the conservative diagnosis of that problem is wrong — dead wrong.
But progressives also lack simple explanations:
What explains the divergence? Public policy certainly plays some role, especially in recent years, as blue states expanded Medicaid and drastically reduced the number of uninsured, while most red states didn’t. The growing gap in educational levels has also surely played a role: Better-educated people tend to be healthier than the less educated.
Beyond that, there has been a striking divergence in behavior and lifestyle that must be affecting mortality. For example, the prevalence of obesity has soared all across America since 1990, but obesity rates are significantly higher in red states.
When you look at individual states, some interesting trends emerge. In 1990, life expectancy by state was tightly clumped between 73.1 in Mississippi and 78.5 in Hawaii. Then there was Washington DC (68.4) looking like a third world country. Suppose that in 1990 you had predicted that life expectancy in DC would be in the middle of the pack by 2016. I’d guess that people would have laughed at you.
If you look at cost of living-adjusted poverty rates, you’ll find two places that are much worse than anywhere else in America, DC and California. And yet California has the second highest life expectancy in the US, trailing only Hawaii. Progressives might point to California’s spending on health care, but I know of no evidence that that impacts life expectancy. More likely, race and lifestyle play a role. California has lots of Asians and Hispanics, who tend to live much longer than whites and blacks. And the whites that do live in California tend to have healthier lifestyles than those who live in Kentucky or West Virginia.
If America wishes to boost its life expectancy to European levels, the most effective way of doing so is to bring in 100 million immigrants from Asia and Latin America, not to spend more money on health care. We already spend far more than European nations—even our government health care spending equals or exceeds Europe’s (in per capita terms.)
Contrary to the claims of President Trump, immigration is the only way to make America great again. Mass immigration combined with China’s plunging birth rate would allow the US to retake the crown of the world’s biggest economy. It would significantly boost life expectancy. It would assure Silicon Valley’s continued dominance of the global information economy. It would help to secure America’s dominance of global arts and popular culture.
Nativism will make America look more like Kentucky, Mississippi and West Virginia, while mass immigration will make America look more like California, Texas, Florida and New York.
Immigrants built New York City. Both of them.
READER COMMENTS
Mark Z
Dec 4 2019 at 2:24pm
I can’t help but note Krugman’s lack of a citation for the claim that medicaid plays a significant role in the gap. And it’s bizarre that he refers to Europe’s higher life expectancy without noticing the obvious issue that Americans consume far more healthcare per capital than Europeans and yet die significantly younger, which points away from ‘access to healthcare’ as an explanation of the difference. It should really be a matter of consensus by now that behaviors like smoking, diet, drinking, drug use, driving, etc. are driving the gap, not the healthcare system.
I’m not sure what the policy implications are. Maybe people are rational and self interested and they don’t see living longer as worth giving up certain pleasures. Maybe they’re irrational and we should impose paternalistic laws against drinking, smoking, drugs, and eating junk food. Either way, building more hospitals is unlikely to make mich difference.
Christophe Biocca
Dec 4 2019 at 2:27pm
Isn’t this the arithmetic fallacy? The common form of that fallacy is “immigration lowers GDP per capita” but this seems just as wrong of an argument, and for the same reason.
Scott Sumner
Dec 5 2019 at 9:19am
I doubt it, unless immigration lowered life expectancy of the native born, which seems quite a stretch.
Christophe Biocca
Dec 5 2019 at 10:44am
The point is that looking at what happens to the average of some metric when you’re adding new people to the population you’re measuring tells you very little about the actual effect on that metric in either the original members or the new ones.
The lower-GDP-per-capita case is especially egregious because we have good evidence that both groups are better off.
For life expectancy, the most likely effect is essentially zero impact on the existing population (if, as seems likely, we’ve reached the point where additional wealth doesn’t seem to help outcomes), and a mixed effect on the new arrivals (improved wealth will help life expectancy for those that move from much poorer countries, but some will start adopting a less-healthy lifestyle as they acculturate), so pointing to the large increase in the average is misleading.
Quite Likely
Dec 4 2019 at 2:57pm
Haha is this real? The way to increase life expectancy is to encourage long lived people to immigrate? Seems like you’re focusing a little too much on the statistic rather than the reason people care about it – people living in the US now want to live longer, not just to live in a country where other people live longer.
robc
Dec 4 2019 at 3:08pm
Do people in this country want to live longer?
I don’t see the evidence of that from their behavior.
IVV
Dec 4 2019 at 3:37pm
If you want to live longer, watch what you eat, get more physical activity, don’t smoke or do hard drugs, hang out with people you like and who like you (loved ones are particularly helpful), and get screened regularly to catch any disease that can sneak up on you.
If you want the government to pass a law that makes you live longer, well, that’s not going to happen.
Lorenzo from Oz
Dec 4 2019 at 6:52pm
How many civil wars do you want? I ask this seriously.
Migration was central to the US dissolving into civil war in 1861. It destroyed the previous Party system, it convinced the South that it would become a permanent minority to a majority determined to destroy the basis of its economy. (See Robert Fogel’s “Without Consent or Contract”.)
The migration of Jews into Palestine, despite being a clear positive for the economy of Palestine, including raising Arab wages, triggered civil strife, followed by a series of wars. Migration of Palestinians into Jordan and Lebanon triggered a short civil war in the former and a long civil war in the latter.
Migration stresses societies, though careful management can minimise the stress. Canada and Australia manage their migration to minimise the stress. The US and Europe do not. Note, the higher migration countries do better at minimising the stress than the lower migration countries. But that is what enables them to be higher migration countries.
The life expectancy problem in the US is, in large part, a “white” male working class life expectancy problem. Migration will do nothing to help that. On the contrary, it is likely to exacerbate it. Then there is the homicide problem in inner cities. Also, not something that will be fixed by migration.
Migration may massage the numbers, but will not fix the underlying problems, may make them worse and will tend to exacerbate current political polarisation. Which, it is perfectly reasonable to argue, migration is substantially driving.
Scott Sumner
Dec 5 2019 at 9:21am
The Civil War was caused by slavery, not migration.
Lorenzo from Oz
Dec 14 2019 at 6:47pm
The American Civil War was over slavery, but why did slavery become such a salient issue? It hadn’t been in the first decades of the American Republican.
Migration fundamentally changed the balance between free and slave States. Migration created a massive surge in nativism which had to be managed. Which the new Republican Party did by a policy of protection and resistance to “the Slave Power”. Migration stressed the American republic in a way which made a pre-existing fault line much more salient.
How about Palestine? Jewish migration raised economic activity, including local wages (because they brought in so much capital). Yet Muslims could not tolerate Jews as equals and the economic changes undermined the existing social hierarchies. We all know where that led. Palestinian migration to Jordan and Lebanon then stressed those societies, leading to a brief civil conflict in the former and a prolonged civil war in the latter. Migration stresses societies: this is obviously true, we can see it all around us. Managing the stress intelligently manages to increase the ability to absorb migrants.
Henri Hein
Dec 4 2019 at 6:57pm
Over on Landsburg’s post, Krugman’s post also came up. Due to the paywall, I cannot read Krugman’s post in its entirety, but nobody.really quoted a passage I had a big problem with:
The problem here is that the populations in Texas and Florida has changed dramatically since 1990. The native and foreign-born populations are growing rapidly in both states. The population in New York is more static. California is a bit in between, but both its native and foreign-born populations are growing slower, compared to TX and FL. So the demographics in those states changed to the point of making snapshot comparisons like that meaningless.
I agree that changing the demographic makeup of the US could improve life expectancy, but I also wonder if the increased immigration rates account for some of the differences in health outcomes that are often cited. European countries and Japan in particular tend to have more static populations, and the US is often compared to those.
I should clarify that I support open borders. I just caution against timeline comparisons of populations with changing demographics.
Andre
Dec 4 2019 at 9:03pm
It’s virtually all diet, people.
Hispanics, despite near-bottom position income-wise, have higher life expectancy than “Clinton states” and higher life expectancy than whites.
It’s diet. Diet. Diet. Diet.
Joseph Hertzlinger
Dec 5 2019 at 3:47pm
So… If we eat Mexican or Chinese food…
Mark Bahner
Dec 6 2019 at 12:51am
Yes, if we want U.S. life expectancy to increase, veganism combined with no alcohol or tobacco is the answer:
https://prime.peta.org/2012/11/longer/
ChrisA
Dec 6 2019 at 2:37am
Mark – Peta might just not be the most unbiased source on the benefits of veganism!
My view is that despite all kinds of motivated reasoning and cherry picked data, if you cannot conclusively prove that red meat or alcohol are deleterious to health, then they probably are not, or at least not worth worrying about. Another factor in favor of meat eating and alcohol is that they have been part of our diet for ever. So any obesity/health risk that has emerged in the last few decades probably isn’t due to consumption of red meat or alcohol. But high loads of refined sugar have not been in our diet except recently, so the likely cause is there for any new health issue.
Alan Goldhammer
Dec 5 2019 at 8:55am
Living close to DC in Bethesda, I well remember the late 1980s. The murder rate, largely a result of the ‘drug wars,’ was very high in DC. Gentrification had not yet taken hold and there were vast areas of downtown DC that had not begun to be re-developed. Since 1990, the fabric of downtown has change markedly and “immigration” into the District has increased significantly with new housing developments. The murder rate is down, population has increased, and life expectancy is up.
robc
Dec 5 2019 at 9:33am
I would guess the 1980s murder rate was especially high in the 15-24 age range, which would have a big effect on the life expectency.
Alan Goldhammer
Dec 5 2019 at 10:59am
Yes, that was the age range for most of the homicides in DC. If one looks at opioid deaths the age range might be a little higher. I put little faith in studies of life expectancy unless one corrects for things such as homicides and drug abuse.
robc
Dec 5 2019 at 12:49pm
Absolutely. Also, how deaths at birth are handled.
stoneybatter
Dec 5 2019 at 10:35am
You argue that Hispanics and Asians live longer than whites, and say that demographics are more important than health care.
How do you square this fact with the separate fact that European countries have better life expectancy but have more white people and fewer Asians and Hispanics?
robc
Dec 5 2019 at 12:59pm
Native Americans: 75.06 years
African Americans: 75.54 years
White Americans: 79.12 years
Hispanic Americans: 82.89 years
Asian Americans: 86.67 years
White Americans are below Western Europe, but not too much.
On the other hand, African Americans, Hispanic Americans, and Asian Americans are outliving the counterparts in their home countries by pretty significant amounts.
The problem becomes much smaller (but does still exist, at least for White Americans) once you remove the mix elements. I bet if you also adjust for diet and homicides and miles driven, the US problem goes away, or becomes a positive.
Jens
Dec 5 2019 at 10:48am
True words Scott Sumner. Libertarian commitment for open borders is the main reason why I keep returning to this blog and listen to all other ideas (with which I often disagree though ^^)
robc
Dec 5 2019 at 12:50pm
The other ideas follow from the same principles that lead to open borders.
Random Critical Analysis
Dec 5 2019 at 11:01am
Although I haven’t exactly analyzed this in the “red” vs “blue” state dimension, you might find my analysis of US life expectancy to be useful. I will say that (1) people are not randomly sorted into states (2) there are strong spatial patterns that aren’t apt to be well explained by which party is currently in control and (3) these patterns correlate quite with the known settlement and migration of American settlers many generations ago.
Self-professed ancestry on the census today (an imperfect proxy is a rather strong predictor of white mortality rates when aggregated at the state level, county-level (ancestry quantiles), and amongst individual counties. These peoples’ ancestors tended to come from the periphery of the current day UK (Scotland, N. Ireland, Wales, and more northerly parts of England), which still substantially lag more southerly parts of England (rather, albeit reversed, strong spatial gradient). I won’t say this is necessarily genetic (arguably subtle deep roots/cultural factors), but this self-identified (“American”) ancestry correlates rather well with UK/Irish genetic ancestry from consumer DNA tests since this group supplied the bulk of UK migrants early on and they apparently had a high TFR. So it’s not very surprising UK genetic ancestry correlates with mortality rates at the state level (though this is mediated by self-professed ancestry and tends to align better with the settlement patterns). These strong spatial patterns and the predictive power of ancestry also substantially survive controls for things like income, educational attainment, insurance status, and the like (despite the fact that some of these covariates are apt to be somewhat downstream of ancestry).
Although I do argue that affluence per se is increasingly a risk factor (-> obesity, drugs, etc), these ancestral components (higher conditional obesity rate, homicide, etc) can likely explain much of the overall US residual and, especially, the strong intranational clustering.
[P.S., Most links supplied from an as-of-yet-to-be-published post of mine. Also, for what it’s worth, at least some of my ancestors came from relatively peripheral regions of the UK, i.e., I’m not trying to cast aspersions or paint with an overly broad brush!]
nobody.really
Dec 5 2019 at 11:17am
Look, I’m as pro-immigrant as anyone–but I strongly suspect that construction of New York City required the efforts of more than two immigrants. Let’s not go overboard here.
Scott Sumner
Dec 7 2019 at 1:59pm
I meant both NYCs, not both immigrants.
nobody.really
Dec 9 2019 at 6:19pm
I know–it’s a joke!
Floccina
Dec 5 2019 at 5:17pm
Sadly, we no loner let people build enough.
Mark
Dec 5 2019 at 6:41pm
Good point about Hispanics and Asians, I hadn’t thought about that when I saw Krugman’s column. The county level map of life expectancy does seem pretty well-correlated with Hispanic and Asian population. It’d be interesting to examine why that is—if it’s lifestyle-related then having higher-life expectancy immigrant groups around could increase life expectancy for natives too.
Another factor that might be at play is urban design. I’ve noticed that a lot of our more liberal cities have gotten more pedestrian and bicycle friendly in the last ten years, whereas cities in red states have lot. Physical activity is a huge factor in health and life expectancy.
ChrisA
Dec 6 2019 at 2:50am
Hispanics tend to work in more labor intensive sectors is my observation, we know there is a strong link between exercise and longevity so perhaps this is why they have better longevity. Of course longevity goes down if you work too much but coming from a rural area, it is amazing how fit the older people are who did manual labor all their lives, and are still able to work a full day in the fields even into their 70’s or even 80’s.
On Asian people, their cuisine must be a big part of their longevity, but I am guessing there are some genetic factors as well. It is a cliche but the Chinese and Japanese don’t see to age as fast as whites despite working hard.
So I don’t know if these factors (except maybe the food) will really impact white americans longevity. I can’t see white americans taking up manual labor again, and they won’t have the same genes as Asians. That all said, a society with lots more older people will probably be a society that is better for older people regardless of ethnicity, in such things as services and medical care.
Mark Brophy
Dec 6 2019 at 4:53am
Asians live longer because they eat less wheat, sugar, and dairy than white Europeans. The second best tennis player in the world, Novak Djokovic, wrote a book about why giving up pizza greatly improved his tennis game. Asians could live even longer if they gave up rice and replaced it with meat, fruit, and vegetables. If you want to maximize your life span, do whatever Roger Federer does.
robc
Dec 6 2019 at 6:28am
Have two sets of twins? I dont see that as conducive to a long life.
Michael Rulle
Dec 6 2019 at 10:59am
Obviously these statistics would be the same had Trump not even run for President. Also, there must be be better studies on the causes of higher life expectancy rather than what States people live in. Nor can I determine what PK’s point was—-other than his need to assign bad outcomes to Trump.
One also must assume Scott was being sarcastic regarding immigration increasing life expectancy for US—-as that is a non-sequitur. I also was not aware Barr invoked lack of belief in God —or “militant secularism”—as a cause of lower life expectancy.
I do agree more immigration would make America better——as long as it is correlated with people believing in America as a place of fair and open equality of opportunity—-which in my experience tends to be the case among the overwhelming majority of immigrants.
While Trump seems to not get this, unfortunately, were he not on the scene at all it is clear that not enough Americans of either party get this either. I have always believed that the desire of others to want to come here is our greatest “comparative advantage” —-in the spirit of the thinking of Julian Simon.
Krugman is a weird dude.
Comments are closed.