Overall health inequality seems to be down

The haves are those who enjoy great health into their 90s. The have-nots are those who suffer from serious health problems and do not live to see adulthood. As we pointed out in a recent study, among those Americans who were born in 1975, the unluckiest 1 percent died in infancy, while the luckiest 1 percent can expect to live to age 105 or longer. Now let’s fast forward to those born in 2012. The bottom percentile of this cohort can expect to survive until age 18. At the other end of the spectrum, the luckiest 1 percent can expect to live to age 108. That’s a much bigger gain in life expectancy among the have-nots than among the haves. Of course, life expectancy is but one measure of health and well-being, but understanding these trends offers a more complete picture than considering income alone.

These findings run counter to headlines noting a widening gap in health outcomes between different demographic groups. For example, a study led by Jay Olshansky of the University of Illinois at Chicago recently demonstrated that the gap in life expectancy between less educated and more educated Americans has widened considerably.

While studies like these are valuable in highlighting disparities between socio-economic groups, they do not tell us much about overall health inequality. That’s because most health inequality occurs within groups. In other words, if we look at a particular demographic group, the best outcomes for people in that group are dramatically different from the worst outcomes for people in the same group. These differences overwhelm any differences in average life expectancy across demographic groups. Thus, while inequality across some demographic groups has increased, it has fallen over the entire population. Overall, therefore, the health have-nots have made progress in catching up to the health haves.

That is from Benjamin Ho and Sita Nataraj Slavov.  I am open to counters on the data side, but so far this seems both a) true and b) rooftop-worthy.  I am reminded of Arnold Kling’s three axes of ideology; perhaps health care inequality attracts attention only when the victims are a group (the poor) who are part of some other narrative of oppression.


Why care about inequality? You gain nothing if someone else loses health, even if this reduces inequality.

The healthiest 1% have 90% of the health!

Health equality may be down but the fact remains that if you are one of the haves you can afford more healthy food choices. This is an important factor to a healthy lifestyle until healthy food prices are lowered in line with processed foods I cannot see and end to health in-equality.

I think Jesus said "The cheap food we will always have with us."

Why would you think processed food is unhealthy?

Clean water is much, much more important in terms of public health.

The poor are now advised toward the inexpensive frozen vegetables, which might even be more nutritious than week old organics at the high end market.

What exactly is your definition of healthy food? Rice, beans, frozen vegetables, canned tomatoes, whole chickens, chicken thighs, pasta,.. these are some of the things that I purchase on a regular basis that are both more healthy and less expensive than fast food or processed packaged foods that you find in the supermarket.

This is one myth that really refuses to die.

Quiet, you. The notion that people make poor lifestyle decisions doesn't let me play moral high ground and call for more wealth tranfers to assuage my guilt.

Eh, I'm not even sure that I buy that aspect of the meme either. There are lots of well-off Americans driving their SUVs through the McDonald's drive-thru, while the best place to get inexpensive produce near me is the market that mostly services less well-off Latin immigrants.

Pasta? Rice?? OMG, those aren't ketonic!!!11one

@Annie Richardson

Have you ever priced carrots, or bananas and they do not even need to be cooked. How about collard greens, have you ever priced collard greens? Whole wheat bread (or whole wheat four - tortillas are very easy to make), oats for oatmeal? Beats? Cabbage?

I could go on but I am now sure that you where just kidding.

The only place near where I live that I could buy most of those things is the organic market, and it would definitely be more expensive than Wendy's.

You can't find carrots, bananas, whole wheat bread, and cabbages at any cheaper store near you? Why?

Inequality is just a metric. Directly it documents the spread. You can double check it with conditions at the bottom. If they are truly good, you are all good. Of course, if conditions at the bottom are bad, a compassionate sort might pause to consider possible responses. A non-Scrooge, that is. Or a post visitation Scrooge.

Roof-top worthy? Seriously?

The fact that some kids are born with a genetic defect or that even rich kids can suffer an accident is used to somehow hide or minimise and otherwise poo-poo the fact that it was "demonstrated that the gap in life expectancy between less educated and more educated Americans has widened considerably".

So, sure, the people among us with the worse genetics and the baddest luck got a sweet deal compared to earlier decades. That's probably even truer when compared with the Dark Ages when only the strongest/luckiest would pass the first five years. That's just technological progress for you and I fully expect it to continue. It's not on cruise control but almost.

OTOH, the sociological aspects are the ones we got the most influence on and they're NOT on cruise control at all. They've been going sharply in reverse.

One way to see that is that, in France, a poor woman has more chances to live longer than a rich man. Genetics trump wealth. In the USA, the rich man got more chances than the poor woman. Wealth trumps genetics.

So all we've got to do to make you happy is kill rich men? Why should genetics be considered a more legitimate determinant of life expectancy than wealth? People certainly don't CHOOSE to be born with particular genes, yet you insist those genes should determine when they should die?

Hi, eggo

I never said we should kill rich men to bring them back in line with the rest of us. That'd be stupid. I'd rather level upward, inasmuch as possible.

As to why genetics should be a more legitimate determinant? Err... because we have no real control over genetics (I'd be against an eugenic program and I'd be wary to the extreme of human genetic manipulation) while we do have a strong control over our societal choices?

NB: I don't insist that genes should determine who dies when, I just observe that they play a crucial role and that it's quite outside our ability to control. Hopefully, great healthcare and progress will make good genes/luck less and less relevant - As per the study quoted, now even the unlucky have better probability of surviving.

Don't we have quite a bit of ability to affect the role that genes play? Eg insulin treatment for type I diabetes, kids with cystic fibrosis now surviving into their 30s and 40s, etc?
Meanwhile there are environmental causes of ill-health and death that are quite outside our ability to control. Eg the common cold.

I didn't realise that insulin treatment intervene at the genetic level and 'cure' the genetic issue of diabetes? What I was referring to was outright genetic manipulation to prevent the genetic expression or 'cure' diabetes (it seems gene therapy cured diabetes in mice and dogs...).

IMO, insulin treatment is an example of the kind of treatment that, as I said, contributes to make 'bad' genes less relevant/less crucial to health outcomes (I use commas because genetic variety is actually important, in and of itself, if I understand the science of evolution more or less correctly) i.e. "Hopefully, great healthcare and progress will make good genes/luck less and less relevant – As per the study quoted, now even the unlucky have better probability of surviving".

Why are you talking about curing now? You were talking about dying before, to quote you:
"I don’t insist that genes should determine who dies when, I just observe that they play a crucial role and that it’s quite outside our ability to control."

But, someone with Type I diabetes will die if they don't get the right amounts of insulin, and can live quite healthily if they do, and that strikes me as being about as controllable as, for example treating a broken leg.

For that matter, as you now want to talk about curing, we can't cure a broken leg either, we can only support it being healed more strongly, by say making sure that the bones are aligned.

You think that technological progress is almost on cruise control? And that we have more influence over sociological aspects than over technological progress? Interesting views.

Firstly, how are you separating sociological aspects from technological progress? Improvements in medical care come about not just from technology, but from using that technology. The fact that Flemning discovered antibiotics by itself doesn't save the life of someone with a bacterial infection, not without those antibiotics being produced on a massive scale, and distributed, and prescribed correctly, and obtained from the pharmacist, and actually taken. Lots of people in poor countries die from a lack of medical care, even though the technology is available somewhere in the world.

Secondly, how do we have more control over the distribution of wealth than technological progress? Technological progress is supported by property rights which we can do something about. Property rights not just in the sense of patents, but in the sense that it's cheap to manufacture the equipment needed for medical treatments, such as stainless steel for surgical instruments, lenses for microscopes to examine blood samples, paper and ink for spreading news about new medical discoveries, electricity to run the monitoring machines used in intensive care, etc. There are plenty of poor countries where needles are reused and surgeries are dirty. Meanwhile, how many countries manage massive shifts in the distribution of wealth in a non-destructive way? (Civil wars/cultural revolutions/hyper-inflation do shift the distribution of wealth in various ways, but not very nicely).

I'd be interested in how you explain your assertions, and what evidence you draw on for them?

Hi, Tracy

Yes, you're describing the magic of a market economy in which property rights more or less work and institutions can cope. It is wondrous but I think that, notwithstanding all the things that need to be changed (imho) and made better (still in imho), it is more or less on autopilot.

Don't we get 2% productivity gains, more or less, year in, year out? This is one of the thing that makes me doubt TGS, regardless of how much I like this blog.

Wealth distribution is entirely in our hands. You cite the case of people dying while the technology to save them is available. Yes. They're dying because they can't afford that technology. This is an issue mostly of wealth distribution (and prioritization of finite resources, maybe) not of creating distribution networks. We certainly managed to sell that technology to emerging markets customers when they became capable of paying for it i.e. as usual it's a problem of getting demand to be solvent...

I find it hard to believe that North Korea or Zimbabwae is getting 2% annual productivity growth from autopilot.

And if wealth distribution is entirely in our hands, then there should be plenty of cases of countries distributing wealth as they wanted, without winding up poor. I can't think of a single case. Can you?

So, sure, the people among us with the worse genetics and the baddest luck got a sweet deal compared to earlier decades

What's 'roof-top worthy' is that this effect is so much powerful than the weak, countervailing widening between rich and poor -- such that the overall gaps between best and worst lifespans has narrowed considerably. And that is what matters most. Prevention of early childhood deaths is *surely* worth much more effort than shrinking a blue-collar/white-collar gap that plays out at chiefly middle age and later. This is especially true since both rich and poor are living longer:

"In 1980-82, Dr. Singh said, people in the most affluent group could expect to live 2.8 years longer than people in the most deprived group (75.8 versus 73 years). By 1998-2000, the difference in life expectancy had increased to 4.5 years (79.2 versus 74.7 years), and it continues to grow, he said."


I don't know. I find the fact that we prevent more childhood deaths now than yesteryear to be entirely unremarkable. It's been like that for a good few centuries.


However your quote by Dr. Singh goes totally the opposite direction. It describes a widening gap between the affluent and the deprived. Furthermore, what really gets me going is that it is entirely preventable. While I am not sure we can do much to hasten the fall of the infancy death rates in industrial countries.

Let's play genetics with the rich and the poor too--the same qualities of high IQ and low time preference, which are highly heritable traits, will lead to both socioeconomic success and good lifestyle decisions.

This conclusion seems to be counter-intuitive. given the fact that healthcare costs have been moving up, up and up for the last two decades. This chart gives the complete story:


Our understanding is that income is probably the wrong variable to base this on. Some other variable is key here, or maybe it is just overall progress and the income argument is just noise?

Healthcare is probably only obliquely related to health, at least past a certain low threshold.

Of course, there is also this perspective -

With increasing life expectancy in the U.S., it is important to know whether a longer life expectancy means a longer healthy life span or a prolonged period of later-life morbidity. This study examines changes in lifetime without diabetes, a leading cause of morbidity in later life.

Using demographic methods and nationally representative data, we estimated changes in diabetes-free life expectancy between 1980–1989 and 2000–2004 for adult men and women in the U.S., estimated the contribution of changes in age-specific diabetes rates, and examined the changing effects of weight status on diabetes risks.

While life expectancy at age 18 for men and women increased between the 1980s and the 2000s, diabetes-free life expectancy at age 18 decreased by 1.7 years for men and 1.5 years for women. The proportion of 18-year-olds who would develop diabetes in their lifetimes increased by almost 50% among women and almost doubled among men. Obese individuals experienced the greatest losses in diabetes-free life expectancy during this period, estimated at 5.6 years for men and 2.5 years for women.

Diabetes-free life expectancy decreased for both men and women between 1980–1989 and 2000–2004, and these decreases are almost entirely attributable to large increases in diabetes incidence among obese individuals.'


'...perhaps health care inequality attracts attention only when the victims are a group (the poor) who are part of some other narrative of oppression.'

Well, I wonder what group the obese/diabetic belong to - and what their narrative of oppression looks like.

Of course, it is also true that the oncoming wave of demographic problems which is bearing down on American society due to obesity seems to be not exactly a burning topic. Well, except for when a Republican 'fascist' (Instapundit cite, by the way) mayor tries to do something trivial about it.

Bloomberg was only a Republican for six years.

Political parties don't cull much.

So, a former Republican has turned into a fascist, according to an Instapundit cite? Well, that explains why Instapundit is always saying 'see, they told me if I voted for Republicans, they would turn out to be fascists later. And they were right.'

Actually, he's never said that.

Bloomberg was a Democrat for decades. In 2001 he ran in the Republican primary to avoid running in the Democratic primary for mayor of NY. Since then he's switched to being an Independent.

However, this line of trolling is better than typical, so congrats on improving your game.

1. Doctors used I higher blood sugar level to define type II diabetes in the past.
2. Hispanics have a higher rate if type II diabetes and are now a larger percent of the population.

I was trying to mentally compare this to wealth inequality and it seems that life expectancies still imply a lot of inequality. An increase at the 1st percentile to 18 years from 0 years of life expectancy is huge. And yet, the remaining difference of 100 years of life between the very bottom and the very top...wow that's a lot too. I would want to see some weighting for quality of life in those years of life but still that's an enormous difference in life experience and flow utility. One may not need money to happy but I suspect being alive is rather important. Of course, one should celebrate the progress they report, while keeping an eye on the inequality that remain. Also on the post, not every presentation of data needs to be ideologically interpretted, right?

Not sure I trust projected life expectancy for people born in 2012.

"Not sure I trust projected life expectancy for people born in 2012." - this is a huge understatement. The healthiest 1% is supposed to expire around 2120! And we all know what medicine is going to be like well into the 22nd century, right: crowded waiting rooms with people leafing through old magazines, that kind of stuff. What a joke...

I think all we as a country can do is try to provide access to healthcare to those who don't have it and let the chips fall where they may. Genetics, diet, and lifestyle (smokers, addicts, thrill-seekers, those with exposure to firearms) have as much if not more to do with health outcomes as access to healthcare.

Well, the government could suck less and put criminals in jail. Even that pro-gun-control paper we reviewed concluded that it is when existing criminals sub out guns in place of knives that gun crime happens. Turns out people really do kill people.

That would be a classic "perfect as enemy of the good" argument, wouldn't it?

Accept no change that does not fix everything.

Ummm, no. DS says we can't do much. I say we can do something on the artificial sub-category of 'gun violence.' Imprison the violent.

BTW, I'm not sure what you mean, but if you mean "we should take guns away from everyone because we can't hope to isolate proven violent criminals" I'd consider that an odd example of 'perfect being the enemy of the good.' What I'm saying is government should demonstrate they can do something right (like their core responsibility) before they start dictating our soft drink size. And back to guns, stop-and-frisk doesn't count. That's not pareto improving when it puts all the burden on freedom. Has anyone determined if they are just pushing the criminals to the edge of enforcement? Where is the study? It may simply be an expensive cost-shift. That's the null hypothesis.

The world is full of examples of gun control which fall far short of "take them all away," and which show good outcomes. They are similar to national health strategies in that regard.

I am now a moderate and independent, but I used to identify as conservative or libertarian. One of the things that drove me away was the broken logic in this cycle: "We in America are number one!" "But we have worse outcomes!"and "That's because (excuse)."

In health care and guns we have worse outcomes, but we are the best, insert excuse (racial for extra credit).

Well, we miss you dearly, as you can tell.

I think moderates are just blind. You see the UK's "good" outcomes on guns and healthcare but they aren't there. You just get less healthcare and riots but you don't tally those correctly.

I see that Forbes, The Capitalist Tool, ranks US health 11th in a survey of nations ... but I'm sure you have an excuse ready.

An excuse for what?

'Turns out people really do kill people.'

Especially people killing themselves, as that is the leading category of gun caused fatalities in the U.S.

And thus should not be counted.

Right. We should not think "preventable death," we should think "martyrs for our freedom."

So teens in gun owning families kill themselves at a higher rate, that is just part of this being "the best of all possible worlds."

Noone counts suicide as a violent crime until they use a gun. It's dumb. You are projecting.

Is suicide a 'health' issue in the way most people construe the notion of 'health'? No. Do guns cause suicides?

(Duggan, 2002)
Nearly 60% of suicides are committed with a gun. Previous research suggests that
individuals who use a firearm when attempting to commit suicide are more likely to be

Here again, what you have is the substitution effect just as with the evidence for criminals. Still, 40% of suicides (almost half) are NOT committed with a gun. So, address the suicidal.

The teen data show that non-purchasers die at a higher rate when a gun is near. Substitution is not actually 1:1.

So what is it?

I think gun control to address suicide is stupid. I think gun control to address crime is nearly as stupid. That's because I don't have a separate category for gun crime and gun suicide because I don't consider imposing new rules on gun people as cost-free especially when these rules don't address the underlying problem.

The underlying problem of suicide and crime is not guns. Removing guns undeniably hits gun owners. It speculatively has accounting affects on suicide and violent crime, but it does not address the fundamental causes. It is an admission that government is impotent to do it.

And now that you have joined the party of science, any comments on the (Brookings) paper that says: "the correlation between guns and suicide is largely selection driven."

What says the party of reality and science about the paper by the party of reality and science?

Yes, as I say it ends with "make an excuse." We start with "we're #1" and then we find obscure papers to explain why we are not, after all. We have these guns, we have these deaths. Someone thinks they can normalize it all away. If you believe a hand-picked excuse, you too can be a modern American conservative.

Suicide is a choice. It's not something that happen so a person, it's something a person decides to do as the owner of his or her life.

If you take suicide away as a legitimate option, everyone is a prisoner on this planet just because they were born, which was never their choice.

An excuse for what?

Why would we want to make it harder for gun owners to commit suicide?

That paper was by Brookings.

So, are people killing themselves due to bad health? Anyway, suicides:

United States[21] (more info) 12.0 2009
35 Sweden 11.9 2009
36 Norway 11.9 2011
37 Ireland 11.8 2009
38 United Kingdom (more info) 23 11.8 2011[22]

Btw, I did not "hand pick" or "find some obscure paper" I googled "Do guns cause suicide?" and pulled up the first paper. It happens to be largely anti-gun and not coincidentally from Brookings, but they find the intuitively obvious answer. Sure, guns work. The facilitate successful suicides. They might even facilitate attempts because a successful method is likely to be preferable if you desire success. But largely that's the role they play. The statistical correlation is just that, a correlation (the author concludes) between the likelihood of living where there are more guns and the desire to commit suicide.

So what? The suicide rate in Japan far exceeds that in the U.S. despite strict gun laws.

Not enough criminals in jail--that is clearly our most pressing problem in the U.S.

No. Non-violent criminals in prison is the problem. And some people want to add non-violent gun owners to the list of criminalized peaceful activities.

I tend to see the suicide issue as diversionary. Someone who wants to kill him- or herself will find another way. It's not as if swallowing a fistful of pills is exactly difficult or hard to come by.

People who want to kill others will similarly kill, the difference being the numbers they can affect. A knife-wielding maniac in a movie theater is going to impact at most a handful of people before a mob tackles him to the ground and beats the living daylights out of him. An AK-47-wielding maniac will get no such resistance. With or without pistol-armed citizen heroes, no matter what the NRA wants you to think.

So yes, let's get the petty drug dealers out of the jails and replace them with violent criminals who deserve long sentences. But can we maybe agree to ban automatic weapons? Or at least assault style weapons? Anyone arguing that he or she "needs" an AR-15 ought to be checked into a pysch ward as far as I'm concerned.

I have nothing against licensed hunters who want to own and use a rifle or a shotgun, as long as background checks are performed. Or someone who feels the need to own a pistol for home security reasons, again with adequate background checks.

But assault weapons? Concealed carry permits? If you feel these are necessary, again, I posit you need to consider moving where the bogeymen can't get you and/or checking yourself into mental care.

Again, the data show otherwise. Possibly that is because humans are not Vulcans, and make emotional decisions. A gun death may visualize very differently from a pill death, and so have a different attraction.

Okay, I won't argue that, especially if data shows otherwise - I'm not exactly pro-gun.

However, my rationale behind wanting more stringent gun control laws has nothing to do with suicide prevention and everything to do with homicide prevention.

I think a person ought to have the right to extinguish him- or herself as long as nobody else is hurt in the process.

The "suicide option" argument assumes a rational consumer, yes? And a well-defined "utility?" If you think human nature is sometimes weird and irrational you might separate "impulse" from "decision." (I'm suddenly reminded of a memoir, The Doryman's Reflection, and young friends of the author, who made the decision, or had the impulse, depending on how you look at it.)

Although I do not usually just dismiss people with differing points of view, I have come to the conclusion that anyone arguing for gun control who focuses on "assault rifles" seems to always be either (a) factually deprived or (b) arguing dishonestly.

One example would be @dead serious who wonders "But can we maybe agree to ban automatic weapons?" dead serious apparently has missed the whole history of gun control including the most well-known gun control act in US history, that National Firearms Act of '34. (dead serious, please start here: http://en.wikipedia.org/wiki/National_Firearms_Act)

Any overview of crime statistics in the U.S. would come to the conclusion that assault rifles are just not a significant contributor to crime problems. Of the 12,664 homicides committed in the US in 2011, only 323 were committed with rifles of ALL sorts, which would include lever-actions, pumps, and bolt-actions, etc.
That's around 2.5%. And, for @dead serious, since '34 the number of homicides committed with automatic weapons is probably less than 10. Total.

But the real issue is that what Senator Feinstein et al term "assault rifles" are hardly any different from other rifles. (No, @dead serious, they are not automatics, which are already tightly restricted by the NFA.)

In fact, they are so similar to other rifles that in proposing her recent bill, Feinstein had to rely on an oddball list of features to distinguish them, hardly any of which had anything to do with lethality. For example, she includes adjustable (telescoping) stocks as a dangerous feature, even though all that does is enable a user to fine-tune the fit of the weapon to his or her personal reach (length of pull, etc.) Her feature list was so confusing that she had to exclude 2,258 long arms specifically just to ensure she hadn't confused the issue entirely.

This is, in fact, par for the course -- gun control advocates that talk about "assault rifles" are either ignorant of the issues, generally confused or are attempting to deliberately confuse others.

Let me be clear: I'm opposed to anything that doesn't require reloading. You should have to think each time you fire. Or at least have some hindrance in mowing down multiple targets.

I would think a hunter or someone concerned about home safety would be okay with that. And as far as I'm concerned I don't give a shit if they're not.

No the data doesn't show otherwise. Guns work. That is your argument.

Mass shootings account for about 100 deaths a year- insignificant in terms of policy.

Death by grenade is, I'm guessing close to or exactly zero per average year.

I guess you'd argue I should have the right to be carrying them around because why not.

I don't know about you guys, but I think we should let the guy who doesn't even know what an automatic weapon is dictate gun policy.

So, you're saying that the government doesn't already put enough people in jail? Are you Rip Van Winkle? Did you just sleep through the last three decades?

It's putting a lot of the wrong people in jail. The fact that they've put an assload of the wrong people in prison and had crime go down indicates that by putting the right people in prison they could reduce crime.

Color me skeptical that the bottom 1% of the health distribution averages a life expectancy of 18, when the infant mortality rate is 0.6%. Less than 0.5% of Americans die between their 1st and 35th birthday? This seems especially suspect since the paper appears to purport to use Quality Adjusted Life Years, which would only be adjusted downward, never upward.

Actually that looks plausible. According to http://www.cdc.gov/nchs/data/dvs/MortFinal2007_Worktable23r.pdf, the death rate for 1-4 year olds is 0.0286%, 5-15 year olds is 0.0153%, that of 15-24 year olds is 0.0799%, that of 25-34 year olds is 0.1049%. So just rawly summing, less than 0.25% of Americans die between their 1st and 35th birthdays.

Aren't those annual rates? I think it sums to roughly 2%.

How are they annual? It says death rate per 100,000 population in the specified group.

How has the US faired relative to other countries, over time, on these health measures, hmmm?

I was all excited by the argument so went to check the whole article for myself. And then I got all depressed. The econometrics are inexistent, the conclusions appear to be completely dependent on long-term life expectancy assumptions and rely on additional assumptions regarding QALY. And on top of that, one of the authors works at the AEI...

Seriously, Prof. Cowen, seriously...

I think there's a leap in logic from "Infant mortality is less than 1%" to "rich and poor receive comparable health services."

What an inane "study". Yes, we already knew infant mortality has fallen.

That statistic indicates a lowering of disparity between haves and have nots through infancy, childhood, & adolescence. Although it adult agegroups might be similarly affected, the statistic doesn't speak to anything on that subject.

I don't think that's necessarily true. The statistic gives scant information on change in longevity post infancy.

Mortality rates plunge for all after infancy. And after the first birthday, pre-adult mortality is driven far more by accidental death than by disease.

It may be that child and teen death rates have fallen, but that's far more likely to be driven by fewer accidents - maybe safer playgrounds or fewer teen drivers? - than by curing more disease. That's mainly because it's accidents, not diseases are the primary killers of the very young.

But the statistics cited don't give us any information along those lines.

Related, though the second bit I quote isn't health related.

Betsey Stevenson
Justin Wolfers

This research contributes to these findings by highlighting the large differences in subjective
well‐being by race. Consistent with the health literature, we also find a large unexplained racial
difference in satisfaction with health, but find, similar to our findings on overall well‐being, that this
unexplained racial difference has declined over the past 35 years.
Turning to income, we see that the well‐being gains were largest for those in the third income
quintile. Blacks in this income category had large well‐being gains, which combined with well‐being
losses among whites to completely eliminate the racial gap in well‐being of nearly half a standard
deviation. The racial gap in well‐being remains largest among those with the most income. While
blacks in the top income category became happier over time, so did those with less income. Those in
the bottom two quartiles and the top quartile all experienced a decline in the well‐being gap of .18 of a
standard deviation. Since the well‐being gap was largest for those with the most income in 1972, it
was also largest for this group in 2008.

my comment: http://bit.ly/13tlZns

What happened to the median? Describing a distribution by its first and 99th percentiles is unusual and suggests cherry-picking. Why is there no graph here?


"The haves are those who enjoy great health into their 90s. The have-nots are those who suffer from serious health problems and do not live to see adulthood."

Did they just define away poor old people?!?!

"Of course, life expectancy is but one measure of health and well-being"

And the first and last percentile do not constitute "life expectancy"!

"the unluckiest 1 percent died in infancy"

I.e. at birth. And the 0.5 percentile life span is *still* 0 years. And the second percentile has barely moved either. All this means is that the infant mortality rate has dropped below the cherry-picked 1% threshold. Therefore inequality is down!

If we compare the most extremely "unequal", the 0.1 percentile still dies at birth, yet the 99.9 percentile probably lives 5 years longer. So inequality has risen by a factor of... infinity! This is ridiculous.

For infant mortality, the right way to see if there is disparity would be to compare rich people infant mortality rate v. poor people infant mortality rate in 1975 v. 2012, which can be done accurately and easily. There is still a big gap and it is bigger in the U.S. than in most other nations.

1) Of those 1% who only make it to 18, what is the breakdown by income? Were the 1% who died in infancy in 1975 equally spread out among income groups, i.e. there was no technology to help those infants at any price?

2) Shouldnt we look at the middle groups and not just the top and bottom? How does this compare with countries that have universal healthcare? Have we advanced less or more?


A generation ago, black male life expectancies dropped due to homicide and AIDS. Now, downscale white life expectancies, especially women, seem to be declining due to obesity and, perhaps, smoking and drugs. I'm not sure how much health care policy matters in this, but life expectancies do matter.

I think it is somewhat amusing that Tyler Cowen cannot think of a good explanation for why certain types of inequality might be considered more noteworthy than other types of inequality other than an explanation involving the need to fit into a "narrative of oppression."

There are at least two kinds of inequality. Inequality that you can do something about and inequality that you cannot do anything about. Yes, I realize that this is really a matter of degree, but for simplicity, lets stick with this binary classification.

Some people are simply going to be born with an incurable disease. They may have rich parents or they may have poor parents. Either way, it is unfortunate to be them. The problem is, given current technology, you simply cannot do much about this. (Although, this is a great reason to invest more money in research and development.)

In contrast, some people may have health problems BECAUSE they do not receive treatment that is technologically feasible. I know a person with a broken finger who did not get treatment right away because they could not afford it. Well, the finger has healed improperly, and now something that could have been easily fixed with timely treatment may not be fixable.

THIS is a sort of PREVENTABLE healthcare inequality. One does not need to depend on some "oppressed group narrative" to think that someone getting treatment for something when it actually is treatable is a good idea. In contrast, healthcare inequality because treatment for a particular disease is technologically infeasible is much less tractable. There is not much you can do about it.

I mean REALLY Tylwer Cowen. You could not think of this simple explanation on your own? REALLY????

The low end improvement in life expectency from less than one year to eighteen years, is a bit of statistical trickery. It is true but misleading. The infant morality rate was slightly more than 1% in 1975 and fell to a bit below 1% in 2012. For about three-quarters of folks in the bottom 1% of mortality there is no change. This makes a moderate decrease in infant mortality (about 25% over thirty-seven years) look like a big deal and ignores intervening trends. A decrease in infant morality half as large would have stil increased bottom 1% morality from 1 year to perhaps 12 or 13 years.

The years from age one to the early teen years are the lowest morality rate years in the life cycle. Teen mortality, which is overwhelmingly from trauma, has declined in substantial part due to trends in murder rates and automobile safety, rather than access to health care. Teen mortality in the early 1990s when murder and other trauma death rates where much higher were worse than in 1975 and then only recently got better with an overall reduction in crime and rise in automobile safety. The is great news but says nothing about the health care system.

The top 1% increasing only three years merely shows that there is a peak possible life span that a lucky few have reached for thousands of years and that at least 1% of the population was already reaching that peak way back in 1975 and still do.

The strong counterevidence is that average life expectency is actually falling for poor rural women and in general for those with less education in the last few years, despite declining trauma death and improved medical technology.

Thanks for the mention. This was actually part of a much larger project, on lots of ways to think about inequality that goes against the current narrative. like happiness, or declining marginal utility adjusted, or adjusted for life time earnings versus annual, leisure adjusted, immigration adjusted, family size adjusted, government transfer adjusted, human capital adjusted, life cycle adjusted, retirement... etc.. this was just the part that seemed easiest to publish.

I can say the that the infant mortality rate is about 1.3 % for balck babies and about .6% for white ones, OR that the unluckiest 1 percent of white babies live on average more than 18 years longer the the unluckiest 1% of Black ones both of which are true but only one is rooftop-worthy

Comments for this post are closed