Pleasantly plump? Or, what if beauty is not truth?

Suppose you were at a fashionable cocktail party talking to well-educated liberal-minded people.  How would they evaluate the following three statements?

1.  Carbon emissions are leading to global warming.

2.  Second hand smoke is a significant health risk.

3.  It’s healthier to be normal weight than overweight or modestly obese.

I think most people would say all three are true.  But why?  One answer is that these people are rational, and scientific studies show these statements to be true.  But I’m not sure that’s the case.  I believe science does show the first statement is very likely true.  But the second statement is far less obvious, and the third really doesn’t have much scientific support at all.

Many years ago I discussed second hand smoke with another economist.  I pointed out that studies hadn’t found much health risk.  He said that common sense suggests it’s a risk.  After all, first hand smoke is very risky, so you’d expect second hand to be somewhat risky.  I actually like that reasoning, but I’m not sure it’s right.  Suppose non-smokers only spend 10% of their time close to smokers, and suppose smoke gets diluted rapidly in the air (isn’t there a formula involving the square or cube of the distance?)  The effects might then be trivial.

Update:  Wikipedia reports that there is a scientific consensus that second hand smoke is harmful.  Oddly, they also report the following, without comment on its accuracy:

Gio Batta Gori, a tobacco industry spokesman and consultant[124][125][126] and an expert on risk utility and scientific research, wrote in the libertarianCato Institute‘s journalRegulation that “…of the 75 published studies of ETS and lung cancer, some 70 percent did not report statistically significant differences of risk and are moot. Roughly 17 percent claim an increased risk and 13 percent imply a reduction of risk.”[127]

That was the sort of article that I recall reading.  Perhaps that information is inaccurate, given the apparent scientific consensus.  I would appreciate if someone could educate me on this question.  We all know about the major problem of confirmation bias in empirical scientific studies.  Just to be clear I find it very likely that second hand smoke produces some risks; my question was whether the effects of a colleague smoking down the hall from you are statistically significant.  Maybe, but I’m still not convinced.

Consider this recent study from the JAMA, with 3 million data points:

Higher levels of obesity associated with increased risk of death

Being overweight associated with lower risk of death

CHICAGO – In an analysis of nearly 100 studies that included approximately 3 million adults, relative to normal weight, overall obesity (combining all grades) and higher levels of obesity were both associated with a significantly higher all-cause risk of death, while overweight was associated with significantly lower all-cause mortality, according to a study in the January 2 issue of JAMA.

“Estimates of the relative mortality risks associated with normal weight, overweight, and obesity may help to inform decision making in the clinical setting,” according to background information in the article.

Katherine M. Flegal, Ph.D., of the National Center for Health Statistics, Centers for Disease Control and Prevention, Hyattsville, Md., and colleagues conducted a study to compile and summarize published analyses of body mass index (BMI) and all-cause mortality that provide hazard ratios (HRs) for standard BMI categories. For the review and meta-analysis, the researchers identified 97 studies that met inclusion criteria, which provided a combined sample size of more than 2.88 million individuals and more than 270,000 deaths. Regions of origin of participants included the United States or Canada (n = 41 studies), Europe (n = 37), Australia (n = 7), China or Taiwan (n = 4), Japan (n = 2), Brazil (n = 2), Israel (n = 2), India (n = l), and Mexico (n = l).

All-cause mortality HRs for overweight (BMI of 25-<30), obesity (BMI of ≥30), grade 1 obesity (BMI of 30-<35), and grades 2 and 3 obesity (BMI of ≥35) were calculated relative to normal weight (BMI of 18.5-<25).

The researchers found that the summary HRs indicated a 6 percent lower risk of death for overweight; a 18 percent higher risk of death for obesity (all grades); a 5 percent lower risk of death for grade 1 obesity; and a 29 percent increased risk of death for grades 2 and 3 obesity.

The authors note that the finding that grade 1 obesity was not associated with higher mortality suggests that that the excess mortality in obesity may predominantly be due to elevated mortality at higher BMI levels.

The researchers add that their findings are consistent with observations of lower mortality among overweight and moderately obese patients. “Possible explanations have included earlier presentation of heavier patients, greater likelihood of receiving optimal medical treatment, cardioprotective metabolic effects of increased body fat, and benefits of higher metabolic reserves.”

The use of predefined standard BMI groupings can facilitate between-study comparisons, the authors conclude.

That’s pretty powerful.  I can imagine alternative explanations; maybe people with alcohol, drug, or mental illness issues tend to be thinner.  But that would have to be shown.  Is there actually any evidence that the overweight or mildly obese are less healthy?  And is the distribution of people’s weight bell-shaped, implying most overweight people are in the “healthier” two categories, not the very obese?

What unites all three beliefs described above?  Coal plants putting pollution into the air are ugly.  Common sense suggests you don’t want to mess with beautiful Mother Nature.  Many consider smoking to be an ugly habit.  And normal weight people often look better than the obese.  You’d think that people who look better are in fact healthier.  I suppose people subconsciously think Keats was right, and link truth and beauty.

I was recently criticized in the comment section for suggesting that unemployment insurance might cause a very small number of Americans to remain unemployed longer than otherwise.  This is certainly not an attractive idea, given the very real suffering being endured by millions of unemployed people.  But scientific studies show that it happens to be true.

Studies also suggest that there are behavioral differences between boys and girls at age one day.  Or even one hour.  That fact doesn’t particularly bother me, but I suppose if I were a very strong feminist I might find reality to be more attractive if this were not so.

Then there is the whole issue of “natural” foods, “local” foods, and all that.

Among conservatives, global warming and evolution seem to cause discomfort.

As a libertarian, I’d prefer it were the case that we’d be better off with a government that did not regulate the environment, force people to save, redistribute income, provide school vouchers, target NGDP, etc.  And maybe we would.  But I’m not convinced, which is why I remain a moderate or pragmatic libertarian.

Mayor Bloomberg should be overjoyed if he read the JAMA study quoted above.  But I’m pretty sure he’d actually be really annoyed.

Of course there’s nothing new in this post, but it’s worth considering these biases every once and a while.

PS.  All you commenters about to say “how dare you call yourself a libertarian” should take a deep breath, and consider the length of the list of things the government is doing that I don’t think they should be doing

PPS.  Here it is . . . pig out folks!!


Tags:

 
 
 

49 Responses to “Pleasantly plump? Or, what if beauty is not truth?”

  1. Gravatar of F. Lynx Pardinus F. Lynx Pardinus
    6. January 2013 at 08:30

    You’re conflating “lower mortality rate” with “less healthy.” Is that actually the case?

  2. Gravatar of F. Lynx Pardinus F. Lynx Pardinus
    6. January 2013 at 08:32

    Sorry, reverse what I said above. “Higher mortality rate” with “less healthy.”

  3. Gravatar of JohnQ JohnQ
    6. January 2013 at 08:43

    I’m not sure who you spoke to that told you that told you a story of secondhand smoking not being a significant health risk. The World Health Organization has long concluded that it does, as does the Surgeon General and other major medical institutions. There is a great deal of statistically significant and consistent evidence that exposure to secondhand smoke leads to lung and other cancers as well as significant respiratory ailments.

    http://monographs.iarc.fr/ENG/Monographs/vol83/index.php
    http://cancercontrol.cancer.gov/tcrb/monographs/10/m10_summary.pdf

  4. Gravatar of Becky Hargrove Becky Hargrove
    6. January 2013 at 08:52

    Government can say that 140 pounds at five foot four lies in the normal range all they want…but in my experience it felt and looked absolutely horrible!

  5. Gravatar of . .
    6. January 2013 at 09:02

    Correlation does not equal causation, even when n=3 million.

    Meta-analyses do not help identify questions like this, they just provide more precise, but still biased estimates of associations.

    Sicker people tend to weigh less, hence they have lower survival. That study cannot account for this reverse causation.

    Sorry Prof, we’ve still got to stay off the pies….

  6. Gravatar of travisallison travisallison
    6. January 2013 at 09:04

    Was it a meta study of tracking studies? If not I doubt the causaion arrow. After a long illness people tend to be skinny.

  7. Gravatar of ThomasL ThomasL
    6. January 2013 at 09:31

    On smoking, Sullum’s book ‘For Your Own Good’ is worth reading.

    As regards (1), I think it depends on whether you qualify it as significant or not. Almost everyone one either side would say it is having some effect, but the magnitude is found to be negligible by one group, and catastrophic by the other.

    The average level of honesty of the scientists in the catastrophic group is so unbelievable poor, it is really a blight on the entire scientific project (or ought to be)–even though that does not prove they are wrong.

  8. Gravatar of Andrew Andrew
    6. January 2013 at 09:32

    Scott;

    If you look at the actual paper, you see that all the decrease in mortality among the slightly overweight comes from those age 65+. Obese is terrible for anyone of any age.

  9. Gravatar of kebko kebko
    6. January 2013 at 09:39

    ” And is the distribution of people’s weight bell-shaped,”—I know mine is.

  10. Gravatar of ThomasL ThomasL
    6. January 2013 at 09:43

    @me

    On the honesty, stuff like this from yesterday:

    “Gleick’s welcome back to… prominence [lecturing at the conference]… was pretty startling, given his admitted identity fraud and distribution (and probable fabrication) of a forged document.”

  11. Gravatar of ThomasL ThomasL
    6. January 2013 at 10:02

    I’m kind of surprised I’m on the cusp of “overweight” at 5’8″, wearing a 38″ jacket and a 32″ trouser. Is this chart for real?

  12. Gravatar of Kevin Dick Kevin Dick
    6. January 2013 at 10:27

    Funnily enough, even (1) is open to question. Sure, it’s trivially true that increasing CO2 will increase the temperature due to it’s UV transmission and IR absorption. But the dose-response relationship for the real atmospheric system is a subject of debate. And whether it matters to human life relative to natural fluctuations in temperature seems rather unclear when you look at at the range of historical Holocene temperatures. But not holding (1) to be true is definitely a pretty unfashionable belief.

  13. Gravatar of Tim Tim
    6. January 2013 at 10:30

    Hi scott,
    What do you think of biases people have for prices that end in .99, like 24.99 rather than 25? Or theres this one by behavioural economist Dan Ariely http://danariely.com/the-books/excerpted-from-chapter-1—the-truth-about-relativity-2/
    btw Im not a fan of behavioural economics, but behavioural economists would say that this sort of irrationality disproves efficient markets..

  14. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    6. January 2013 at 10:48

    Anyone going to be in San Diego tomorrow;

    http://delong.typepad.com/sdj/2013/01/work-and-search-in-recessions-theory-history-and-measurement-elizabeth-ballroom-f-manchester-grand-hyatt.html

    ‘Between 1985 and 2007–the period of the “Great Moderation”–the Federal Reserve and the rest of the U.S. government on the west edge and the central banks and institutions of the European Union on the east edge of the Atlantic Ocean provided a broadly stable macroeconomic environment within which private-sector businesses, workers and investors could make their economic plans. In the U.S., on an annual basis: the rate of nominal GDP growth dropped below 4% for only 3 of those years and rose above 7% for only 2 of those 22 years; the rate of consumer price inflation rose above 5% for only 3 and fell below 2% percent for only 2 of those 22 years; and the civilian adult employment-to-population ratio remained between 60% and 64% for that entire period. And Western Europe experienced a similar “Great Moderation” with low inflation, relatively smooth growth, and diminishing unemployment.’

    Other than J. Bradford being wrong about ‘1985’–it was from the end of 1982–the above is essentially what I (and others) tried pointing out to him, to no avail, years ago. Not only did he not want to hear it, he didn’t want anyone else reading his blog to do so either.

  15. Gravatar of ssumner ssumner
    6. January 2013 at 11:06

    John Q, I added an update.

    Dot, You might actually consider reading the post before commenting on it.

    Maybe the JAMA editors are idiots, but before assuming that I’d like to see studies showing normal weight people who are healthy end up living longer than modestly overweight people who are healthy. Please cite such studies before providing inane comments like “correlation doesn’t prove causation,” an issue already discussed in the post.

    Tim, Is that a joke?

  16. Gravatar of Jake Jake
    6. January 2013 at 11:06

    Scott, I’ve taken a deep breath and “how dare you” is a much harsher tone than I consider warranted, but I think “pragmatic libertarianism” is kind of a false brand of libertarianism.

    I interpret this philosophy to mean that it’s okay to use aggression, as long as overall utility is increased for some set of people. But I think most libertarians would balk at the kind of forced property redistribution that this implies. It’s probably more accurate to describe this philosophy as utilitarian, not libertarian.

    I hope I’m not coming off as too critical here, because it’s not my intent. I think you are one of the good guys, and given a choice between your policy preferences and what we currently have, I’d take yours in a heartbeat.

  17. Gravatar of ssumner ssumner
    6. January 2013 at 11:13

    Jake, Yes, I am also a utilitarian, and do favor forced redistribution of property.

    But only in moderate amounts. As I’m sure you know many of the most famous libertarians in history favored redistribution. (Friedman, Hayek, etc.) In any case, it doesn’t much matter what term is used, as long as people understand what pragmatic libertarianism means.

  18. Gravatar of Richard A. Richard A.
    6. January 2013 at 11:16

    This is Dr. Joel Fuhrman’s rebuttal to the assertion that overweight people live longer–

    http://www.diseaseproof.com/archives/news-overweight-people-live-longer-yeah-right.html

  19. Gravatar of Becky Hargrove Becky Hargrove
    6. January 2013 at 11:41

    I consider myself utilitarian as well but perhaps would word it differently, as we also have forced redistribution of knowledge property which is different from taxation of wealth, as it limits entry to knowledge use and utilization. For instance I know that second hand smoke increases my chances of getting bronchitis, but were I to set up a blog and elaborate on such “experience settings”, government might ask me to cease and desist. Whereas, studies can be funded by, say, providers of corn syrup and yet still get taken seriously.

  20. Gravatar of Jon Jon
    6. January 2013 at 11:53

    Scott, no one debates the global warming statement as you phrased it. The debate is about the magnitude of the effect. The basic physics is for 1C per doubling. A small clique claim this number should be much smaller, most support somewhat higher. The IPCC report claimed 2,5-3.5C. The bulk of the dissent favors a range of 0.7-1.8C. Everything hinges on how you model the rest of earth system. Observations show something around 1C but there are various ways in which this argued around.

    Regarding your other claims, the issue is that the standards of significance are too low in medicine. I’ve met many people who don’t grasp that if you try twenty unrelated correlations one of them will show significance against their standard.

    Until the medical profession internalizes this and enforces publication discipline, they will produce mostly crap. At a minimum they need to adopt an ethical rule against press releases touting 2 sigma results. Geez, don’t go to the public, until you can get something five sigma.

  21. Gravatar of Tim Tim
    6. January 2013 at 12:02

    Is what a joke? Theres obviously some behavioural reason why firms often price things at 24.99 instead of 25 for example. If consumers are irrational and suffer from biases, then certainly crowds can as well, and so markets can also be irrational. Thats the behavioural economics argument. My question is simply do you think these sort of biases, including the one mentioned in the link about the economist subscription fees, have any practical implications? Personally i think that despite some examples of irrationality and bias, and consumer theory models being a very overly simplified model of the real world, behavioural economics still cant just explain away the basic points of consumer theory like the income and substitution effects.
    Its also how i see the EMH. Its not 100% true all the time, but that still doesnt mean that its not useful and even though markets arent always perfectly efficient, that doesnt disprove that markets price in relevant public information, that stock prices represent the PV of all future cash flows and that you cant accurately and consistantly beat the market.

  22. Gravatar of Tim Tim
    6. January 2013 at 12:06

    Btw it can be hard to articulate an argument and write a decent comment when all youve got is a mobile phone and youre typing on it while sitting in a hostel in prague!

  23. Gravatar of Full Employment Hawk Full Employment Hawk
    6. January 2013 at 13:11

    “Gio Batta Gori, a tobacco industry spokesman and consultant”

    Consider the source. Anything that a person with such a blatent conflict of interest needs to be treated with a high degee of scepticism.

  24. Gravatar of Full Employment Hawk Full Employment Hawk
    6. January 2013 at 13:20

    “that unemployment insurance might cause a very small number of Americans to remain unemployed longer than otherwise”

    As a progressive, I agee with the proposition that A VERY SMALL NUMBER of Americans would do that. But people on the right ignore the VERY SMALL NUMBER qualification and generalize about all of the unemployed from these unrepresentative special cases. The implication of the finding is that most of the cyclically unemployed are involutarily unemployed and that therefore unemployment insurance is not only equitable, but does not SIGNIFICANTLY increase the unemployment rate.

  25. Gravatar of mbk mbk
    6. January 2013 at 16:23

    There is actually a body of evidence that suggests that perceived beauty is linked to ease of perception. This also applies to “logical” entities to be perceived, e.g., theories that are simpler = easier to understand, appear more elegant, more beautiful, more … true.

    Esthetics as processing fluency (go to pp.377):

    http://homepage.psy.utexas.edu/homepage/faculty/swann/docu/fall395pdocs/reber%20perceptual%20fluency.pdf

    What’s really “true” though is likely to be a complex amalgam of partial results and for fear of confusing a public deemed not as smart as themselves, experts are often made to push a simple “message” that just isn’t correct in its simplicity.

    Examples. On Global Warming the juiciest part is not that its reporting has been exaggerated for advocacy reasons, something that I do too believe to be true, but that its current and near term effects seem to be economically positive. This is due to an under reported but already occurring northward extension of agriculture in the Northern Hemisphere. Example,
    http://www.sciencedirect.com/science/article/pii/S1161030102000047
    Even biodiversity is set to increase, as from standard ecological theory it typically correlates positively with temperature. Similarly, the most significantly helpful impact to reduce US CO2 production came from the use of shale gas ( CH4=methane gas, is the most hydrogen rich and carbon poor of our fossil fuels) and not from Kyoto, Copenhagen, carbon taxes or any other pious measure proposed by the experts. Poor Michael Crichton who predicted just this in the early 2000s was widely ridiculed for it, and fracking is already widely demonized in Europe as I hear, which goes to show that ye shalt not deviate from simple and widely believed to be true messages, in this case, “fossil fuels bad, painful policies good”.

  26. Gravatar of bellisaurius bellisaurius
    6. January 2013 at 16:42

    The aesthetic read of science. Yep, that’s how it comes across to me too. It’s not that I don’t think there’s some science to back up the statements, but hell if they don’t feel like aesthetic value judgments as opposed to cool headed rationalistic cost/benefit decisions.

    Then again, I don’t really mind about being called callous or cold hearted if the effect is generally positive, so I’m probably the one in the wrong. Or perhaps Robin Hanson would say I’m signalling something different than those around me, and I’m just as shallow…

  27. Gravatar of Suvy Suvy
    6. January 2013 at 17:18

    I’d like to add one note about global warming and the environment. We don’t know what the consequences of our actions are when it comes to the environment. Contrary to what many liberals say, cutting down trees for something like paper really isn’t a big deal. However, something like deforestation has had unintended consequences. The problem is that we don’t know what these unintended consequences are and I highly doubt anyone can predict them. It’s the same thing that happened with SO2, CFCs, DMT(which I think is highly overblown), etc. So we’ve had issues dealing with chemicals that we’ve pumped into the atmosphere.

    Now, we’re pumping massive amounts of CO2 into the air. This is a compound that has been associated with shifts in temperature. The problem I’m worried about isn’t the shifts in the ocean’s temperature, I’m worried about all of the other things that it could have an impact on. Another thing that people really do not understand is that the impact of these things isn’t linear. So you could see no effect for a long, long time and then see a massive effect. The problem with nonlinear systems is that they are highly sensitive to the conditions, so if the conditions shift slightly, you could be screwed.

    For example, the oceans are still absorbing CO2 from the atmosphere; however, there’s a threshold where they will stop absorbing CO2 and could start releasing massive amounts of CO2. How in hell can anyone know the consequences of this stuff? There’s a massive risk here and we’re just starting to grasp the risks. There are also second and third order effects that we don’t understand and don’t know.

    Let me give another example of an ecosystem. If you take one animal(a seemingly small part of the ecosystem) out of an ecosystem, you see massive changes in the ecosystem itself. This is because there are inter-dependencies and cause and effect factors that we simply cannot know. For example, take away the main predator from an ecosystem and the entire ecosystem could fall apart due to the impact on the populations of the prey. These kinds of effects have been noted before. Now, imagine doing this with some compound that is related to global temperatures.

    To say that the environment shouldn’t be regulated reminds me of those LTCM guys that thought they understood the markets and took massive amounts of risk. Then, those idiots blew up and went completely broke. They took massive amounts of risk playing in a game that was too complicated for them and I fear we’re doing the same thing now.

  28. Gravatar of Suvy Suvy
    6. January 2013 at 17:21

    “I remain a moderate or pragmatic libertarian.”

    Too bad there are zero of these in government. That’s the real sad part. It gives us terrible policy and leaves us with idiots in charge.

  29. Gravatar of Suvy Suvy
    6. January 2013 at 17:33

    I would also like to add a comment about the way obesity and overweight are measured. They are measured by BMI which is a terrible way to measure health. For example, if someone was 5’10” 185 with 5% body fat, they would be considered overweight. However, if they were 5’10” 140 with 20% body fat, they would be considered normal. However, the overweight person is much healthier in this scenario.

    This same comment applies to the way a lot of these things are measured and calculated. I am also highly skeptical towards the way a lot of these health studies are interpreted. People have a tendency to think correlation is causation. For example, there was a study that linked eating meat to certain types of cancer. By looking at this study, you would think that eating meat might give you cancer. The problem doesn’t account for the fact that those who eat meat might also eat other things worse for health.

  30. Gravatar of Don Don
    6. January 2013 at 17:35

    I hate statements like “CO2 causes global warming”. They are stated like they are meaningful, but they are not. It is like saying, “coins lost in sofas cause global unemployment.” Both are true, but both are meaningless.

    Assume that global warming is meaningful (I don’t). If it is 50% manmade, then we are in just as much trouble if humans stop all emissions. But if we use lots of CO2 to develop a more prosperous world, we are probably better off. As of today, the big winner in the CO2 debate is Al Gore who is up about $300mil.

  31. Gravatar of Darren Darren
    6. January 2013 at 18:10

    (3) is not earth-shattering news. The new study’s conclusions have been observed for over a decade, it’s called the obesity paradox. And I would be very hesitant to draw any firm conclusions from that, I suspect that it has more to do with a statistical anomaly than actual causal relationship.

    Even though the sample size is high, confounding variables that cannot be easily sorted out can still skew the data. For example, older people who lose a lot of weight after fighting off illness who then have much higher mortality rates. There is also evidence to suggest that people who are overweight are treated more aggressively following hospitalization and thus have higher survival rates.

    Simpson’s paradox may also play a role, depending on the statistical analyses used. I’m just speculating on this but it may have to do with the fact that, simply put, people with cardiovascular disease and are normal weight outnumber those who are slight overweight but without CVD.

    We simply do not have a causal, mechanistic explanation for the obesity paradox. However, we DO know that being overweight is casually linked (as in we have more than just cross-sectional data) to atherosclerosis and other illnesses such as high blood pressure and hypercholesterolemia. That study should really come with a nice big “CORRELATION IS NOT CAUSATION” label attached.

  32. Gravatar of Darren Darren
    6. January 2013 at 18:30

    You’re also wrong on (2), there is a scientific consensus on second-hand smoke. Maybe not the guy from down the hall but I challenge you to believe that someone blowing cigarette smoke into your face day after day isn’t going to increase your risk of lung cancer, keeping in mind the risk-ratio for developing lung cancer in smoker vs. non-smokers is ~10, one of the highest risk ratios in modern medicine.

    Although I agree with your overall point, that people don’t make decisions strictly through rationality but often rely on heuristics and social factors.

    But if you believe that, then logically, you should be skeptical of EMH ;).

  33. Gravatar of Geoff Geoff
    6. January 2013 at 19:24

    “As a libertarian, I’d prefer it were the case that we’d be better off with a government that did not regulate the environment, force people to save, redistribute income, provide school vouchers, target NGDP, etc. And maybe we would. But I’m not convinced, which is why I remain a moderate or pragmatic libertarian.”

    “All you commenters about to say “how dare you call yourself a libertarian” should take a deep breath, and consider the length of the list of things the government is doing that I don’t think they should be doing”

    “Jake, Yes, I am also a utilitarian, and do favor forced redistribution of property. But only in moderate amounts.”

    It seems, and correct me is I am wrong, that “pragmatic libertarianism” is being used as a euphemism for “indefinable political tool to be used as needed, when needed, to enact one’s desired anti-libertarian looting of others”, i.e. “pragmatic socialist.”

    The justification is that zero looting is impossible. So why not join in the looting…but only modestly.

    Pragmatic socialism = Pragmatic libertarianism.

    Suppose I am a pragmatic socialist. I could say the following:

    “As a socialist, I’d prefer it were the case that we’d be better off with a government that increased regulation of the environment, force people to save, redistribute income, provide school vouchers, target NGDP, etc. And maybe we would. But I’m not convinced, which is why I remain a moderate or pragmatic socialist.”

    “All you commenters about to say “how dare you call yourself a socialist” should take a deep breath, and consider the length of the list of things the government is not doing that I do think they should be doing”

    “Jake, Yes, I am also a utilitarian, and do favor forced protection of property. But only in moderate amounts. As I’m sure you know many of the most famous socialists in history favored property. (Proudhon, etc.) In any case, it doesn’t much matter what term is used, as long as people understand what pragmatic socialist means.”

    If pure libertarianism and pure socialism are impossible, and an individual favors forced redistribution of, and forced protection of, property, then it follows that individual is both a pragmatic libertarian and pragmatic socialist, depending from which end of the (libertarianism – socialism) spectrum is the standard of measure. From the libertarian end of the spectrum, every pragmatic socialist is a pragmatic libertarian. From the socialist end of the spectrum, every pragmatic libertarian is a pragmatic socialist.

    I don’t see how

    “Pure libertarianism is impossible, so that’s why I favor some looting.”

    is any different from

    “Pure socialism is impossible, so that’s why I favor some protection of property.”

    They are two ways of saying the same thing.

    ———————

    Regarding the data that shows overweight people have a lower mortality rate, I think every person is different. Even if 99% of the world’s population became healthier if they became a little overweight, that doesn’t mean it will apply to me. I could become more unhealthy.

    The BMI is bogus anyway. Most muscular people are overweight according to the index. When I was a mega buff in my tweens, I was always considered overweight, even with only a 32 inch waist.

  34. Gravatar of Full Employment Hawk Full Employment Hawk
    6. January 2013 at 20:16

    “On Global Warming … that its current and near term effects seem to be economically positive.”

    Tell that to the people who live in the northeast, like in lower Manhatten.

    And part of the increased biodiversity is that the nasty diseases that exist in the tropical regions are spreading northward.

  35. Gravatar of Full Employment Hawk Full Employment Hawk
    6. January 2013 at 20:16

    “On Global Warming … that its current and near term effects seem to be economically positive.”

    Tell that to the people who live in the northeast, like in lower Manhatten.

    And part of the increased biodiversity is that the nasty diseases that exist in the tropical regions are spreading northward.

  36. Gravatar of Full Employment Hawk Full Employment Hawk
    6. January 2013 at 20:23

    “Being overweight associated with lower risk of death”

    I would CONJECTURE that natural selection caused humans to evolve to carry a significant store of food on their body to tide them over during periods of famine, so that being somewhat heavier than the current weight standards is actually normal and being slim is abnormal and therefore subjects the body to additional strains.

  37. Gravatar of Browsing Catharsis – 01.07.13 « Increasing Marginal Utility Browsing Catharsis – 01.07.13 « Increasing Marginal Utility
    7. January 2013 at 05:03

    […] -Scott Sumner, “Pleasantly Plump? Or, what if beauty is not truth?“ […]

  38. Gravatar of StatsGuy StatsGuy
    7. January 2013 at 05:09

    Just a quick note – it could be that all 100 studies show no effect of second hand smoke, but a well conducted meta-analysis shows massively significant effects because of the much larger implied n-size.

  39. Gravatar of Matt Matt
    7. January 2013 at 05:13

    Scott,

    What kinds of evidence exist that a “very small number” of the unemployed use unemployment to delay looking for work? I’m only asking because all of the anecdotal evidence I have seems contrary to this idea. I only talk to a limited number of demographics so I suppose that might be the reason (I mostly talk to unemployed people in lower income brackets in one region of the country).

    I’m probably not as academic as most of the commenters on your site so a link to a study or a previous post would be greatly appreciated.

  40. Gravatar of J.V. Dubois J.V. Dubois
    7. January 2013 at 05:45

    Actually I knew that bit about being overweight or slightly as an indicator of better health. I also heard second-hand stories that that being having BMI below 20 makes your insurance more costly then having one above 25, which corroborates this.

    Anyways I completely agree with your general observation that liberals have their own biases. I recall recent article by Matt Yglesias who was even better at showing this. He was wondering about such a large push against GMO food, where liberal groups used the very same tactits they so disdain if used by “climate change denialists”. So liberals definitely have their very own skeletons in their closets – besides being obsessed by fiscal stimulus and denouncing monetary stimulus.

  41. Gravatar of ssumner ssumner
    7. January 2013 at 06:36

    Richard, Isn’t that pretty much what I said?

    Becky, Good point about intellectual property.

    Jon, In the comment section of the next post I just criticized the 2 sigma cutoff, which I agree should not be viewed as “significant.”

    Tim, I agree with most of your new comment. I was wondering if you were joking in claiming that the 99 cents issue tells us anything useful about the EMH.

    FEH, I should have been more specific. I meant the current 73 weeks limit probably creates unemployment among a small number of Americans, say 1%.

    MBK, Good points.

    Darren, You said;

    “You’re also wrong on (2), there is a scientific consensus on second-hand smoke. Maybe not the guy from down the hall but I challenge you to believe that someone blowing cigarette smoke into your face day after day isn’t going to increase your risk of lung cancer, keeping in mind the risk-ratio for developing lung cancer in smoker vs. non-smokers is ~10, one of the highest risk ratios in modern medicine.”

    I agree with everything you say here except that I’m wrong on point two.

    Statsguy, Good point.

    Matt, Academic studies do show that unemployment compensation increases joblessness–I don’t think that is disputed. The most famous examples show a big spike in people finding new jobs right after their UI runs out (unusally 26 weeks.) And I have heard anecdotal cases where this was true. Of course it’s also a strong prediction of economic theory. Were it not true, we’d have to revisit the fundamental tenets of economic theory.

    Check out Casey Mulligan’s posts, he would have some links.

    JV. Yes, both side do it.

  42. Gravatar of Arthur Arthur
    7. January 2013 at 08:08

    The question of what is second hand smoking is important.

    I’ve seen some studies showing strong effects of second hand smoking, but they considered only people working in places where there’s a lot of smoking: bars, pubs, nightclubs.

    If you constantly work on a place which is always filled with cigarette smoke it seems reasonable to think you’ll have health problems. If I remember well the effect was sometimes stronger than first hand smoking.

    Also I think if you define overweight right, for sure it causes health problem. If your weight does not affect your health you are not overweight.

  43. Gravatar of Doug M Doug M
    7. January 2013 at 09:41

    It is actualy very difficult to scientifically measure the long-term helth consequences of something like obesity or, for that matter, second hand smoke.

    To do so requires assembling a rather large group of people an tracking that group for many years. To measure the influence of any one variable there are many others to control for. So, much of health science and nutrition depends upon intutition and conventional wisdom.

    It is not entirely unlike economics.

    Regarding obesity, it is my personal and unsceintific theory that wieght gain after middle-age likely has very different health consequences vs. being heavy one’s entire life. Late in life, a few extra pounds can make the difference between surviving a bout with pneumonia and death.

  44. Gravatar of Errorr Errorr
    7. January 2013 at 10:11

    The problem with second hand smoke studies is that they focus on cancer risk which is probably negligible compared to other problems associated with nicotine and smoke. I believe the most major studies show that cancer risk takes at least 15 years of a pack a day before it ever increases. That is alot of smoking and suggests excise taxes could probably lower mortality by reducing the number of cigarettes smoked a day. Of course just because the cancer risk has been overstated by anti-smoking groups doesn’t mean that the real health problems of smoking are not major. Heart disease caused by hardened arteries probably kills an order of magnitude more people than lung cancer. The think is that ‘cancer’ is a more effective tactic to scare people and thus has been over emphasized while the real dangers of smoking are not really talked about.

    I also would not be surprised if lung cancer was much higher in the past among the long lived because wood smoke from campfires is hundreds of times more damaging than cigs. OF course nobody lived that long.

  45. Gravatar of bmcburney bmcburney
    7. January 2013 at 10:11

    The truth is that none of the three propositions listed in the post are actually supported by real science. All of them owe their support to “cargo cult” science which adopts the superficial forms of science without being based on actual science. What we have in all three cases is unsupported and untestable opinion. That does not mean they are wrong, it just means they are not science.

    Science requires that the theory be capable of testing and be tested. There is more science in my opinion that the Seattle Seahawks are going to win the Super Bowl this year than in any of the three propositions above. My theory is testable (and is being tested). Those theories are untestable. Or, at least, they are not testible to the satisfaction of their proponents. In fact, all three propositions have failed some tests. Of course, the propoents of these theories reject the tests as being improper or inadequate but the point is they have not suggested other tests which would meet with their satisfaction.

  46. Gravatar of bmcburney bmcburney
    7. January 2013 at 10:38

    Full Employment Hawk,

    As best we can tell, severe storms have caused damage to the Northeastern US at approximately the same rate since the end of the last ice age. Indeed, recent years have seen a decrease in hurricanes making landfall in the continental US. Damage caused by Sandy, which was not even a hurricane at landfall, merely tends to demonstrate that the Northeast has become “soft” over the years as a result of the absence of storms. In any case, the fact that a severe storm occurred recently is not evidence of anything. Weather is not climate.

    Under the circumstances, it is easier to attribute the spread of tropical diseases to increased air travel than to AGW. For example, if West Nile virus suddenly appears in Manhattan without first becoming common in Europe and the Near East, it is difficult to see how this could possibly be the result of changes in the world climate. Among other things, the total increase in surface temps caused by AGW cannot be more than .6 degrees C. It is difficult to see how this can possibly cause important health effects today.

  47. Gravatar of PrometheeFeu PrometheeFeu
    7. January 2013 at 13:50

    I don’t understand the point of the BMI. Well, I do. It was invented as a population statistic to look at malnutrition. But I don’t understand how it has any validity in calling somebody “obese” or “overwheight”. Do we actually have a curve which shows that the health level (whatever that could be) plateaus at the “appropriate” BMIs and drops on either side of it? And I mean plateau, not peak somewhere in that range. What possible objective test could there be to determine “healthy” vs “unhealthy” as opposed to more vs less healthy.

  48. Gravatar of FormerSwingVoter FormerSwingVoter
    7. January 2013 at 15:18

    I haven’t read all the comments, so I apologize if this has been mentioned, but I think I know why you would get a confirmation from most liberals on those three statements: a perceived consensus among experts.

    Most people don’t look for research to support those perceptions.

  49. Gravatar of ssumner ssumner
    8. January 2013 at 07:11

    Doug, Good points, but why wouldn’t the following test tell us something useful:

    Look at 100,000 healthy normal weight people who are in their 50s, and 100,000 healthy but moderately overweight people in their 50s. See which group lives longer. That’s a statistically significant sample size isn’t it? And you avoid the bias caused by sickly people being thin.

Leave a Reply