Obesity and Causal Empiricism

I see that I do have some new readers. Gene Callahan says

Your “ridiculous” answer on obesity seems obviously correct to me — maybe insufficient, but fine as far as it goes — and the answer to the “puzzle” also obvious — adjust calories in and calories out until you are at a stable weight that you like. What is “ridiculous” about that?

The first issue is that saying one ate too much or too exercised too little is a non-answer. It simply takes the equation of motion and applies a normative frame to it.

I could say that rocks fall because they have too little resistance to the propensity to fall. This is a non-answer. It simply takes the analytical notion that resistance impedes motion and the observation that things fall and applies a normative frame to it.

How would we know if the resistance is too little? Because the rock fell. How would we know if the resistance is enough? Because the rock did not fall.

This tells us little, except that you are aware that things fall.

What we would like is something that takes parameter values and then tell us what result we should expect.

Now, the second part of Gene comment is slouching towards a model. He words it as instructions but if we recast it as a general framework we get something like this.

1. Weight change is the residual of the choice variables calories-in and calories-out
2. When people are unsatisfied with their weight the adjust one or both of these variables to make the residual positive or negative.
3. When they are satisfied with their weight they adjust one or both of these variables to make the residual zero.

This has the virtue of being the beginning of a model. It has the vice of being at odds with the facts.

Lets set aside for a moment the issue how would we know if people are satisfied with their weight because I think it is too emotionally loaded. Instead, lets investigate the dynamics of this model with exogenous shocks.

First, imagine a young teenager who is temporarily at some stable weight and so presumably is satisfied and has chosen a residual of zero. Then the teenager hits a growth spurt. Under our model the teenager’s weight does not change.

He or she will grow taller but will not grow heavier. And, so BMI will fall and at the square of height change. If at this new BMI the teenager is unsatisfied with his or her weight then he or she may alter calories-in and calories-out to achieve a new weight, but in the absence of such conscious alteration weight will be fixed as height extends and so BMI will fall.

Does this appear to be what’s going on? Are teenagers accidently getting thinner as they grow taller and then moving to correct that?

Second, imagine a woman who gets pregnant. Again, under our model her weight will not change. Thus in the absence of conscious alteration of her caloric balance her core body will lose at least as much weight as the fetus and placenta gain. Indeed, she will lose more because the fetus and the placenta produce metabolic waste.

Now she may become unsatisfied with the loss of core body weight and move to alter caloric balance but in the absence of such unsatisfaction she will lose core body weight. Moreover, once the fetus and placenta exit the residual will exogenously rise and her core weight will move back to its original amount.

Does this seem at all like the patterns we observe?

Not to give away the show, but one might suspect that there is a mechanism which works to adjust either calories-in or calories-out in response to growth spurts and pregnancy. Our questions might be, does such a mechanism exist? How does operate? Is it present even when these events are not occurring?

Obesity and Genetics, Ctd

To emphasize, I do not actually believe that assortative mating is the cause of the obesity epidemic.

A major challenge to any theory, however, is that it has to operate as a multiplier. The heritability of obesity is seemingly constant. That is, what we mean when we say “Whatever is happening, its seems to be happening to all of us”

If that were not true then presumably some people would be protected not only by their genes but by their environment. In which case heritability would fall.

For example, suppose that it is fast food. Well some people may be adopted into families which don’t eat fast food. This person should have a large degree of protection that is not genetic. Yet, the data don’t bear that out. Environment has not become more important.

Suppose it was watching TV, or drinking Soda, or any of those things. Then some sort of environmental protection should show up in the data but it doesn’t seem to.

This is why I have argued before that I suspect it is a ubiquitous small molecule. Its something that’s in everything. Maybe a universal preservative. Or, a legally mandated fire retardant. Something like that.

Some of the people I am fans of suggest that it is sweeteners like High Fructose Corn Syrup. This is not nutty as there is a clear mechanism and HFCS is more universal than you might think. However, its still a big hurdle. Surely some parents restrict their kids exposure to HFCS. Why don’t we see that?

Further my point is that a wholly genetic theory of the obesity epidemic can be constructed and to my knowledge no one has laid down solid evidence that it is in fact wrong.

That’s different from thinking its right, but it does highlight the state of our ignorance.

Also, for there are a couple of go-to studies on Obesity and Heredity. I’ll pull a chart from one of my favorite hereditary studies, the Sacerdote’s “What Happens When We Randomly Assign Children to Families” because this is the type of study we are typically tying to emulate statistically.

This shows the relationship between the child’s outcome and that of their adoptive parents and their biological parents. The children are on average about 30 or so I think.

Some outcomes are of course going to be influenced by neither. However, you can get a sense of the relative contribution of “role model” vs “genetic” effects by looking at the ratio.

On this measure BMI and Height are about the same. Studies usually find height is about 90% genetic and BMI is between 70% and 90% genetic. Usually measures of obesity as a binary variable find lower genetic effects but because “unexplained” variation is categorized as environmental this is at least partially and possibly entirely due to binary cut-off.

That is, if you have a BMI of 29.9 and your twin has a BMI of 30.1 then you are obese and he is not. However, this would look like unexplained variation when really the variation is an artifact of the cut-off.

Lastly, you can build a model that says its just self-control. One’s self-control is genetic and what has changed is the opportunities for indulgence. Maybe, but its hard to square that with childhood obesity which is under control of the parents and thus we should see a strong affect of strict adoptive parents.

There are also some other issues with self-control being the mechanism that I’ll just mention. One, self-control seems to be rising by almost all other measures yet obesity is rising. Not a killer problem for the theory but perplexing.

More importantly its hard to square with the stability of weight, which is probably the biggest or second biggest next to heritability, stylized fact about obesity. That is, if you always over eat you don’t get fat, you get fatter. Presumably, you will get fatter and fatter over time.

However, non-dieting obese folks have a weight stability that is roughly the same as thin people. Its as if they tried to go exactly to a particular too heavy weight and then stop and act like a thin person with 50 extra lbs.

Why does it stop like that?

That’s a puzzle.

It also *suggests* that there is some regulating mechanism that is set to the wrong parameter. We can also measure regulating hormones and they are systematically off in obese people, but this is enough for one post.

Obesity and Genetics

Sarah Kliff writes

But more recent research, particularly a highly cited study in the American Journal of Clinical Nutrition, suggested that was more a question of genes than habits. The study, which followed 5,092 pairs of British twins ages 8 to 11, found that the most influence parents have on obesity is actually genetic, a factor of inheritance, rather than the environment. “Contrary to widespread assumptions about the influence of the family environment, living in the same home in childhood appears to confer little similarity in adult BMI [body mass index],” authors Jane Wardle, Susan Carnell, Claire Haworth and Robert Plomin write.

The research raises one key question: If obesity is genetic, why has it increased so much in recent years? The researchers explain that, some of the genetic predispositions for obesity have existed for awhile. But changes to our food environment, like the proliferation of fast food, have enabled those predispositions to become more powerful.

“Although contemporary environments have made today’s children fatter than were children 20 years ago, the primary explanation for variations within the population, then and now, is genetic differences between individual children,” they write.

One thing, just as a comment on how public health is done in America, is that “everyone” knows this.

That is to say, its nearly impossible to spend more than 6 hours on Google Scholar researching obesity and not come away knowing that obesity is high heritable. You might question whether the heritability has declined over time, but looking this up will point you to immediately to studies which suggest, no.

You’ll hear offhanded references to this fact when you read certain obesity researchers who write phrases like: “whatever is happening is happening to us all”

However, few people come out and say it. The fear I think is that if people believe that obesity is genetic they will stop trying to do anything about it. The problem is that the implicit lying to people causes all types of immediate human suffering.

Those who have been reading me for a while know that this is also the case with exercise. Its well known that exercise as an intervention is unlikely to cause a reduction in body fat percentage. Yet, this fact is shoved under the rug seemingly because exercise has extensive health benefits and the public health authorities know that lots of people would stop exercising if they were aware it was unlikely to cause them to loose weight.

On many counts the way public health is treated in America borders on intentional deception.

Secondly, no one knows for sure why obesity has exploded. Indeed, though I doubt this is the answer, I have not seen anyone convincingly show that it is not genetic.

That is, that it is not assortative mating. Under this theory fat genes are synergistic. If you have fat gene 1 you are 20 lbs overweight. If you have fat gene 2 you are 10 lbs overweight. If you have fat genes 1 and 2 you are 50 lbs overweight.

When overweight people marry overweight people they then have morbidly obese children. This would help explain the bimodality of obesity but off the cuff you would expect the thin peak to expand not contract, if this was the case.

Still, differential reproduction rates might get you around this.

In any case I don’t think that’s the likely answer, but the point is that no one actually knows for sure what the answer is and whatever it is it has to amenable to the fact that not only is obesity highly heritable but its heritability has not changed – up or down – since the onset of the epidemic.

Slippery slope of paternalism watch: advertising edition

Some time ago I challenged those who don’t believe that paternalistic regulation is characterized by a slippery slope  to provide some examples of regulation that would prove them wrong. The problem I saw was that paternalism fans always deny the slippery slope exists by claiming that new regulations are just reasonable policies. But of course this is how the slippery slope works, as today’s new policies will be used to justify future policies and to make them look reasonable. After all, every new step is only a small distance from where we are currently standing, but what are we walking towards? Nobody took the challenge, but pivoting off of San Francisco’s Happy Meal ban I did makee some predictions about future likely paternaism:

Making fast food less attractive may protect parents when they happen to be near a McDonalds with their kids, but it doesn’t protect them from having McDonalds reach out to children in the first place and getting it into their heads that their food and toys are awesome. If you’re going to stop this problem, it must be at the root. One way to do this is to ban advertising of fast food targeted at children. This would probably start with children specific magazines and TV shows, but move to a general ban.

Now regulators are helping to make my predictions come true, as they attempt to place limits on advertising by food companies to children. Here is how Ad Age describes the guidelines:

…the rules would start in 2016 and only allow foods that contain no trans fat and not more than one gram of saturated fat and 13 grams of added sugar per “eating occasion” to be marketed to children. Also, the foods could not contain more than 210 milligrams of sodium per serving. The sodium restrictions would tighten by 2021. In a concession to industry, the rules do not include “naturally occurring” nutrients. Additionally, the foods must provide a “meaningful contribution to a healthful diet,” including from at least one major healthy food group such as fruit, vegetables, whole grain, fish, eggs and beans.

The guidelines are said to be “voluntary”, but as Ad Age points out this is a little murky:

Although not binding, whatever emerges in the final report to Congress will likely be adhered to in some fashion because the rules are put forth by a quartet of agencies that have strong sway over marketers, including the FTC, Food and Drug Administration, Centers for Disease Control and Prevention and Department of Agriculture. “Despite calling these proposals ‘voluntary,’ the government clearly is trying to place major pressure on the food, beverage and restaurant industries on what can and cannot be advertised,” the ANA said in a statement.

I would be interested in reading more about the “strong sway over marketers” that these agencies have, and exactly how the nominally voluntary guidelines would be non-voluntary in practice.  This will probably come to light, as Ad Age says that this announcement is only an “opening salvo in what will be a lengthy debate between government and industry on how to solve the growing childhood obesity crisis”.

If paternalists truly were concerned about reducing childhood obesity and not simply trying to make themselves feel good, then they should be willing to include in these regulations a sunset provision that repeals them if they don’t have a demonstratable impact on childhood obesity rates in 5 years. My guess is that paternalists wouldn’t go for this, because deep down they know this isn’t going to make much if any difference in children’s health and are really interested in banning something they find distasteful.

The slippery slope from here is pretty obvious: strictly non-voluntary guidelines that require any food packaging or advertising of must be approved by a regulatory agency and subject to standards similar to those above. But we know that advertising isn’t the only way that companies influence purchasing decisions. Why shouldn’t the color of packaging be regulated? I’m sure behaviorlists can tell us which colors children like most, and I’m sure regulators would be happy to insist on gray boxes for unhealthy foods. Children are also probably more drawn to items low on grocery shelves or in the checkout aisle, so why shouldn’t regulators determine where in a store products can be placed?

I’ll repeat my challenge to paternalists: if this isn’t evidence of the slippery slope of paternalism, then what would be?

Is This Why You Are Fat?

Matt writes a post which displays American diet composition in 1970 vs 2008.

He concludes

If everyone ran an hour a day at eight miles per hour, that would actually make up for the increase, but obviously that’s not what’s happening.

The experimental and historical evidence suggests that this is not the case. The experimental evidence says that if we induce people to exercise more and let them eat freely, they will on average not lose weight. Most people will change weight but some will lose and some will gain.

As always, no one disputes that wedging will result in weight loss. That is driving calories-in and calories-out in opposite directions will lower the caloric content of the body. Excluding water there is a rough relationship between caloric content and mass. For most fat, which is our primary concern in obesity, the relation is about 3500 calories per pound.

The tendency of all animals is to try to get calories-in and calories-out to move in harmony. If you have to wedge then that means that this system has failed. At a minimum we would like to know why.

Historically people worked a lot more than they do now. They also ate a lot more than they do now. From the year 1400 to 1970 average calories expended fell dramatically but so did average caloric intake. Obesity was never a severe problem. The system did not fail.

Then from 1970 to 2010 average calories expended actually rose but calories consumed rose more and obesity exploded. The system failed.

If you go and look at the actual graph though, you can see Gary Taubes’s thesis on display. This is natural since the dataset used to make the graph is one of Gary’s favorites. First you see meat falling, then fat falling/stalling, then sugar falling as various healthy eating theories rose to prominence.

The only thing that rises consistently is grains. Gary insinuates and sometimes outright says that obesity was caused by the encouragement international health authorities for people to eat more grain. The naturally tendency is for people to eat more meat as calories become easier to obtain. He suggests that consciously overriding this mechanism led to an excess of insulin and possibly deficiency of peptide YY, which are key regulators of caloric balance.

I am skeptical of this theory, but it is one that at least recognizes the underlying theoretical problems.

Oh and just for uber-nerds, Andrea Jezovit, the chart creator says

By calculating such food losses, the USDA data closely approximates the amount of food that actually makes its way from the farm into the average American stomach.

This is hotly debated. The general charge is that the USDA loss adjustments are outdated at best and methodologically suspect in the first place. The charge that comes up the most is that the USDA is overestimating fat consumption because they are not properly accounting for the way fat is used as a cooking medium.

Healthy Living Vouchers

Many things Modeled Behavior come up in Matt Ridley’s suggestion that we use vouchers to combat obesity.

After all, as Friedrich Hayek pointed out, the true genius of markets is that they discover things. Perhaps the answer to obesity is to spend money not on the producers (of gyms, diets, surgery, vegetables) but on the consumers.

Well somewhat.

The genius of the market is the way it aggregates masses of information that no single individual could possibly possess and subjects scores of untestable hypothesis to the forces of economic evolution. This allows us to do without knowing what we do and to design without a designer. Great and beautiful things.

However, raw scientific and administrative knowledge are powerful things as well. Indeed this is revealed by the emergence of socialism as a doctrine. It was in large part the ability of technocracy to win wars, purify water and cure diseases that inspired folks to believe that technocracy could do anything.

Drawing a direct analogy with the effect of vouchers in the education system, Messrs. Seeman and Luciani suggest “healthy-living vouchers” that could be redeemed from different (certified) places—gyms, diet classes, vegetable sellers and more. Education vouchers, they point out, are generally disliked by rich whites as being bad for poor blacks—and generally liked by poor blacks. A bottom-up solution empowers people better than top-down government fiat.

This is certainly true, though I am not sure it really gets you anywhere. Lack of empowerment doesn’t seem to be the core problem here. The number of private weight loss attempts that fail every year far exceed those that succeed. People can and do try wacky individualized weight loss programs. Entrepreneurs can and do promote all manner of weight loss products.  The overwhelming majority just fail.

After all, the root causes of obesity are multifarious and new ones are being added all the time—such as diet sodas, gut bacteria, genes, sleep apnea, leptin levels, medication, depression, poverty and peer pressure. So the solutions need to be multipronged, too. What works for you may not work for me.

The underlying notion here – that the obesity epidemic is a multipronged problem with lots of individual causes – is likely wrong. There are lots of levers with which one can attack obesity. There are lots of failsafe systems that exist in the regulation of appetite and activity and they can be overridden in different ways. However, the odds that an epidemic with the steady widespread march of obesity is multi-factor are slim. There is probably a single cause and it probably operates directly on the endocrine system. Of course at this point we don’t know what it is, but I suspect that it is a particular molecule or class of molecules.

People in the future will think it as toxic and find it amazing that we were so careless with it. The same way we think it maddening that children used to play with mercury.

In due course, the obesity problem will be solved, I suspect. The ultra-rich have already solved it. Most of them are very thin these days, quite unlike in ancient times. That’s because they can afford the solutions that work for them, from low-carb diets to personal trainers.

In due course I expect it to be solved as well. However, the notion that the ultra-rich have solved it is wrong. First, off there are obviously selection effects that decrease obesity levels among the rich. You are more likely to become rich, by business success and especially by marriage, if you are thin.

Second, we know that susceptibility to obesity is highly heritable. This means that the children of a rich spouse and a thin spouse are more likely to be both rich and thin. A similar effect happens with height. The rich are likely to be tall. Yet, this isn’t because they’ve found a solution to shortness.

Third, as Ridley suggests the rich are spending lots of time and money on combatting what obesity they have. This is not solving the problem. This is managing the symptoms of the problem. Whatever underlying condition is forcing you to expend all of this effort still exists. Its that underlying condition that we want to cure.

For example, managing Tuberculosis by moving to a dryer climate is better than dying in the sewers of Paris, but its not the same as curing TB. Tuberculosis in some ways provides an analog to obesity. It is a disease that has been with us since antiquity yet saw an enormous, at the time inexplicable, spike in the 18th century.

Theories on the cause of tuberculosis ranged from vampirism to masturbation to impure air. It was argued that a variety of maladies could lead one to succumb to Consumption, as it was then known. Consumption was also endemic among the urban poor, a fact sometimes attributed to their low moral character.

Yet, in the end it was a single disease with a single cause and streptomycin was the cure.

We’re always at the bottom of the slippery slope

I have argued that the slippery slope of paternalism is real, and we are sliding down it. Defenders of paternalism argue there is no slippery slope because a) where would we fall from here? and b) why haven’t we begun sliding yet? I rush to point out paternalism that targets sugar and salt, and the defenders argue “Well, that’s just good policy. Let me know when we’ve actually started sliding down the slippery slope”. What happens is paternalists are forever moving the goalposts, and declaring the newest ban or tax just reasonable policy. Their burden of proof demands that that we slide two, three, or four steps down the slope at once instead of one step at at time, since the one step we’re taking now is just reasonable. We’re always seemingly at the bottom of the slope, and things aren’t so bad from here are they?

Well folks, we’ve reached a new slope bottom: San Francisco has banned the McDonalds Happy Meal.

Since paternalism defenders will surely claim this is “just reasonable policy, and if there is a slippery slope then where could we possibly slide to next?”, let me repeat what I wrote awhile ago:

I think it would be useful to for critics of the slippery slope theory of paternalism to demarcate now what future policies would constitute evidence that they are wrong, because my guess is the point of demarcation will move right along down the slope with policy. Several years ago many of todays critics of slippery slope theory would have said that an attempt to regulate salt would constitute evidence. But now, farther down the slope, salt regulation is just sensible policy.

Behavioral Economics in the Cafeteria

There is a very interesting article at the Atlantic from David Just and Brian Wansink from the Cornell Center for Behavioral Economics in Child Nutrition Programs. They discuss their work on improving healthy eating in school cafeterias using the subtle wizardry of behavioral economics. There have been some impressive results:

One school in upstate New York was able to increase consumption of salads by close to 300 percent by simply moving their salad bar six feet from the wall and placing it near a natural bottleneck in the check-out line. Another school increased fruit sales by 105 percent by moving the apples and oranges from stainless steel bins into a well-lit and attractive basket.

It is encouraging to see behavioral economics being put into creative use like this. The authors argue that “It is difficult to teach a high school student how to make healthy choices in the real world if only escarole and tofu on are on the school lunch menu”. But is this teaching them to choose better or tricking them into choosing better? After all, if behavior is so amenable to subtle tricks like this then what hope can there be that any behavioral changes will actually last? I can’t tell if articles like this should make us more or less hopeful. Yes, the good scientists here are making a difference, but are we really so impressionable?

A Can of Soda A Day

Kevin Drum asks

Now, there are a bunch of things you might say about this right from the start. Maybe governments shouldn’t be in the business of running nanny state ads about personal nutrition. Maybe this particular ad was disgusting and shouldn’t have been released. Maybe obesity isn’t really that big a deal in the first place. But those weren’t the issues at stake. Rather, it was this single sentence in the ad:

Drinking 1 can of soda a day can make you 10 pounds fatter a year.

What, I thought, could be wrong with that? A can of sugared soda contains about 150 calories, and adding 150 calories a day to your diet would almost certainly produce a ten-pound weight gain over the course of a year or so. There are some caveats, of course:

So I’m curious: what do you all think of this? I’m open to argument here, but it seems crazy to me, less a politicization of science from the health commissioner than a case of geekdom run amok among the scientists. I mean, if you can’t tell people that adding a bunch of calories to your diet will make you gain weight, what can you tell them?

The problem is that the calorie balance interpertation implies a completely false understanding of what is going on. There is an extent to which geekdom can tolerate this level of nonsense and there is a point at which it must be combated.

I will compare to something I know Kevin gets. The calorie balance logic is equivalent to saying.

Government deficits drain savings. Savings are the engine of growth. Therefore, cutting the deficit immediately is our best shot at growth.

In both cases you are taking an accounting identity

1. Private Savings – Public Borrowing = Net National Investment
2. Calories-In – Calories-Out = Calories Contained in the Body

And, treating it as if it were a model of the world.

You have to be aware that public borrowing might effect private savings. In particular if public borrowing stimulates the economy it will increase private income which in turn will increase private savings.

You also need to be aware that Calories-In affects energy and hunger levels which not only feeds back to Calories-Out but also to other Calories-In.

I used to post this thing a lot, but since the blog has new readers it might be worth our while to look at how a properly functioning metabolism responds to a rapid increase in Calories-In

The big question we have is why does this stop working in some people? Just to note, there are many, many other feedback loops that are important as well. I point out this one because it so obvious both that it works in the healthy metabolism and that it fails in the unhealthy one.

You are probably aware of the relationship between diabetes and obesity. It is commonly assumed that obesity causes diabetes. This is in part because even some scientists are fixated on the accounting identity. However, there is a reasonable case that diabetes may cause obesity.

That is, the resistance of the muscles to insulin causes the breakdown in the “sugar rush” response (and other loops) which then breaks down the feedback from calories-in to calories-out.

Now, if it is in fact the case that sugary drinks induce insulin resistance this connection may still hold. However, it is almost certain that the simple minded thinking that in general dropping a 150 calorie item from your diet will not feedback on other metabolic components promotes a fundamental misunderstanding of what’s going on.

_____________________________________________________________

For the geeks. Yes, in truth even what I have written here is a gross oversimplification and ignores central facts such that an increase in obesity from sugar consumption must be proximately caused by an increase lipogenesis or a decrease in lypolysis both of which are hormonly regulated processess. That is, just as there is no such thing as immaculate transfer there is no such thing as immaculate obesity.

You can’t just throw organic matter at a metabolism and get fat. You actually have to create fatty acids and bind them up into triglycerides. Any model that assumes that you can is going to wind up disappointing you and of course there are a fair bit of disappointed dieters. We need to do better as intellectuals.

Scientists against salt regulation

When the Institute of Medicine recommended broad, draconian regulation of salt last year, I pushed back against the idea, one might say, obsessively. Now, via Marion Nestle, comes a new paper in The American Journal of Clinical Nutrition arguing that the current level of sodium intake is not a problem for the population. The article comes with an accompanying editorial titled  “Science trumps politics: urinary sodium data challenge US dietary sodium guideline” that closes with this appeal:

The analyses of extensive measurements of 24-h UNaV, which these 2 reports have collected from the medical literature over the past 5 decades, are compelling. They provide plausible, scientific evidence of a “normal” range of dietary sodium intake in humans that is consistent with our understanding of the established physiology of sodium regulation in humans. This scientific evidence, not political expediency, should be the foundation of future government policies, thus respecting the known and unknown scientific complexities surrounding sodium’s role in health and disease. Guidance for sodium intake should target specific populations for whom a lower sodium intake is possibly beneficial. Such an approach would avoid broad proscriptive guidelines for the general population for whom the safety and efficacy are not yet defined. An appropriate next step is not to lower the sodium guideline further.

You can find my previous coverage of salt regulation here, here, here, here, here, and here.

Soda, food stamps, and paternalism

Michael Bloomberg and David Paterson announced a proposal yesterday to ban the purchase of soda with food stamps. It is sure to be controversial, but is it a good thing?

At the very least, if the government is determined to try and reduce decrease public expenditures on health care by reducing soda consumption, than this is a preferrable approach to a soda or sugar tax. A first best approach would be to tax individuals who a) are drinking enough soda that it increases their risk of illness, and b) with some probability part of their health care costs will be born by public.

Since foods stamp recipients seem like a likely target for b), this at least meets one criteria. In contrast a general soda tax falls on everyone, and meet neither criteria. Even with the more targeted food stamp approach, people whose soda consumption is at safe levels or who have private insurance will be inefficiently restricted by this.

I have not looked at the data myself, but the conventional wisdom and the contention of the proposed law is that a) is very much true.

The whole discussion of course presumes that the law actually reduces soda consumption. For one thing, if individuals are paying some non-soda food costs with cash they can just shift to spending that cash on soda. There are also ways to trade around this: I buy $10 worth of soda with cash, you buy $8 worth of food with food stamps, and we trade. In either case though, transaction costs have been raised, although in the former the amount may be very slight.

Another problem is that individuals may respond to the lower calories by simply consuming more calories. While researching the health effects of soda for my recent defense of diet soda, the literature appeared mixed as to whether switching from regular to diet soda caused weight loss because of the calorie substitution problem. Perhaps Karl will chime in on this; he is much more knowledgeable about all things obesity.

The final question to ask is whether this policy is simply too paternalistic? I have to say I don’t think it is. Food stamps by themselves are already highly paternalistic. Essentially they tell low-income people that on average they will not spend cash in a way that best benefits them and their family. The government defines a subset of goods and tells them “you will be better off if you stick to these goods instead of buying what you want”. The marginal paternalism of reducing the size of the goods the government allows is slight compared to the paternalism food stamp recipients are already enduring. If your significant other tells you that you have to go to bed between 9:30 and 10 that is highly paternalistic. If they further refine that and decide it has to be between 9:40 and 10 it’s not that much more paternalistic. Perhaps this graph will help:

Another issue that you are free to reject, as it just reflects my diet soda bias and my love of delicious aspartame, is that I suspect much of the continued preference for regular soda over diet soda despite the health advantages is motivated by a lack of understanding of the safety of diet soda. Fear not New Yorkers, despite the email chain letters and urban legends, diet soda is not bad for you.

Overall though we should be weary of this kind of paternalism, and the desirability depends on how effective it would actually be. However, given the high level of paternalism in food stamps already I don’t consider this marginal paternalism to be that troubling.

The Demand For Ignorance

Jodi Beggs, aka Economists Do It With Models, argues that paternalism need not be justified by assuming irrational agents, but can simply be an efficient response to an informational problem:

Any Economics 101 course will tell you that a required condition for markets to be efficient (read, value-maximizing for society) is that consumers have full information about the products they are considering consuming. In this way, the calorie-labeling legislation is helping to push the fast food market in the direction of efficiency as much as anything else. What’s so behavioral-y about that?

One counter to this is that markets should be supplying the amount of information that consumers prefer, and that the reason we don’t observe a lot of menu labeling and other information from restaurants is because consumers don’t want to know. Of course, you could argue that they don’t want to know only because of what they don’t know…Wait, what?

Ok, bare with me. Pretend I had a sealed envelope that contained a letter from someone telling you telling you exactly why they hate you. But say you believe that everyone who you care about doesn’t hate you, therefore you assume it’s from someone you don’t care about, and since you don’t want the annoyance of reading hate mail from someone you don’t care about, you choose not to open the letter. But, say that letter is from your wife, who secretly hates you. Well you would want to know why your wife hates you, but since you believe your wife can’t possibly hate you, you won’t get information you want. Basically what I’m saying is that your current information set determines your demand for information.

So what does this have to do with menu labeling regulations? If we assume markets are working, then the level of information we observe is the amount demanded by consumers, which efficient. In this case menu labeling laws would make people worse off by giving them information they don’t want. That is unless the amount of information they are demanding is based on their assumption that restaurant food is kind of unhealthy, but not as unhealthy as it really is. If they had any idea how bad it was, they would want to know. In this case menu labeling laws could make people better off.

Determining the source of the lack of information is critical to knowing whether menu laws are efficient or not. This is especially important because the amount of information can affect demand, which contra Jodi, can change the choice set. She writes:

…the consumer has exactly the same set of choices available to her regardless of whether calorie counts are on the menus or not. Because of this feature, it’s hard to argue that this sort of legislation is significantly bad for anyone- here, the worst-case scenario is that some people keep eating unhealthy food but are no longer blissfully ignorant and instead feel guilty.

But what could happen is that when people are no longer able to be blissfully ignorant, which they prefer, they consume a healthier but lower utility set of products. This in turn could change restaurant supply decisions, which would mean a different choice set.

So what is it:  is our demand for ignorance efficient, or is our ignorance causing us to demand an inefficient amount of ignorance?

Eat Skinny Boy Eat

Amid the hand wringing over eating as a “sport” I would be remiss if I didn’t point out that the obesity rate among elite professional eaters is lower than the obesity rate overall.

In particular legends Joey Chestnut and Kobayashi are pretty fit guys

 

But of course its just calories-in minus calories-out. No equilibrium modeling required.

HT: Appel

USDA Soda Report Worse Than I Thought

I wanted to get something out quick about the USDA’s soda report and its implication that a modest soda tax could cause children to loose 4.5 lbs.

I assumed that the 4.5 number came from an equilibrium METs analysis. Basically to do that you note that living one’s life requires burning some number of calories per pound of body weight. The units we use are Metabolic Equivalent Tasks (METs).

Then the embarrassingly simply model says that if we reduce calories by X we will get weight loss. However, as the body looses weight it will burn fewer calories in daily activities. Thus calories-out will be reduced. This rate is determined by your METs assumption.

At some point calories-in and calories-out will re-equilibrate and we think of that as net equilibrium weight reduction. Now setting aside all of the endogeneity / partial equilibrium problems with this simplistic analysis, the USDA report doesn’t even go that far.

No what it does reads to me as unthinkable. They multiply 43 calories per day times 365 days a year and divide by 3500 calories per pound of fat to suggest that children will loose 4.48 ~ 4.5lbs of fat per year!

Per year my friends. Per year!

I often sense from the comments that the extremity of these types of claims fails to sink in. Lets just do a little abstraction.

So, child is loosing 4.5lbs per year. Mind you child’s weight is nowhere in this analysis even implicitly. 4.5lbs a year – that sounds pretty good. Indeed at that rate an eight year old formerly soda guzzling kid could loose 45lbs of fat by her 18th birthday.

But wait a second. What if our child only has 15 lbs of excess body fat. What then. Does she become gymnast level ripped, incapable of menstruating?

Does she fail to grow? We know that mass balance is a fundamental law of nature not a social convention. The laws of physics don’t know if she’s “too thin.” If the calories don’t come from fat, of which she only has 15 lbs, they have to come from somewhere. Perhaps, her growth is permanently stunted? Perhaps, her brain development is impaired? Perhaps she suffers all of the maladies associated with underweight childhood development?

All of these thing are possible from persistent caloric restriction but they seem a bit strong as a result of a soda tax?

“Well that’s taking it to extremes.” you say. This analysis doesn’t work at the extremes. Its just about averages and point estimates.

Here is the important point, however: this analysis makes no distinction between moderate and extreme extrapolation. There is no “distance from baseline” component.

The fact that you know this analysis doesn’t work in extremes means that at minimum the model is imprecisely constructed with relation to scale. Further, I would add, there is simply no reason to assume that the model works over “reasonable” scales but simply fails over extreme ones.

The model can, I would argue does, begin to breakdown as soon as the first gram of fat is lost.

This is a large part of why weight loss science looks and acts like voodoo.

Someone takes a really complex equilibrium system. They indentify a property or set of properties. Re-inserting that property into the entire system is mathematically intractable and indeed, we don’t completely understand the system anyway.

So the analyst linearizes the assumption. If all else is held equal they say. Yet, the scale over which the human metabolism will match a linear approximation is tiny. The body immediately acts to undue whatever effect you tried to create.

So maybe, maybe if you are lucky and you hit the body with the equivalent of sledge hammer’s worth of adjustment you can temporarily squeeze out five, ten maybe even 20 lbs of pure adipose tissue. However, the metabolism soon adjusts and acts to overcome even that enormous effort.

This is the core of the problem we face. Any effort to address this problem that does not recognize this difficulty is doomed to failure.

I cannot beat this drum enough, because people continue to try these types of methods, continue to fail and continue to be shocked at that failure. And, even if that type of hamster wheel insanity was ok as a private choice, if definitely shouldn’t form the basis of public policy.

Let me say will all the force I can: We must not pick winners and losers based on analysis that fails to recognize key elements of the public problem.

It is deeply, deeply irresponsible to do so.

Note: A previous version of this post attributed the report to the FDA when in fact it was from the USDA

USDA: Human Metabolism Will Greet Soda Tax as Liberators

Sullivan points us to an USDA study suggesting that a 20% soda tax could cause the average child to loose 4.5 lbs. In particular the study suggests that such a tax would work by reducing caloric intake by 43 calories a day.

Now before we get into the details of the study lets run this by the well established “Hooey Limit Test.” I say to you, “hey looking to loose that last stubborn five pounds, just consume 1/3 less of a soda per day. Not a whole soda mind you, that’s worth a full 15 lbs. Just leave the bottom third in the can.”

Does that sound reasonable or like a bunch of Hooey to you?

If that’s the world we live in then how are we to make sense of the endless struggles that millions of women (and men) go through to loose five pounds, say nothing of a greater battle against obesity.

As a side note, this is what makes me skeptical of explanations like “sugary drinks” for obesity. Wouldn’t someone have noticed by now that people who don’t like sugary drinks are never fat? Wouldn’t that be everywhere? Its not a hard observation to make.

Whatever the secret to weight loss is, it must be elusive. It must be something that is either difficult to find or difficult to execute. Otherwise people would have succeeded by now.

Now, on to the study. What it does is pretty straight forward. They look at consumer behavior. They then estimate the change in consumer purchasing behavior that would come from a change in soda prices. How less soda would consumers buy. How much more bottled water, milk and juice.

They then net out the calories. Since, soda is more calorie dense on average, substituting away will cause a reduction in calories. They then turn caloric reduction into a equilibrium reduction in weight.

Regular readers will note that the immediate problem is that they assume consumption is exogenous. That humans choose their caloric intake for some set reasons unrelated to body composition. They then independently choose their caloric expenditure. We subtract one from the other and composition pops out as the residual.

Now, obviously no one, not even the authors of the study believe that this is accurate in real life. At a bear minimum humans are both aware of and care about their body composition. Getting fatter is costly and so, as we observe, will cause people to take actions to loose weight.

Indeed the FDA waives all these concerns away in typical academic fashion

Assuming that everything else remains equal (e.g., constant physical activity level and no shift to foods other than beverages), a reduction of 3,500 calories leads to a 1-pound loss in body weight

Which is akin to saying that assuming no insurgency the wars in Iraq and Afghanistan should go swimmingly.

Personal attempts to control body composition obviously affect the results in a feedback mechanism that is familiar to economists. However, even deeper than that body composition is critical to the survival of all animals from nematode worms to Blue Whales. In response animals have evolved mechanisms for regulating that composition. One of the most powerful we are all familiar with, its called hunger.

But hunger is not the only one. There is satiation or fullness, restlessness, exhaustion, and all the variants there of. These emotions are balanced through a complex set of hormones which in a healthy adult causes calories-in to be matched to calories-out with a less than .5% error.

When we see obesity, something must have gone wrong with that mechanism. Now Bob Lustig suggests that drinking soda is what goes wrong, but because of the effect of sugar on insulin levels. Gary Taubes says carbs generally. Seth Roberts says our physiological association of calories with taste. David Kessler says food industry generated hyper-palatability. Until repeated large scale studies showed otherwise the official government line was high percentage intake of fat.

My readers know that while I am sympathetic to the carb, read insulin, based theories I am not even convinced that the culprit is nutritional. Every possibility from endocrine disrupters to biological agents to autoimmunity needs to be taken seriously.

What we do know, however, is that something has gone wrong metabolically. And the key to ending obesity is finding out what that something is and fixing it.

In any case its not as simple as adding up beverage changes and then netting out calories as if a change in beverage calories has no feedback mechanism against composition or calories expended.

Note: A previous version of this post attributed the report to the FDA when in fact it was from the USDA

Poverty and Health Food

Subbing for Ezra Klein, Mike Konczal analyzes food prices and purchasing decisions. He concludes that the poor have it worse than some would believe

The poor have more purchasing power because, in part, they are buying food that isn’t very healthful. And the important thing about this different inflation rate quantification for income inequality is that nobody
gets diabetes. The long-term health costs of "choosing" a different inflation rate for your food isn’t estimated, nor are they included to see if it all balances out economically.

There are important points here that I am sensitive too. However, I simply must drive home the point that the underlying logic is built on a mountain of speculation. We don’t have a really solid explanation for why people are poor. We don’t really have a really solid sense for what food is healthy nor how food contributes to long term disease. We have mountains of speculation. However, recognizing them as such is important because we have seen such mountains before.

Here are a few links to some promising treatment to a form mental illness recognized during the 20th Century

The application of anticipatory avoidance to the treatment of homosexuality

A learning approach to the treatment of homosexuality

Treatment of Homosexuality by Individual and Group Psychotherapy

The authors of these well published scientific studies developed methods which they sincerely believed to be effective in treating a psychological disorder.

We now know that not only is homosexuality not a disorder but it is almost certainly not “psychological” in the sense meant by these authors. That is, at this point it seems all but certain that people are either born gay or straight and that no series of life events can affect this.

The drama here is easily lost, so let me repeat. These were scientific papers reporting the successful psychological treatment of what we now know to be congenital traits. To be even more blunt -  they were reporting empirical success in accomplishing something that is not even logically possible.

Moreover, at the time they were well within the conventional wisdom and strongly supported by a theoretical framework. This was not quackery – yet it was deeply, deeply wrong.

To wit – it is very, very easy to get carried away with a theory of disease, disability or affliction. Passions run high. Prejudices run deep. Really useful data is frustratingly hard to come by. And fueling it all is the conviction on all sides that “something” must be done. Even if that something is to justify away the problem.

So I am strongly urging everyone involved to be careful assembling our very shaky models of health on top of shakier models of nutrition and then combining those with our almost complete ignorance of the deep causes and consequences of poverty. Such an assemblage quickly becomes an edifice of supposition and the foundation of potentially hurtful policy.

I am have my own prejudices on these issues and in full disclose they run slightly counter to Mike’s. However, this isn’t about who is right. This is about the need for everyone to tread lightly.

They came first for the sugar, then they came for the salt…, ctd.

A friend writes:

I’m really glad you posted that piece on paternalism and salt in food. This strikes a personal chord with me, as you could imagine. Especially as someone who consumes pickles as much as most people consume cereal and bread, and as someone who regularly eats anchovies and puts mustard on potato chips. I could go on.

A friend from college recently posted a facebook message saying something to the effect of “about time!” in support of mandating a reduction in salt in food. Earlier this year he posted something about the dangers of vaccines.  I think he might be my greatest enemy.

I have personally witnessed him eat handfuls of cold meat at probably every single hour of the day, so I can vouch for the fact that salt regulations would disproportionately affect him.

They came first for the sugar, then they came for the salt…

The ongoing debate at Cato Unbound on soft paternalism has focused a lot on the issue of slippery slopes. There are two directions one can slide down a slippery slope: an increasing scope of paternalism, and an increasing degree of paternalism. The danger of the former is captured in Glen Whitman’s example of smoking. First they banned it on airplanes, then it was bars and restaurants, and now it is increasingly in all public places.

The other slippery slope occurs when the presence of paternalism in area A makes it more likely in area B. This type of slippery slope is evident in the spread of regulation from sugar, which is becoming more popular, to salt, which is on the horizon:

Citing 40 years of failed efforts to voluntarily reduce the amount of salt in food, an advisory panel Tuesday recommended that the government regulate sodium for the first time… The proposals outlined by the Institute of Medicine envision step-by-step efforts that would both ratchet back Americans’ desire for salt and mandate the maximum amount that could be added to various types of foods.

The proposed policy is for the FDA to begin setting a gradually decreasing maximum amount of sodium that could be legally added to foods and beverages. The level of regulation and control of individual choices that proponents sound comfortable with is really ridiculous to me:

“It must be done very thoughtfully,” said panel chair Henney, a former FDA commissioner and now a medical professor at the University of Cincinnati. Pickles, for instance, “are very high in salt content but are not eaten that often,” she said, “so what you get with pickles might be quite different than something that is eaten more frequently, like bread, or cereal.”

Can I please spend a month with this woman first, checking the label of everything she eats and granting or denying her permission to eat it? Seriously though, if this is not evidence that there is a slippery slope out there begging for us to slide down it, then I don’t know what is.

I hope that most people would consider the letting the FDA deciding how much salt we can have an egregious encroachment into personal freedom, but I fear that the growing presence of sugar taxes has warmed people up for this. The general unpopularity of the proposed Philadelphia sugary drink tax, however, does provide me with some optimism.

Overall, I’m really curious to see what the next absurd thing will be that someone tries regulate. Food spiciness? Temperature? How many people suffer from a burned mouth every year? It’s clearly market failure and information problem, as well as evidence of consumer irrationality and potentially time varying preferences, as surely nobody would choose to burn their mouths. And can we do something about the scourge of brain freeze the nation is facing? I’d like to see a minimum temperature and size of milk shakes. Or maybe we can just provide a “nudge” by mandating narrower straws. Or on the more deadly side, how many people are killed by tired drivers every year? Can we get a mandatory nap-time policy to fix this?

Score One For Gary

Taubes fans will be happy.

The added sugars in prepared and processed foods are threatening Americans’cardiovascular health, lowering levels of protective HDL cholesterol, raising levels of potentially dangerous triglcerides and possibly making people fatter, a new study finds.

from Health.com.  Bob Lustig is smiling as well.

Obesity and the First Law of Thermodynamics

This is the second part of my response to Brad’s comment

Your post scratches at some scientific-sounding ideas of the involvement of genetics, but fails to cite any definitive research into inheritable weight-control issues. The mathematics behind maintaining a healthy weight are startlingly simple, and the mythology of “fat genes” and thyroid problems and impossible to shed pounds is merely there to sell books and gym memberships.

The body is a closed system and cannot create additional weight without an over consumption of calories. Joel S is correct — it’s literally impossible to gain weight by eating your “maintenance” number of calories without violating the first law of thermodynamics.

Previously, I talked a little bit about the scientific evidence for a genetic basis for obesity and gave a few links to some big papers on genes and obesity.

Now I’ll address the first law. So, first a minor nit to pick. The commonly repeated statement

it’s literally impossible to gain weight by eating your “maintenance” number of calories without violating the first law of thermodynamics

conflates energy balance and mass balance. This is somewhat important if you actually want to predict and measure weight loss. It is true that calories-in minus calories-out must equal the calorie content of the organism. However, not all tissue and body fluid have the same calorie/mass ratio. Thus, it is technically possible to gain mass while having a negative calorie balance.

As a super nerdy aside, mass balance must hold as well. That is mass-in minus mass-out must equal mass of the organism. So when the organism looses mass where does that mass go? Answer at the end.

So, yes calorie balance must hold. Calories-in minus calories-out must equal the calorie content of the organism.

Doesn’t this imply that we can control weight by controlling calories-in?

No.

For several reasons. First, is that calories-in and calories-out are not independent. If you feed a load of sugar to a five year-old you will witness one of the natural calorie balancing processes the body engages in.

Tolerance to a glycemic spike is actually one of the earliest breakdowns in the calorie regulation mechanism and I suspect that it is in part responsible for the mild weight gain that most adults experience and have experienced for generations.

However, even in adults one will find that consistent calorie restriction results in exhaustion, drowsiness, and a reduction in body temperature. Thus restricting calories-in reduces calories-out.

Secondly, you have no direct control over the composition of the weight loss. It could be fat, muscle, bone, hair, blood, etc. In general it is a combination of all of the above. Even in bariatric surgery, which shows great success in fat reduction, muscle loss is common and hair loss is not infrequent.

Muscle loss in particular is counter-productive because muscle is much more metabolically active than fat. Thus a loss in muscle reduces the regular amount of calories expended and thus reduces calories-out.

The combination of these factors is likely responsible for the plateau that most dieters experience where continued calorie restriction does not result in additional weight loss. Surprisingly the plateau appears even in Very Low Calorie diets of 500 kcals a day or less. 

There is an intense debate over whether this could possibly be due to a reduction in the maintenance level to only 500 kcals. I am not sure where I side on this debate but it does seem that the plateau appears.

We should also note that hormone changes can induce changes in weight. The most obvious is human growth hormone. Since the first law must hold in all cases how is it possible that growth hormone can do what only a change in calorie balance can accomplish?

The short answer is that changes in body composition induce changes in calorie intake and expenditure. Teenage boys are famously hungry and they are all adding mass. The same is true for pregnant women.

Could it be that the same is true for the obese? I don’t know how we could rule it out.

Perhaps, a hormonal change induces the body to accumulate fat. The accumulation of fat then increases hunger to satisfy calorie balance. The individual then attempts to eat more. In addition, like pregnant women and teenage boys we should not be surprised that such an individual develops cravings for calorie dense food that would sicken a normal person.

Indeed, that’s one of the observations that makes us think this is hormonal. The behavior of many obese persons is not desirable to the non-obese. Many thin people do not want to eat ultra-high fat food.

Odd desires point to hormones. Again see the behavior of teenagers, both make and female, and pregnant women for notes on the strange desires that raging hormones can induce.

A more important observation, however, is that both thin and obese people seem to know how many calories they “need.” That is both the thin and the obese are typically in caloric equilibrium, neither gaining nor loosing weight. A person will often be 50lbs overweight for years.

How does their body “know” exactly the right number of calories to eat to maintain 50 lbs extra fat. It would be one thing if their weight was constantly fluctuating with the season, the opening of new restaurants, etc.

But that’s not what we observe. We observe either solid gaining trends or solid holding patterns. How does the body know that it needs exactly maintenance plus 500?

Moreover, how does the thin body know that it needs exactly maintenance? Most people have no idea what their maintenance level is and even if they did caloric estimates on food are only accurate to around 90% and not universally available. Are we really to believe that all thin people are counting up every calorie in and every calorie out to make sure they balance? 

No, almost certainly their hormones are telling them. They are telling them they desire food or that they have had too much and now food is repulsive. Their hormones are telling them they really need a hardy steak or they are just “in the mood” for a salad. Their hormones are monitoring body composition and issuing feedback to the brain to tell the brain what to eat.

Thus when someone becomes obese our first suspect should be a breakdown in this feedback loop. This is where I believe that research needs to focus. What is the nature of this feedback breakdown and how can it be fixed?

Answer: When we loose weight were does the fat mass go?

It turns into air and water. Fat is a primarily a chain of carbon and hydrogen atoms. The carbon combines with the oxygen we breathe to produce carbon dioxide, which we breathe out. The hydrogen combines with the oxygen we breathe to produce water, which we urinate out.

Most of the mass, however, is carbon and so we loose weight primarily because the air we breathe out is slightly heavier than the air we breathe in.

Obesity and Genetics

Brad comments

Your post scratches at some scientific-sounding ideas of the involvement of genetics, but fails to cite any definitive research into inheritable weight-control issues. The mathematics behind maintaining a healthy weight are startlingly simple, and the mythology of “fat genes” and thyroid problems and impossible to shed pounds is merely there to sell books and gym memberships.

The body is a closed system and cannot create additional weight without an over consumption of calories. Joel S is correct — it’s literally impossible to gain weight by eating your “maintenance” number of calories without violating the first law of thermodynamics.

So there are a couple of issues here. One is on whether or not genetics can make a difference. Are there “fat genes?” Does the thyroid matter? The second is whether or not First Law solves the obesity question. I’ll do this in two posts.

On Genetics

I’ll post a couple of links on genetics of obesity and one chart

The chart comes from a paper I am writing on obesity and is based off of data organized by Bruce Scaredote and first used in his paper What Happens When We Randomly Assign Families.

The red dots show the the BMI of mothers plotted against that of their birth children.  The blue dots show the BMI of the mothers plotted against that of their adopted children. As you can see the correlation for adopted children is very slight. While the correlation for birth children has a steeper slope.

In both cases we are dealing with heights and weights for all persons reported by the mother. This will tend to introduce measurement error and cause the correlations to be less than the true correlation.

I think its important to note that almost all of the adopted children of morbidly obese mothers, BMI over 40, were normal weight. While the birth children of morbidly obese women tended to be at least overweight if not morbidly obese themselves.

We should also consider the possibility of response anchoring. That is giving a high or low weight for one child will influence the mother’s estimation of the weight for the other child. This will tend to introduce correlation between adopted and birth weights. Which will mean that the two correlations with Mother’s BMI will be closer together than reality. Because the adopted correlation is slight and the birth small, this should lead us to believe that the adopted correlation is even smaller and the birth even larger.

The take home here is that the tendency for whole families to be overweight or normal weight at least seems to work through genetic as opposed to environmental channels.

Other work on obesity and heritability

http://www.ajcn.org/cgi/content/full/87/2/275

http://diabetes.diabetesjournals.org/content/57/11/2905.short

http://www.nature.com/nrendo/journal/v2/n8/full/ncpendmet0220.html

http://www.nature.com/ijo/journal/v32/n7s/full/ijo2008239a.html

Quick Note on Gary Taubes

While I am preparing the First Law of Thermodynamics and Obesity post I thought I’d mention:

A commenter asked why I don’t  discuss Gary Taubes’s book Good Calories, Bad Calories more as it addresses many of the issues I bring up.

My failure to cite Taubes is probably a mistake. Taubes’s lectures on the subject of obesity were instrumental in my intellectual journey. He was the first to shake me from my dogmatic slumber of calories-in vs. calories-out.

Where I break with Taubes is over the solution. Taubes seems fairly convinced on the carbohydrate-insulin hypothesis. I am more skeptical. For me carbohydrate-insulin is one of the leading candidates for a Theory of Obesity but there are a lot of questions it fails to answer.

Not least of my objections is that carbohydrate restriction is not universally and completely successful. It is more successful than calorie or fat restriction but it doesn’t have the track record of say bariatric surgery. A complete Theory of Obesity needs to explain that.

Moreover, we are rapidly learning more about the hormone regulation mechanisms involved in obesity. Leptin was discovered in 1994. The fat regulation effects of grehlin and peptide YY were discovered in 1999 and 2006 respectively.  All three seem to have an importance on par with insulin.

Its important that we don’t hang our hats on a single hormone or regulation chain. We are not even close to uncovering all of the hormones involved. The best we can say right now is that while obesity seems strongly related to hyperinsulinemia we still cannot say for sure what the cause of the obesity epidemic is.

What’s in a Disease

Dan comments

All of this talk of “obesity epidemics”, genetic factors … it’s as if people weren’t deliberately choosing to be obese! They are ALL choosing it. Every time an obese person eats anything other than steamed spinach and egg whites with no salt or butter, every instant they spend on their ample bottom rather than running, swimming, or doing pushups – every instant of self-indulgence is a CHOICE to remain massive.

Suppose there was a condition that required that you had to eat steamed spinach and egg whites only and spend every leisure hour of the day working out or else you would become horribly disfigured, your organs would shrivel and your chance of death would rise. We would rightfully call such a condition a disease and if millions of people had it, it would be a major priority to understand what caused this disease and how to ameliorate it.

At best this is where we are with obesity. Indeed, Very Low Calorie diets often fail to work. Why such failure occurs is a question of hot debate, but the evidence suggest they do.

On the other hand for example, Fen-Phen had an excellent track record. If Fen-Phen treats obesity more effectively than near starvation then its important that we understand why and what the heck is going on.

Brad, you’re on deck for tomorrow.

Keep the comments coming. If you think I am an idiot, off my rocker, don’t know the first thing, etc then please let me know. This is a conversation worth having.

Hoisted from the Comments: Obesity Mystery Already Solved?

Joel S comments

The genetic argument doesn’t hold water: how many of our grandparents were obese? Not many, and they had the same genes.

We search endlessly for a cause for obesity when it is common for an adult man in America, whose caloric requirement (to maintain the same weight with a sedentary lifestyle) might be 2000 or 2500, to be ingesting 4000 or more per day.

There is no mystery: a person who is ingesting only their maintenance number of calories per day will not be obese. Our caloric intake, combined with sedentary lifestyles, explains the obesity epidemic entirely. It’s not sugar, per se, and it’s not fats, per se.

I appreciate Joel bringing these issues up. I never know whether discussing these basics will insult my readers or not.

On genetics: How is it possible that obesity can be strongly genetic if our grandparents weren’t obese?

The nerdy argument is that while the incidence of obesity has risen over time the variability in obesity attributable to genetics has actually remained constant. This is one of the remarkable stylized facts that a Theory of Obesity would need to explain.

In more common terms it works like this. In our grandparent’s day very few people were obese but a few were overweight and in general some were pudgier than others. Well, it turns out that if you carry the genes that produced overweightness in our grandparents time then you are almost certainly obese today. If you carry the genes that produced pudigness in our grandparent’s day then you are very likely overweight today.

The general level of fatness is rising but the relative fatness of people in the population is as genetically determined as ever. Moreover, genetics has a lot to say.

There is a very interesting parallel with height. In the same way as obesity, height has been increasing over time. Yet, height is strongly genetic. Indeed, height seems to be only slightly more genetic than obesity.

Even more fascinating, the rise in height slowed down just as the rise in obesity was speeding up. If you look at the rise in body mass you actually see a smooth trend that extends at least back to the early 20th century. However, for the first part of that trend people were getting heavier because they were getting taller. Now they are getting heavier because they are getting fatter.

One radical hypothesis that I have toyed with is that obesity is the result of improving nutrition. In short, we see that height, intelligence and obesity are rising over time at similair growth rates. We think the first two are due to increases in nutrition. Is it possible the same is true for the third?

Its clear that being obese is undesirable in the modern world but its not immediately clear that its less evolutionarily maladaptive than being 6’2” or having an IQ of 135.  In all cases you have resources going to create tissue that probably wouldn’t have conferred much of an advantage in our evolutionary past.

I don’t endorse this hypothesis, but I think its important to keep even radical suggestions in mind.

On Calories: Don’t we just eat more and exercise less?

We definitely eat more. Its not clear that we exercise less or that sedentary behavior can explain anything. In fact households that make their living doing manual labor are more likely to be obese than those who make their living as professionals.

We do eat more though. Doesn’t that explain it?

No.

The question is why do we eat more. Eating behavior in all animals is actually fairly tightly regulated by numerous feedback loops. Most animals do not “watch what they eat” and most do not get obese. At least not those who are not genetically prone to do so.

More importantly humans in the 1950s and 1960s were mostly normal weight yet most of them were not on tightly regulated diets. Most thin people today probably could not tell you how many calories they ingested each day nor how many they expended. This regulation is carried on subconsciously by hormone mechanisms through out the body.

The regulation mechanism is also highly fine tuned. In order to stay within one pound of your current weight over a year, calories-in have to match calorie-out with 99.7% accuracy. The calorie testing equipment that determines what goes on food labels is not even that accurate, so it seems implausible that people are achieving this through conscious equation of calories-in with calories-out.

So, for thousands of years the regulation mechanism worked despite people living in environments that were quite different from our hunter-gatherer past. However, in the past 30 years it has broken down completely. That begs for an explanation.

For the nerdier, I would also suggest that the sudden breakdown is why I am skeptical of “many independent causes of the obesity epidemic” theories. Why should it be the case that all of sudden lots of independent forces came together all to breakdown the calorie regulation mechanism in the same direction?

That is, if we were seeing just as much spontaneous anorexia as spontaneous obesity then maybe I could buy the “things just went haywire” hypothesis. However, it looks like something particular went wrong with the down-regulation mechanism of calorie management. This was probably a single cause or at least a single complex of causes. Not independent events.

There is a lot more to say but this is enough for one post.