The Food Desert is a Mirage

Let’s recap.  

Since humans developed civilization and had the wealth to allow a small portion of their population to consume excess calories, we have noticed a consistent and near universal trend for members of these groups to have more adipose tissue.  In addition to members of these groups tending to have more Rubenesque or rotund physiognomies, we have also noticed that they tend to have a much higher chance of having particular diseases.  Further, we have noted that similar disease processes can also occur within these historical groups in relation to the consumption of certain “rich” diets, even in the absence of obvious superficial changes in body size.  

In summary, for all of civilization, there is a near linear correspondence on the group level between an increase in adiposity (subcutaneous and we now know also visceral) with the types and amounts of food one ate.  Yet, for almost all of civilization, these changes in body type and health status were reserved for royalty and the very rich.  

Given these historical phenomena, what would happen if the general population ate in the ways that only the rich and royal could previously?  The answer is obvious.  The body types and medical pathologies of the general population would resemble those of the rich and royal of old.

And what would happen if everyone could eat in ways that not only resembled the ancient diet of the aristocrats, but one that had vastly more energy dense delicacies than what they had, combined with all other food also becoming vastly more available?  We would expect that the body types and medical pathologies of the general population would surpass those of the rich and royal of old in both rotundness and severity of disease.  

And this is exactly what we find.  

The prevalence of insulin-resistant diabetes mellitus and obesity (as defined by a body mass index of 30 or more).  Body mass index is calculated by taking someone's weight in kilograms and dividing it by a proxy of their surface area, calculated by height in meters squared.  From a population standpoint, BMI is adequate to determine obesity rates.  BMI is not as accurate in some subpopulations in approximating adiposity, however.  Yet, the primary population subject to this discrepancy are body builders and other advanced athletes.  In these individuals, an abundance of skeletal muscle may register as an increased BMI even without adiposity.  Yet, since this subpopulation comprises a very small percentage of the whole (much less than one percent), BMI is adequate as a marker for adiposity.  

The prevalence of insulin-resistant diabetes mellitus and obesity (as defined by a body mass index of 30 or more).  Body mass index is calculated by taking someone’s weight in kilograms and dividing it by a proxy of their surface area, calculated by height in meters squared.  From a population standpoint, BMI is adequate to determine obesity rates.  BMI is not as accurate in some subpopulations in approximating adiposity, however.  Yet, the primary population subject to this discrepancy are body builders and other advanced athletes.  In these individuals, an abundance of skeletal muscle may register as an increased BMI even without adiposity.  Yet, since this subpopulation comprises a very small percentage of the whole (much less than one percent), BMI is adequate as a marker for adiposity.  

The trend in the United States and the rest of the world is that all demographics are becoming more overweight, more obese, and all demographics are suffering from the diseases associated with excessive energy consumption.  

Increasing body mass from adipose is not just the privy of the rich, or even just the United States.  Adiposity and the diseases that go with it are going up everywhere.  

Increasing body mass from adipose is not just the privy of the rich, or even just the United States.  Adiposity and the diseases that go with it are going up everywhere.  

However, while adiposity and its associated diseases are going up across the board, there has been an inversion of demographics in regards to the obesity and its associated diseases.  While previously it was only the richest who had the honor of storing extra fat, currently it is individuals in the poorer demographics who tend to have the greatest prevalence of high levels of adiposity. 

There is a negative correlation between income and obesity, in Utah and throughout the developed world.   https://ibis.health.utah.gov/indicator/complete_profile/Obe.htm

There is a negative correlation between income and obesity, in Utah and throughout the developed world.   https://ibis.health.utah.gov/indicator/complete_profile/Obe.htm

It should be no surprise that everyone is getting more adipose and more diseases associated with said adipose.  This is the natural consequence of the combination of the human desire for energy-dense food and its easy availability.  

What may (or may not) be surprising, however, is that the people in the poorer demographics seem to be affected by this phenomenon the most.  Why is this?  

One hypothesis for why this is that has gained popularity since the mid-1990s is the “Food Desert Hypothesis”.  

The Food Desert Hypothesis has two main assumptions.  

  1. The Economic Assumption: Even though food availability has gotten tremendously better on average, the availability of high calorie/low nutrient density foods has increased even more.  Since nutritionally poorer calories are cheaper, poorer people will tend to purchase them.  

  2. The Logistics Assumption: Certain areas have a relative deficit of more expensive, healthy foods (such as those described in part 1.2 – recall milk and honey, eggs, pork, beans, rice, etc).  This relative deficit may occur because there are not sufficient food outlets for poorer demographics, and/or the food outlets present do not carry “healthy” foods, and/or the afflicted demographics lack transportation to get to whatever food outlets are present.  Because obtaining healthy foods implies more difficult logistics, poorer people are forced to go to food outlets that do not carry healthy food.   

The “Food Desert” arguments deserve merit because they are an attempt to identify demographic trends so as to hopefully rectify them and improve the fate of people who are frequently marginalized.  

Yet, though it is meritorious to try and help the sick and poor, or to at least understand our struggling fellow humans, the Food Desert Hypothesis should immediately raise suspicion.  If some people in the United States, perhaps even those living close by, are in a Food Desert, then what about the astoundingly less expensive rice, beans, pork, eggs, veggies, fish, and fruit that we spoke about earlier?  If people are opting to get high energy/low nutrient food because of safety concerns, what about the declining rates of getting murdered across the board?  And if transportation is an issue: sure, a lot of people who are poor lack cars, but prior to the early twentieth century, no one had cars.  And if you have to walk, isn’t walking good for you anyway?

Thus, if the Food Desert Hypothesis is correct, it seems to immediately bring up some discrepancies that don’t seem to pass the sniff test.  

And the reason these discrepancies don’t pass the sniff test, and as will be seen shortly, slightly more in depth inspection, is because the Food Desert Hypothesis is wrong.  It is likely that many of those who support the Food Desert Hypothesis do so because they are genuinely trying to understand and help.  Yet instead of helping the poor, this hypothesis has a greater potential to hurt the poor.  Instead of clarifying the real causes of increased obesity in any demographic, the Food Desert hypothesis does little more than rehash unscientific political dogma.  Instead of helping, it confuses and inverts this problem.  

Let us look at each assumption and see why.

The Economic Assumption

The first part of the economic assumption says this:

Even though food availability has gotten tremendously better on average, the availability of high calorie/low nutrient density foods has increased even more.

This is actually partially true.  Most modern delicacies are some combination of mainly four things. Refined sugar, salt, vegetable oil, and refined flour.  These together vaguely sound like a music album that came out in 1991.  Each of these ingredients are currently relatively inexpensive.  None of them have particularly great nutrient density.  Let’s look at them in more detail:

  1. Refined sugar: either from beets, cane, or more commonly from corn in the form of high fructose corn syrup (HFCS).  From whichever the source, this is roughly 50% of the simple carbohydrate glucose (remember from part 1.3?) combined with around 50% of the simple carbohydrate fructose, which is about four times sweeter and likely even more problematic physiologically than glucose.  Glucose attached to fructose makes the dimer sucrose.  HFCS is around $0.28 per pound on average.  As an aside, consider what happened to the population in general in Great Britain when sugar production started to take off.  

  2. Salt: salt was previously so precious that it was considered as to food what precious stones are to jewelry.  To say someone is “the salt of the earth” meant not just that they were wholesome (though it also meant that), but that they were beyond value.  Today, a pound of salt is about $0.60.  Definitely not beyond value anymore.  

  3. Vegetable oil: this is typically from corn or soy.  Less common varieties include from peanut and crystallized cottonseed (aka Crisco) as well as as few others.  These oils are very non-viscous and primarily of the omega-6 polyunsaturated fat variety (more on this in part 2 in which we dig further into the causes of obesity).  These should not be confused with other oils which come from vegetables, which, for one reason or another, are not typically referred to as vegetable oils.  These include coconut oil, olive oil, palm oil, and a few others.  A more appropriate name for soy, corn, and other oils of their ilk is “industrial seed oils” as they were not typically present until an involved process was made (very) commercially viable in the early twentieth century.  Sixteen ounces of corn or soybean oil costs around $0.90.    

  4. Refined flour: this is typically a bleached and ground form of either the dwarf variant of the triticum genus of wheat or of a type of cultured maize (i.e. corn).  Rachel Laudan in her book, Cuisine and Empire, describes the unimaginably tedious process by which our species has tried to make refined grains for nearly all of history.  Her book is excellent.  You should read it. In the meantime, you can buy a pound of bleached, ground wheat flour for about $0.55.  

With these four things (and a few trace other ingredients, like lecithin, artificial flavors and colors, etc) you have everything you need to make Funyuns, Cheetos, Cap’n Crunch, donuts, tortilla chips, and the majority of food products found in any grocery store, whether that store is in the wealthiest or the poorest area.  And indeed, these ingredients have seen the most precipitous drop in price since antiquity.  This price drop is both because these four things are so inexpensive now and because they were among the most expensive things previously.  

Thus, as stated above, the first part of the economic assumption is sort of true.  Most modern delicacies are both sometimes less expensive than the already inexpensive non-delicacy foods (recall a few mentioned in part 1.2) and vastly less expensive than they were before advances in the food industry over the last 150 years.  

However, the second part of the economic assumption is patently false, for both empirical and logical reasons.  

The second part of the economic assumptions states:

Since nutritionally poorer calories are cheaper, poorer people will tend to purchase them.  

The empirical reason why this statement is false should already be clear.  While Funyuns and Cheetos are inexpensive, so are rice, beans, eggs, pork, sardines, and a variety of other nutrient dense foods that humans have used to both survive and thrive.  It should be obvious as seen in part 1.2 that an adult should be able to eat well on 1-3 dollars a day.  This is what I spend on myself typically (though, in truth, closer to $3-5) and I require extra calories given that I lift weights, walk regularly, and all that.  Furthermore, don’t just take the examples I gave earlier or the fact that I do this myself as proof.  The whole misplaced notion that healthy food is more expensive than unhealthy food has been shown to be present mainly because people have been systematically comparing the wrong foods and nutrient profiles.  Consider the graph below and the USDA study from which it is derived.  

Source: www.ers.usda.gov/media/600474/eib96.pdf

Source: www.ers.usda.gov/media/600474/eib96.pdf

For another example of an economic non-sequitur, consider sugar water.  A tremendous amount of unhealthy calories are consumed not as food, but as liquid in the form of soda and other sweetened beverages.  And the demographics with the highest obesity, such as those with low income, drink the most sweetened beverages.  From an economic perspective, this makes absolutely no sense.  If you were on a budget and trying to maximize your nutrient to energy ratio, why would you pay for a drink with no nutrients, loads of energy, and that was relatively expensive, when modern technology has granted us with nearly free, unlimited water?  Of course, you might consume soda if you were worried about dying from caloric deficit, but this is not the case (and I am getting ahead of myself because now I am speaking of the logical absurdity of this argument).  

As an aside, when you ask people about their eating habits long enough, as I do in my medical practice, you’d be surprised to hear how many simply say, “I don’t like water.”  Or, sadly, how many tell me stories of how they either were given soda via bottle as an infant or give it to their infant currently as a parent.  But I digress.  While my experience is anecdotal, there is nonetheless a non-anecdotal or empirical direct correlation of poverty and likelihood of soda consumption.  Just as there is with cigarette consumption.  Like soda having no to negative nutritional value (unless you were starving) and being expensive, cigarettes have negative nutritional value and are on average $5.51 a pack (in Alabama, where I live, at least).  

Furthermore, if it is for monetary reasons that people of limited means seek high calorie/low nutrient density foods, you would expect that they would be eating the least expensive, highest caloric foods at home.  There is a very large markup when buying food at a restaurant.  This is one of the reasons why I cook and eat at home.  I get healthier food for vastly less money.  Plus, eating out frequently requires having a car (costing money), buying gas (costing money), using public transportation (costing money), and it takes longer than making food yourself (costing time, which is like costing money, given the time-value of money).  Yet, those in the lowest socioeconomic strata have seen the largest increase in eating out among all other demographics until the 1990s, after which it has plateaued.  Of course, richer demographics still eat out more than poorer demographics.  Which would be expected, since they have more money.  But, that still should lead one to wonder, why was the eating out trend going up at all if it is so much more affordable to eat at home?  

So, by looking at the data, the Economic Assumption of the Food Desert hypothesis fails.  There are a wide variety of healthy foods that are easy to prepare at home and that are inexpensive.  Further, if economics were a major factor, self defeating and/or expensive habits such as soda consumption and eating out should drop to negligible levels.  Instead, as obesity has gone up, eating out has also gone up or at least failed to go down, and soda consumption has gone up in those populations most affected.   

The logical reason that the Economic Assumption fails has already been alluded to but will be now be spelled out some more.  Again, recall the Economic Assumption:

Even though food availability has gotten tremendously better on average, the availability of high calorie/low nutrient density foods has increased even more.  Since nutritionally poorer calories are cheaper, poorer people will tend to purchase them.  

We have already seen why the assumptions in this statement are either half truths or full falsehoods. Healthy food can be acquired for comparable cost and tedium to unhealthy food.  And, when viewed from a nutrient to price ratio, many “healthy” foods are actually cheaper.

The data shows that the Economic Assumption is wrong.  Now, let’s look at why we don’t even need data to prove the statement to be nonsense.   

The prime rule of logic is that you cannot rationally make both a statement that says “A” and and a statement that says “not A”.  This is called the Law of Contradiction.  

How does this apply here?  Because the dilemma is “why are certain demographics more obese than other demographics (who are also getting more obese)?”  And the stated answer is that unhealthy calories are cheaper (a statement that we have already shown to be misleading and wrong).  Yet, even if healthy calories were always more expensive, the problem is not that poorer people are getting too few calories.  They are getting too many.  

It would thus make perfect sense if the outcome of healthy food being more expensive was to have poorer people consuming fewer calories so as to get the less energy dense, but healthier foods.  But this is not what we are seeing.  We are seeing poorer demographics spending more on foods with little or no nutrients in the context of already having too much body fat.  

Let’s make this look fancier.  We cannot have both “A” and “Not-A”:

A: People are seeking nutrients.  Only inexpensive food has a higher energy to nutrient ratio than expensive food (we’ve already shown this to be false, but let’s go with it).  Thus poor people get too many calories in the search for nutrients.  Since people want nutrients, this is why they get obese.

Not-A: People are seeking nutrients.  Only expensive food has a high nutrient to energy ratio (again wrong but, again, go with it).  Thus poor people get too few calories in the search for nutrients.  People want nutrients.  This is why they get lean.  

Both A and not-A are fair logical deductions.  If we choose A, we must also choose not-A.  

But we cannot have both A and not-A.  This is thus a logically invalid argument.  Since its premises are also false, it is both invalid and unsound.    

So much for the Economic Assumption.  Now let’s consider the Logistics Assumption.  

The Logistics Assumption

In review, the Logistics Assumption is that certain geographical areas and groups of people in different geographic areas have a relative deficit of access to healthy foods.  These relative deficits may occur because there are not sufficient food outlets, and/or the food outlets present do not carry “healthy” foods, and/or the afflicted group of people lack transportation to get to whatever food outlets are present.  This assumption figures that because obtaining healthy foods implies more difficult logistics, poorer people are forced to go to food outlets that do not carry healthy food.   

Yet, the logistics assumption simply does not play out in real life.  In antiquity and even relatively recently, there truly were food deserts in most of the world.  Even in many parts of the globe currently, it is a several mile trek in war-torn areas just to get dirty water.  

But there is a big difference between the past and the present in developed countries and between the present in developed countries and the present in war zones and kleptocracies.  Trekking across a war zone to get dirty water means you likely live in a food desert.  Having to walk past a 7-11 to get to the grocery store does not mean you live in a food desert.  It is not a food desert unless your main goal is to call something what it is not in order to make a play on words and fool people.  

One of the ideas behind a food desert is that poorer or more obese people have a lower concentration of supermarkets around them.  The idea is that supermarkets tend to have healthier food (like fruits and vegetables).  And while certain academic papers pronounced this to be the case, these studies had major methodological errors.  Plus, they fly in the face of common sense (for instance, you can actually can eat healthfully at 7-11).  For instance, in 2012, researchers in California found:

“…no evidence to support the hypotheses that improved access to supermarkets, or that less exposure to fast-food restaurants or convenience stores within walking distance improve diet quality or reduce BMI among Californian youth. There are isolated significant coefficients, but would be expected due to chance.”

Another 2012 study, looking at nationwide data, found that while poorer demographics tend to have more fast food restaurants, they also have more grocery stores.   

“…children who live in residentially poor and minority neighborhoods are indeed more likely to have greater access to fast-food outlets and convenience stores. However, these neighborhoods also have greater access to other food establishments that have not been linked to increased obesity risk, including large-scale grocery stores. When examined in a multi-level modeling framework, differential exposure to food outlets does not independently explain weight gain over time in this sample of elementary school-aged children. Variation in residential food outlet availability also does not explain socioeconomic and racial/ethnic differences. It may thus be important to reconsider whether a salient factor in understanding obesity risk among young children.”

So, healthy food is in fact more available in poorer areas, on average, than in wealthier areas.  

That’s all well and good, but what if the problem isn’t that the healthy food is far away?  What if the problem is that it is close, but that for one reason or another you can’t get to it?  One commonly made argument in this regard is that poorer people lack easy transportation.  Yet, if transportation was the issue, the rise in eating out could not have occurred.  Further, if transportation was the issue, than that should effect soda consumption and all food consumption, as these need to be purchased from somewhere.  Also, if in reality supermarkets are as prevalent as fast food and convenient stores in poorer areas, then it should be as easy to get to these locations as it would be a convenient store and the whole topic is moot.  

A related argument to transportation availability is transportation safety.  But, as alluded to earlier, not only is transportation easier now, nearly all trends of criminal violence and homicide are going down.  And while there are certainly discrepancies and setbacks, this is the case even in poor neighborhoods.  Further, the transportation safety argument fails in the same way that the general transportation argument fails.  If it is dangerous to get the main course, it is equally dangerous to get the desert.  This doesn’t therefore explain why people opt for the desert.   

The Pretense of Rationality

There is one more essential point that needs to be made about the two main assumptions of the Food Desert Hypothesis.  

We have implicitly granted in some of the above arguments that people choose food based on some type of consideration for what is healthy.  Yet everyday experience of course tells us that this is usually not the case.  

Most people do not consistently buy or eat food for nutrient content.  And most people do not consistently buy or eat certain foods because they are the most convenient or available.  Most people eat certain things because those things taste or make them feel good when eaten.  Hence soda, chips, ice cream, Funyuns, Pop Tarts, and Cap’n Crunch.  Hence the relative absence of kale and sardines.  Regarding the latter, as you likely know, almost every gas station and 7-11 sells sardines.  But only I buy them.   

Besides its bad data and logic, perhaps the most fatal flaw of the Food Desert Hypothesis, and as will be seen with the Food Insecurity Hypothesis, is that it assumes that people act for rational reasons.  But as almost everyone already knows, people, by and large, do not act for rational reasons.  Sure, it helps if you understand the flaws of the concept of the Efficient Market Hypothesis, or if you have read the works of the psychologist Dan Kahneman, but all it really takes is that you interact with other people on a regular basis.  Interact with any group of people (rich, poor, or in-between) and you’ll quickly learn that the default status for most of humanity is a thought process that is largely subcortical and has little to do with cool, calm, logic (or, as Kahneman would say, we are usually in the system one mode of thinking).  Consider unprotected casual sex, alcohol abuse, cigarettes, bar fights, lottery tickets, daytime television, gossip, and almost the whole gamut of human action except the very rare cases when we actually do act rationally.

To skip ahead to part 2 in which we talk more of the subtleties of obesity, the real reason people eat too much of the wrong food is because people like pleasure.  The part of our bodies, particularly the brain, responsible for the sensation of pleasure is called the hedonic system.  We will talk a lot more about this later.   

In summary, the main reason people buy unhealthy food has little to do with economics.  It has little to do with logistics.  It has nothing to do with a Food Desert and nothing to do with Food Insecurity.  It has everything to do with pleasure.  

Oasification  

We live not in a Food Desert but a Food Oasis.  And the world is becoming ever more oasified (a word I just made up) over time.  Our problem is not one of scarcity but one of abundance.  The consequences that we are seeing in regard to body size, composition, and health are the clear results of abundance, not scarcity.    

And now we are pretty much done talking about Food Deserts.   Let’s close by talking about the somewhat related topic of the misplaced notion of Food Insecurity.  

 

Click here to continue onward to part 1.5, Homeland (In)security.