Mark Palko quotes Justin Fox:
On Monday, software engineer Rob Rhinehart published an account of his new life without alternating electrical current — which he has undertaken because generating that current “produces 32 percent of all greenhouse gases, more than any other economic sector.” Connection to the power grid isn’t all Rhinehart has given up. He also doesn’t drive, wash his clothes (or hire anyone else to wash them) or cook anything but coffee and tea. But he still lives in a big city (Los Angeles) and is chief executive officer of a corporation with $21.5 million in venture capital funding.
That corporation is Rosa Labs, the maker of Soylent, a “macronutritious food beverage” designed to free its buyers from the drudgery of shopping, cooking and chewing. In the 2,900-word post on his personal blog, Rhinehart worked in an extended testimonial for Soylent 2.0, a new, improved version of the drink — algae and soy seem to be the two most important ingredients — that will begin shipping in October.
Fox’s piece is headlined, “Soylent Is Weird, But It’s Good Weird.”
But is it really “good weird”? Or, if so, what kind of “good” is it?
According to Palko, Soylent is not so nutritious.
Here’s the comparison of 115 grams of Solyent:
to comparable servings of black beans:
and nonfat Greek yoghurt:
And I think it’s safe to say it’s not so delicious.
Nor is it so amazingly convenient. Palko writes:
Nor do you have to cook to do better than Soylent. I did a quick check at the grocery store last night and I found lots of frozen entrees that gave you more nutrition for less calories than Rosa Lab’s product.
In summary:
Basically, when you cut through all of the pseudo science and buzzwords and LOOKATME antics, Rhinehart is simply peddling a mediocre protein shake with the same tired miracle food claims that marketers have been using since John Harvey Kellogg gave C.W. Post his first enema.
The paradox . . . or is it?
At first this seems like a paradox . . . Silicon Valley genius, $21 million in venture capital funding . . . how could it be just a scam?
But then you realize that nutrition has nothing to do with it (other than as a marketing concept).
Recall that the goal of the people who invested 21 million dollars in this product, is not to give people healthy and satisfying meals, it’s to have the image of something healthy and satisfying.
Is Soylent a scam? Yes and no. It’s a scam to the people who are being sold the product, but maybe not to the investors.
Perhaps the whole Silicon Valley thing is a distraction, and the right analogy is to something like the movie Battleship, which was universally agreed to be crap but still sold jillions of dollars worth of tickets.
So, when business writer Justin Fox writes that Soylent is “good” and that it is “an interesting product,” this would be like a movie reviewer saying that Battleship is a good movie. It was good to its investors, I assume!
And for a business writer to credulously take Rinehart’s word on the health benefits of “macronutrient balance” and “glycemic index” of the products he’s selling, without just going to the supermarket and comparing to the label on a can of black beans and a tub of yoghurt.
But is Solyent a good model for a business? I guess that depends on whether potential consumers view it as a sugary, fatty, bad-tasting alternative to beans and yoghurt; or as a healthy processed-food alternative to a breakfast of cornflakes and Coca-cola.
And that in turn must depend on part on press coverage. As Palko has written elsewhere on his blog, A Statistician Walks into a Grocery Store, journalists typically don’t seem to have a good framework for writing about food and nutrition, especially when it comes to low budgets.
So, in that sense, the credulous news reports on Soylent (and it’s not just Justin Fox; see, for example, this gee-whiz article by Lizzie Widdicombe in the New Yorker, subtitled, “Has a tech entrepreneur come up with a product to replace our meals?”) are just part of the larger picture.
Food and nutrition reporting have little context. Imagine if entertainment reporting were the same way:
Battleship: The Hamlet for the 21st Century
I’m not going to argue that anyone should eat Soylent all day and nothing else, but the point of it is to in fact do that. It looks like they intend for a person to have 4 servings of Soylent a day, which is 2000 calories. If you had 2000 calories of the black beans + yogurt, you would have something like 10% of your recommended fat intake. And it’s recommended for a reason: there are essential fats that the body needs to work properly (https://en.wikipedia.org/wiki/Essential_fatty_acid). It isn’t obvious to me that beans and yogurt all day is the better option. But since the average US male needs more like 2500 calories a day (and actually eats more than that), you’d probably lose weight either way until you got sick of eating only one thing all day every day and went crazy on some cake.
Yeah, without offering nutritional expertise here, Palko and this post seem to be notably missing the angle of Soylent. The entire point is that one serving of Soylent provides a proper nutritional balance (per calorie) at minimum cost of money and time (per calorie). Cutting out the fat (which is what causes Soylent to “score” low in Palko’s conceptual nutrition-per-calorie metric) would yield a less nutritionally-balanced product that is less effective at its overall goal. I can’t evaluate the critiques of the professors he quotes, but those actually address what the product is trying to do, unlike his whole post and embarrassing snake-oil comparisons.
(The 2nd reply is a different Alex)
Without knowing its micronutrient composition it’s impossible to say how nutritious the product is, but based on the partial information included on the nutrition label, the the product has good points, namely, a high unsaturated:saturated fat ratio, a high potassium:sodium ratio, and protein adequacy without excess. However, one nutrition scientist quoted by Palko states that the product would have a high glycemic index, whose potential problems can be googled.
What do the dietitians have to say about Mark Palko’s black beans and yoghurt diet? That’s what I want to know.
From http://io9.com/could-soylent-really-replace-all-of-the-food-in-your-di-510890007
Susan Roberts, Professor at Tufts University’s Friedman School of Nutrition Science and Policy, likened Soylent to already available nutritional shakes. While there might be some benefit to Soylent’s low saturated fat content, she said, there are certain risks inherent in a non-food diet. “[T]here are so many unknown chemicals in fruits and vegetables that they will not be able to duplicate in a formula exactly,” she said in an email. She says that, if Soylent is formulated properly, a person could certainly live on it, but she doubts they would experience optimal health. She fears that in the long-term, a food-free diet could open a person up to chronic health issues.
Tracy Anthony, Associate Professor of Nutritional Sciences at Rutgers University, speaking to us in an email, criticized the formula specifically:
[T]he fast-digesting soluble protein source (whey) in combination with the simple sugar source (maltodextrin) creates a product that is very high glycemic. This product would not promote healthy glucose control, satiety and cognitive well-being with insulin spiking at each point of Soylent consumption.
She also echoed Roberts’ sentiment, that ingesting the minimum nutrition required by the human body is not the same thing as maintaining a healthy diet. Could a person live off of Soylent for a while? Certainly, she says, for quite a while; but that doesn’t mean that they are taking in the nutrients necessary to prevent disease, manage disease, or live a long life.
I believe this product has no whey in it (contrary to what Prof. Anthony is quoted as writing), though I have no intentions of drinking it.
No one is suggesting a beans and yoghurt diet. See the reply to Phil below.
> No one is suggesting a beans and yoghurt diet.
That is, unless you share an office at work and wish to induce your officemate to move out.
If you are not suggesting a beans and yoghurt diet, why did you include it? The only logical purpose that point could serve is as an example of a cheap simple diet healthier than Soylent. If that’s not what you meant then you’re implying that your entire post is an overblown, illogical sham.
It’s pretty clear that both you and Andrew have set aside your statistical expertise on this issue and are instead trying to rationalize your feelings that Soylent is weird/stupid or whatever. Otherwise we all know that you wouldn’t be committing obvious logical errors, such as comparing apples and oranges (food replacement vs. one serving of beans) or ridiculous nirvana fallacies (if someone doesn’t eat Soylent, the alternative is that they eat the optimal balanced diet recommended by nutritionists).
I’m a busy college student who doesn’t enjoy cooking. I lived on mostly Soylent for more than 6 months, from Soylents 1.1 to 1.3. I can assure you that it was better than the alternative. The only deficiencies I ran into was that it didn’t include enough fat or salt or protein. The fat and salt part, I presume, is because they had assumed people would be eating fatty and salty foods in addition to Soylent. The protein problem was a result of me working out regularly (when I slacked off it wasn’t an issue).
So once every few days or so I did have to eat non-Soylent foods. Which is fine.
The only reason I stopped is the price. One package (now 4 servings, it used to be 3) is about 1 day of Soylent for me. At current prices that’s $10 a day for Soylent. I can’t justify that on my budget, so for now it’s back to my alternative cheap diet, which is uncomfortably close to your black beans and yoghurt.
So Soylent does have some issues worth writing about, if you could set aside your prejudices and use rigorous statistics. Right now I think nearly everyone who has actually used Soylent is cringing at how much both of you are making fools of yourselves (especially because I respect a lot of Andrew Gelman’s work, and own two of his books, and use some of his work in my own work).
And you’d be amazed by how much time you can save by gulping down a few cups of Soylent for dinner. It’s a mind-bending, and ultra-convenient, experience.
Will:
You may be right here. I’ll let Palko respond on the details (if he’d like) but I gotta say, I feel much more confident on my recent Breaking Bad post than on the topic of the health benefits of Soylent.
My impression from reading Palko on this is that he is interested in Soylent as an example of media hype, where journalists run uncritical stories about some high-tech innovation (whether it Soylent, Mars One, charter schools, moocs, or Netflix) while showing a remarkable ignorance about available low-tech alternatives (in the Solyent case, this being what is readily available in discount supermarkets). But that said, even if Soylent is being hyped uncritically by some, that doesn’t mean it has to be an out-and-out scam.
The reporting reminds me of the late great “Chocolate is good for you” hoax. Don’t reporters have any ability to do some critcal thinking any more?
Solyent may not be, strictly speaking, a scam but it is silly/stupid and likely dangerous. Might there be a reason why no one over the age of about 1 year lives on fluids alone?
I imagine most people would be a lot better off with a baked potato and a bit of cheese (or milk). Except, gasp, they would actually have to cook the potato. Most of the population of Ireland lived almost exclusively on potatoes and whey during much of the 18th Century and a goodly part of the 19th. Probably a much more balanced,if almost as boring) diet than this Solyent.
It’s a smoothie or shake for people with more money than sense. A bit like all those people who buy bottled water when they don’t need to.
“I imagine most people would be a lot better off with a baked potato and a bit of cheese (or milk).”
Wrong!
If you had read the first comment you might have caught the fact that Soylent is meant to be a completely balanced meal no matter what size you eat. A baked potato w/cheese is not remotely balanced.
Disclaimer: I use Soylent as a meal replacement at work to ensure that I never miss a lunch and sacrifice my health/diet due to work constraints on my time.
Like Alan and James, I have a friend who drinks Soylent at work. He claims that although it’s probably not optimally healthy, it’s certainly healthier than what he was eating otherwise, which sounds about right to me. counterfactuals, counterfactuals…
If you had read the first comment you might have caught the fact that Soylent is meant to be a completely balanced meal no matter what size you eat.
Oh, I did. That does not mean I believe it. Nor, apparently, do some experts whom Mark Palko consulted.
And you have any evidence of your assertion that A baked potato w/cheese is not remotely balanced?.
I certainly am no dietitian but just for a minor point I’d refer you to something as simple as a paper by William McNeil, How the potato changed the world’s history pg. 75 regarding the standard Irish peasant diet. Again McNeil is not a dietitian but a distinguished historian who, apparently did his dissertation on the subject : In fact, a single acre of potatoes and the milk of a single cow turned out to be enough to feed a whole family; and such a diet, however monotonous, was nutritionally adequate to sustain what became an exceptionally healthy, vigorous (and desperately poor) rural population. http://www.jstor.org/stable/40971302
I don’t think that the Irish poor subsisted totally on potatoes and milk but, more likely, potatoes and milk made up 90 to 95% of their diet, probably supplemented with a small amount of fruits, vegetables, seaweed (dulse), and perhaps some fish–depending on game laws/poaching skills and/or how close one lived to the sea.
A couple of years ago I ran across a report to parliament about potato consumption in rural Ireland in the early 19th C (probably in the 1820’s IIRC) where it was reported that a reasonable quantity of potatoes per day for a healthy adult male was approximately 1.0 stone (i.e. 14 lb or roughly 6.5kg). I’suggest that potatoes were a key element in the diet.
I use Soylent as a meal replacement at work to ensure that I never miss a lunch and sacrifice my health/diet due to work constraints on my time.
This sounds like an excellent use of the product although the failure to take time for a decent lunch break does not say anything good about the (I am assuming where you live) North American work culture where we are expected to do this.
Again I am not a dietitian and really cannot vouch for the following link but it is interesting.
http://nutritiondata.self.com/facts/vegetables-and-vegetable-products/2770/2.
Maybe Silicon Valley genius millionaire is less than what it seems. I have been tossing 57 pennies, and one of them came up heads six times in a row, but I am unwilling to come to the conclusion that it is a heads-only penny. Many people have tried entrepreneurial undertakings. Few made it to millionaire status. Some who made it were actual geniuses, and some got through the sieve because of being in the right place at the right time. Each new endeavor by anyone should be looked at as a new toss of the coin. It is unwise to place too much confidence in someone based on their past performance especially in an unrelated field.
“And for a business writer to credulously take Rinehart’s word on the health benefits of “macronutrient balance” and “glycemic index” of the products he’s selling, without just going to the supermarket and comparing to the label on a can of black beans and a tub of yoghurt.”
This quote bugs me Daniel, as you obviously didn’t take the time to research Soylent and understand that it didn’t start as a VC topic, but as a kickstarter that validated the business model before any VC firms were involved.
“But is Solyent a good model for a business? I guess that depends on whether potential consumers view it as a sugary, fatty, bad-tasting alternative to beans and yoghurt; or as a healthy processed-food alternative to a breakfast of cornflakes and Coca-cola.”
This quote also bugs me because it shows you don’t actually have an understanding of nutrition. I’m not an expert but trying to live off of beans and yoghurt would mean your intake would consist of way more sodium than you should have and essentially no fat, which you need. Here’s the actual link to Soylent nutrition facts, which also includes vitamins and minerals not discussed in either article.
http://files.soylent.com/pdf/soylent-nutrition-facts-1-5-en.pdf
Alan,
The sodium point seems to be causing a lot of confusion. Read the reply to Phil below.
Andrew,
Those tables of nutritional information should really be graphs!
As Alex says, it makes no sense to compare a 500-calorie serving of something to a 110-calorie serving of something else. For example, if you eat/drink 2000 calories of Soylent, you’ll get about 65% of your RDA of sodium; if you get just HALF of your daily calories from those cans of black beans, that’s 135% of your RDA of sodium…and that’s bad (sodium is one of the things where it’s better to be under the RDA than over). I wouldn’t call these “comparable” at all, in spite of Palko’s claims.
But more importantly, Soylent evidently really does have a better nutritional balance than a diet like beans and yogurt would give you. nutritiondata.self.com has what claims to be a nutritional profile. Black beans have no vitamin C, D, E, K, or B12, and 1000 calories of black beans has less than 30% the RDA of Riboflavin, Niacin, B6, and Pantothenic acid. (And if you’re looking for yogurt to bail you out, forgettabout it: according to that nutritiondata site, Breyer’s low fat strawberry yogurt has no vitamin C, D, E, K, Thiamin, Niacin, B6, Folate, Pantothenic acid, etc.)
Whether Soylent is worthy of any attention, I don’t know (I, for one, do not intend to start eating it); but Palko’s specific beans-and-yogurt argument is full of beans. In fact, it’s so bad that you can’t even say he’s ‘lying with statistics’ since on the whole he doesn’t even give statistics to support his claims about the near nutritional equivalence of yogurt and black beans. Those claims are false, at least if we believe the nutritiondata site. I also question Palko’s claim that he was easily able to find frozen entrees that provide nutrition as good or better than Soylent; I searched around nutritiondata a bit and was unable to find anything with a vitamin and mineral mix that came anywhere close to providing as close a match to recommended allowances. Indeed, I was a bit surprised and disappointed to see that a strawberry banana smoothie is deficient in so many ways!
And finally, Palko cites Susan Roberts of Tufts as saying Soylent is similar to other nutritional shakes. She’s in a position to know, so I think that’s probably true, but how big an indictment of a product is it to say “it’s similar to other products”?
Soylent may be overhyped, it may taste bad, it may even not be all that good for you (although it’s hard to believe it would be worse than the typical American’s diet), but I’m much more appalled by Palko’s coverage of it than of the product itself!
Phil,
Sorry I wasn’t more clear on this but no one is suggesting a “beans and yoghurt” diet. That would be insane. It would leave huge nutritional gaps, it would be unappealing and it would actually have too much protein and fiber. Having spent an entire post railing against superfoods and mono-diets, I certainly wouldn’t propose one of my own. These foods are a good part of a balanced diet but the operative word is balanced.
If we were talking about a mono-diet, most of your points would be valid except for the point about sodium. There seems to be some confusion on this point but those numbers come from salt added for flavor. You can get beans without the added salt. Some people even prefer them.
Finally, when citing Roberts, you left out the most important part:
‘While there might be some benefit to Soylent’s low saturated fat content, she said, there are certain risks inherent in a non-food diet. “[T]here are so many unknown chemicals in fruits and vegetables that they will not be able to duplicate in a formula exactly,” she said in an email. She says that, if Soylent is formulated properly, a person could certainly live on it, but she doubts they would experience optimal health. She fears that in the long-term, a food-free diet could open a person up to chronic health issues.’
Mark,
I guess reading Andrew’s post set me up to read things into yours that you didn’t intend. You say in this comment that no one is suggesting a “beans and yoghurt” diet and I’m sure that’s true, but Andrew did say (right in the title!) that beans and yoghurt are better than Soylent, and you did imply that bean soup plus yoghurt is at least in the same ballpark as Soylent when it comes to nutrition. And you did say that about some frozen foods you were able to find…which may be true, but I wasn’t able to find such foods in a quick search.
Saying someone eating Soylent probably won’t experience ‘optimal health’ and could be open to ‘chronic health issues’…those just seem like very weak claims that I easily believe, but so what? You could say the same thing about everyone who eats fast food five times a week, or doesn’t eat enough fruits and vegetables, or is 50 pounds overweight, and I’ve just described half the American population!
To sum up: (1) If I read too much into your “supplement your bean soup with yoghurt” comment, I apologize…but I also think you need to take some responsibility for this interpretation. By juxtaposing those food labels you are implicitly comparing them straight up, without noting the factor of 4.5 difference in calories, and I do think you’re implying that bean soup + yoghurt is as good as Soylent (although that is not the same as saying it’s a complete diet!). And (2) I certainly believe Soylent isn’t “optimal” and I can’t imagine eating it myself except for maybe trying it once out of curiosity, but there’s nothing in your post to support the claim that the Soylent guy is a ‘snake oil salesman.’ I’m not saying he isn’t, I’m just saying you haven’t given us evidence for it. Maybe his malfeasance would be more obvious to me if I knew something about Soylent. Finally (3) I think Andrew is way too quick to jump to conclusions in _his_ post! Soylent appears to be a product that delivers a pretty comprehensive mix of most of the vitamins and minerals for which there’s a recommended daily allowance. It may not healthy to live on it all by itself, but based only on what I know (which isn’t much) I wouldn’t call it a scam.
Phil,
The difference is that the chronic health issues of fast food & gross obesity etc. are well known.
Soylent is pitching itself as health, perhaps uber healthy. It’d be like drawing an equivalence between a medical food supplement & a greasy corn dog.
There’s a difference between making no pretenses of healthiness versus making questionable claims.
Are the claims questionable, or are they definitely wrong? In your post, you seem to me to come down heavily on the side of Soylent being bad for you. Andrew amplifies that even more. But neither you nor Andrew provide any evidence for that. The closest is a nutrition researcher who says that eating NOTHING BUT Soylent will probably not be optimal.
From the information I was able to find in a quick search –and I found it in the Internet so you know it’s true! — Soylent looks like it provides pretty close to a balanced diet, which cannot be said of the typical American meal. I think if you want to call the guy promoting it a snake-oil salesman you need more than “if you don’t eat anything else at all, it might not be optimal.”
I should say again that I don’t know what claims the Soylent guy makes so he may indeed be spouting loads of BS. I just haven’t seen it.
Oops, I said “your post”, I meant Palko’s post.
Mark: I agree with Phil; I’m struggling to see what point the black beans and yogurt examples serve after your clarification. If the point is “a balanced diet is nutritionally superior to Soylent,” which I agree with, surely pasting the nutrition facts of two foods that are, by themselves, unbalanced, doesn’t add much to that point? It’s hard to interpret “For less than a fourth of those calories, here’s what you get with a bowl of black beans,” as fair to Soylent’s design: it’s not a weight loss shake, and *of course* isolating the nonfat part of any balanced diet is going to “buy” more nutrition per calorie.
It seems to me that heuristics are leading the discussion astray. Most products in the vague category of processed meal replacements do, in fact, make overheated health claims, hide behind pseudoscience, and claim to be healthier than normal balanced diets without them. But as far as I can tell, Soylent does nothing of the sort, and therefore most of the post serves as a takedown of something that doesn’t really exist. You say “Rhinehart is simply peddling a mediocre protein shake with the same tired miracle food claims that marketers have been using since John Harvey Kellogg gave C.W. Post his first enema.” These are strong words. Where are these claims? The CEO describes Soylent’s goal as “seeking to provide full daily value requirements in a sustainable fashion,” which seems to me to be a) literally accurate, in that the product is explicitly designed top-down to *try* to do that and b) as far away from a “miracle food claim” as I’ve ever seen from a company.
IMO you should strongly reconsider whether the personal attacks on Rhinehart you include in your post are appropriate and fair.
I love the site at self.com. On the page for any particular food, it has a feature that suggests things to eat whose nutrition profiles are complementary. I went there one day when I’d harvested a lot of kale from the garden, and was looking for ideas. It suggested a small pepperoni pizza.
This is a confusing post. I drink Soylent because it’s very convenient (important for depresso fucks like me who may not always have the executive function to make a frozen meal) and because it’s definitely healthier and not much more expensive than what I’d actually be eating instead. I sigh at and then ignore virtually all press coverage of it, positive or negative.
In what way am I being scammed, Andrew?
So do they make Soylent without using any electricity either?
Probably not, but I bet they would if they could.
1. Who wants not to eat? Eating is one of the great pleasures of life.
2. There is no one food that’s sufficient. If there is, it’s probably Caribou meat (i.e. Inuit diet in winter is restricted to a small number of sources).
3. J H Kellogg and the early Battle Creek guys probably did more good than harm. It’s not JHK who decided to put sugar on everything, and degrade those whole grains into white flour. [Still, it’s hard to get past those enemas at the Sanitarium. And there certainly have been / are a lot of food scams.]
That’s a good point about Battle Creek. They had a tendency to make excessive pseudo-scientific claims about some of their food but many of their suggestions weren’t bad at all.
1: If your diet is restricted (main experience for me: picky eating) in some ways, eating can be more of a stress factor than a pleasure thing. If you don’t have time or simply do not want to worry about food, coordinating food with other people or any other reason why food isn’t always a pleasure. From what I understand all-Soylenters do social eating like sunday roasts or whatever still, they just dont deal with everyday food choices and stress.
If your dietary restrictions make your diet rather unhealthy, this can also at least improve your situation.
A world where food has been replaced by a food-like-substance (¿”Fud”?) is boilerplate dystopian novel/movie stuff. If you have to eat Fud, you are probably living in a terrifying fascist future where everything has gone to hell. I mean, can you invent a more obvious super-villain than a rich white guy with bionic implants* who spews overblown rhetoric about how the world is heading towards self-destruction and wants to convince you that he can save it?
Also, the difference between Soylent and Battleship is that Rihanna is very, very pretty.
* “The walls are buzzing. I know this because I have a magnet implanted in my hand and whenever I reach near an outlet I can feel them. I can feel fortresses of industry miles away burning prehistoric hydrocarbons by the megaton. “
I’ve often thought that it would be a great idea to have the USDA or someone like that create some kind of “standard healthy diet”, but primarily for the purposes of dealing with inflation, and cost of living.
Suppose for example that the USDA figured out and cooked say 10 breakfasts, 10 lunches, 10 dinners all together they’re enough to feed some kind of theoretical family of 2 healthy adults and 2 children of median child-age for 10 days. Then they list out the ingredients you’d need to buy, and some economic bureau could go price that basket of food at different stores in different regions, and then attach the prices to the lat/long or UTM location where the store is, and then publish that dataset in Sqlite3 format.
jrc, do you think we could write this as a grant? ;-)
Perhaps it would be good to also subset the ingredients by “type” such as “fish”, “chicken”, “pork”, “beef”, “green vegetables”, “root vegetables”, “condiments”, “dairy”, “whole grains”, “baked goods”… or whatnot
and then give both the total mass for the basket, total price for the basket, and a breakdown of the fraction of the mass for each type, as well as a fraction of the basket price for each type…
the national average cost for the basket * 3 would be a good starting point for a monthly Universal Basic Income per adult over 18, I’d argue ;-)
Daniel,
That is actually not too too far from what our inflation index is actually based on: “FOOD AND BEVERAGES (breakfast cereal, milk, coffee, chicken, wine, full service meals, snacks)” along with a bunch of other stuff: http://www.bls.gov/cpi/cpifaq.htm#Question_6
It also plays some role in the Purchasing Power Parity comparisons of prices across countries.
More and more as I think about nutrition measurement though (as an economist), I think it is basically a lost cause as either a good measure of “productive inputs” into humans or as a standardization measure (><= 2000 calories, or whatever). Specific changes in diet/intake – those might have measurable effects on health and human capital outcomes in the long term, but teasing apart the whole "basket" is just impossible. The idea that all calories, or all micronutrients, or all sugars are the same is just too wrong to be a useful model.
Oh, and related to your last comment, the $2/day international poverty line is sort of based on thinking like that: enough to feed/shelter a family. But really it is just a convenient number that people can understand and is probably within the ballpark of what is needed to not die (or at least not die very young or right this minute).
All that said – food stamps have essentially replaced cash welfare in this county, so some analysis of the type you suggest that looks at how well food stamps allow you to get your "standard healthy diet" – if at all – might be a pretty interesting study. I guess we could go ask Bill & Melinda for some $, but I think poor Americans are still too rich for them to be interested in.
The CPI average over the country, or even over states, or even over metropolitan areas, is a far poorer dataset than a spatial dataset for a large selection of stores. There’s a lot of talk about “food deserts” and there’s a lot of variability in cost of basic nutrition from place to place. That’s what I’d really be interested in.
I think the point of nutrition based indexes is that there’s a big difference in outcome between not even being able to get your required calories, vs getting 2000 calories a day in the form of the very cheapest food you can find vs 2000 calories in the form of something considered quite healthy, so I’d expect a lot of marginal productivity value for a dollar in that range. As a denominator in a dimensionless ratio for income, it seems like a very good one.
also, are you kidding me, the foods they track don’t even make up a single decent meal, except maybe breakfast cereal + milk + coffee.
My index would be things like: “salmon steak + broccoli + flavored rice + 1 glass white wine” and “breaded chicken + baked potato + salad (lettuce, bell pepper, tomato, olives, red onions, olive oil + balsamic vinegar and one bottle of beer” or “oatmeal + raisins + pecans + maple syrup” or “grilled chicken burrito with rice, beans and avocado”
The point is to measure how to eat healthy, not to see if there’s any fluctuations in the price of a few commodity goods you can get in large sacks off a boat :-)
Hmmmm… I’m starting to see your point and the value of your index. The project would be pretty high on the cost, so would have to be pretty high on the benefit. You might be right that it is a good idea in principle, I just don’t know about in a cost-benefit sense.
I’d add: your measurement requirements are going up. Now you need specific food prices, and very small geographic levels, and you need transport costs for middle-low income people in each of those areas. In some places, everyone has a car, and 5 miles to the grocery doesn’t matter. In others, 2 miles might take 2 different buses and 3+ hours, since you can’t walk home 2 miles lugging groceries.
I guess I am starting to see the value of your index, but I’m more and more not seeing the feasibility. Maybe that means I lack imagination (after all, I think the UC has scanner data for like a decade from a major Supermarket chain in CA… or something like that, so I hear), but no doubt it is a huge project. I’m trying to think of what I would do with that index once it was created… I only have some vague ideas at the moment.
I’ve been playing around with this for a while here in LA
http://astatisticianwalksintoagrocerystore.blogspot.com/2015/05/journalists-cant-shop-i-was-way-too.html
http://astatisticianwalksintoagrocerystore.blogspot.com/2015/05/black-bean-soup-recipe.html
and I can tell you that there is tremendous store-to-store variation that will be lost in the aggregate. For example, a dish of chicken, carrots and rice might cost about the same at Ralphs and Trader Joes but you could knock maybe fifty percent off your bill by buying some ingredients at one store and others at another.
Obviously, this supports your point about transportation, but I’d go further and add in time, dietary constraints, and kitchen resources (both for cooking and storage). A working mother with no car, a small refrigerator and a child with food allergies would need a much larger budget to feed her family.
That store-to-store variation supports what I’ve been intuitively doing, which is I split up my grocery shopping between Super King (a local cut-rate market that caters to Armenian and Mexican populations, and has armed guards at the exits, but the produce is really cheap typically because it’s not “high visual grade” so misshapen peaches etc), Trader Joes, and either Ralphs (Kroger) or Vons (Safeway). The typical method is to go and get the things that each store does cheaply, so for example canned ripe olives, or whole bean coffee or cereal is something you can probably get cheaper at the Safeway/Kroger place, and Jarlsburg cheese and sourdough bread is best at TJ’s (price and quality) and the fresh produce is optimal at Super King. But the only way you can do this is if you can shop with a fairly large basket (and wallet) to get enough quantities so that you go to say only one store per week, or maybe two on some weeks.
People who have to carry things on the bus are going to go more frequently, and get smaller quantities, and probably shop at fewer stores, and so have to sample more of the less cost-effective items at each store.
Also, there are two perspectives here. One is to fix the menu as a wide ranging set of things, and then sample the prices for those items to see how expensive different kinds of things are in different places. This is the sort of “cost of living based on a fixed set of healthy choices”
the other way is to, based on the price spectrum of a local area, design alternative healthy choices that are cost effective for the area. That’s also a very good and useful thing to do, but has a different purpose I guess.
Imagine the following input vector:
Foods you like/can eat: this would be a list of foods that you eat (or assume you eat everything, but you specify exemptions)
Geography: Location; Transportation Method (back-ended to google maps)
Budget: how much money can you spend; how many hours/which hours per week you can shop
Stock: A list of things you have at home already (flour, sugar, oil, spices); how much dry/cold storage you have
You build an algorithm that matches this information to a dataset of recipes, and then spits out 10 or 20 options for the week (or longer horizon) that could be purchased on that budget constraint, and ranked by predicted “how much you’ll like it”, which learns from your behavior over time.
If you could get people to use it, I’d bet that would be a really interesting project. You’d get the benefit of your static index, the public good of the information dissemination, and interesting research on behavioral economics (price saliency, effort, revealed preferences, habit formation). And it could all be done in… 10,000 hours of labor? 100,000? I have no idea. Maybe you cut that in half if you can find a genius way to scrape the internet.
I dunno. I really like the idea making \emph{useful} information on food prices, quality and nutritional value available to people in a convenient manner, especially one that makes it easier for them to be healthy if they want. It is expensive to eat well (money and time and effort), and that disproportionately hurts poorer households and likely contributes some to the inter-generational persistence of poverty.
But right now I guess I think the research value is secondary to the public goods aspect. Possibly the same with your original index proposal.
@jrc
The crucial part is “if you could get people to use it”!
Is this a solution in search of a problem? Is there even a latent demand for such a website / app that can be stimulated?
Most food decisions tend to be highly impulsive & subjective. Do people even want to make these more rational / optimal?
Rahul, you have no idea how many times I’ve wanted such an app as jrc mentioned, but of course I’m a data nerd and a father of 2 small children who does most of the shopping and food prep for our family.
Still, the popularity on Google Play for apps like this suggests there is a latent demand. The big issue with the apps is that they don’t do a good job of personalization, and they tend not to be flexible. If you say for example that you don’t eat black pepper, they might just cut out any recipe that says “salt and pepper to taste” even though 0 pepper is a totally valid amount in such a recipe (or whatever, this is just sort of an example of the general issue).
Still, I think we’re talking about two different issues.
1) is to help people via a “public good” approach to disseminating useful information about cost-effective high quality nutrition, especially to poor people. This is Mark Palko’s main interest clearly, and is a great thing to do.
2) which is what I was mainly interested in, is measuring what it means, nutritionally and economically, to have a high quality diet with fairly standard foodstuffs, not Boutique Organic Kombucha-Kale smoothie from Whole Foods with imported Kobe beef sandwich on handmade German Style Artisan Rye bread with baby red lettuce heirloom tomatoes, and boutique Napa Valley extra virgin olive oil for $32.18
Damn, I’m hungry now.
Where’s a good Jamaican Meat Pie when you need one? Na’a’amsayin?
Daniel – maybe it is time for a trip south of Pico.
http://www.yelp.com/biz/wi-jammin-carribean-restaurant-los-angeles
http://www.yelp.com/biz/natraliart-market-and-restaurant-los-angeles
Natraliart. Real ting dat.
Thanks. I incorporated some of this into my response to Daniel below.
You don’t have to actually BUY all the items, just price them. You’d design the menus, and then from the menus design a shopping list sorted by area of the store, and then go and write down say the most cost-effective item that fits the bill.
But, yes, I think this is the kind of thing that some govt stats bureau should have been doing since at least the 1950’s but hasn’t.
But the UC data from scanners sounds awesome! that’d be another way to do it, you just filter down the scanner data to get prices of the items you want through time, and then maybe do smoothing/time-series. Of course you’d only get the one chain.
I recall MIT had a Group tracking prices online for a huge set of consumer goods to come up with a check on the official CPI. Someone like that would easily have access to the food prices you’d want.
https://en.wikipedia.org/wiki/MIT_Billion_Prices_project
Might be a fun project especially if it requires field sampling. On the other hand I think you are going for too small a number of breakfasts given the ethnic diversity you are likely to encounter. I may be wrong but my impression is that dietary habits often are more lasting than something like language spoken.
Do not click on this link if hungary
http://www.buzzfeed.com/ailbhemalone/breakfasts-around-the-world#.dh9V6exWE
Oh, and don’t forget the sizable youth/student population who consider cold pizza a mainstay for breakfast.
It’s not supposed to index what people actually eat, it’s more an index of what a large basket of healthy food costs. Generating the basket by designing 30 healthy meals doesn’t mean you expect people to eat only those 30 meals, since you could easily make other alternative meals from the same ingredients. The point is to get a lot of good ingredients in there.
Even if it doesn’t index all the different types of actually healthy breakfasts, lunches, dinners etc it’s got to be a better index of healthy food than CPI: “breakfast cereal, milk, coffee, chicken, wine, full service meals, snacks”
For example if the price of fish is skyrocketing due to overfishing, the CPI doesn’t know a thing about it. If fresh fruit is going through the roof because of water shortages in CA…. yep, no luck. etc etc.
Oh, okay, got it.
Related reading:
1) Warren Ellis, “Drinking Soylent With The Last Of The California War Boys” – http://morning.computer/2015/08/drinking-soylent-with-the-last-of-the-california-war-boys/ (Reminds me a little of Hunter S. Thompson; more specifically, the intro to “Generation of Swine”)
2) Rob Rhinehart, “How I Gave Up Alternating Current” – http://robrhinehart.com/?p=1331 (It starts off well enough but, from my perspective, soon drifts off the reservation.)
It literally starts off with “The walls are buzzing. I know this because I have a magnet implanted in my hand…”
And then it got weird.
21st Century Hunter — When the going gets weird, the weird open a Kickstarter account.
I’m surprised no one has put Michael Pollen into this discussion:
Eat food. Not too much. Mostly plants.
Don’t eat anything your great-grandmother wouldn’t recognize as food.
http://www.goodreads.com/author/quotes/2121.Michael_Pollan
This has got me wondering if any of my great-grandmothers would recognize an avocado as food.
Why are people ok treating themselves with drugs their great grandma didn’t know, & wear fabrics their great grandma wouldn’t have known.
Why this fetish about food? What’s so special about foods our great grandma would have known about? For all I know people died of food borne typhoid and cholera back then and also of scurvey and diabeties.
I sure want to inherit my great grandparent’s generation’s average mortality!
I think it’s because since our great grandparents, industrialized food production has given us entire AISLES devoted to corn syrup + flavor + carbonated water, and the Cheeto. Pretty much everyone knows damn well they shouldn’t be eating that crap, but it’s scientifically optimized to set off primitive fruit + salt cravings.
It’s not that you should eat *like* your great-grandmother, it’s just that you shouldn’t eat things that were only invented in the last 100 years by industrialized food processing plants.
The important things are that Soylent is a tech start-up with an ironic name! How can it not be a great idea that nobody in history ever came up with before?
The perception of health is pretty interesting. One shocking thing we found while doing market research is that a lot of people perceive Clif bars to be healthy despite the fact that they have similar levels of sugar to a snickers bar on a per calorie basis. Fun fact: mealsquares contain both beans (garbanzo beans, more nutrient dense than black beans) and milk (same micro content as yogurt mostly). We’re also not quite sure how to honestly signal healthiness given that the market is saturated with all sorts of health claims. We’re working on adding a lot more citations to our nutrition page for starters.