« Assorted links | Main | Did we *really* need that bail-out? »
Wal-Mart and obesity
Here's Charles Courtemanche and Art Carden:
We estimate the impacts of Wal-Mart and warehouse club retailers on height-adjusted body weight and overweight and obesity status, finding robust evidence that non-grocery selling Wal-Marts reduce weight while grocery-selling Wal-Marts and warehouse clubs either reduce weight or have no effect. The effects appear strongest for women, minorities, urban residents, and the poor. We then examine the effects of these retailers on exercise, food and alcohol consumption, smoking, and eating out at restaurants in order to explain the results for weight. Most notably, the evidence suggests that all three types of stores increase consumption of fruits and vegetables while reducing consumption of foods high in fat. This is consistent with the thesis that Wal-Mart increases real incomes through its policy of "Every Day Low Prices," making healthy food more affordable, as opposed to the thesis that cheap food prices make us eat more.
Of course, not everyone likes Wal-Mart.
Posted by Tyler Cowen on September 10, 2008 at 01:15 PM in Food and Drink | Permalink
Comments
Is there a selection bias here? People who tend to shop more groceries tend to eat at home more than outside. Which natural lends to greater control over the quality and nutrition content.
Posted by: at Sep 10, 2008 1:43:21 PM
It's "Walmart" now, not "Wal-Mart."
Posted by: gottabekidding at Sep 10, 2008 1:47:27 PM
Haven't we had enough of WalMart bashing? Why don't these people get a life?
Posted by: jorod at Sep 10, 2008 1:55:17 PM
Walmart forces efficiency on suppliers by demanding low prices. This competition raises the value provided by the suppliers, in effect increasing our society's overall wealth. Walmart then distributes that extra wealth among its employees, shareholders, and customers. And since its customers are more typically lower income, it raises the wealth (not in terms of money spent, but in value received) of the less privileged segments of society. Arguably, Walmart has done more for poor people than all the government programs in the last 20 years.
Posted by: steveSC at Sep 10, 2008 2:03:08 PM
"Is there a selection bias here? People who tend to shop more groceries tend to eat at home more than outside. Which natural lends to greater control over the quality and nutrition content."
If you look at the cited excerpt, you'll note that people who shopped at Walmarts that did not have groceries lost weight. The idea being that they saved money on other things, and chose to spend more, all things being equal, on fruits and vegetables.
In other words, fatty bad for you food can be an inferior good, and people substitute fruits and vegetables when they have more money and junk food when they have less. Note that the argument implies that too much of an emphasis on organic fruits and vegetables, if they raise the prices, can be bad for health because too many people will switch from cheap fruits and vegetables to junk food rather than staying with more expensive organic produce.
Posted by: John Thacker at Sep 10, 2008 2:06:41 PM
It's true, whenever I shop at a WallMart I swear off fatty foods for a least a couple of weeks..... It's like eating at Denny's, you glance around at your fellow patrons and then order the fruit salad.....
Posted by: patski at Sep 10, 2008 2:11:02 PM
When I read this, I thought "damn, that makes sense". Back in the day (10-15 years ago), I used to do most of my grocery shopping at Costco. The main driver of that was the 2 gallons of milk deal, because grocery stores weren't matching that. I drink a lot of nonfat milk. It's my way of making up for not being a smoker or heavy drinker. I now do most of my grocery shopping at a local Ralphs. Costco is a 10 mile round trip and is always very crowded, even in the mornings when business members get their special hour.
But when I go to Costco occasionally, I load up on frozen fruit, fresh fruit, vegetables, etc. Their stuff is generally much better quality than the grocery store. Bananas for example... At Costco, I can usually find a nice bunch that's still green and needs a week to get ripe. At Ralphs, they're almost always yellow, too ripe for my tastes, and two days away from going bad. The other thing that Costco and Sams Club have is diet food in bulk at significant discount over Ralphs. When I focus on eating right, it's not just difficult when shopping at Ralph's, it's expensive.
Posted by: BoscoH at Sep 10, 2008 2:39:48 PM
Adding specious suppositions to hand-cranked survey data, no wonder they call it the dismal science.
Posted by: Adrian Monck at Sep 10, 2008 2:50:03 PM
Time saved could be important too. If I have to drive to fewer stores and spend less time waiting at the cashiers, then I'll have more time to do other things, like exercise. As long as those other things burn more calories than driving multiple places, there's part of the weightloss explanation.
It could go the other way too. People could spend more time watching tv. But, it looks like the data suggests the former.
Posted by: Scott Wentland at Sep 10, 2008 3:28:20 PM
As far as I can tell this paper is based on very little real evidence and the reasoning makes very little sense. It rings of corporate propaganda.
"All three types of stores increase consumption of fruits and vegetables while reducing consumption of foods high in fat."
Increase consumption of fruits and vegetables compared to what? Shopping elsewhere? To say that shopping at Walmart actively reduces weight with no explanation strikes me as misleading and deceptive. They may sell cheap diet food but the fact is that "diet" food -- foods which have been chemically manufactured in laboratories rather than grown from the ground -- are not promoting health. They may contain fewer calories but they contain chemicals that have never been tested as carcinogens or promoters of other serious illnesses.
The only way that Walmart's promotion of organic food could result in low-income shoppers choosing junk food over fresh produce is if they ONLY sold organic food, which they never will. There will always be less expensive produce options.
I find this kind of baseless propaganda being promoted as scientific research dishonest and disturbing. Obesity in this country will go down when people start eating real food and education around nutrition and health is more accessible in the public school system.
Posted by: Sarah at Sep 10, 2008 4:10:44 PM
I've slimmed down since I started shopping at Walmart. It all has to do with the long walk from the parking lot to the store. I live in a town of about 80,000 and about 40,000 cars are parked in the lot every time I go.
Posted by: glh17 at Sep 10, 2008 4:12:22 PM
"As far as I can tell this paper is based on very little real evidence and the reasoning makes very little sense. It rings of corporate propaganda."
I'm curious. Did you bother to read this paper before condemning the authors' research as "baseless propaganda"?
Posted by: Some Random Economist at Sep 10, 2008 5:10:10 PM
"The only way that Walmart's promotion of organic food could result in low-income shoppers choosing junk food over fresh produce is if they ONLY sold organic food, which they never will. There will always be less expensive produce options."
Sarah,
Besides noting that you did not appear to read the paper at all before dismissing the research as based on little real evidence, I'd like to offer another point.
The "only way?" Really? You don't think that promotion of organic food as "safer" and "more healthy" could make some low-income shoppers somewhere believe that regular produce is less healthy and thus less likely to purchase it? You don't think that someone out there could decide to, instead of buying 2 pounds of regular apples, buy 1 pound of organic apples and 1 pound of junk food for a negative return on healthfulness? I certainly wouldn't guarantee that these things would happen, but they seem within the realm of possibility.
Posted by: John Thacker at Sep 10, 2008 5:28:48 PM
What happened to all the comments on this post? I could have sworn that earlier today there was a very interesting series of about 30 comments here, now it's down to just 11.
Posted by: John S. at Sep 10, 2008 6:29:26 PM
What happened to all the comments on this post? I could have sworn that earlier today there was a very interesting series of about 30 comments here, now it's down to just 11.
Posted by: John S. at Sep 10, 2008 6:30:12 PM
We are talking about a lower weight of a couple hundred grams. On the other hand, smoking and drinking go up a couple percent while exercise decreases.
But I don't find the hypotheses that Walmart causes these statistically significant but tiny effect sizes unconvincing.
What I still can't understand, is why some libertarians dwell in veneration for big corporations and do the work of corporate apolegetics. I reject that, just like I reject the worhship of mom and pop stores and farmers markets on the other side of this continuing argument.
Posted by: Ethnic Austrian at Sep 10, 2008 7:23:12 PM
Big corporations, mom and pop stores, and farmers markets are all equally deserving of veneration, when they are not taking government handouts. It's just that big corporations do more good than any mom and pop store or farmer's market because... they're bigger. More voluntary transactions = greater welfare.
Posted by: Cliff at Sep 10, 2008 8:46:39 PM
And thank goodness the most obnoxious blogger in the world is blogging again after a hiatus.
Posted by: at Sep 10, 2008 8:58:36 PM
Was it ever absolutely certain Kathy G was the blogger Tyler was talking about?
Posted by: Brian J at Sep 11, 2008 12:28:31 AM
Cliff: "It's just that big corporations do more good than any mom and pop store or farmer's market because... they're bigger. More voluntary transactions = greater welfare."
Big retail corporations - just like big farms and big passenger airlines - take advantage of huge economies of scale. The efficiencies of those economies of scale provide consumers with real increases in standard of living. It's not the volume of transactions, but the sharply lower cost per transaction that "bigness" provides.
Posted by: John Dewey at Sep 11, 2008 3:57:59 AM
OK, I did a first pass on the actual paper.
It looks to me like it says that counties where women self-report that they're less obese are very slightly more likely to get a new WalMart.
There is no evidence provided that the women who claim they're less obese actually shop at WalMart, only that there are more WalMarts in counties where they say they weigh a bit less compared to their claimed height.
While changes in diet tend to have slow results -- you gradually gain or lose weight on a new diet -- the results shown are immediate. The small weight loss happens precisely when the new WalMart opens. For SuperWalMarts the result gets smaller with time, for price clubs it gets larger.
The paper includes a bewildering variety of mixed hypotheses, together with claims for indirect data that might affect them. Very unclear whether they tested the hypotheses that matter. For example, they note that people tend to lose weight when they're recently unemployed, and they asked whether new Walmarts resulted in higher unemployment, but it looks to me like their data doesn't show that -- it doesn't test single counties monthly, it tests them yearly and you can estimate percentage unemployment after new Walmarts open by looking at one county the month before, another county during, a third county the month after, etc.
Here is something I didn't understand from a quick read, that maybe somebody else might read more carefully and explain to us all?
"Including county trends increases the standard errors considerably while slightly reducing the magnitudes for Wal-mart and Super Wal-Mart and causing the sign of the coefficient on warehouse clubs to become positive. In these estimates, all coefficients are statistically insignificant. Note that county trends explain 58%, 57%, and 54% of the variation in Wal-marts, Super Wal-marts and warehouse clubs remaining after controlling for fixed county effects and the other regressors in the baseline model. Moreover, county effects and trends, along with the other regressors, explain 95%, 88%, and 96% of the total variation in per capita Wal-Marts, Super Wal-Marts, and warehouse clubs. Therefore we suspect that the inclusion of county trends eliminates too much of the meaningful variation in the variables of interest to accurately identify their effects in these models. In any case, for all three types of stores the baseline estimates are well within the 95% confidence intervals of the estimated coefficients from regressions that include county trends. We therefore cannot conclude that the baseline fixed effect estimators are inconsistent. Because we find no evidence that our baseline estimates are inaccurate, we emply the county fixed effects estimation strategy throughout the rest of the paper."
I sort of thought I understood that, but I could be wrong. Maybe it isn't as damning as it looks?
Posted by: J Thomas at Sep 11, 2008 10:44:23 AM
Big retail corporations - just like big farms and big passenger airlines - take advantage of huge economies of scale. The efficiencies of those economies of scale provide consumers with real increases in standard of living.
How much of that is economy in scale of production versus economy in scale of negotiation?
I've read that WalMart auditors can go to suppliers and say "Show me your books" and if they determine that the profit is too much -- above, say, 3% -- they can adjust the price down to something more reasonable. A mom-and-pop store can't do that, they couldn't afford the auditors even if they had the bargaining power. They have to depend on competition among suppliers eager for their business to keep prices reasonable.
This suggests several ways that modern small businesses might work with each other, given the new communications systems. You could have transport companies that generate economy of scale transporting other people's property, and markets that match up small buyers and small sellers, and of course they could build relationships with businesses they've had successful dealings with before. It isn't necessary to have giant middlemen who buy from small suppliers and sell to small retailers.
Posted by: J Thomas at Sep 11, 2008 11:02:24 AM
I'm not much of an economist, nutritionist, activist, environmentalist... so what I say may have very little substantive value in this thread.
If you are really all that concerned with the truth, not some nice theory packaged up in cute statistics, get two cameras. Take one to walmart and photograph the first 24 people you see. Take one to Whole Foods, Central Market, a Farmer's Market etc.
Compare the pictures. Even take the time to ask the people about their health and well being.
There is no comparison. There is no debate.
Posted by: Adam at Sep 11, 2008 2:23:10 PM
I'm not much of an economist, nutritionist, activist, environmentalist... so what I say may have very little substantive value in this thread.
If you are really all that concerned with the truth, not some nice theory packaged up in cute statistics, get two cameras. Take one to walmart and photograph the first 24 people you see. Take one to Whole Foods, Central Market, a Farmer's Market etc.
Compare the pictures. Even take the time to ask the people about their health and well being.
There is no comparison. There is no debate.
Posted by: Adam at Sep 11, 2008 2:25:15 PM
I think it is of the utmost importance for people to understand that weight gain is something that can be overcome even if everyone in your family has yet to defeat it.
Posted by: natural diet pills jim at Sep 11, 2008 3:27:30 PM
