Wall Street Journal September 4, 2014
How the Feds Distort Their ‘Food Insecurity’ Numbers
This isn’t about hunger. It’s about feeling that the ‘quality and variety’ of available food isn’t what you’d prefer
by James Bovard
On Wednesday the Agriculture Department released the results of its annual Household Food Security in the United States survey for 2013. According to the USDA survey, 14.3% of U.S. households—some 49 million Americans—were “food insecure at least some time during the year in 2013.” The decrease from 14.5% of households in 2012 was “not statistically significant.” Yet if the past is any guide, the survey will be wrongfully invoked by politicians and pundits as proof of a national hunger crisis.
Is being “food insecure” the same as going hungry? Not necessarily. The USDA defines a “food insecure” household in the U.S. as one that is “uncertain of having, or unable to acquire, enough food to meet the needs of all their members because they had insufficient money or other resources for food” at times during the year. The USDA notes: “For most food-insecure households, the inadequacies were in the form of reduced quality and variety rather than insufficient quantity.”
The National Academy of Sciences urged the USDA in 2006 to explicitly state that its food-security survey results are not an estimate of nationwide hunger. The USDA responded by dropping any mention of “hunger” in the survey’s response categories. Nevertheless, the survey’s results continue to be pervasively misrepresented as an accurate measure of hunger in America.
Democrats have a long history of demagoguing the survey’s findings. In 2009, after a spike in the number of “food insecure” households, President Obama announced that “hunger rose significantly last year” and promised to reverse “the trend of rising hunger.” Mr. Obama’s comment spurred a Washington Post headline, “Hunger a growing problem in America, USDA reports,” while the New York Times chimed in with, “Hunger in U.S. at a 14-Year High.”
In 2013 the USDA reported that federal food programs—most notably food stamps provided by the Supplemental Nutrition Assistance Program (SNAP)—”increase food security by providing low-income households access to food, a healthful diet, and nutrition education.” But food insecurity was more widespread in 2013 (14.3%) than in 2007 (11.1%), while food-stamp recipients rose to 47 million from 26 million.
Paradoxically, rising government dependency may in part explain rising food insecurity. A 2007 Journal of Nutrition study concluded that families receiving food stamps are over 50% more likely to be “food insecure” than similar households not on food stamps. In 2010, the Government Accountability Office reported that food-stamp participants “tend to be more food insecure” compared with eligible nonparticipants. A 2013 Harvard School of Public Health study also found that enrolling in the food-stamp program failed to significantly boost participants’ food security or dietary quality.
Though the food-security survey results are often touted as evidence of widespread hunger, another USDA survey debunked that conclusion. The agency’s Agricultural Research Service conducts periodic surveys of “What We Eat in America.” The most recent survey (2009-10) revealed that children ages 2 to 11 in households with less than $25,000 in annual income consume significantly more calories than children in households with incomes above $75,000.
Unfortunately, the Obama administration is resisting disclosures that could explain the paradoxical relationship between food stamps and food insecurity. Seeking some clarity, Rep. Tom Marino (R., Pa.) introduced the SNAP Transparency Act last year to compel the USDA to disclose more information about what food stamps are used to buy. Mr. Marino said that “Congress has virtually no information to ensure that the program is operating effectively.” The Association of Health Care Journalists, the National Association of Science Writers and other organizations have endorsed Mr. Marino’s legislation, but Democratic lawmakers made sure it went nowhere.
Some Americans do indeed suffer from hunger, but the federal government has shed little light on the challenges they face. The National Academy of Sciences urged the USDA in 2006 to develop measurements of individual hunger instead of household “food security,” but the USDA hasn’t done so. More than 40 years after President Nixon declared war on hunger, the federal government still doesn’t care to accurately measure the problem.
Mr. Bovard, the author of “Attention Deficit Democracy” ( Palgrave Macmillan, 2006), testified before President Reagan’s Food Assistance Task Force in 1983.
OMG!!! I have long been unable to find pumpernickel / onion pretzel sticks (by Snyders) and cherry-vanilla Dr Pepper here, and have had to settle for plain sourdough and ordinary cherry Dr Pepper! I’M FOOD INSECURE!!!! I’M PART OF THE PROBLEM, NOT PART OF THE SOLUTION!!!!
This is no different, really, than the use of the word “poverty.” Relative poverty means the kids are wearing WalMart brand sneakers instead of Nikes. Absolute poverty means the kids have to go barefoot because their parents can’t afford shoes at all. But both are trumpeted as “living in poverty.” Technically correct – factually irrelevant.
When you have a large population of quasi-literate (at best) individuals who have neither the interest nor the inclination toward critical thinking, this sort of linguistic bait and switch is easy to pull off.
See further comparisons at “global warming” versus “climate instability” or squishwords like “assault weapon.”
Actually, this is another one of those cases that almost makes me lose faith in government statistics.
Almost? Seriously? What ever gave you the impression you could trust them at all?
Government statistics have one function only: to advance the interests of the ruling class and their bureaucrats. Thinking they are for anything else sounds awfully naive.
I’m ashamed to admit it but naiveté on govt. shenanigans has always been my downfall.
My grandmother developed a taste for tabloids like The National Enquirer (“Best reporting on the planet,” says Mr. Jones) and it was always fun to go through the articles and look for The Nugget. The Nugget was the sentence, sometimes two, containing the factoid around which the story was developed – like the grain of sand in a pearl. For instance, there was a story captioned “Musket Ball In Baby’s Chest Proves Reincarnation,” Buried about 2/3rds of the way through the breathless story was the admission that “infants are occasionally born with roundish calcium deposits in their bodies.” If I recall correctly, such deposits were either harmless or routinely removed.
Likewise government “statistics.” Or any statistics for that matter. The methodology by which they’re gathered, the questions asked, the *way* those questions are asked, and the compilation and presentation of the results – not to mention the interpretation (or spin) placed on them all go into the overall scheme of things. Much like The Nugget, the raw data is frequently there, even if the raw data is nothing more than “crap methodology yields crap results.”
Any survey that uses self-reporting is inherently suspect. Not necessarily inaccurate or wrong, but suspect. Following Joe Snuffy around to observe what he buys, weighing his portions, and tallying every bit of caloric intake will give you a far better idea of what Joe Snuffy’s food consumption is, rather than a phone call asking “whadja have for dinner last night?”
Then, of course, there’s the interpretation. If the Great And Awesome Benevolent Food Program gives me and Jim $100 for food for a month, and I spend my $100 on fine chocolate to go with my bourbon, while Jim carefully purchases beans, rice, eggs, second-hand meat, and day old fruits to keep him well fed, my food insecurity (heck, out and out hunger) on day two is not a fault of the program, nor is the “solution” to give me more money to spend on food. But that’s the default for the “we need more money for the poor people” crew. The, as I think Walter Williams may have described them, “poverty pimps.”
In short – dotgov statistics *can* be useful, so long as you know their background. But never forget that statistics is a decidedly prostitutional science. It’s willing to do pretty much anything for anybody, so long as the price is right…
Lawhobbit- what is “second-hand meat”? That sounds pretty damn spooky.
Your riff on govt stats is good stuff. You should repost it on your blog – and elsewhere.
You make a good point on the surveys. 30 years ago, nutritionists relied heavily on 24 hour dietary recalls – but those have been shown to be very unreliable.
I expect that a 365 day recall – the tacit measure of the USDA food security phone survey – is even shakier.