USDA's Pesticide Data Program

Each year, USDA's Pesticide Data Program publishes data on pesticide residues, principally in fruit and vegetable crops, selected in part because of their frequent consumption by children.  It is difficult to know how worried consumers should be about the results.  Here are some thoughts and questions on the most recent 2008 residue detections for fruits and vegetables.

For some pesticides, EPA sets safety tolerances for the maximum amount of residue that should show up on food.  For other pesticides, EPA sets no tolerance, meaning that there should not be any residue of that pesticide at all.

First, we look at data on total residue detections.  Many fruit and vegetable samples have multiple pesticide residues, but the amounts may be small, usually far within the Environmental Protection Agency's safety tolerances.  Let's say for the moment that we are not very worried about these total detections, but instead want to know about detections that violate EPA standards.

		All samples		Residue detections
		#		#		per 100 samples
Asparagus		372		39		10.5
Blueberries		726		1736		239.1
Broccoli		554		797		143.9
Celery		741		3821		515.7
Green Beans		741		1392		187.9
Green Onions		186		272		146.2
Greens, Collard		240		540		225.0
Greens, Kale		318		622		195.6
Nectarines		672		1603		238.5
Peaches		616		2155		349.8
Potatoes		744		1410		189.5
Spinach		747		1850		247.7
Strawberries		741		3703		499.7
Summer Squash		554		1050		189.5
Sweet Corn, Fresh		152		1		0.7
Sweet Potatoes		184		92		50.0
Tomatoes		740		903		122.0
Total		9028		21986		243.5

Second, therefore, we look at residue detections that exceed EPA's established tolerances, for pesticide uses that have a tolerance.  These detections are more worrisome when they happen, but they do not happen very frequently.  Fewer than 1% of samples had this type of residue detection exceeding an established tolerance.

		All samples		Residue detections exceeding an established tolerance
		#		#		per 100 samples
Asparagus		372		0		0.0
Blueberries		726		3		0.4
Broccoli		554		0		0.0
Celery		741		2		0.3
Green Beans		741		2		0.3
Green Onions		186		0		0.0
Greens, Collard		240		11		4.6
Greens, Kale		318		10		3.1
Nectarines		672		0		0.0
Peaches		616		0		0.0
Potatoes		744		7		0.9
Spinach		747		16		2.1
Strawberries		741		2		0.3
Summer Squash		554		5		0.9
Sweet Corn, Fresh		152		1		0.7
Sweet Potatoes		184		0		0.0
Tomatoes		740		1		0.1
Total		9028		60		0.7

Third, we look at residue detections for pesticides that have no EPA tolerance.  The lack of a tolerance may mean that the chemical is judged to be of greater safety concern.  The USDA found many such residue detections -- almost 5 such residue detections per 100 samples, which seems like a lot of detections in violation of EPA standards.  However, the tests are quite sensitive, and the residue amounts may be very small.  USDA is not greatly worried: "In most cases, these residues were detected at very low levels and some residues may have resulted from spray drift or crop rotations."

		All samples		Residue detections for pesticides with no tolerance
		#		#		per 100 samples
Asparagus		372		1		0.3
Blueberries		726		21		2.9
Broccoli		554		17		3.1
Celery		741		123		16.6
Green Beans		741		23		3.1
Green Onions		186		9		4.8
Greens, Collard		240		34		14.2
Greens, Kale		318		45		14.2
Nectarines		672		2		0.3
Peaches		616		61		9.9
Potatoes		744		45		6.0
Spinach		747		33		4.4
Strawberries		741		5		0.7
Summer Squash		554		15		2.7
Sweet Corn, Fresh		152		1		0.7
Sweet Potatoes		184		1		0.5
Tomatoes		740		1		0.1
Total		9028		437		4.8

Environmental groups are not convinced. The Environmental Working Group uses these data as one of several sources in constructing its "dirty dozen" and "clean fifteen" lists of fruits and vegetables.

I wish the USDA's PDP reports did a better job helping readers to understand the implications of residue detections for pesticides that have no EPA tolerance. If USDA's position is that these are negligible detections, attributable to inconsequential pesticide drift, then it should explicitly set a threshold for these negligible detections. On the face of it, without that type of interpretation, I am reluctant to accept that these detections are all inconsequential. For example, notice that 16% of celery samples and 14% of many leafy greens had this type of violation.  Could those really all be spray drift?  That would seem surprising.  It would help to have a more blunt assessment from USDA experts: "These detections are negligible, but we want these other detections to fall into compliance in the near future."

Also, the Environmental Working Group throws a good heavy punch this week regarding the produce industry's government-funded information campaign to convince people not to worry about pesticides.  Agriculture departments, including the California Department of Food and Agriculture and USDA, should steer clear of anything that smacks of misleading propaganda on this topic.  For example, now that I have had time to study the residue data, I suspect last November's report from the Alliance for Food and Farming is overconfident that these residues are harmless.  A better approach might be to undertake some vigorous enforcement, drive down the frequency of violative residue detections, and then boast about the results. Indeed, if industry leaders took a long-term perspective, I think they would encourage USDA to do so.

Pesticide residues are not my leading food safety concern -- they rank behind foodborne illness on my list of things to worry about -- but I do take them seriously.  Among agricultural economists, I sometimes hear an outright dismissal of concern about pesticide residues as completely silly, but I think scientists who study cancer risks and toxicity take these concerns more seriously.  For example, here are the conclusions of the 2008-2009 report from the President's Cancer Panel, issued by the federal government's National Cancer Institute:

The entire U.S. population is exposed on a daily basis to numerous agricultural chemicals. Many of these chemicals are known or suspected of having either carcinogenic or endocrine-disrupting properties.

The report identified several problems that hinder policies to address environmental chemical contaminants: (a) inadequate funding and staffing, (b) fragmented and overlapping agency authorities, (c) excessive regulatory complexity, (d) weak laws and regulations, and (e) undue industry influence.

Sharing information about food safety

Food safety depends on both market incentives and government regulation.  If consumers and intermediaries (such as supermarkets) knew more about the safety of different food categories and food producers, they would be better able to defend their own interests in the marketplace.  Without such information, strong government oversight is essential.

This blog post is about two kinds of food safety information: the first has recently been in the news, and I think the second has not been covered much in the news.

Recent research on food safety priorities

First, the Emerging Pathogens Institute last week released estimates of the Top 10 riskiest food/pathogen combinations. The riskiest food category was poultry, with annual estimated costs of $2.46 billion and 180 deaths. The riskiest food/pathogen combination was campylobacter in poultry, with annual costs of $1.26 billion and 55 deaths. Salmonella in poultry had annual costs of $0.71 billion and 81 deaths.

The Institute discussed implications for the USDA's Food Safety Inspection Service (FSIS) and the Food and Drug Administration (FDA):

Salmonella causes more disease burden than any other foodborne pathogen, and according to FoodNet surveillance data, is one of the few foodborne pathogens that has not significantly declined over the past 10 years....  Our analysis also shows Salmonella disease burden as being associated with a wide variety of foods regulated by both FSIS and FDA, with significant risks associated with poultry, produce and eggs. This suggests that reduction of the national burden of salmonellosis will require a coordinated effort by both agencies addressing a broad array of foods. We recommend the agencies convene a national cross-agency initiative in collaboration with CDC that looks across the entire food system to target opportunities for risk reduction.

FSIS data on salmonella in particular food plants

In recent years, FSIS has begun to make public information about particular poultry plants where random samples tested positive for salmonella.  I am still learning to read and interpret these reports.  I think the major media may be having similar difficulties, because I do not know that these reports get any media coverage.

FSIS provides three categories: Category 1 is good, Category 2 is intermediate, and Category 3 is the worst category (for poultry plants that failed at least one recent test and did not do very well on a second test either).

For example, in the most recent report on Category 3 establishments (.pdf), the Tyson Foods plant in Temperanceville, VA, nearly failed one test (more than 10% of samples had salmonella) and then failed a second test (more than 20% of samples had salmonella).  In 2010, according to the most recent FSIS annual progress report, only 5 plants out of 172 did poorly enough to fall into this Category 3 status, so clearly most poultry plants find the FSIS criteria to be a reasonable standard to achieve.

Economists see many food safety problems as a type of information failure.  If consumers and institutional buyers only had good information, market incentives would solve many food safety problems.  So, I find these FSIS reports intriguing, but I see them as a promising work in progress.

Can anybody tell me if I interpreted the reports correctly?  Has there been any media coverage at all of these reports?  What would it take to make this type of information a major factor in food safety improvement?

Update May 9, 2011:  First, USDA's Food Safety and Inspection Service (FSIS) has just this year updated the Salmonella standards discussed in this blog post and established new standards in the same spirit for another pathogen, Campylobacter. Second, a 2009 report from USDA's Economic Research Service suggests that better information sharing could lead to stronger market incentives for food safety protection:

The forces driving management-determined actions lead to the conclusion that USDA’s FSIS could increase incentives by providing consumers and buyers with more information about the meat and poultry food safety control of particular plants and fi rms. USDA’s FSIS records plant performance on Salmonella spp. tests and noncompliance with process regulations. Making this information public should encourage greater food safety investments by meat and poultry producers.

Joseph Gallo Farms

As you travel through the food producing regions of the United States, every farm and food manufacturing business has its own story.  Yet, in the most intensive food producing regions of California, some of these stories seem to have a particularly epic scale, appropriate to the setting.

The story of  Joseph Gallo Farms is like a bottled up distillation of every manner of U.S. food policy issue.

The other week, while visiting the Salinas Valley to learn about the environmental challenges of concentrated animal agriculture in close proximity to large-scale vegetable production, I took these roadside photographs of one of the large-scale Joseph Gallo Farms dairies.

Joseph Gallo Farms produces Joseph Farms California Natural Cheeses.

You may ask, why does this dairy company not produce "Joseph Gallo Farms" cheeses?  Why drop the last name?  Well, now, there indeed is a story.

Joseph Gallo was the younger sibling of Ernest and Julio Gallo, who founded the largest exporter of California wines and one of the wine industry's most famous brands.  To protect the wine brand, Ernest and Julio successfully sued their younger brother to prevent him from using the Gallo name on cheeses at the retail level.

The judge who authored the 1992 court decision clearly got caught up in the operatic narrative and let loose his inner novelist.  Here is just one early section.  If this blog post were a movie, the screen would fade on the current decade and you would see an antique automobile puttering along a dusty dirt California road in the early 1900s.

This lawsuit arises out of a tortuous family history apparently involving sibling rivalry on a grand scale. Because Joseph's counterclaims concern his parents' estates, the relevant facts date back nearly a century.

I. The Rise of the Gallo Family, the Establishment of the Winery, and Ernest and Julio's Guardianship of Joseph

The individual parties to the action, Ernest, Julio, and Joseph, are the children of Joseph Gallo ("Joseph Sr.") and Assunto ("Susie") Bianco, immigrants to Northern California from Italy in the early 1900s. Joseph Sr. and Susie married in 1908. Ernest was born in 1909, Julio in 1910, and Joseph in 1919. Following their marriage until the advent of Prohibition in 1919, Joseph Sr. and Susie operated various boarding-houses and saloons, in connection with which they served and sold wine purchased from other California wine dealers. Evidently they stenciled the family name GALLO on the ends of the wine kegs, although they did not make the wine themselves. Throughout the 1920's, the family purchased a series of vineyards, where they grew their own wine grapes, bought wine grapes from other local growers, and shipped the grapes to the midwest and the east coast, where customers made wine with them for their home use under an exception to Prohibition. Ernest and Julio became involved in this shipping business during the mid- to late-1920s. While Joseph Sr. did have a brush with the law for bootlegging during Prohibition, there is no other evidence that he and Susie sold wine after 1919.

The Great Depression caused the grape business to suffer. Prices dropped; the 1932 season was a financial disaster for Joseph Sr. and Susie. On June 21, 1933, Joseph Sr. took Susie's life and his own.

The court decision goes on to recount that Joseph was raised by his brothers as guardians, his part of the inheritance was used as an early source of capital for Gallo Wines, he sued his brothers and was repaid, and, many years later, they prevented him from using his full name as the brand name for his cheeses.

A layperson may be astonished that Joseph Gallo could not use his own real name as his brand. But, several key features of this particular dispute favored the older brothers. The Gallo wine brand is a "strong" and widely known brand. Because wine and cheese are consumed together, and it is plausible that a wine company could sell cheese, a consumer might really be confused about the connection between the cheese brand and the wine company. Consumer research showed that a written label disclaimer did not suffice to clear up the confusion.

If you buy Joseph Farms California Natural Cheeses in your grocery score, there is more history to that food label than you might otherwise know.

Soda, SNAP (food stamps), and New York City

Last October in U.S. Food Policy:

New York City this week petitioned USDA for permission to disallow soda purchases with food stamp (SNAP) benefits. USDA may well disapprove the proposal, having in previous years turned down requests from other states for similar waivers from federal rules.

This week, the New York Times and Marion Nestle both cover recent developments in the resulting controversy.

Here are several thoughts, addressed to health policy advocates such as the American Public Health Association, NYC officials, and others.

1. You never should have called the proposed policy a ban. Has there ever been a controversy over the "ban" on SNAP spending for hula-hoops and yo-yos? The "ban" on SNAP spending for gasoline? No, you could say SNAP has made a sensible bureaucratic division between eligible goods and ineligible goods. People can use cash for all the ineligible goods. The proposed policy would slightly narrow the class of foods and beverages eligible for taxpayer support through SNAP benefits.

2. You should have emphasized the pilot nature of the proposal, and the research questions that will be investigated. In the NYT article, the quote from Kelly Brownell at Yale gets this right. The pilot should ask two questions: (a) does the proposed policy improve food and beverage choices?, and (b) did the proposed policy annoy participants or make them less likely to participate?  Some, but not all, SNAP participants will like the new policy. Remember, low-income parents struggle with their children over food choices in the grocery aisle just like middle-income parents do. The soda companies have no evidence that this proposal will cause any problems whatsoever for SNAP participants.

3. You should have negotiated with the anti-hunger community before taking this proposal public. The quote in the NYT article from the Food Research and Action Center is a bad sign.  Some health advocates have already written off the anti-hunger community as hopeless on health and obesity issues, but this is a mistake in political strategy. Key leaders in anti-hunger policy advocacy and the charitable emergency food system watch public concern about health and nutrition very closely, and calibrate their own programs accordingly. Health advocates have leverage with the anti-hunger community, which cannot afford to be indifferent to health issues.  I think a bargain should have been possible.

Beyoncé's Let's Move video

I liked it. It doesn't get an "A" on every plank of the checklist, but it does well on most. Friends have asked, does the video feed the notion that exercise alone without food is enough? The cafeteria tray at the start and apple at the end earn the video a B+ for food, not worse. Friends have asked, what about sexist stereotypes and high heels in the workout? Although she is a bit ... um ... unrestrained, Beyoncé seems like a star in command of her world happily flaunting good looks, very different from the nearly naked bodies explicitly at the service of a male star's gaze, which is more typical of the genre. In related workout videos, people wear more realistic clothes. Again, a B+ not worse. Making a school cafeteria and its neatly dressed students look cool and appealing gets an A. Message that physical fitness is cool and appropriate for kids of all sizes gets an A. Comments welcome.

NRDC Growing Green Awards

The Natural Resources Defense Council (NRDC) today announced its 2011 Growing Green Awards, in four categories. The awards went to:

Food Producer:  Jim Cochran, founder and co-owner of Swanton Berry Farm, started the first commercially successful organic strawberry operation in California and the first 100 percent unionized organic farm in the nation. Jim not only helped jumpstart the organic strawberry industry, but is also encouraging fair labor practices among sustainable growers.

Young Food Leader (new category):  Molly Rockamann, founder of the non-profit organization EarthDance Farms in Ferguson, MO, won for her efforts to cultivate a new generation of sustainable farming stewards in the Midwest. Through Molly’s unique apprenticeship program, urban St. Louis residents -- with ages spanning five decades -- learn the complete cycle of organic farming from seed to market.

Knowledge Leader: Chef Ann Cooper is an outstanding leader in the national movement to ensure that all children have access to nutritious food at school. Known as the “Renegade Lunch Lady,” Ann started the Food Family Farming Foundation, which provides schools with the tools and guidance they need to transition from over-processed meals to healthy, fresh ingredients on a limited budget. Ann’s day job is in the trenches of the Boulder Valley School District, where she is transforming lunch menus in the entire 48-school district, just as she has done before on the east and west coasts.

Business Leader: Pam Marrone is the CEO and founder of Marrone Bio Innovations, a leading developer of environmentally-responsible biopesticides based in Davis, California. MBI’s products use naturally occurring materials like plants and microorganisms to help growers control pests, maintain yields, and reduce risks to human health and the environment frequently associated with conventional pesticides. Having launched three successful companies to develop bio-based pesticide alternatives, Pam has created a strong business model in a rapidly growing industry.

From the NRDC's press release:

“With the double whammy of rising food demand and growing environmental challenges, we need a food system that can produce more while using fewer natural resources,” said Jonathan Kaplan, Senior Policy Specialist at NRDC. “These Growing Green Award winners are providing the leadership needed to make that happen. They inspire all of us.”

High crop prices are good for farmers

This headline might seem obvious.  Farmers get paid more when crop prices are high.

So, I was surprised to read the title of the recent report by Tim Wise of the Global Development and Environment (GDAE) Institute here at Tufts:

Still Waiting for the Farm Boom:
Family Farmers Worse Off Despite High Prices

The GDAE frequently does good work advocating for the environment and for the interests of the world's poor, but let's give this claim a skeptical look.

Driven in part by high energy prices, and the use of crops for biofuels, world food crop prices are once again at or near record levels. This trend is worrisome for consumers, especially in developing countries, but it is good news for farmers around the world.  For U.S. farmers, net farm income is expected to be 20% higher in 2011.

It is wise to draw some distinctions within the broad category of U.S. farmers, because the boom times could be reaching some farmers more than others.  Table 1 shows USDA's 2009 estimates for two groups of farmers that sometimes get muddled together under the heading "family farmers."  This is the same most recent year and data source that Wise's report uses.  In this table, following USDA conventions, the small family farms have sales of $100,000 to $250,000.  The large family farms have sales of $250,000 to $500,000.  The table excludes very large family farms and non-family-owned corporate farms on the one hand, and small hobby farms or "rural residence" farms on the other hand.

Table 1 shows that the small family farms had somewhat lower average household income than the U.S. average in 2009, and they would not be able to get by on farm income alone. Many farm families include at least one person working in the paid labor market. The large farms, by contrast, had much higher average household income than U.S. average household income.

There might be some grounds for saying that the small farms are still waiting for the boom, but the large farms seem to be doing well.  This distinction is important. If we describe large family farms as "still waiting for the boom," it could be a recipe for bad policy. It could contribute to a rhetoric of pity for U.S. farmers at a time when most U.S. farmers -- even many family farmers -- are doing well, while other people's hardships are neglected by governments.


Table 1.  Farm income and selected characteristics in 2009 (USDA data).

Informal name		Small family farms		Large family farms
Formal USDA terminology		Small, farming occupation, higher-sales		Large, excluding corporate farms
Average annual operator household income
From farming sources ($)		19,274		52,193
From non-farming sources ($)		35,859		51,909
From all sources ($)		55,133		104,102
Percent of U.S. average household income		81		153
Total value of production ($ billions)		19		34
Total acreage (million acres)		122		140
All products
Number of farms		110,034		94,909
Average acreage per farm		1,106		1,477
Corn
Number of farms		17,963		26,857
Average acreage per farm		533		849
Soybeans
Number of farms		9,747		6,490
Average acreage per farm		597		1,157
Wheat
Number of farms		4,969		3,376
Average acreage per farm		2,098		3,344
Cattle [Update: previously "beef"]
Number of farms		21,210		11,055
Average acreage per farm		3,247		5,530

Who are the large family farmers in the second column? The largest number are corn farmers. If you picture a typical corn farm of 849 acres, or a typical soybean farm of 1,157 acres, you are picturing one of these large family farms. These are serious businesses, and they typically have some hired labor, but they are owned by the family that operates the farm. As you drive across the U.S. rural landscape, most of the cropland you pass belongs to farms that are this large or larger. The largest fraction of total food production value comes from farms that are this large or larger.

Who are the small family farmers in the first column?  The largest number are cattle operations, with large acreage but frequently with fragile finances.  Among other things, many of the livestock businesses face rising costs for animal feed, due to the very same crop price increases that allow crop farmers to prosper. Annual sales of less than $250,000 is quite small for a full-time crop farm business, and most of the crop farms in this category are quite small.  The average corn farm in this category has only 533 acres, and the average soybean farm has 597 acres, a scale that may have only marginal economic success.

In the GDAE report, Wise only tabulates the small family farms in the first column. The report does not include the data I have provided for the second column, nor does it distinguish different farm products within each column as Table 1 does.

Yet, Wise describes the small farms in a way that seems to characterize a larger class of farmers, breezing over the fact that many family farmers are doing well. Consider this description of the scale of the farm: "In 2009, the average farm household in this group farmed 1,104 acres, but earned just $19,274 from farming." By quoting such a large acreage, the report implies larger crop farms than the report's data really describe. Though the GDAE report does not mention it, this large acreage is an average of the comparatively small acreage for many crop farms (533 acres for corn, 597 acres for soybeans) and much larger acreage that is typical for wheat farmers and cattle ranchers.  The report seems to be discussing commercial family crop farm businesses, but does not point out the much smaller acreage of most crop farms in this category.  Wise's blog post introducing the report drives the point home: "These are not small farms: average size is 1,100 acres and gross sales are $100,000-$250,000."

There is a long tradition in U.S. farm politics of telling stories about the plight of small farmers and then subtly slipping large family farms under the same umbrella, in order to pass subsidy policies that provide the greatest economic benefits for large farmers. It seems misdirected for GDAE to say of the first column, "these are not small farms" when the data come precisely from USDA's "small farm" category, and other family farms are not tabulated.  The sales cutoff of $250,000 will sound lucrative to many non-farm readers, who are accustomed to thinking about labor market income rather than sales, but these are quite small farm businesses that are "still waiting for the boom."

Most mid-sized to large family farms, and especially most family-operated crop farms with 800-1,000 acres or more, are doing quite well when prices are high. I imagine that on reflection Wise would agree that farms on this scale were not the intended subject of his concern.

[Update: Tim Wise has a thoughtful response in the comments section.]