Photo from Socially Responsible Agricultural Project and CARE, Washington State.

Not So Fast, Purveyors of Junk Science: Factory Farms Are Not Green

Something's been bothering me since Thanksgiving, festering deep in the gut like an angry conglomeration of undigested yams and cranberry relish.

The disturbance: junk science. The offending bit of ivory tower mediocrity: Demystifying the Environmental Sustainability of Food Production, a paper by Jude Capper, Roger Cady and Dale Bauman published in the Proceedings of the Cornell Nutrition Conference 2009. I didn't want to respond, tried to ignore it, but ultimately, I couldn't resist.

Off with the gloves (a note to the paper's authors):

Jude Capper, Roger Cady and Dale Bauman: your paper demonstrates either a lamentable misunderstanding of the impacts of livestock production practices, or a willful effort to misrepresent the facts. Or perhaps a little of both.

Hmmm (a note to readers):

Surprise, surprise - Capper, Cady and Bauman are Big Time Big Dairy proponents; they've all been pushing rBGH for years - in fact, Cady is actually employed by Elanco Animal Health (the division of Eli Lilly to which Monsanto sold rBGH in 2008). But I'm sure this had no impact on their analysis.

Junk Debunked

The thrust of the authors' argument is that the "efficiency" of industrial ag enables factory farms to produce a given quantity of meat, eggs and/or dairy products with a smaller adverse environmental impact than less efficient, traditional livestock farms. As you'd expect, there are a number of Major Flaws in their analysis - read quick descriptions of each flaw below; find detailed assessments after the jump.

Major Flaw #1 : Grossly inadequate assessment of "environmental impact"

Greenhouse gas emissions aren't the only environmental impact of factory farms.

Major Flaw #1a: Total manure production doesn't necessarily reflect environmental impact

Not all manure is created equal.

Major Flaw #1b: Net feed, water, and land use doesn't necessarily reflect environmental impact

Not all resources are created equal. Efficiency can create new environmental problems. Resource consumption stats should include resources destroyed by pollution.

Major Flaw #2 : Grossly inadequate assessment of "livestock"

Dairy cows and beef cattle aren't the only animals raised on factory farms.

Major Flaw #3 : Dubious analysis of greenhouse gas emissions

It's fine to use stats from papers you've already published - as long as your original analysis is accurate. Theirs was not.

Major Flaw #4 : Weak analysis of food miles

Just because food miles are an imperfect indication of transportation emissions doesn't mean you should start buying eggs from the other side of the world.

The Details:

Warning : if junk science makes you violently ill, stop reading now. But if you enjoy exploring the murky depths of misguided analysis, read the detailed critique below.

Major Flaw #1 : Grossly inadequate assessment of "environmental impact"

Throughout the paper, Capper et al. repeatedly assert that industrial livestock production has a smaller "environmental impact" than traditional livestock production. Although the authors briefly state that certain industrial livestock sectors use fewer resources and generate less waste per unit of food (more on this later), the only environmental impact they chose to describe in any detail is the emission of greenhouse gasses (GHGs).

Newsflash: Greenhouse gases aren't the only pollutants generated by livestock production. Even the most cursory analysis of contemporary livestock production practices would quickly reveal that industrial operations emit a host of additional pollutants that degrade the environment and threaten human health.

A very brief, far-from-comprehensive list of pollutants emitted by industrial livestock operations:

It's worth noting that many of these pollutants pose a more significant, less easily mitigated threat to the environment than GHG emissions. In any case, the many environmental impacts of industrial livestock production are described extensively in the scientific literature; any expert in this field should be well-aware that GHG emissions are only a small component of livestock's environmental footprint.

Major Flaw #1a:

Total manure production doesn't necessarily reflect environmental impact

Although the paper's assessment of environmental impact is based primarily upon GHG emissions, the authors do state (in passing) that industrial dairies generate less waste per unit of milk than traditional dairies. The implicit assumption is that there's a direct correlation between a livestock production system's waste output (as measured by manure production), and the magnitude of its adverse environmental impact. Here's why this assumption is invalid:

Location matters

As legions of earnest freshmen learn in Environmental Studies 101, a potential pollutant's location can affect the degree to which it impacts the environment. For instance, when a cow on a traditional farm excretes manure in a field, the manure serves as fertilizer, providing nutrients for the pasture - no problem! When a cow on a factory farm excretes manure, it's collected in a manure lagoon with a tremendous volume of additional waste. Since the quantity of manure exceeds the carrying capacity of surrounding land, it becomes a hazardous pollutant - Big Problem!

Manure storage/handling affects environmental impact

When waste is collected and stored in manure lagoons (the standard procedure on factory farms), the process of decomposition releases a chemical cocktail of hazardous air pollutants that impair human health and cause surrounding towns to smell like cesspools. Oh yeah - the lagoons are also prone to leaks, overflows, and catastrophic collapses, which increases the capacity of factory farm manure to pollute soil and water. Since traditional farms don't use lagoons, their manure doesn't generate these adverse environmental impacts.

Manure composition varies

All manure is not created equal; animal waste from factory farms contains pollutants that aren't present in manure from sustainable farms (e.g., antibiotics and/or antibiotic-resistant bacteria, hormones, heavy metals, etc.). Again, this means that a given volume of industrial manure has a greater potential environmental impact than the same volume of manure from a traditional farm.

Major Flaw #1b:

Net feed, water and land use doesn't necessarily reflect environmental impact

Capper et al. suggest that compared to traditional dairies and beef cattle systems, industrial operations require fewer "resources" (defined narrowly as feed, water and land requirements in the case of dairy, and land use and energy requirements in the case of beef cattle) to produce a unit of milk or beef.

The implicit assumption is that there's an inverse relationship between a production system's feed, water, land and animal energy requirements and the system's environmental impact (i.e., less resource use = smaller environmental impact). Of course, in the real world, this isn't always the case. For instance:

The production of different types of animal feed has different environmental impacts; e.g., maintaining pasture for grassfed animals has minimal impact but the production of feed crops for industrial livestock causes a wide range of problems (e.g., impact of pesticides, synthetic fertilizers, erosion, feed transportation, etc.). Thus, while a factory farmed animal might consume fewer calories, the environmental impact of producing its industrial feed might be greater than the impact of producing a larger quantity of non-industrial feed (e.g., grass) for an animal on a traditional farm.

Major Flaw #2 : Grossly inadequate assessment of "livestock"

Although Capper et al. present the paper as an analysis of food animal production systems (and specifically mention chicken, poultry, eggs and pork in the opening paragraph), they only devote significant attention to the GHG emissions of dairy and beef production; the paper includes no analysis of the impacts of producing pork, poultry or eggs.

Just so we're all on the same page: the paper's title is "Demystifying the Environmental Sustainability of Food Production" - but actually, it only addresses the environmental impact of livestock production... and by "environmental impact," Capper et al. really mean greenhouse gas emissions... and by "livestock," they mean only dairy cows and beef cattle.

Major Flaw #3 : Dubious analysis of GHG emissions

The comparative analysis of GHG emissions mentioned above is derived from a few papers that Capper and the gang published before this one. Not surprisingly, the earlier work contains substantial flaws. There's not enough space here to provide a thorough critique, but it's worth noting two glaring problems:

Major Flaw #4 : Weak analysis of food miles

At the end of the paper, Capper et al. slap on a somewhat incongruous section about food miles in which they note that the GHG emissions generated by the transportation of food cannot be accurately assessed by simply measuring the linear distance between the food's point of origin and its point of consumption. In other words, you can't say that the transportation of one food emitted fewer GHGs than the transportation of another food just because the first food was produced closer to your home.

They're absolutely right about this; GHG emissions generated during food transportation are affected not only by the distance from farm to fork, but also by the efficiency and carrying capacity of the vehicle (or vehicles) used for transportation. So if you hop in your Sherman tank, drive to a farm and buy one carrot, the transportation GHG emissions per carrot are much higher than if you drive to the farm in a compact hybrid and buy 20 carrots.

Capper et al. make this point in order to suggest that industrial foods shipped long distances to the supermarket might actually generate fewer transportation GHG emissions than local foods purchased at the farm or farmers' market. The authors demonstrate this possibility by introducing a totally fictitious hypothetical scenario in which eggs are purchased from a grocery store, a farmers' market and a farm.

In this example, the transportation fuel consumed per dozen eggs is smallest for the grocery store eggs - even though they're shipped 802 miles. This counterintuitive situation arises because the transportation of grocery store eggs is very efficient (they're shipped in a truck that carries 280,800 eggs at once), while the transportation of eggs to the farmers' market is much less efficient (eggs are shipped in a smaller, less fuel-efficient pickup truck) and the transportation of eggs purchased at the farm is even less efficient (a car is used to transport just 12 eggs at a time).

But before you start buying your eggs from the other side of the globe: