Widely diverse information exists on the size of the yield penalty associated with organic crop production. Some authors/spokespersons – often connected with organic production/marketing – claim organic yields are typically 80-100% of non-organic. (I prefer the term “non-organic” over “conventional,” because so much of modern agriculture is anything but conventional.) Other sources say 50-70% is more common.

One might ask, “Why does this matter?” Just let farmers grow what they believe they can grow profitably, sorting out yield and price-premium relationships for their individual farms, crops and market environments. And if profit expectation is too low relative to risk and management needs, then organic buyers can raise the price to stimulate production – or import organic produce from afar – just as occurs with any other farm commodity.

But the question is often voiced in more fundamental terms. Many in organic production/marketing/advocacy portray organic as morally superior and more sustainable, not withstanding some small reductions in yield. Others (example here) argue the reverse: that organic agriculture is bad ethically because of markedly lower yields and the attendant major increase in land needed to produce food.

There is more to sustainable agriculture than yield. Quality is important and so too are the long-term adequacy of input needs, energy usage and effects on environment. Organic may or may not be environmentally superior; it depends on which analysis, commentary or assumptions you read or use. But yield is highly important too.

Thus it was with pleasure that I read a quality analysis published in August 2016, by Drs. Kniss and Jabbour at the University of Wyoming and Dr. Savage in San Diego. It’s entitled, “Commercial Crop Yields Reveal Strengths and Weaknesses for Organic Agriculture in the United States.” A popularized column based on the paper was published simultaneously by Kniss. The strength of the paper is that, unlike any before, it is based on many thousands of actual on-farm records. The paper involves data only from the United States but it’s a major contribution and you must read it – ideally both the paper and column.

The following consists of a few comments on their analysis and will make more sense if you have some knowledge of what Kniss et al have written.

In briefest terms, the authors compared yield data provided for the year 2014 through a special USDA survey of more than 10,000 organic growers, with similar yield data collected for all US crop farmers through the 2014 USDA-NASS December Agricultural Survey (available here). They found that organic crops normally yield less than non organic but with a huge range in yield ratios across crops and across states. Notable exceptions are hay and haylage crops where organic crops yield as much or more – and up to 60% higher for haylage. On average, Kniss et al concluded that organic crops average 80% of non-organic.

I had serious doubts about both the 60% and 80% numbers and contacted the authors asking questions and providing further calculations using their data. I am pleased that the errors have been corrected. The stats properly show that organic haylage yielded 76% as much as its non-organic counterpart, on average. The average yield ratio (organic/non-organic) for all 65 crops included in their analysis is now 67%. Worded otherwise, 1.5 acres of land in organic production is needed, on average, to produce as much food as 1 acre of non-organic land, according to the USDA survey.

With the corrections, the Kniss et al paper produces results equivalent to those published by co-author, Steve Savage, one year ago. Here’s one graph from Savage’s web site for row crops. (Savage reported the organic yield gap as percent lower yield, rather than as a percent of non-organic crop yield as preferred by Kniss et al.) The Savage web site contains similar graphs for a range of other crops.

The yield depression for organic corn and soybeans is similar to that reported by crop insurance officials for those crops in Ontario. However, the yield depression is greater with organic winter wheat in Ontario (an average of 42% lower over eight years) than shown in the USDA data.

The authors highlight a conclusion in the USDA organic report that 40% of organic farmers reported using no-till or minimum tillage practices. Observing organic practices in Canada, I simply don’t believe this statistic and think it is a result of a USDA survey process which involved self-reporting. Vast numbers of farmers in North America likely believe that they practice “minimum tillage” – with “minimum” generally meaning less than what they did in times past or less than what they might have done. I am aware that organic researchers and some farmers are experimenting with no-till seeding using crimped cover crops to control weeds (with mixed success). But this still represents a minute percentage of total organic acreage.

Kniss et al cast doubt on claims that high-yield agriculture allows land to be diverted out of arable crop agriculture into conservation or other purposes. They cite a 2014 US report showing a decline in number of acres in the US land conservation reserve in years after 2007. In my view, any meaningful analysis of the effects of yield enhancement on land usage has to include a much longer time frame.

The following table copied from a USDA-ERS summary of historical agricultural statistics, shows the total acreage planted to principal crops from 1983 to 2015. (The USDA-ERS report contains a related table for the years 1909-1990 but with a different inclusion listing of “principal crops.”)

The acreage numbers show effects of poor crop prices in the mid 1980s, good prices around the year 1996 and after 2007, as well as anomalies such as the US “PIK” land set-aside program of 1983. However, the overall trend in principal crop acreage is down. For those statistically inclined, the slope of the linear regression line is -490 thousand acres/year with R² = 0.24 and P<0.01. Some that diverted land went into urban development for sure, but I expect much was land conversion into non-cropped rural landscape. Principal crop acreage declined even as usage for both food and non-food uses (biofuels and biomaterials) grew.

A final comment: It’s a common practice for some organic farmers to plow under a soil-building crop (for example, perennial legume or buckwheat), without harvesting, in year one to provide better growth for the crop in year two. When that’s the case, the harvested crop is actually the product of two years of growth and the reported yield per acre should be halved to calculate yield/acre/year. This adjustment is not included in the USDA data and I expect no one knows how large the adjustment should be. That factor does mean that the 67% is a slight over-estimate. (The same practice can occur in non-organic agriculture, though I believe to a smaller extent.)

With that noted, I do compliment the authors on an excellent and highly useful paper containing farm-level stats on the performance of specific crops.

And from a practical standpoint, it is likely more important for farmers to know that they might expect hay yields (though not haylage) comparable to non-organic with organic production, but corn, soybean and wheat yields which are 30-35% lower, and organic grape yields 50% lower – than it is to know what the US average is for all crops is 67% or whatever. Yield data for all 65 crops are contained in supplementary tables in the Kniss et al paper.

Thanks Drs. Kniss, Savage and Jabbour for a valuable contribution.