Observations on Competition in the U.S. Seed Industry

(Submitted December 2009)

  1. The seed market is competitive today
  2. Grower choice is increasing
  3. Farmers have benefited from changes in the seed industry
  4. Investment is driving a competitive pipeline
  5. Endnotes



Introduction

The U.S seed industry experienced two main periods of technology-driven change in the 20th century, first with the development of hybrid seeds in the 1920s, then with the introduction of seeds improved through modern biotechnology in the 1990s. American Cyanamid (now part of BASF) launched herbicide-tolerant corn in 1991 and DuPont launched herbicide-tolerant soybeans in 1993. Both of those traits were developed through chemical mutagenesis. The first transgenic crops followed shortly thereafter – herbicide-tolerant cotton from Rhone-Poulenc (now part of Bayer) in 1995 and insect-protected corn from Novartis (now part of Syngenta) and Mycogen (now part of Dow) in 1996. Monsanto launched herbicide-tolerant soybeans and insect-protected cotton also in 1996.

The 1990s also saw the acquisition of some of the larger independent seed companies by companies that had invested in the research and development of biotechnology traits. For example, in 1998 Monsanto acquired DeKalb Genetics Corporation, which accounted for about 11% of U.S. corn seed sales in that year, and in 1999 DuPont acquired the much larger Pioneer Hi-Bred International, which had accounted for about 39% of U.S. corn seed sales the previous year.i

Additional acquisitions by these and other companies have since occurred, along with numerous cross-licensing agreements between most of the leading technology providers. As a result, many biotech trait providers are vertically integrated with seed companies, enabling them to directly take on the risk of quickly introducing new traits in the seed they sell, and many of those seed companies also offer traits developed by someone other than their owner.

On the farm, biotech seeds have been rapidly adopted, especially in the major row crops such as corn, soybeans and cotton, due to the significant benefits they provide to growers. These include improved weed and insect control, greater yield, convenience, environmental sustainability, and increased profits. In 2008, there were approximately 156 million acres of biotech-improved crops planted in the United States.ii

Along with this period of rapid technological innovation and change have come questions about competitiveness in the industry. Specifically, it has been asked if trait providers have abused their market position to constrain farmers’ ability to choose a range of seed products at different price and quality levels. In this discussion, it is important to distinguish what products farmers actually choose from whether or not they have robust, meaningful choices. In other words, the fact that farmers purchase high performing products that may cost more does not mean that they were forced to make those decisions; based on the data presented, it appears that farmers had many choices along the price/quality spectrum.

This paper examines the question of competition through the lens of three large-scale row crops that have been improved through modern biotechnology: corn, soybeans and cotton. These three crops account for the vast majority of the 150-million-plus acres of biotech-improved crops planted in the United States last year. Four propositions appear to be true and will be defended: 1) the seed market for these crops is competitive today in terms of company shares, number of choices, and prices paid by farmers; 2) farmer choice is increasing with time; 3) farmers have benefited economically as a result of change and innovation in the seed industry; and 4) the leading seed companies are continuing to invest in new products that they plan to offer to farmers in the future. The four sections below address each of these propositions and, for the benefit of readers with limited time, each section ends with a bullet point summary of its conclusions.

I.The seed market is competitive todayiii

Shifting Shares

A look at the three charts below showing shares of U.S. sales of corn, soybean and cotton seed from 1997-2009 (2000-2008 for cottoniv) reveals these industries were diversified and dynamic. There is substantial movement in shares, with no single player achieving a dominant position. Note that where several companies or brands are currently owned by a single large company – including as the result of an acquisition – share has been combined and assigned to that company for all the years represented.v

Historic Share, U.S. Corn Seed: 1997-2009
Source: dmrkynetec


The corn seed data represented above illustrates the share held by the four largest companies (Lines A, B, C and D) and 169 smaller, independently owned seed companies (Line I). Company A and Company B exchanged position between 2006 and 2008. Company C has lost share overall but shows a small increase in 2009. Of particular note is line I, the independently owned seed companies, which grew from 1997 to 2007 and still have significant share.

Historic Share, U.S. Soybean Seed: 1997-2009
Source: dmrkynetec


The soybean seed data shown above depicts the share held by the four largest companies (Lines A, B, C and D) and approximately 153 smaller, independently owned companies (Line I). Company A has increased its share and now enjoys roughly the same share of the market as Company B and the independent seed companies (I), which have lost some share but have largely held their own. Company C and Company D have smaller shares and have been relatively unchanged.

Historic Share, U.S. Cotton Seed: 2000, 2002-2008
Source: dmrkynetec


In cotton seed, shares of the two largest players, Company A and Company F, have converged, while the independent seed companies (I) have maintained share over the time period following initial share loss between 2000 and 2002. Company D has also seen steady share growth over the time period.

Return to top





Many Seed Companies to Choose From

Competition for the farmer’s seed dollar is robust at the local level as well. In 2008, farmers were likelyvi able to buy corn seed from at least four different companies in most USDA Crop Reporting Districts (CRDs). In some CRDs in Illinois, Iowa, Minnesota, Nebraska, South Dakota and Wisconsin, farmers could probably buy corn seed from as many as 30 different seed companies. Again, seed companies with common ownership are treated as a single company.

Farmer Options for Corn Seed: 2008

A very similar picture is seen for soybeans, where farmers could probably purchase seed from at least four different seed companies in most CRDs in 2008. In some CRDs in Arkansas, Illinois, Iowa, Indiana, Kentucky, Minnesota, Ohio and North Dakota, farmers could probably buy soybean seed from as many as 22 companies.

Farmer Options for Soybean Seed: 2008

Farmers in most cotton CRDs could probably purchase seed from at least three cotton seed companies in 2008 and farmers could probably buy seed from as many as seven or more seed companies in some CRDs in Arizona and Texas.

Farmer Options for Cotton Seed: 2008

Return to top





Many Seeds to Choose From

The competition in the corn, soybean and cotton seed business is further evidenced by the large number of brands, hybrids and varieties that farmers reported buying from seed companies in 2009 (corn and soybean) and 2008 (cotton, where 2009 data are not available).

Farmer Reported Seed Choices: 2008-2009

The hybrids and varieties purchased in 2008-2009 featured not only a wide array of germplasm and maturity groups but also a number of traits developed through both biotechnology and traditional breeding.

For example, in corn, farmers reported buying hybrids with 30 different trait profilesix in 2009. This included a mix of herbicide tolerant traits, rootworm protection traits and corn borer protection traits. There were 22 different “stacks” that combined herbicide tolerance and/or insect protection traits often supplied by different companies. Stacks have been on the majority of U.S. corn acres since 2007. Approximately 10% of the corn seed farmers said they bought in 2009 was conventional, non-traited seed.

The chart below illustrates the aforementioned products, including those developed through licensing agreements.

Corn Trait Profiles
Source: dmrkynetec and Monsanto 2009


In soybeans, farmers purchased varieties with six different trait profiles in 2009, including herbicide-tolerance traits and one “stack” with tolerance to two different herbicides. These traits were developed using transgenic and non-transgenic breeding. Interestingly, the reported number of acres planted in conventional, non-traited soybeans grew from 3.8% of the total in 2008 to 5.3% in 2009 – one out of every 19 soybean acres.

The chart below illustrates the aforementioned products, including those developed through licensing agreements.

Soybean Trait Profiles
Source: dmrkynetec and Monsanto 2009


In cotton, farmers reported purchasing seeds with 12 different trait profiles in 2008. The biotech seed choices included herbicide tolerant and insect protection traits. There were six “stacks” purchased. Conventional, non-traited cotton accounted for 3.7% of reported seed purchases.

The chart below illustrates the aforementioned products, including those developed through licensing agreements.

Cotton Trait Profiles
Source: dmrkynetec and Monsanto 2009


Return to top





Many Prices to Choose From

With such a large number and extensive array of seed products, average price ranges are, to a certain extent, meaningless. They would be of no use to someone actually wanting to make a purchase decision. However, when considering the level of competition in the marketplace, this information becomes significant. With thousands of seed products (hybrids and/or varieties), hundreds of companies and brands, dozens of traits (both individual and stacked), and enormous variation in plant genetics, one would expect to see huge variations in the average price of corn, soybean and cotton seed. As the chart below shows, this is indeed the case.

Farmer Reported Average Price per Bag of Seed: 2008/2009

Growers are able to compare higher-priced seed and lower-priced seed in order to evaluate whether the higher-priced seed offers additional value, such as higher yield or the ability to reduce other inputs, which would justify the higher price. Farmers also have several information sources available to them as they make their purchase decisions, including the real-time data collected from their yield monitors (commonly included in harvest equipment such as corn and soybean combines, as well as cotton pickers) as well as varietal and hybrid yield data from third-party sources (e.g. universities, seed companies, etc). Based on such an evaluation they would be expected to choose the seed offerings that bring them the highest profitability on their farms.

Traits are Broadly Licensed

Patented traits developed through biotechnology are not available only through the seed companies that are owned by the developers of that trait. Monsanto, for example, has elected to pursue a strategy of broadly licensing its patented traits to competitors large and small. Other trait developers have come to follow Monsanto’s lead and have also licensed traits to seed companies they do not own. In addition, the trait developers have many cross-licenses that enable their seed companies to use each other’s traits in combination with traits developed by others. As the number of traits has grown, many seeds include (or are “stacked” with) more than one trait and it is common to find seeds stacked with traits from two and sometimes three separate companies.

Several things should be kept in mind with respect to this situation. First, traits are only a part of the total value of the seed, which is why there is substantial competition and variation in price even among seeds that contain the same trait combination. (While the traits may be identical, the underlying genetics of the seed are not, and the quality of the underlying genetics is the most important component of the value of a seed.) Second, even though the patent holder can keep others from commercializing its discovery for the duration of the patent, broad licensing can facilitate competition by enabling a much larger number of competitors to offer the new technology. Third, the farmer benefits from this arrangement by virtue of the larger number of product choices presented to him than would otherwise be the case. In sum, a business climate that encourages broad licensing achieves a desirable balance between the rights of the innovator (the patent holder) and the benefits of increased variety in the marketplace.

Return to top





Conclusions

  • No single company has a dominant share of seed sales in corn, soybean or cotton. Shares have changed over time. Independent seed companies, numbering in the hundreds, have held their own and have significant share in corn, soybeans and cotton.
  • Farmers have the choice of several seed companies at the local level. Corn and soybean growers were able to purchase seeds from a minimum of four companies in most Crop Reporting Districts (CRDs) in 2008. In several corn CRDs, farmers were probably able to purchase seed from more than 20 companies.
  • Farmers can choose from among hundreds of seed companies, hundreds of brands, and thousands of hybrids and varieties. These choices include seeds with individual and stacked biotech traits, traits developed through traditional breeding, and conventional seeds.
  • As would be expected in such a diversified marketplace, average prices paid by farmers per bag of seed vary enormously, as much as nine-fold in some cases. Farmers may prefer the higher-value seeds but they still have a multitude of choices at many quality and price levels.
  • The broad licensing strategy pursued by Monsanto and, to a lesser extent, by other companies with patented traits has facilitated competition and brought greater choice to the farmer.

Return to top





II. Grower choice is increasingx

Seed Offerings Are Large and Include Both Traited and Conventional Seed

The chart below shows the total number of seeds with one or more traits (including those developed by non-transgenic breeding) and the total number of conventional seeds offered to the market in 2005 and 2009/10. The chart below is based on a broad review of seed company catalogs and marketing materials for those years.

Despite the smaller number of corn and/or soybean seed companies included for 2010 (i.e. 178 companies in the 2010 data set vs. 252 companies in the 2005 data set), the number of traited corn seeds offered increased. The number of traited soybean and cotton varieties declined. Conventional seeds for all three crops also declined but still accounted for 12.8 percent of the total in 2009/10. The number of offerings is broadly consistent with the figures presented above regarding farmer reported seed choices, and overall, as was reflected in the ranges in the pricing data described earlier, these data suggest farmers do, in fact, have a wide range of seed choices.

Seed Catalog Offerings: 2005 and 2009/10

Return to top





Trait Mix Becoming More Diversified

The chart below shows which trait providers had at least one traitxi in the seeds offered in catalogs in 2005 and 2009/10. For example, at least one trait from Company A was present in 81% of the 6,079 traited corn hybrids offered in 2010, at least one from Company C was present in 11%, and so forth. (The number totals to greater than 100 because multiple traits appear on “stacked” products.) Note that despite the fact that the seed companies in this analysis are Monsanto licensees, their offerings do include traits from other technology providers. It is also important to note that licensing agreements enable seed companies to use the trait provider’s technology in their seed offerings, but do not (at least in the case of Monsanto’s licenses) require such use.

The table shows significant changes in the “trait mix” between 2005 and 2009/10, with greater diversification and increasing presence for many trait providers.

In corn, the leading trait provider had at least one trait in 86% of the hybrids offered in 2005; this declined 5 percentage points in 2010. Company F’s percentage more than tripled to approximately 24%, Company D went from three to 13%, and Company C went from one to 11%.

In soybeans, the picture changes somewhat less, although the leading trait provider’s presence has declined by 5 percentage points, while Company F has entered the market with at least one trait in 7% of the biotech varieties offered. Company B’s presence has grown from five to nine percent.

In cotton, there are fewer varieties offered which may reflect the significant decline in the number of cotton acres planted in the United States. However, there has been a steady change in the trait mix as all of the trait providers have grown.

Changes in Trait Mix in Seed Catalogs: 2005 and 2009/10

Return to top





Conclusions

  • Farmers had a wide range of choices offered to them in seed company catalogs in 2009 and 2010, for both biotech seeds and conventional hybrids/varieties. In corn alone, there were more than 6,000 traited hybrids and over 1,000 conventional ones offered for 2010 planting.
  • There has been a change in seed company offerings of biotech traits, with greater diversification and increased presence for many trait providers.
  • Conventional seeds continue to be offered and represent a meaningful choice for farmers.
  • These data indicate a dynamic situation with new entry bringing new value propositions to farmers at a rapid pace.

Return to top





III. Farmers have benefited from changes in the seed industryxii

The U.S. seed industry has undergone significant change since the introduction of the first genetically enhanced seeds in 1995. The data discussed above highlights that farmer choice in terms of the number of seed types is broad and has not fallen meaningfully, and the data discussed below suggest that the improved quality of the seed they buy has greatly benefited them economically. In total, biotech corn, soybeans and cotton are reported by Brookes and Barfootxiii to have increased farm income by almost $20 billion between 1996 and 2007.

Brookes and Barfoot: U.S. biotech crop farm income benefits - 1996-2007
- In billions of dollars -

Return to top





Herbicide tolerant corn

The authors state that the main economic benefit to farmers has been to reduce costs and thereby improve profitability levels. Average profitability improved by $49-$62/acre, resulting in a net gain to farm income in 2007 of $392 million. Cumulatively, since 1997 the farm income benefit has been about $1.4 billion. In added value terms, the increase in farm income in just 2007 was equivalent to an increase in production of 0.81%.

Brookes and Barfoot: National farm income impact of using biotech herbicide-tolerant corn in the United States - 1997-2007
 

Return to top





Corn borer protected corn

According to the authors, the primary impact has been increased average yields of about 5%. The annual total national farm income benefit from using insect-protected corn has risen from $8.76 million in 1996 to $1.14 billion in 2007. The cumulative farm income benefit between 1996-2007 was about $3.9 billion. In added value terms, the increase in farm income in just 2007 was equivalent to an increase in production of 2.28%.

Brookes and Barfoot: Farm level income impact of using biotech insect-protected corn in the United States - 1996-2007

Corn rootworm protected corn

Corn rootworm protected (CRW) corn has been planted commercially in the United States since 2003. The authors state that the main farm income impact has been higher yields of about 5% relative to conventional corn. At the national level, farm incomes increased by $4.6 million in 2003, rising to $548 million in 2007. Cumulatively since 2003, the total farm income gain from the use of CRW technology in the U.S. corn crop has been nearly $900 million.xiv

Return to top





Herbicide tolerant soybeans

According to Brookes and Barfoot, the annual total national farm income benefit from using biotech soybeans has risen from $5 million in 1996 to $1.36 billion in 2007. The cumulative farm income benefit between 1996-2007 was $10.4 billion. In added value terms, the increase in farm income in recent years was equivalent to an annual increase in production of between 5-10%.

Brookes and Barfoot: Farm level income impact of using biotech herbicide-tolerant soybeans in the United States - 1996-2007

Return to top





Herbicide tolerant cotton

The authors state that the primary economic benefit to farmers has been to reduce costs and thereby improve profitability, with annual average profitability increasing by between $7-121/acre, resulting in a net gain to farm income in 2007 of $16 million. Cumulatively since 1997 the farm income benefit has been $800 million. In added value terms, the effect of increased farm income in just 2007 was equivalent to an increase in production of 0.31%.

Brookes and Barfoot: National farm income impact of using biotech herbicide-tolerant cotton in the United States - 1997-2007

 

Return to top





Insect-protected cotton

Brookes and Barfoot state that the primary economic benefit has been increased yields (from 9-11%), although small net savings in costs of production have also been obtained (reduced expenditure on insecticides being marginally greater than the cost of the technology). This resulted in a net gain to farm income in 2007 of $274 million. Cumulatively, since 1996 the farm income benefit has been $2.2 billion. In added value terms, the effect of the increased yields and reduced costs of production on farm income in just 2007 was equivalent to an increase in production of 5.1%.

Brookes and Barfoot: Farm level income impact of using biotech insect-protected cotton in the United States - 1996-2007

 

Return to top





Benefits to non-adopters

In addition to the financial benefits enjoyed by farmers who adopt biotech seeds, economic studies have indicated that significant benefits also accrue to farmers who choose not to adopt them. The case of glyphosate tolerant soybeans is illustrative. From 1995-2000, the percentage of U.S. soybean acres treated with each herbicide class except glyphosate declined. During this period, use of imazethapyr decreased by 32%, use of trifluralin by 16%, and use of chlorimuron by 6%. Prices declined as well: chlorimuron and imazethapyr declined by 40-50% in 1997 and 1998. The conclusion reached by two economists who studied this phenomenon was: “Release of a GM [genetically modified] variety impacts prices of competing pesticides used on the conventional varieties, making the conventional variety less costly than prior to introduction of the GM variety. This causes an increase in surplus for those farmers who adopt the GM variety, as well as those who plant the conventional variety….” xv

Return to top





Conclusions

  • Farmers in the United States have enjoyed great benefits from the introduction of biotech seeds – research has calculated that use of biotech corn, soybeans and cotton increased U.S. farm income by nearly $20 billion in the period 1996-2007.
  • Even non-adopters of biotech seeds have benefited as competition has led to a decrease in the prices of pesticides used on conventional varieties.

Return to top





IV. Investment is driving a competitive pipeline

As noted at the beginning of this paper, the U.S. seed industry entered a period of intense technology-driven change during the mid-1990s with the introduction of seeds improved through modern biotechnology. This period was preceded by many years of research and development, involving hundreds of millions of dollars and a high degree of risk.

This investment has continued. The available evidence suggests a clear expectation of future profits from new products on the part of trait developers. The charts below depict the wide array of new products that have advanced far enough to be identified by grower organizations as part of the “pipeline” of traits they are anticipating. As noted, commercialization depends on many factors, including successful completion of the regulatory process. But the pipeline is sufficiently robust as to virtually guarantee many new product offerings over the next decade – many of which, if successful, will bring great value to farmers, the environment, and society at large.

The U.S. Grains Council has identified a total of 20 new products in the corn pipeline starting in 2010, ranging from various new forms of insect protection and herbicide tolerance to “breakthrough” applications like drought tolerance, improved nitrogen utilization, and improved feed quality. The major technology providers are all well represented, both alone and in collaboration with each other.

Soybean industry associations have identified 27 new products in the soybean pipeline starting in 2010, including a wide range of both quality and agronomic traits. Similar to the industry’s corn pipeline portfolio, the major technology providers have all invested in one or more of these products, either alone or in collaboration. Note also the involvement of trait developers from the public sector (Virginia Tech, USDA).

Return to top





Conclusion

  • Ongoing investment on the part of trait developers is reflected in a robust new product pipeline featuring a wide array of traits coming from diverse developers. The farmer will be the primary beneficiary of this increase in new product offerings as companies compete with each other to bring him increasing value, but the environment and society at large will benefit as well.

Return to top





Endnotes

i Nicholas Kalaitzandonakes and Marvin Hayenga, "Structural Change in the Biotechnology and Seed Industrial Complex: Theory and Evidence," presented at the International Consortium on Agricultural Biotechnology Research Conference (June 17–19, 1999), p. 219.

ii International Service for the Acquisition of Agri-Biotech Applications (ISAAA) Brief 39-2008.

iii The information in this section comes from dmrkynetec, the nation’s leading provider of market research information for agriculture. The dmrkynetec data are obtained from farmer surveys in which respondents report what they purchased, how much they paid, how many acres they planted, etc. dmrkynetec treats the information as confidential and proprietary and only allows it to be cited with permission. dmrkynetec has placed restrictions on the manner in which this information can be presented; for example, specific company market shares may not be publicly disclosed. Information about how to obtain access to dmrkynetec data can be found at that company’s website: www.dmrkynetec.com.

iv dmrkynetec data for cotton are only available for 2000 and 2002-2008.

v For example, Syngenta acquired Garst and Golden Harvest in 2004, so those companies would be aggregated under Syngenta in the data for the entire period covered, not just post-2004.

vi The dmrkynetec data represent purchases as reported by farmers, and so would reflect the minimum number of seed companies that a farmer could have reasonably turned to for seed in any given area. The corn and soybean maps have been adjusted to reflect Monsanto’s assumption that at least three of the four major vertically integrated seed companies would have been available to the farmer in all CRDs even if, due to a small sample size, they were not always reported. The corn and soybean maps are available on Monsanto’s website in interactive format showing which seed companies have been included in each CRD.

vii Percentages on all three maps are from dmrkynetec data.

viii Note that dmrkynetec does not determine whether hybrids/varieties offered by one seed company are the same as those offered by another and as such the figure cannot be understood as representing unique hybrids/varieties.

ix “Trait profile” as used in this section includes single traits, stacked traits and conventional, untraited seed.

x The information in this section comes from an analysis of company seed catalogs and other marketing materials conducted for Monsanto by Precision Agricultural Services, Inc. The analysis includes data from 252 corn and/or soybean seed companies and eight cotton seed companies in 2005, 178 corn and/or soybean seed companies in 2010, and seven cotton seed companies in 2009. Three things should be noted: 1) the data set does not include all seed companies; 2) the data set was based on the catalogs and marketing materials published by companies licensed by Monsanto and 3) the data reflect what was offered for sale by these seed companies, not (as with the dmrkynetec data in the previous section) what farmers actually reported purchasing.

xi Includes non-transgenic traits.

xii Other than as indicated in the subsequent notes, all data in this section are from Brookes and Barfoot, “GM crops: global socio-economic and environmental impacts 1996-2007.” Originally published in AgBioForum 11(1):21-38 as “Global Impact of Biotech Crops: Socio-Economic and Environmental Effects, 1996-2006” and subsequently updated by authors.

xiii Brookes and Barfoot, op cit.

xiv Brookes and Barfoot did not publish a table summarizing these data.

xv Huso and Wilson, “Producer Surplus Distributions in GM Crops: The Ignored Impacts of Roundup Ready® Wheat,” Journal of Agricultural and Resource Economics 31 (2):339-354

xvi U.S. Grains Council

xvii American Soybean Association, U.S. Soybean Export Council, United Soybean Board

Return to top