2. An overview of global seed markets

This chapter presents key figures about global seed markets, including breakdowns of market size by region and crop, data on the importance of genetically modified (GM) seed across countries and crops, and information on trade, prices, and R&D spending in the industry.

2.1. Sources of seed

Seed used by farmers can come from three sources: farm-saved seed; purchased seed derived from public plant breeding, or purchased seed from the private sector (Heisey and Fuglie, 2011[7]).1 Originally, all seed was farmer-saved. Over the past century and a half, seed originating in the public sector has played an important role in many countries, including the United States (Kloppenburg, 1988[14]), Australia and Canada. Over time, the importance of the private sector has grown and private-sector seed now dominates global markets, especially in high-income countries (Heisey and Fuglie, 2011[7]).

The share of farm-saved seed varies across regions and crops.2 Figure 2.1 shows how estimated rates of farm-saved seed vary from less than 10% of the total volume of seed used in North America to more than 60% in the developing regions of Asia and the Middle East and Africa.

Similarly, Figure 2.2 shows how the estimated rate of farm-saved seed varies from close to zero for sugarbeet to more than 60% for wheat and barley, rice, and potato.

For some crops, farm-saved seed (as well as public seed varieties) continue to play an important role even in developed economies.3

2.2. Size and growth of the global commercial seed market

The size of the global commercial seed market (i.e. excluding farm-saved seed but including public commercial varieties) was estimated at around USD 52 billion in 2014 (Figure 2.3).4 In the past years, global seed markets have grown strongly in (nominal) value, driven by the expansion of GM in particular, although conventional seeds also registered growth.

2.3. Seed markets by region

Estimates of the regional split of seed markets are presented in Figure 2.4. The United States is globally the largest seed market, followed by the People’s Republic of China (hereafter “China”). The next most important national seed markets, led by France, Brazil and Canada, are considerably smaller. Regionally, North America is the largest market with an estimated one-third of the global market by value.

Taken as a whole, the European Union is the third-largest seed market in the world after the United States and China, accounting for 20% of the global total (Ragonnaud, 2013[17]). France by itself accounts for nearly one-third of the EU total. France, Germany, Italy, Spain, and the Netherlands combined account for two-thirds of the EU market.5

2.4. Seed markets by crop

While the global seed market is diverse, a small number of crops dominate total seed sales (Figure 2.6). Of an estimated USD 52 billion in 2014, almost 40% is made up of maize and an additional 14% of soybeans. The large size of maize and soybeans is driven by the Americas where GM varieties are commonly used. Globally, 78% of the area planted with soybeans uses GM varieties. The typically higher prices for GM seed automatically lead to a larger estimate of market sizes in terms of value.

Rice is estimated to be the third largest seed market by value, driven by demand in Asia Pacific where around 90% of the global area planted with rice can be found.

The global market for vegetable seed is estimated to be around USD 4.7 billion (Figure 2.6). Vegetable seeds typically have a high value and account for a much larger share of the global seed market by value than would be expected from their relatively modest volumes. Within vegetable seed, an estimated 43% of the market consists of Solanaceae seeds – a crop category which includes tomatoes, peppers, and eggplants (Figure 2.7).

2.5. Genetically modified (GM) seeds

Genetically modified seeds have had a drastic impact on the structure and evolution of global seed markets.6 Advances in genetics led to the development of the first genetically modified (GM) plant in 1982, and the first commercialisation of GM plants took place in the early 1990s with the introduction of the Flavr Savr tomato variety (1994). Today, at a global level 190 million hectares are planted with GM crops. Since 2012, the area planted with GM crops in developing countries has exceeded that in developed countries. In 2017, developing countries accounted for 53% of the global area under GM crops, a share which is expected to grow further (ISAAA, 2017[18]). In 2107, the countries with the largest GM area are the United States (75 million hectares, or 40% of the global total), Brazil (50 million hectares, 26%) and Argentina (24 million hectares, 12%) (Figure 2.8).

In the European Union, only one GM “event” has been approved for cultivation: a type of insect-resistant maize grown on 130 000 hectares in 2017.7 Spain (124 000 hectares, representing around 28-30% of the total maize area) and Portugal (7 000 hectares) are the only countries where this GM maize variety is planted (ISAAA, 2017[18]) (European Seed Association, 2016[19]). Other EU Member States which used to have GM cultivation are the Czech Republic, Slovakia, and Romania, but none of these countries planted GM crops in 2017.

The main GM crops globally are soybeans (94 million hectares in 2017), maize (60 million hectares) and cotton (24 million hectares) (Figure 2.9). GM adoption rates vary among these commodities. Seventy-seven per cent of hectares planted with soybeans are planted with a GM variety. For cotton, this share is 80%, but for maize only 32%.

Three types of GM crops can be distinguished: crops with enhanced input traits (e.g. herbicide tolerance; resistance to droughts, pests, diseases); crops with enhanced output traits (e.g. crops with better micronutrient availabilities); and crops for non-traditional uses (e.g. crops that produce pharmaceuticals or bio-based fuels). The adoption of GM crops is mostly limited to the first category (Fernandez-Cornejo, 2004[1]) and have two main traits: herbicide tolerance and insect resistance.

A leading example of organisms with genetically modified herbicide tolerance traits are crops with genetic modifications making the plant resistant to the herbicide glyphosate, such as Monsanto’s Roundup Ready GM trait.

A leading example of organisms with genetically modified insect resistance traits are crops that incorporate a gene of the soil bacterium, Bacillus thuringiensis (Bt). Bt produces a protein which is toxic to certain insects. By incorporating the gene into the genetic material of plants such as maize or cotton, the plant produces the protein in its leaves, thus providing protection against insects such as the European corn borer (for maize) and the bollworm (for cotton).

In the past, most GM crops were either herbicide tolerant or insect resistant, but stacked traits (combining different traits) have recently been gaining importance. Stacks can include traits for tolerance to several herbicides and/or traits for resistance against different insects (e.g. corn borer and root worm in maize). Globally, single-trait herbicide-tolerant seed occupied 89 million hectares in 2017, or 47% of the total hectares of GM crops worldwide. Single-trait insect resistant seed was used on 23 million hectares (12% of the total). The area covered with single-trait GM crops has been declining in recent years, but the area covered with stacked traits has registered strong growth and at present occupies 78 million hectares. Stacked traits currently account for 41% of global GM hectares (Figure 2.10).

2.6. International trade in seeds

Seeds are widely traded. In 2015, around 3.9 million tonnes of seed were traded, representing a value of more than USD 10 billion according to statistics gathered by the International Seed Federation (ISF). This compares to an estimated value of the global seed market of around USD 50 billion. Although caution is needed in comparing these numbers, trade seems to represent about one-fifth of the value of the global seed market.

Figure 2.11 compares global exports of field and vegetable crops, by volume and by value, in 2009 and 2015. In terms of volume, the majority of trade (>95%) is in seeds of field crops. However, vegetable crop seed has a much higher value per weight. As a result, vegetable crop seed represented 35% of exports in value terms, but less than 5% in volume terms in 2015.

International trade in field crop seeds has been expanding strongly in both volume and value terms. Between 2009 and 2015, volumes grew by almost 80% while the value of exports rose by 37% (indicating a decline in unit values). By contrast, volume growth for vegetable seeds has been almost flat. Nevertheless, the value of exports of vegetable crop seeds grew by 32%, driven by an increase in the unit value of exports.

Figure 2.12 shows the main exporting countries (by value) in field crop seed and vegetable crop seed. For field crop seed, France, the United States and Germany are the main exporters, accounting for 40% of global exports by value. For vegetable crop seed, exports are more concentrated; the main exporters are the Netherlands, the United States and France, accounting for more than 60% of exports by value.

Figure 2.13 shows the main importers of seeds for field crops and vegetable crops in 2015. Imports are more dispersed than exports for both categories. Interestingly, there is considerable overlap between the list of main importers and exporters, even within the same category. For both field and vegetable crops, no less than six of the top-ten exporters are also among the top-ten importers. For field crops, the top importers are France, the United States, Germany, Hungary, the Netherlands, and Italy; for vegetable crops, these are the Netherlands, the United States, France, China, Italy, and Japan. This pattern is explained by a high degree of re-exporting as the seed value chain may cross borders multiple times. For instance, Hungary imports the parent seeds of maize and sunflower, breeds the hybrid crosses, and exports the resulting hybrid seeds back to markets such as France and the United States. The Dutch seed industry tends to outsource vegetable seed multiplication to many countries. Final processing occurs in the Netherlands and seeds are re-exported.8

International trade in seeds depends on an efficient and reliable system to certify the identity of the varieties traded. This is made possible through international co-operation in the OECD Seed Schemes (Box 2.1).

Box 2.1. The OECD Seed Schemes

The OECD Seed Schemes were established in 1958 and provide an international framework for the varietal certification of agricultural seed in international trade (OECD, 2012[20]).

Once a country has been accepted, its certification standards for the identity and purity of seeds are considered equivalent to those of other member states in the same Seed Scheme. The goal is to facilitate trade by reducing technical barriers, improving transparency, and lowering transaction costs.

Membership is voluntary and includes many non-OECD states. Currently, 60 countries participate in various Seed Schemes, although with varying participation per Scheme. Participation in the Cereals Seed Scheme, for instance, is nearly universal, while that in the Vegetables or Beet Seed Schemes is limited.

In total, around 50 000 varieties and 200 species are covered under the OECD Seed Schemes. The quantity of seed certified continues to grow and currently exceeds 1 million tonnes out of a total estimated international seed trade volume of around 4 million tonnes (OECD, 2016[21]) (ISF, 2017[22]).

Recent work at the OECD in the context of broader research on trade-related international regulatory co-operation has demonstrated the positive effect of the OECD Seed Schemes on international trade in seeds. Results show that the value of seed exports can increase by more than 12% if a country joins the OECD Seed Schemes. If both the exporter and the importer are members of the same Seed Scheme, results show that trade is about 30% higher (OECD, 2017[23]). This demonstrates the important role the Schemes play in facilitating international trade in seeds.

2.7. Prices

Seed for sugar beets, vegetables, maize, and soybeans are typically more expensive while the seed price for wheat is typically lower (Bonny, 2017[12]). In addition, seed prices vary depending on which GM traits are included (if any), as well as on local market conditions. Giving a full account of seed prices is therefore difficult even if data were easily available. However, some broad findings can be presented.

United States

Figure 2.14 shows estimates of US maize seed prices (in USD per acre) in 2011, highlighting the pricing differentials between conventional seed and different types of GM seed. Conventional maize seed cost around USD 54 per acre (USD 133 per hectare), while different single-trait GM seeds cost around USD 67 per acre (USD 167 per hectare).9 Double-stacked traits are more expensive, costing on average around USD 77 per acre (USD 190 per hectare), while triple-stacked traits are more expensive still at around USD 91 per acre (USD 226 per hectare). Unsurprisingly, GM seeds are more expensive than conventional seeds, although the premiums vary from around 25% for single-trait GM seed to around 70% for triple-stacked traits (stacks with more than three traits are also commercially available).10

Figure 2.15 shows the evolution of US maize seed prices between 1996 and 2011. In real terms, seed prices have grown considerably during the late 2000s. Conventional seed prices increased by 54% between 2001 and 2011, after correcting for inflation. The price of glyphosate-tolerant maize seed increased by 74% in real terms over the same period. Most of the increase appears to have taken place between 2007 and 2010. The high output prices during this period may be a partial explanation for the observed price evolution as these increased farmers’ willingness to pay while simultaneously increasing seed production costs (Ciliberto, Moschini and Perry, 2017[24]).

To put these developments in context, Figure 2.16 compares seed costs for US maize producers with the gross value of production and other operating costs since 1975. All values are expressed in 2009 USD per planted acre. Panel (a) shows that operating costs in the long run broadly follow the trend of gross value of production: declining between 1975 and the early 2000s, an increase during the 2000s, and a decline in recent years. Panel (b) looks in more detail at the relationship between seed costs and gross value of production since 1996. Seed costs and gross value of production both increased strongly between 2005 and 2011, a period during which the US maize price almost tripled in real terms. After reaching a peak in 2012, the real US maize price had fallen by half by 2016. Panel (b) shows that growth in seed costs slowed down after 2011, and declined between 2015 and 2016.

Likewise, Bonny (2014[25]) shows that the seed cost for soybeans in the United States has remained relatively stable between 2001 and 2013 when measured as a percentage of the gross product per hectare. She points out that prices of both GM and conventional soybeans increased over this period as seed prices are often based on soybean prices at the Chicago Board of Trade. In addition, higher seed costs were generally compensated by lower costs for pesticides.

European Union

A study by Wesseler et al. (2015[26]) on the agricultural inputs sector in the European Union looks at the evolution of seed costs as a share of total farm costs. Wesseler et al. (2015[26]) analyse the evolution of the seed cost share across the NUTS2 geographical regions of the European Union and find that the seed cost share increased modestly over time between 1989 and 2009. Estimates indicate that the seed cost share for the average farm increased from around 5.2% of total costs in 1989 to around 6.3% of total costs in 2009 for the EU-15. Moreover, the estimates indicate that since 2004, seed costs in the EU-15 have been declining as a share of total costs (although the change appears modest).

Figure 2.17 presents additional data on the evolution of seed prices in the European Union, using Eurostat data. As data availability differs by country, this figure shows information for 11 countries for which a seed price index is available since 2001. Panel (a) shows the evolution of the real price of seed in these countries between 2001 and 2016. There is no clear trend in seed prices. In some EU Member States seed prices have increased strongly (e.g. Czech Republic, Latvia, Malta), in others there is evidence of a decrease (Slovakia, Finland), and in yet others, changes have been modest (e.g. Luxembourg, United Kingdom).

Panel (b) shows an unweighted average over time of the price indices for these countries, as well as indications of the “spread” around this average. Over the period as a whole, there is evidence of a modest increase (6% over the 15-year period), but with variations over time. Prices increased in real terms in most countries between 2006 and 2008, declined between 2008 and 2010, and increased in 2011-2013. This development corresponds to the path of cereal prices in the European Union over the same period (Figure 2.18). As in the United States, seed prices in Europe appear to be strongly influenced by price developments in output markets.

2.8. Research and development

The growth of private R&D

Following the development of hybrid maize in the 1930s and the strengthening of intellectual property rights, plant breeders could expect a greater private return from investments in research and development (R&D). As a result, private R&D in plant breeding has been growing significantly over time. In real terms, private R&D expenditures on plant breeding in the United States increased almost fourteen-fold between 1960 and 1996 (Fernandez-Cornejo, 2004[1]), and private R&D spending has continued to increase in recent years (Fuglie, Clancy and Heisey, 2017[27]).

The growth in private R&D for crop improvement is also seen at a global level. Figure 2.19 presents estimates of global private sector R&D in agricultural input industries between 1990 and 2014, using data from Fuglie (2016[28]). In real terms, private sector R&D for seed and biotech has grown by a factor of three over this period, with most of the increase occuring after 2004.

The growth in private R&D for seeds and biotech is remarkable in comparison with other agricultural input sectors. In 1990, private R&D for seeds and biotech, farm machinery, and animal R&D (including animal health and animal genetics) was considerably lower than R&D for fertilizer and crop protection. In the past quarter-century, private R&D grew faster for seeds and biotech than for any other agricultural input industry. Real R&D spending grew 200% for seed and biotech compared to 190% for farm machinery, 88% for animal R&D, and 22% for crop protection and fertilizers.

Seeds and biotech also have a higher research intensity than other agricultural input industries (Figure 1.2). R&D for crop seed and traits represented more than 10% of sales in 2009, considerably higher than the R&D intensity in, for example, crop protection or farm machinery (although these numbers predate the increase in farm machinery R&D visible in Figure 2.19, which may be tied to the development of precision agriculture).

The large increase in R&D is driven to an important degree by traits for genetically modified crops, but indirect evidence suggests that R&D intensity is also high for conventional (non-GM) plant breeding. For instance, the German seed company KWS obtains about half of its revenues from the European market where the share of GM is negligible, yet spent EUR 190 million on R&D in 2016-2017. This corresponds to an R&D intensity of 14%. Similarly, 44% of the revenues of the French seed company Limagrain/Vilmorin originate in the European market; Vilmorin invested EUR 240 million on research in 2016-2017, corresponding to an R&D intensity of 15%.11 These data are only suggestive, as most of this expenditure may be on traits research for those markets where GM seeds are in wide use. Nevertheless, qualitative information offered by both firms in their annual reports confirms sustained efforts using conventional breeding for both firms. Similarly, in the Dutch vegetable plant breeding industry, R&D expenditures are estimated to be 15%-30% of sales in the absence of GM technology (Schenkelaars, de Vriend and Kalaitzandonakes, 2011[29]).

The evolving roles of private and public R&D

The development of plant breeding has been shaped by the interplay of public and private efforts. While the relative contribution of private and public plant breeding differs by country, in general the public sector has played an important role in improving plant varieties.

In some countries, efforts originated in the private sector and were subsequently supplemented by research in the public sector (Kingsbury, 2009[30]). This was the case, for instance, in Sweden, the United Kingdom, and the Netherlands. In Sweden, private efforts led to the creation of the Svalöf Institute at the end of the 19^th century, which became part of a state system by the early 20^th century. In the United Kingdom, despite the establishment of public plant breeding institutions, only 40% of the funds for plant breeding came from public sources between 1910 and 1921. However, following World War II, public plant breeding received a major boost, and wheat varieties bred by the public Plant Breeding Institute (PBI) dominated agriculture by the 1970s, as did potato varieties originating in the Scottish PBI.

In the Netherlands, small seed firms, hobby breeders and farmer cooperatives have played an important role in plant breeding, possibly because there was a form of legal variety protection at an early stage. The Netherlands set up a public institute for plant breeding in Wageningen in 1912, the Foundation for Agricultural Plant Breeding (Stichting voor Plantenveredeling) in 1948, as well as the Institute for Horticultural Plant Breeding (Instituut voor de Veredeling van Tuinbouwgewassen). Nevertheless, private plant breeders remained important as public institutes concentrated on research rather than breeding and were not allowed to release commercial varieties of crops where private breeding companies existed.

In other countries, the role of the public sector was even more prominent. This was particularly the case for the United States. Already in 1819, naval and consular personnel were encouraged to collect plants which could be useful to US agriculture, and the US Patent Office played an important role in collecting and distributing plant material throughout the 19^th century. Since the establishment of the United States Department of Agriculture (USDA) and land grant universities in 1862, and State Agricultural Experiment Stations (SAES) in 1887, these institutions have devoted considerable efforts to plant breeding and distribution of improved varieties (Kingsbury, 2009[30]) (Kloppenburg, 1988[14]).

The strong growth in private R&D has led to a change over time in the relative importance of private and public R&D (Fernandez-Cornejo, 2004[1]). In 1960, public spending accounted for 60% of total R&D on crop improvement for maize in the United States. By 1984, this share had fallen below 40%. For soybeans, practically all crop improvement research in the United States was conducted by the public sector in 1960; by 1984, the share of the public sector had fallen to about three-quarters of the total. Comparable data for recent years is not available, but given the emergence of genetically modified seeds during the 1990s the private share of crop improvement R&D for maize and soybeans has almost certainly continued to increase.

R&D expenditures by large private companies now dwarf the R&D budgets of the largest public sector agricultural research agencies. In 2007, the R&D budgets of Bayer (USD 978 million), Syngenta (USD 830 million), Monsanto (USD 770 million), BASF (USD 655 million), and DuPont (USD 633 million) exceeded the crop science research budget of the USDA’s Agricultural Research Service (USD 456 million) and that of CGIAR (USD 178 million) (Fuglie et al., 2011[9]).

Given the rising importance of private R&D, public and private R&D tend to play complementary roles. A detailed analysis for the United States in the 1990s found that private sector R&D consisted more of short-term applied work focused on varietal development. In contrast, the USDA Agricultural Research Service focused more on long-term basic research, e.g. developing new breeding techniques, with the State Agricultural Experiment Stations falling somewhere in-between (Frey (1996[31]), Fuglie et al. (2017[27])).12

International agencies continue to play an important role in plant breeding for developing countries. The International Rice Research Institute (IRRI, founded in 1960) and the International Maize and Wheat Improvement Center (CIMMYT, founded in 1966) were key actors during the Green Revolution. They were later joined by the International Institute for Tropical Agriculture (IITA, 1967), the International Centre for Tropical Agriculture (CIAT, 1967), the African Rice Centre (WARDA, 1971), the International Centre for Agricultural Research in Dry Areas (ICARDA, 1977), and the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT, 1977). To coordinate the research efforts of these agencies, the Consultative Group for International Agricultural Research (CGIAR) was established in 1971. Several of these centres have significant public-private partnerships with leading international seed companies, although with limited effects on food security, poverty reduction or agricultural development (Spielman and von Grebmer, 2006[32]).

In addition to specific agricultural science research, plant breeding benefits from research within the broader scientific community. This link has become stronger over time with the emergence of genetic modification and New Plant Breeding Techniques (NPBT), including genome editing techniques, such as CRISPR/Cas (Scheben and Edwards, 2017[33]).13

Annex 2.A. Selected data tables

Annex Table 2.A.1. Evolution of the global commercial seed market, 2001-2014
Year	Conventional	GM	Total
2001	11.5	2.3	13.8
2002	11.0	2.8	13.8
2003	11.5	3.4	14.9
2004	12.4	4.3	16.7
2005	13.3	5.1	18.4
2006	14.9	5.8	20.7
2007	16.9	6.8	23.7
2008	20.5	8.5	29.0
2009	21.2	11.1	32.3
2010	22.6	13.4	36.0
2011	24.4	16.0	40.4
2012	26.8	18.3	45.1
2013	28.2	19.6	47.8
2014	30.7	21.1	51.8
Note: In billions of USD, not adjusted for inflation. Includes public and commercial seeds; excludes flower seeds and farm-saved seed.
Source: Syngenta (2016_[15]).

Annex Table 2.A.2. The global seed market by crop and by region, 2014
Billion USD	North America	Asia Pacific	EMEA	South America	Total
Maize	8.7	3.7	3.9	3.9	20.2
Soybean	4.4	n.a.	n.a.	2.9	7.4
Rice	n.a.	5.1	n.a.	n.a.	5.1
Vegetables	0.5	1.6	2.2	0.5	4.8
Cereals	n.a.	1.9	2.3	n.a.	4.2
Cotton	0.9	1.2	n.a.	n.a.	2.0
Rapeseed	0.9	n.a.	0.3	n.a.	1.2
Sugar beet	n.a.	n.a.	0.8	n.a.	0.8
Sunflower	n.a.	n.a.	0.6	n.a.	0.6
Others	1.7	1.2	1.4	1.0	5.3
Total	17.1	14.7	11.4	8.4	51.6
Note: EMEA stands for Europe, Middle East and Africa. North America refers to Canada, United States, and Mexico.
Source: Estimates based on Syngenta (2016_[15]).

Annex Table 2.A.3. Top 20 domestic seed markets in 2012
Rank	Country	Value (USD billions)	Share of global market (%)
1	United States	12.0	27%
2	China	10.0	22%
3	France	2.8	6%
4	Brazil	2.6	6%
5	Canada	2.1	5%
6	India	2.0	4%
7	Japan	1.4	3%
8	Germany	1.2	3%
9	Argentina	1.0	2%
10	Italy	0.8	2%
11	Turkey	0.8	2%
12	Spain	0.7	1%
13	Netherlands	0.6	1%
14	Russian Federation	0.5	1%
15	United Kingdom	0.5	1%
16	South Africa	0.4	1%
17	Australia	0.4	1%
18	Korea	0.4	1%
19	Mexico	0.4	1%
20	Czech Republic	0.3	1%
	Rest of world	4.3	10%
	Total	44.9	100%
Source: Based on International Seed Federation data cited in ISAAA (2016_[16]).

Annex Table 2.A.4. Global area of GM crops in 2017, by country
Country	Area under GM crops (million hectares)	Share of global GM area
United States	75.0	40%
Brazil	50.2	26%
Argentina	23.6	12%
Canada	13.1	7%
India	11.4	6%
Paraguay	3.0	2%
Pakistan	3.0	2%
China	2.8	1%
South Africa	2.7	1%
Uruguay	1.1	1%
Other	3.9	2%
Total	189.8	100%
Source: ISAAA (2017).

Annex Table 2.A.5. Global area of GM crops in 2017, by crop
Crop	Area under GM crops (million hectares)	GM as % of total area
Soybean	94.1	77%
Maize	59.7	32%
Cotton	24.1	80%
Rapeseed	10.2	30%
Alfalfa	1.2	n.a.
Sugar beet	0.5	n.a.
Papaya	<1	n.a.
Other	<1	n.a.
Total	189.8	n.a.
Source: ISAAA (2017).

Annex Table 2.A.6. Global exports of field and vegetable crop seeds, 2009-2015
	Volume (Million tons)		Value (USD billions)
	2009	2015	2009	2015
Field crops	2.10	3.74	4.92	6.75
Vegetable crops	0.11	0.12	2.75	3.63
Total	2.21	3.86	7.67	10.38
Source: International Seed Federation.

Annex Table 2.A.7. Main seed exporters, 2015
Field crops, 2015		Vegetable crops, 2015
Country	Exports (USD billions)	Country	Exports (USD billions)
France	1.20	Netherlands	1.22
United States	0.90	United States	0.62
Germany	0.58	France	0.41
Hungary	0.40	China	0.16
Canada	0.28	Chile	0.13
Netherlands	0.24	Israel	0.13
Argentina	0.24	Italy	0.11
Romania	0.23	Japan	0.10
Denmark	0.23	Thailand	0.09
Italy	0.21	Germany	0.07
Rest of world	2.24	Rest of world	0.58
Total	6.75	Total	3.63
Top 10 share	67%	Top 10 share	84%
Note: Statistical discrepancies exist between global export and import data.
Source: International Seed Federation.

Annex Table 2.A.8. Main seed importers, 2015
Field crops, 2015		Vegetable crops, 2015
Country	Imports (USD billions)	Country	Imports (USD billions)
Germany	0.56	Netherlands	0.42
United States	0.54	United States	0.38
France	0.54	Mexico	0.30
Russian Federation	0.36	Spain	0.21
Italy	0.34	Italy	0.18
Netherlands	0.31	China	0.17
Hungary	0.24	France	0.14
Spain	0.23	Japan	0.13
United Kingdom	0.23	Turkey	0.11
Ukraine	0.23	Canada	0.10
Rest of world	2.95	Rest of world	1.39
Total	6.53	Total	3.52
Top 10 share	55%	Top 10 share	61%
Note: Statistical discrepancies exist between global export and import data.
Source: International Seed Federation.

Annex Table 2.A.9. Costs and returns for US maize, 1975-2016
Year	Gross value of production (USD per acre)	Seed costs (USD per acre)	Total operating costs (USD per acre)	Output price (USD per bushel)
1975	694.3	29.7	259.4	8.1
1976	566.1	28.7	234.8	6.5
1977	513.2	31.5	228.0	5.8
1978	601.6	30.9	232.6	6.0
1979	678.6	30.5	245.5	6.2
1980	629.7	32.1	269.0	7.0
1981	536.0	33.3	278.4	4.9
1982	472.4	31.9	259.1	4.2
1983	472.1	31.1	239.6	6.0
1984	492.1	32.5	239.4	4.7
1985	440.6	32.3	239.0	3.8
1986	286.0	33.0	206.2	2.4
1987	311.6	31.5	196.0	2.6
1988	350.5	30.5	197.8	4.2
1989	396.6	32.6	207.2	3.4
1990	385.4	30.7	201.0	3.3
1991	369.6	31.3	199.8	3.3
1992	388.8	31.3	197.7	2.9
1993	314.3	31.1	192.3	3.2
1994	401.6	30.7	199.3	2.8
1995	427.5	31.8	209.9	3.7
1996	482.3	34.7	209.9	3.7
1997	424.8	36.8	208.0	3.2
1998	333.4	38.1	200.0	2.4
1999	288.2	37.8	196.0	2.1
2000	301.2	36.7	201.5	2.2
2001	318.7	38.6	193.8	2.2
2002	367.9	37.4	171.1	2.7
2003	368.5	40.2	185.8	2.5
2004	406.6	41.3	197.4	2.4
2005	283.1	44.0	202.6	1.9
2006	371.1	45.9	217.3	2.7
2007	481.8	50.4	235.3	3.4
2008	634.1	60.5	297.9	4.4
2009	561.2	78.9	295.0	3.6
2010	681.1	80.6	283.0	4.3
2011	811.0	81.7	321.7	5.5
2012	762.8	87.5	332.3	6.5
2013	673.9	91.3	332.6	4.3
2014	554.3	92.8	328.0	3.3
2015	556.9	92.4	303.4	3.3
2016	517.9	88.5	277.1	3.0
Note: All prices converted to 2009 USD using the US GDP Deflator; excluding government subsidies.
Source: OECD analysis using USDA Commodity Costs and Returns reports for maize.

Annex Table 2.A.10. Global private R&D spending by agricultural input sector, 1990-2014
Year	Seed & biotech	Animal R&D	Fertilizer and crop protection	Farm machinery
1990	1,431	1,188	2,708	1,065
1991	1,466	1,217	2,649	1,080
1992	1,479	1,265	2,650	1,058
1993	1,550	1,261	2,713	1,041
1994	1,664	1,256	2,767	1,053
1995	1,716	1,344	2,840	1,090
1996	1,789	1,401	2,970	1,224
1997	1,960	1,407	3,021	1,227
1998	2,189	1,404	2,948	1,264
1999	2,146	1,379	2,600	1,312
2000	2,317	1,348	2,314	1,325
2001	2,160	1,275	2,139	1,321
2002	1,969	1,288	2,083	1,282
2003	2,026	1,384	2,417	1,276
2004	2,095	1,444	2,534	1,337
2005	2,133	1,484	2,547	1,416
2006	2,286	1,543	2,435	1,525
2007	2,493	1,659	2,514	1,742
2008	2,897	1,778	2,695	1,990
2009	3,096	1,775	2,701	2,247
2010	3,426	1,880	2,848	2,363
2011	3,796	1,999	3,020	2,705
2012	3,911	2,087	3,053	3,017
2013	4,074	2,172	3,203	3,152
2014	4,290	2,229	3,291	3,091
Annual real growth	4.7%	2.7%	0.8%	4.5%
Note: In millions of constant 2005 PPP dollars.
Source: Based on Fuglie (2016[23]), Tables 3 and 4.

Notes

← 1. The focus in this study is on seed markets relevant to food and agriculture. Most of the data and discussion relates to field crops and (to a lesser extent) vegetables, excluding, for example, ornamental crops, grasses, or forestry.

← 2. Data on farm-saved seed needs to be interpreted with caution, as this number is likely to be underestimated (Heisey and Fuglie, 2011[7]). On a global level, farm-saved seed is most common in those countries where data is the least available. On the importance of farm-saved seed in developing countries, see van Etten et al. (2017[120]) and Spielman and Kennedy (2016[154]).

← 3. Bonny (2014[25]) calculates that farm-saved seed accounted for 45% of the total wheat crop planted in France between 1981 and 2012, without a clear downward trend.

← 4. As pointed out by Bonny ( (2014[25]), (2017[12])), different sources and methods lead to different estimates of market sizes, and figures provided by some market research agencies tend to give lower estimates as they may underestimate sales by small and medium-size enterprises. In particular, often-cited market estimates by Phillips McDougall (of USD 35 billion in 2015) only account for two-thirds of global crops. If the four-firm concentration ratio is calculated using such low estimates, the degree of market concentration is automatically overstated. This is particularly the case for the widely-used estimates of ETC Group (2013[8]). The Syngenta estimates used here are in line with estimates of the International Seed Federation and the bulk of the market research estimates surveyed in Bonny (2017[12]).

← 5. The European Union forms a single market for seeds as it has a harmonised regulatory environment across all Member States. However, agro-ecological differences across countries imply that the relevant market for farmers may have a narrower geographic scope, depending on the crop.

← 6. See Bonny (2014[25]) for an introduction to the genetically modified seed sector.

← 7. A note on terminology: a GM trait refers to a (phenotypic) characteristic such as herbicide tolerance (see further). A GM event refers to the underlying genetics and is defined by the DNA sequence that has been inserted into the host genome and the site(s) where this DNA has been inserted (Mumm, 2013[248]). The terms GM and biotechnology/biotech are used interchangeably in this report.

← 8. Personal communication with Szabolcs Ruthner and Marien Valstar.

← 9. One acre is around 0.4 hectares, so multiplication by 2.5 gives the approximate equivalent in USD per hectare.

← 10. The higher cost of GM seed may partly reflect market power and the need to recover high R&D investments, but production and post-harvest processing costs may also be higher. To prevent low-level pollen contamination, GM seed production may use greater isolation distances or other isolation mechanisms, which raise the cost of producing seed. Another factor influencing seed prices is the degree of post-harvest processing, conditioning, and seed treatment.

← 11. These R&D intensities are calculated using revenues for KWS and Vilmorin including their share of the AgReliant joint venture, following Vilmorin’s own presentation of its R&D numbers. The Vilmorin R&D number of EUR 240 million represents the total investment in R&D; for accounting reasons, only EUR 192 million are included in Vilmorin’s profit-and-loss statement. For KWS, the EUR 190 million represent the R&D outlays included in the profit-and-loss statement. Both companies are discussed in more detail in the next chapter.

← 12. The complementary roles of public and private agricultural R&D spending are explored in greater detail in Chapter 7.

← 13. The precise impact of broader scientific research on plant breeding is difficult to assess, as scientific research in one field may find unexpected applications in another field, often with long lags. For instance, CRISPR was first identified in research on the genome of E. coli in 1987 (Bortesi and Fischer, 2015[169]). Recent research on the impact of public funding of science through the National Institutes of Health in the United States shows that public funding positively affects private-sector patenting activity, although lags are long (up to twenty years after the initial NIH grant approval) and roughly half of the ultimate impact takes place in different research areas (Azoulay et al., 2015[254]).

End of the section – Back to iLibrary publication page