WORKING
DOCUMENT Rev. 2
Directorate-General
for Agriculture
Economic Impacts of Genetically Modified Crops on the Agri-Food Sector
5. Markets: Segregation, Identity Preservation and Labelling
The introduction of GM crops has until now mainly addressed the supply side of agricultural crops and food markets. The development of efficiency enhancing GM crops dominates the agricultural applications in most countries where GM crops are grown. The EU debate on GMOs, on the other hand, has been dominated by demand factors, such as food safety concerns. In the EU, consumer demand for a continuous supply of agricultural raw materials and processed products at a certain price and a certain quality is seen as the underlying force for the agricultural sector to adapt and to innovate production techniques. Furthermore, the recent reforms of EU Common Agricultural Policy provide several incentives to adapt production quantities to market demand and to put emphasis on quality aspects, both of products and production methods.
Further technological developments and continued increase in GM crop production could affect the future competitiveness of conventional non-GM production. Nevertheless, consumer reaction to GM food has given rise to uncertainty about market developments, in particular the short term prospects for GM products. As a result to consumer concerns, the regulatory framework concerning GMOs has developed and is partially still under review not only in the EU but also in many other countries including the USA.
Labelling has been recommended as a tool to enable consumer choice between products and to avoid further market and trade disruptions. However, labelling systems in which consumers have confidence would require at least segregation of product lines throughout the processing system. Moreover, Identity Preservation would be required to distinguish the different types of products according to their contents of GM material or the way they have been produced whether using GM technologies or not. Segregating and Identity Preservation are attempts to create and establish a separate market for a "new" product, a specific crop. The success of such attempts will depend on supply and demand concerning the new product.
GM crops with enhanced quality traits are most likely to supply niche markets. They are addressing a specific demand and the opportunities for supply are highly dependent on innovations, e.g. new varieties, which provide enhanced quality. On such markets competition and market transparency are generally less advanced than for commodity markets.
Since segregation and Identity Preservation appear to be means to offer choice between GM and non-GM products to the consumer, this chapter will start with a discussion of the key features of these systems compared to the commodity trade system (section 5.1). Three systems for Identity Preservation and labelling in the GMO context, based on current EU legislation, will be identified in section 5.2. The available studies and information about additional costs of IP have been summarised in section 5.3. Furthermore, the distribution of these additional costs along the food production chain is discussed in section 5.4. The following section 5.5 provides some background information about EU markets for soybeans and corn, about the supply to serve potential non-GMO demand and about the different stance on food and feed use. Finally, some trade issues are briefly outlined in section 5.6.
5.1. Key features of agricultural trade systems
rade of agricultural products today is based on the commodity system. Any system of identification which goes beyond the common specifications would require additional handling effort and would thus create additional costs. Segregation and Identity Preservation are possible responses to consumer demand for specific products.
- 5.1.1.
Commodity System
The bulk commodity system works on the basis that crops from different farms are sufficiently alike to be traded at a common price and to a common grading specification21. Usually, commodities from different origins are blended to meet specific grades. For example in the case of wheat protein content, moisture, falling number, specific weight and percentage of extraneous material are taken into account. On its journey to a milling plant, the wheat can be sampled and blended several times and there is no traceability back to the producer. Commodity prices are fixed on spot markets, futures markets or by contracts.
International trade of agricultural products, in general, is based on the commodity system, which covers about 200 Mio t of grains per year. In the oilseeds sector about 50 Mio t of soya, sunflower and rape seed are traded annually across borders, in addition to 13.5 Mio t of oil from the different seeds and 43 Mio t of meal. Furthermore, many more millions of tonnes are traded on domestic markets under the commodity system.
Bulking up the produce of many producers means that transport and handling costs can be reduced. For example, Cargill has calculated that ocean transport from the US to Europe may only add 13 € to the price of a tonne of soybeans (180 - 225 €) if 50 000 tonnes are shipped at a time. The total cost of transportation from a US mid-west farm to European harbours is estimated at 10% of the farm-gate price of soybeans (Cargill 1999).
Furthermore, bulk transport enables a continuous flow for processing, since taking a processing plant down and firing it up again can be time consuming and costly.
Segregation refers to a system of crop or raw material management which allows one batch or crop to be separated from another (House of Commons 2000).
Segregation is an attempt to create and establish separate markets for differentiated products or to set up a "new" market for a "new" specific product. This corresponds to a dis-aggregation of the supply and demand. Some possible economic effects of market segregation are shown in figures 5.1 and 5.2.
- In figure 5.1 it is assumed
for simplicity reasons that the aggregated supply for a certain crop would
be subdivided equally among the specific markets A and B (dotted lines
SA1 and SB1). Assuming further that demand would
follow the same pattern for both sub-markets (DA = DB),
the price on market A should be the same than on market B (pA1
= pB1).
However, due to lower quantities produced and traded, potential economies of scale may not be used and production cost per unit might be higher than on the aggregated market. In figure 5.1 this effect is captured by shifting the supply curves from SA1 to SA2 and from SB1 to SB2. The effect will be a reduction of quantity produced (qA2 and qB2) and an increase in prices on both markets (from pA1 to pA2 and from pB1 to pB2). In general, losses in economic welfare can be expected because the potential for trade and specialisation gains will remain partially unused.
Moreover, the assumption of equal pattern of demand on both sub-markets will be unrealistic. More realistic would be the situation as shown in figure 5.2 with different demand functions on the respective sub-markets (DA3 and DB3).
In our example, the price increase caused by segregation would be outweighed by a price reduction due to a low demand on specific market A. This effect would be accompanied by a reduction in quantity supplied compared to figure 5.1. On specific market B, a high demand (DB3) would lead to a further increase in price to pB3 and an increase in quantity supplied to qB3.
In addition, the application of new cost-saving or output enhancing technologies on one of the specific markets would result in a rightward shift of the supply curve. New technologies thus result in price reduction and in higher equilibrium quantity on that specific market. Biotechnology is expected to provide such technological effects, at least in the long-run (see chapter 3).
Segregation implies that specific crops and products are kept apart, but does not necessarily require traceability along the production chain. In the GMO context, this may pose major problems of liability and consumer confidence. A French investigation identified the absence of labelling requirements at all the stages of a production chain to be the most important difficulty to apply segregation along the production chain and to operate the current labelling requirements for GMOs (Ministère de l'Economie, des Finances et de l'Industrie, 1999).
5.1.3. Identity Preservation Systems
Marketing experts have stated that "Identity Preservation programs are the best alternative and the most economical way to meet customer and regulatory requirements" (Young 1999). Identity Preservation (IP) is a system of crop management and trade which allows the source and/or nature of materials to be identified (Buckwell et al. 1998). Thus it goes beyond segregation, since it implies a stronger positive desire to know about the origin of a crop or a product.
The objective of IP is to
ensure that a particular crop is monitored throughout the food chain and
thus to guarantee certain traits or qualities which might command a premium
(House of Commons 2000). IP requires a set of actions to allow traceability
and is usually communicated to the consumer by a label. Thus, IP causes
additional
cost in supply which are illustrated in figure 5.3.
- Figure 5.3 Economic
effects of Identity Preservation cost
Introducing Identity Preservation on specific market A would result in a further shift of the supply curve from SA2 to SA4. The effect will be a reduction of quantity produced from qA3 to qA4 and a reduction in price to pA4 on market A.
Currently IP is used to identify crop varieties which provide additional features concerning the content or composition of products (eg, protein content, starch level, oil content). In addition, IP is also applied for features which are not related to the contents but to the method of production (organic food or animal welfare standards) or the geographical origin of the product.
A common example of an IP grown crop under contract is the production of certified seed. Contamination by foreign pollen or other seed varieties has to be avoided and inspections take place to verify purity. The premium for seed wheat production is about 15-20% of the price of a normal wheat crop. This premium should cover the extra work involved for identity preservation.
Other examples for IP systems already in place are related to high erucic acid rapeseed, grown for technical use, waxy corn for starch production and flint corn for breakfast cereals. Identity preservation systems have also been established for certain other specialised (niche) markets, organic produce for example or special varieties of soybeans for tofu production. These products are transported in smaller quantities, reserved trucks or reserved holds of smaller ships.
Compared to the main commodity markets, the quantities currently traded under IP systems are small. Organic food is for instance representing a market share of less than 5%, often less than 1%, in most EU Member States (Michelsen et al., 1999). The highest market shares are obtained for dairy products in Denmark (14.2%) and in Austria (8-10%).
In the US, about 100 000 tonnes of soybeans are identity preserved, compared to 75 Mio t harvested under the commodity system (Rawling 1999).
However, variety choice through IP is seen as contributing more than any other factor to improve the market value of grains (Clarkson 1999). A comparison of recent US prices shows that the premium paid for certain quality traits and for organic products is much higher than the current premium for conventional non-GM crops (s. annex). In the health food sector the price for IP grains and soybeans is about 200 - 300% of the commodity price (Cargill, 1999).
The following analysis will concentrate on IP systems since their degree of compliance with consumer concerns appears to be higher than for segregation.
5.1.4. Some specific issues of Identity Preservation Systems
Testing and control: An important element to establish IP systems is the technical possibility to test samples for the preserved identity (e.g. its physical or chemical contents). Random or regular tests can be carried out for the final product delivered to the consumer or the processor. To enhance the performance, control mechanisms might be applied not only to the final product but also at different stages of production and transportation.
For IP relating to production methods or regional origin, testing of the final product is generally not possible and the consumer has to rely on the integrity of the supplier and the robustness of the IP system (Buckwell et al., 1998). Controls would then have to verify this integrity at different stages of production in order to establish consumer confidence.
Tolerances: Ensuring absolute purity of a food product would be related to prohibitively high costs in practical processing and handling chains. The principle of fixing a tolerance level (threshold) in purity standards is therefore a long-established feature for IP systems throughout the food industry.
Tolerances have for instance been fixed for organic food. Because of the difficulty of eliminating all commingling throughout the production chain, a 5 % tolerance level of non-organic material is allowed in some processed food derived from and labelled as being made from organic ingredients.
The costs of an IP system can be expected to increase with a reduction of the tolerance level. Thus, the setting of a certain tolerance level will be an important factor to determine the costs of an IP system.
Contracts: Identity preservation often involves advance contracts with farmers who commit themselves to keep the crop separate during harvesting or to produce only under certain rules (quality labels, organic farming). Furthermore, seed varieties, growing specifications, chemical treatments or handling and storage requirements may be subject to specific contracts. With an increasing degree of specification for an agricultural product, which is reflected by a price difference, the likelihood of establishing a contract can be expected to rise.
- This section first summarises
the reasons to consider IP systems in the GMO context, then reviews the
current EU legislation on labelling and finally - with this background
- identifies three approaches for IP related to the introduction of GMOs.
5.2.1. Reasons to consider IP systems in the context of GMOs
The fear of consumers that GMOs could have negative impact on their personal health can be a reason to require traceability. This would allow the identification and if necessary eradication of a harmful modification or product and could be a way to increase confidence in the new technologies.
Most crops are living organisms which are able to reproduce a plant. Biosafety considerations require traceability of GM crops to avoid uncontrolled gene transfer and possible danger for biodiversity. (see box on Biosafety Protocol in section 5.6).
There is a need for processors and traders to meet emerging mandatory GMO-labelling requirements in certain countries, in particular the EU, but also in Switzerland, Australia, New Zealand, Japan etc. The tolerance levels for labelling may differ among countries or still have to be decided. EU legislation on labelling is summarised in the following section.
The set of GMOs approved in different countries is not the same. For instance, some corn varieties grown in the US include transformation events not yet approved in the EU. Thus, IP could help to avoid trade disruptions due to differences in the approval status.
Consumer demand for non-GM or GM free food provides an economic incentive for farmers, processors and distributors to supply such products which require IP to be accepted by the consumer.
Furthermore, with the development of the second and third generation of GMOs, i.e. specific traits addressing the consumer and the procession industries, IP will become necessary to ensure providing the specific traits to the consumer and to enable a premium for the enhanced value.
5.2.2. EU legislation concerning GMOs, in particular labelling
The release of GMOs in the environment and their placing on the EU market are governed by Directive 90/220. GMOs can only be introduced onto the market after having been assessed and authorised to this end. In February 1998 the Commission tabled a proposal to amend this Directive, which is subject to the co-decision procedure between the European Parliament and the Council. The main objective of this revision is to increase the efficiency and the transparency of the decision-making process whilst ensuring a high level of protection for human health and the environment. With this view, Member States will be required to ensure labelling and traceability of GMOs at all stages of the placing on the market.
Sector-based legislation covers products derived from GMOs, in particular GM food. This legislation has to be further specified and extended, in line with the revision of Directive 90/220. The Commission will table a proposal on Novel Feed, including GM feedstuffs, in the second half of 2000. The White Paper on Food Safety22 identified a number of actions to re-establish public confidence, in particular completing and harmonising labelling requirements.
The Novel Food Regulation23 and the seed legislation24 already provide for mandatory labelling of food and seeds containing or consisting of GMOs. Two GM varieties, one of corn and one of soya, and their derived products were already on the EU market before the Novel Food Regulation came into force. Therefore these two varieties were not subject to additional labelling requirements. In order to ensure the labelling of these varieties it was necessary to ensure the appropriate labelling through a specific regulation25.
With the adoption of this labelling regulation, the Council invited the Commission to study the practicability of setting down a de minimis threshold which takes account of the problem of adventitious contamination. In response to this request the Commission has adopted a regulation26 which fixes a tolerance level of 1% for each single ingredient on the condition that the operator has taken appropriate steps to avoid the use of GMOs as a source.
A (non-exhaustive) list of food ingredients or foods comprising a single ingredient in which neither protein nor DNA resulting from GM is present, shall be drawn up. The negative list is a concept applicable to processed GM products in which no genetically modified material can be detected any more. These products would be exempted from compulsory labelling.
In January 2000 the Commission also introduced labelling requirements for additives and flavourings that have been genetically modified or have been produced from GMOs27.
The Commission White Paper on Food Safety proposes the harmonisation of labelling rules for food, additives, flavourings, clarification of the authorisation procedures in the Novel Food Regulation and the establishment of a legislation concerning food and food ingredients produced without genetic engineering.
Figure 5.4 Labelling and Identity Preservation
1. Voluntary IP of specific GM traits: IP systems are common practice for crops that have a specific value to their consumer, for example through improvements in nutritional value, colour, texture, flavour or processing properties. With the development of new traits by biotechnology, the economic incentive for IP would increase. In addition to the labelling requirements under the novel food regulation, there would be a clear incentive on the supply side (farmers, processors and retailers) to introduce IP and thus to preserve the additional value or quality of such a GM crop through the processing chain. IP would distinguish a product for which consumers are expected to pay more than for a conventional product.
Tolerance: GMOs offering specific qualities to the producer and the final consumer will only be accepted if these qualities can be guaranteed within a certain tolerance level. Tolerances will have to be fixed in accordance with the purity expectations of the buyers of these products.
2. Voluntary IP of GMO-free products: The second approach for IP is to preserve and label GM-free products in order to enhance consumer choice. Current EU legislation already requires compulsory labelling for food containing GMOs. Thus, the introduction of labelled GMO-free food would in theory enable the choice between three categories of foodstuffs: novel GM food; conventional non-GM food and GMO-free products (figure 5.4).
Some European food trade companies which are trying to serve the non-GMO market niche are claiming that they are supplying non-GMO products. However, the explanations and guarantees given to the consumer sometimes lack a sufficient transparency. A Wall Street Journal article (October 26, 1999) stated both confusion and legal risk in the current labelling which can be found in supermarkets. An investigation of 94 companies by the French Ministère de l'Economie, des Finances et de l'Industrie (1999) revealed that more than 50% of the enterprises had modified the composition of their products to avoid GMO labelling. Most of them had attestations by their suppliers, 14 enterprises were able to present traceability documents and 19 got analytical certificates.
However, it can be expected that the share of conventional food will diminish over time, since the pay off for GMO-free products can be expected to be higher than for conventional non-GM products. If producers decide to make an effort to segregate, the additional costs to comply with GMO-free standards might be low compared to the additional premia achieved on the market. On the other hand, if at least part of the consumers accept labelled GM food, some conventional raw material would enter into GM-labelled final products.
Testing: Reliable testing will be necessary to prove that a product does not contain or contains only a limited percentage of GM material. A workshop on GMO research perspectives held in the context of the EU Fifth Framework Programme identified the "development of rapid, reliable detection methods for GM foods and their derivatives" as one of the top priorities for further research in the GMO context (External Advisory Groups, 1999).
For international trade a standardised test would help to avoid liability conflicts and trade distortion in case of different labelling requirements and approval status (Brookins 2000). The American National Grain and Feed Association, for instance, has called for the introduction of an accurate, repeatable and low cost test to distinguish between conventional and GMO products. USDA has recently announced that it will establish a reference laboratory to evaluate the validity of analytical procedures and to establish sampling procedures for use in testing bioengineered grains and oilseeds (USDA, 2000c).
The more expensive testing for GMO, the more likely will be arrangements based on declaration of honour concerning the GMO-free status of a product. This would imply a certain system of field and production control to satisfy consumer confidence.
Example for
IP system based on producer declarations and testing: Champagne céréales,
non-GM corn (Oustrain 1999)
|
The debate about tolerance levels has raised other questions concerning the non-GM status of a plot or farm on which GM crops have been grown. The argument is that inherited modified genetic sequences are likely to persist on the farm - for example in the case of rape - even after the crop has been harvested and sold. Standards for non-GM or GM-free lines will have to address this question.
Contracts: To sell IP crops farmers will have to agree on terms of contract with their trading partners. Such contractual arrangements always imply the question of liability. Some proposed voluntary certification procedures have been developed for producers wishing to segregate non-GMO commodities in response to a premium offered.
Currently most US extension services have warned farmers to be careful when signing a contract to supply non-GM or GMO-free products, since accidental contamination cannot be excluded (Charpentier, Hazouard, 1999).
3. Compulsory IP for GM products (GM traceability): Trading GM crops as part of a commodity system would result in losing their track within the transportation and processing chain. Thus any commodity sample originating from a region or country where GM and conventional crops are grown in parallel might contain GM crops. Traceability, i.e. a compulsory IP system, has been introduced as a strategy to re-establish consumer confidence in the EU beef sector following the BSE crisis. Traceability could also be a strategy to monitor the environmental and health effects of GMOs and to enable choice to those consumers who want to avoid GMO consumption.
According to the EU Council Common Position with a view to amending Directive 90/220/EEC on the deliberate release of GMOs28 traceability will be required: "It is necessary to ensure traceability at all stages of the placing on the market of GMOs ." (Common Position, recital No. 40). Member States are invited to ensure traceability at all stages.
Testing: In general the testing requirements for GMO traceability would be the same as for GM-free products. However, the objective will be to detect the presence of specific modifications and not to measure the quantity versus a threshold. Reference material and genetic sequence information will be needed to develop reliable tests.
Tolerance: The tolerance approach can be expected to be very strict if the objective is to ensure traceability. Every bunch containing only a minimum trace of a certain GMO would have to be identified.
- Additional costs of IP arise
with the additional work involved in growing, handling, storage, transport,
processing, cleaning, and administration (Buckwell et al. 1998).
They would apply to all three IP approaches identified above, independently
of their voluntary or compulsory character. However, the magnitude of IP
costs will depend on several factors which will be summarised at the end
of this chapter.
Many opportunities for mixing and contamination exist along the production, processing and distribution chain of an IP product. Thus, IP costs arise on different stages of the chain: seed production, farm, transport, further storage, processing, labelling and distribution. The following overview of additional costs corresponds to the structure suggested by Buckwell et al. (1998). Some empirical experience has been added to illustrate the magnitude.
- 5.3.1.
IP costs for seed production
Already under conventional systems basic and certified seed is normally distributed separately bagged and labelled. No difference would occur for an IP system.
The two main sources of mixing seed varieties are through pollen and through other seeds. Avoiding such contamination is a usual feature of seed breeding. The EU has fixed minimum distances from neighbouring crops of different varieties or inbred lines of the same species (table 5.1). For instance, certification of basic seed requires a minimum distance of 400 m for cross-pollinating oilseeds, 300 m for rye and of 200 m for corn.
The minimum varietal purity for basic seed of oats, barley, wheat, spelt and rice has been fixed at 99.9% and at 97% for soybeans. Several other purity criteria are provided by the seed directives concerning the minimum germination rates, analytical purity and the maximum content of seeds of other plant species.
Purity criteria applied for seed multiplication (certified seed) are partially less restrictive than for basic seed (table 5.1). The cordon sanitaire for the production of certified oilseeds for instance amounts to 250 m for certified rye seed and to 200 m for corn and for cross-pollinating oilseeds. The minimum varietal purity for oats, barley, wheat, spelt and rice has been fixed at 99.7% for first generation certified seed and at 99% for second generation.
At least for certified seed of soybeans and beets, the current EU standards for varietal purity might conflict with the tolerance levels for GMO labelling. For soybeans, the EU seed marketing standards require a minimum purity of 95%. This means that seed could contain up to 5% of other varieties, possibly including GM varieties. During the last years, the EU has imported soybean planting seed from the US, where the purity norm for soybean seed runs about 98% (Blumenthal, 1999). For beet seeds, the varietal purity has been fixed at 97%.
The crucial variable to determine the additional costs in seed production is the tolerance level applied for IP. Currently farmers obtain a premium of 15 to 20% for the extra work required for the production of wheat crop for seed compared with growing normal wheat crop for commercial sale (Cargill 1999). Representatives of the seed industry have confirmed that they could provide seeds at any desired tolerance level. However, costs would rise following rather an exponential than a linear function with a tolerance level approaching zero percent.
- contamination of seed used by the farmer (within the limits of genetical purity);
- crop mixing with volunteer GM plants that are already present in the soil when the crop is sown;
- mechanical commingling in sowing, harvesting and storage;
- cross pollination with other varieties which varies with the distance, sexual compatibility between crops and the method of pollen transport (wind, insects).
The number of volunteers can be reduced by cultivation practices or by herbicides. In fields where rapeseed has been grown, volunteers are likely to grow during a period up to seven years. Volunteers of herbicide tolerant variants should be treated by alternative non-selective herbicides (SCIMAC, 1999).
To avoid mechanical commingling, the planting and harvesting equipment must be thoroughly cleaned before use. Furthermore, the on-farm storage facilities must be cleaned or new facilities must be provided to separate IP crops. The costs of cleaning, in particular the amount of time spent on this mainly depends on the required tolerance level. Due to cleaning breaks, there may be additional cost associated with a reduction in work time during which the harvest machine is operational. Moreover, a particular low tolerance could require the use of separate machinery for each crop.
Physical distance between the pollen donors and the crop is the most important factor to avoid cross pollination among specific varieties. The amount of cross pollination also depends on the amount of outbreeding in the crop, the overlap of flowering periods and the area of the crops grown (Moyes and Dale, 1999).
In the UK context, SCIMAC has set up guidelines for good agricultural practice for growing herbicide tolerant crops which provide minimum distances from certified seed crops, organic crops and conventional crops of the same species. Basic guidelines for growing GM crops with specific agronomic traits are currently under development (SCIMAC, 1999). On the other hand, the standards for organic farming provided by the UK Soil Association require minimum distances from GM crop plantings which are significantly higher than the SCIMAC provisions (Soil Association, 1999).
Cross pollination furthermore may affect the relationship between neighbour farms. GM cropping on one plot may affect the non-GM status of another plot, and more controversially, the GM status of other farms. The possibility of litigation with neighbours could also influence the economic considerations of a farmer (Griffiths, 1998). SCIMAC's guidelines propose that "the onus lies with the GM grower to notify neighbouring farms in writing of his planting intentions." This issue is of particular importance if the neighbour is growing organic food, where GMOs are prohibited in general. Failure to reach agreement must be notified to SCIMAC and has to be solved by further consultation or through normal legal channels (SCIMAC, 1999).
Cross-pollination and commingling raise a number of legal and economic issues concerning the coexistence of three production systems: GM, conventional, and organic.
IP products would be very likely to be grown at a contractual basis. Contracting requires certain transaction costs for all contracting parties involved, such as the time devoted to negotiations and probably some fees. Keeping accurate records is essential to ensure IP and traceability. Record keeping might result in additional work for the farmer. Cargill (1999) has pointed out that farmers growing IP crops also face additional price risks and their options for selling the crop might be reduced.
Table 5.2 indicates some examples for additional cost at the farm level and the available information about premia currently paid to farmers. However, the premia may not only reflect the additional costs of segregation but also the additional value of a certain trait or a certain production system.
The examples for soybeans indicate that US producers have received a premium of 5 - 9 €/t for non-GM soybeans in the last years. This amount corresponds to the IP costs for GM soya with specific traits and represents about 4% of the farmgate price. More recent sources signal a lower premium level of 3 - 7.5 €/t which corresponds to 1.5 - 4.4% of the average price received by US farmers. European farmers are offered a slightly higher premium of 11 - 12 €/t for non-GM soya and in 1998 some buyers also seemed to be willing to pay a premium up to 24 €/t above the conventional US price to get Brazilian non-GM soya.
However, according to US grain handlers, the premium paid for food quality soya was much higher than the non-GM premium. The average price of food use was estimated to be 35 €/t higher than the commodity price for soybeans (Bender et al., 1999). In autumn 1999, the IP premium for quality traits ranged from 20 €/t for medium high protein contents to more than 140 €/t for sugar balanced soybeans compared to an average commodity price of 170 €/t (Clarkson, 1999).
|
|
|
|
premium |
|
|
| GM quality traits: low linolenic, high oleic, low saturate, high protein, high sucrose | USA | (1997) | 8 - 9 €/t | 4% | (1) |
| non-GM herbicide resistant (DuPont STS programme) | USA | 1998 | 5
- 8 €/t
(premium) |
(2.4 - 3.8%) | (2)
(3) |
| non-GM
herbicide resistant |
Brazil | 1998 | 24 €/t **) | 10% | (1) |
| non-GM | France | Spring 1999 | 11 - 12 €/t (premium) | (4) | |
| non-GM herbicide resistant (ADM) | USA | 1999 | 6
- 7 €/t
(premium) |
(3.5
-
4%) |
(3) |
| non-GM
commodity grade US#1 |
USA | Autumn1999 | 7.5
€/t
(premium) |
4.4% | (5) |
| non-GM | USA | Sept
99
Feb 00 |
3.6
€/t
2 - 3 €/t (premium) |
(2%)
(1 -1.5%) |
(6) |
| non-GM | USA | (1999/
2000) |
3.8
- 5.7 €/t
(premium) |
(2
-
3.2%) |
(7) |
| *)
farmgate price (percentages in brackets have been calculated by DG Agriculture)
**) due to higher average price for Brazilian soybeans Sources: (1) Buckwell et al. 1998; (2) Bender et al. 1999; (3) Deutsche Bank Alex. Brown 1999; (4) Circuits culture 1999; (5) Clarkson 1999; (6) Brookins 2000; (7) Lin 2000 |
|||||
The premium for organic soybeans was estimated at 245 €/t (commodity quality). This means producers of organic soybeans received a premium of almost 150% of the commodity price (Clarkson, 1999). Thus, farmers who are thinking about entering into non-GM production might consider as well to switch to organic farming in order to realise the higher market price.In contrast, GM soybeans (without specific quality traits) are being discounted by up to 10% of the farmgate price in many parts of the USA because foreign buyers and some US companies have announced not to buy GM material. Therefore many grain elevators are discounting not only GM varieties but all varieties because they cannot separate due to a lack of facilities to handle both types.
While quality trait premia (high oil contents) for corn range between 4 and 6 €/t, non-GM premia appear to be slightly lower. They range between 1.8 and 5.6 €/t. IP premia range between 2.5 and 9% of the farmgate price for corn. However, when these price differences per tonne are translated into price differences per hectare the farmer will have to take account of yield differences. Yields of quality trait varieties are often lower than average, while several studies have found evidence on yield gains for Bt corn compared to conventional varieties (see chapter 3).
|
|
|
|
premium |
|
|
| Quality
trait (conventional)
high oil contents |
|
|
5.3
€/t
(premium) |
5% | (1) |
| Quality
trait (conventional)
high oil contents |
|
|
4.2
€/t
(premium) |
(5%) | (2) |
| Quality
trait (conventional)
high oil contents (Optimum Quality Grain) |
|
|
6.1
€/t
(premium) |
(7.5%) | (7) |
| Non-GM |
|
1998 |
1.8
- 2.8 €/t
(premium) |
(2.5
-
4%) |
(3) |
| Non-GM
commodity grade US#2 yellow |
|
1999 |
5.6
€/t
(premium) |
(9%) | (5) |
| Non-GM |
|
2000) |
2 - 4 €/t (premium) | (3
-
4.5%) |
(7) |
| *)
farmgate price
Sources: (1) Buckwell et al. 1998; (2) Bender et al. 1999; (3) Deutsche Banc Alex. Brown 1999; (5) Clarkson 1999; (7) Lin 2000 |
|||||
Some examples for other crops, i.e. sunflower and oilseed rape, unveil that a premium of 3.5 to 5% of the farmgate price is paid to the farmer for cropping (conventional) quality trait varieties (Buckwell et al. 1998). However, a Canadian example for GM herbicide-tolerant oilseed rape (Canola) shows that farmer's costs for separate storage and handling can be as low as 0.5% of the farmgate price.
- 5.3.3.
Costs for testing
The easiest and probably cheapest way to segregate different grain varieties would be to use grain confetti for identification. Nevertheless, qualitative and quantitative testing may be required to control for particular specifications and GM contents. For GM crops providing quality traits testing will refer to these specific modifications. GMO traceability would extend the need for testing to all genetic modifications, including agronomic traits. For GMO-free products, the testing would not be limited to determine the presence or absence of GMOs, but would also have to confirm that the tolerance levels have been respected.
| GMO testing methods
A Genetically Modified Organism can be distinguished from a non-GMO by the fact that it contains either unique novel deoxyribonucleic acid (DNA) sequences and/or unique novel proteins not present in its conventional counterpart. Two methods are actually applied: a PCR (Polymerase Chain Reaction) test based on DNA detection and the ELISA (Enzyme Linked ImmunoSorbent Assay) test based on protein detection. Validation programmes for both methods are currently exercised by the EU Joint Research Centre (Lipp et al. 2000). PCR The polymerase chain reaction is based on the detection of DNA fragments that are inserted in the plant genome. This method allows amplification in a few hours of specific DNA fragments to a degree that they can be analysed qualitatively and quantitatively by common laboratory techniques (e.g. electrophoresis). However, it requires specialised equipment and training. PCR testing is applicable and extremely sensitive in the case of unprocessed food where the DNA is still intact. This is not the case for processed food where it is more difficult to isolate high quality DNA and where GM material from more than one GM species can be present. In the latter, the method is laborious and costly. PCR requires little reagent development time compared to immuno-logical assays, but it can still take 1 to 3 days to receive results from a testing laboratory. The test is estimated to be about 99.9% accurate. ELISA This method is able to detect and to quantify the amount of a certain protein which is of interest in a sample that may contain numerous other dissimilar proteins. ELISA uses antibodies to bind specific proteins. Antibodies are soluble proteins produced by the immune system of animals in response to exposure to a foreign substance (called antigen). For GMOs, the antigen can be the newly synthesised protein. A colorimetric or fluorometric reaction can visualise and measure when the antigen and specific antibody bind together. One restriction for using the ELISA test is the denaturation of proteins in some food processes. Similar to PCR, the ELISA method requires trained personnel and specialised equipment. This method also requires high investments to develop the assay and to generate antibodies and protein standards. However, once reagents are developed, the cost per sample is low. The test is reported to be 95% reliable. |
The DNA-based PCR test takes 1 - 3 days, at a cost of 104 - 310 € per test. The ELISA test takes only 2 - 8 hours and may cost up to 10 € per test. A faster and simpler ELISA dipstick test to provide a "yes or no" result takes 5-10 minutes and costs only 3.6 € per test (ACPA 1999, Lin 2000).
- In order to compare the
different cost elements, testing costs have been calculated per tonne,
although testing is not only applied to raw products but also to processed
foodstuffs. The additional cost per tonne of soya or corn for testing the
presence of a specific biotech trait by the ELISA technique has been estimated
at 0.4 € (Lin, 2000). However, since current ELISA testing methods
require a separate test for detection of each unique trait, several tests
may be required to determine if a shipment is free of biotech material,
in particular for corn. At subterminals and export elevators, PCR testing
is more common than ELISA because it is more sensitive and can be used
to detect presence of several genetic modifications by one set of tests.
Furthermore, it becomes more efficient with larger volumes of grain to
be tested (Lin, 2000).
Cost for an IP testing system have been estimated to range from 1 €/t for a simple checking to as much as 20 €/t for the most disciplined systems of overlapping documentation, field inspections, product sampling and laboratory testing by third parties (Clarkson, 1999). A 1996 Canadian IP example for herbicide resistant GM oilseed rape indicates a total cost for testing, administration and monitoring the IP system of 2.9 €/t (Buckwell et al., 1998, p.65). An alternative to expensive tests could be the introduction of additional genes that provide visual markers to facilitate identification. However, IP documentation is likely to reduce the need for testing compared with, for example, on the spot testing of commodities for GMO contamination or specific traits.
5.3.4. IP costs for transportation and further storage (merchandising)
Additional costs will occur with the need to find separate storage at local elevators and with possible restrictions in the delivery schedule. An IP system for non-GM crops would require traditional elevators to handle at least four types of grains - two types of corn (more likely three incl. high oil corn) and two types of soybeans. This reduces their capacity to quickly and efficiently receive grain at harvest time and will reduce their effective storage capacity. If transportation and storage facilities in silos, trains, trucks or ships cannot be fully used by IP crops, further costs might occur per unit.
- For IP crops the transport
and storage means have to be cleaned. Avoiding any co-mingling during
the loading or unloading process would require cleaning the equipment and
would entail labour downtime costs during cleaning. The costs incurred
would mainly depend on the tolerance level.
Another cost element for IP appears for seasonal crops. EU soybean imports generally come from Brazil and Argentina during the summer and from the US in winter. If it would not be possible to set up IP chains both in the US and in South America, some material would require storage to ensure a constant stream of supplies (Cargill, 1999). However, storage would be more expensive than transportation and might add 15 - 25% to the price of the raw material (Cargill, 1999).
| US survey on firms handling
speciality crops (Bender
et al. 1999)
A spring 1998 survey of 84 US firms trading speciality corn and soybeans (of a total of about 200 US firms) reports that 56% of the speciality (IP) crops traded by these firms were from local origin (max. 15 miles away) and only 5% originated more than 250 miles away. The data suggest that as the percentage of speciality crops handled by a firm increases, they must be collected from an increasingly larger radius. On average, 96% of speciality crops were delivered by truck, 3% were delivered by rail, and 1% by other methods. Speciality crops handled include 61% stored on farm, 23% stored at the country elevator, and 14% received at harvest. The average percentage of speciality crops purchased through farmer contracts was 85%. Contracts with country elevators accounted for 8% of specialty crop purchases and only 5% were purchased through the open market. The contracts varied between basic contracts with quality adjustments (26%), flat price contracts (23%), basic contracts (20%), acreage contracts (16%) or forward contracts (12%). Quality tests are made at delivery (93%), at the farm (56%), for the required variety (83%) or at seal bins (18%). About 80% of the speciality crop was shipped in bulk, 20% in bags, in particular soybeans for food. The primary market for speciality crops was the export market (47%); 33% went to processors (STS soybeans and food corn), 6% went to brokers and 7% to livestock feeders (in particular high oil corn). The additional costs incurred in handling speciality corn have been estimated to be 6 €/t. Average cost increase for handling was less for high oil traits (2.1 €/t) than for food use corn (13.7 €/t). For soybeans the additional costs of handling has been estimated to average 15.8 €/t. The additional costs for food use was 20 €/t and for non-GM STS soybeans it was 6 €/t. Distribution costs to different cost items shows that all of them were higher for speciality soybeans than for corn, except for the analysing and testing cost which were at the same level (see table in annex B). |
- The additional transport
cost range from 1 to 9 €/t for the different products and IP approaches.
These costs represent about 0.5 - 5% of the farmgate price. Lin (2000)
reports the results of an ERS survey that the cost for segregating non-GM
crops could be higher than for speciality crops but does not present any
data.
The key factors will be the amount of crop traded under the different IP systems and the tolerance level for contamination.
Internet marketing: Several actors are offering trade contracts on their websites. Buyers are thus asked to submit requests in good time to allow farmers to adjust their planting decisions and order the appropriate seed (Young, 1999 for DuPont).
Electronic trading and the internet would shorten the chain from the producer to the end user. This would allow multiple IP and marketing systems to exist.
The internet is also used to call for a buying networks of farmers to combine their negotiation power. Registration of farmers and quantity indication by each member would allow to concentrate selling negotiations and organise transportation needs. (Progressive Ag Marketing, 1999).
5.3.5. Additional Costs for the processing industry (feed and food)
Storage tanks of processing plants have to be cleaned prior to use for IP products. Very low tolerance levels might require dedicated storage facilities. A feed mill would probably not want both GM and non-GM supplies of the same ingredient, because of the difficulty of keeping them apart.
Table 5.5: Some examples for IP costs at the processing level
The capacity of larger US processing plants for soybeans and corn is between 2000 and 8000 tonnes a day (Cargill, 1999). Normally, they are run continuously except for annual cleaning or repair breaks. Stopping production and cleaning the facilities would cause additional cost. Therefore, the solution for the processing plants could be to use a certain quantity of IP grains to "clean" the plant and to sell the product mixed with non-IP output. Only after a certain period of IP grain use, the IP supplies run through would be guaranteed to retain their identity. The cost of this solution clearly depends on the quantity of IP supplies put through.
The cost of IP processing would further depend on the number of secondary products produced from the raw material. If only one of the output products is required to be IP, e.g. the soya oil, it will bear the whole cost of IP. If there is a market for all the products of IP however, then the costs of IP will be spread across all end products.
If there is sufficient IP supplies of a crop, it may be possible to dedicate a plant to processing such supplies, in which case there would be no additional costs involved from separate processing and storage.
Samples and tests might be necessary to ensure quality specifications or to check for the required level of tolerance. Ensuring correct product labelling would require additional time and costs as well as the re-setting, re-designing and printing of labels.
The examples given in table 5.5 indicate additional costs of 1.5 - 9 €/t, which is about 0.5 - 3% of the farmgate price of the product concerned.
5.3.6. Total costs for IP systems
Summarising the different costs along the production chain allows the total costs of IP to be estimated. According to the examples available, they range from 5 to 25 €/t depending on the different grains and the IP systems. Thus, IP would increases the grain price by 6 - 17% compared to the farmgate price. These results confirm the conclusions of Buckwell et al. (1998) for quality traits. Since such a range corresponds to the experience with well established IP systems for value added market segments, it can be taken as a reliable estimation of IP costs.
For modifications that focus on agronomic traits, Buckwell et al. stated some difficulties to assess the representativity of the examples. However, the more recent examples confirm a similar range of additional costs compared to IP systems for quality traits.
Summarising the main factors which determine IP costs, the following have been identified:
- Tolerance: The more stringent the purity requirements, the more expensive will be the IP system. For the farmer, the size of the premium will also vary with the degree of purity required in the crop (Cargill, 1999). The tolerance level is an important cost factor for all three IP approaches discussed in this report. Fixing a threshold will particularly concern the cost of seed production, the costs for testing, storage and transportation and the decision to switch a whole farm and a whole processing plant to specific (IP) production.
|
|
|
|
|
|
|
|
| Soybean | GM quality traits: low linolenic, high oleic, low saturate, high protein, high sucrose | USA | (1997) | 15 - 22 €/t | 6
-
9%*) |
(1) |
| Soybean | non-GM:
herbicide resistant |
USA | 1998 | Soyameal
protein:
119 €/t |
50%
**) |
(1) |
| Soybean | non-GM | Italy | 1999 | Soyameal
> 23 €/t |
(9) | |
| Soybean | non-GM | UK | (1999) | 17.2 €/t | (8) | |
| Soybean/ corn | Any type of identiy preservation | USA | 1999 | 4.7 - 21.4 €/t | (4) | |
| Corn | post harvest chemical free | USA | (1997) | 14 €/t | 16% *) | (1) |
| Corn | high oil content | Europe | 1997/98 | 17.6 €/t | 17% *) | (1) |
| Oilseed rape | GM: herbicide resistant | Canada | 1996 | 10.4 - 13.3 €/t | 6 - 8% *) | (1) |
| Oilseed rape | GM herbicide resistant (limited acreage:5% of total acreage in CAN) | Canada | 1996 | 19.7 - 21.4 €/t | 9.5%
*)
8.5-9% **) |
(3) |
| Sun-flower | high oleic | USA | 1997/
1998 |
16.0 - 23.0 €/t | 7
-
10% *) |
(1) |
| *)
farmgate price **) commodity price
Sources: (1) Buckwell et al. 1998; (3) Van Wert (AgrEvo) 1996; (4) Clarkson 1999; (8) House of Commons 2000; (9) Brookins 2000 |
||||||
- Agronomic traits: The genetic disposition for cross pollination and for volunteers will determine in particular the costs on the farm.
- Market volume: Economies of scale can be expected for any IP system. The more crops are traded under such a system, the higher will be the potential to reduce costs. Furthermore, if an entire stream can be devoted to an IP system, additional costs should be quite low.
- Seasonality: A strong seasonality of market supply could increase the storage costs of an IP system, in particular if the IP crop is grown only in a particular region or country.
- Derived products: IP costs per unit depend on the share of all processing products which can be marketed as IP. If only one of a whole range of the output products is to be identity preserved, it will bear the whole costs of IP.
- Price responsiveness (own-price elasticity): Depending on the responsiveness of demand and supply to price at each of the stages additional costs can be shifted - at least partially - to the previous or to the following stage of the production chain. Generally the less price-responsive demand is at a certain stage, the more of the additional costs will be absorbed by the consumer at this stage. Equally, the less price-elastic is supply, the more of the additional costs have to be absorbed by the producer (Buckwell et al., 1998).
- Availability of substitutes: The more substitutes are available, the more responsive would be the price. Thus for products, which can easily be substituted, additional costs will hardly be shifted to the processor or the final consumer. In this case, it will be the farmer who has to bear most of the additional costs of IP. On the other hand, if a product is difficult to substitute, it will be the consumer who has to bear the IP costs.
- Market structure: Price-responsiveness can be affected by the competitive structure of the industry. The more concentrated the structure, the more likely that any additional costs are passed over to the previous or the next stage of the chain. In the food sector, the market power is in general stronger at the food processing and retailing levels compared to the farmer and consumer level. Thus IP costs are very likely either to be passed back to the farmer through lower prices for his products or to be passed forward to the consumer in the form of higher food prices.
- Agricultural price policy: Agricultural policy measures, in particular those established to control agricultural prices may have an adverse impact on the transmission of additional costs to the consumer. On the other hand, price policy may also reduce the transmission of benefits of cost reductions by new technologies and thus reduce the economic incentives to apply these innovations.
These factors apply to
all three IP approaches which have been identified in the context of GMOs.
1. Voluntary IP of specific GM traits: If GM crops have a specific value to the consumer, these crops have to be handled separately, in order to preserve their value through the chain. Price elasticity of supply can be expected to be high. On the demand side, the new trait will create a situation in which the scope for substitution is limited and thus demand gets fairly price inelastic. The effect will be that most of the additional cost can be passed on to the consumer. The market will be a niche market - at least in the beginning - for each of the new traits introduced by genetic modifications.
Thus it is very likely that the consumer will be charged a premium which covers not only the intrinsic additional value of the new product, but also the costs to handle them separately through the food chain.
2. Voluntary IP of GMO-free products: If GMO-free products have a specific value to consumers, they are willing to pay a premium for these products, which are handled separately or identity-preserved.
With a voluntary IP system for GMO-free products, additional costs will be borne by the producers, processors and consumers of these GMO-free products. The scope for passing over the costs of IP for a GMO-free product will depend upon how strong the demand for GMO-free products will be. The stronger the demand, the less responsive will it be to price change. This would increase the scope for suppliers to pass over the costs of IP in the form of higher prices (Buckwell et al., 1999). Thus it will be more likely that the consumer bears the costs than the farmer of GMO-free crops.
For the short-term development, however, some impact on the market for GM crops cannot be excluded. In a short-term analysis supply of GM and GM-free products is assumed to be fixed. Consumers without specific preference for non-GMO products will not care whether they consume GMO or GMO-free products. However, GMO-free demand will not accept GMO supply. So there will be one-way situation for substitution and the magnitude of demand for IP products relative to the demand for commodities will be the crucial factor to determine the distribution of the additional costs as well as of the price of GM and GMO-free crops (see also section 5.3.2).
To analyse the short term market effects, two scenarios can be distinguished:
| Scenario 1:
The share of total demand for GMO-free crops is greater than the share
of GMO-free market supply.
In this case, severe market disruptions may occur as processors strive to locate and purchase GMO-free crops. With a high demand for GMO-free crops, their prices would increase rapidly and a surplus of GM products is likely to be build up. Substitution of GMO-free by GM products would in general be rejected by consumers or processors which are looking to avoid GMOs. However, the increasing price gap might be an incentive for some of them to change their minds and accept purchasing GM products. Furthermore, a surplus of GM crops could only be avoided by offering a discount which makes customers buy more GM crops. Processors will be forced to develop a price schedule that reflects the relatively low value of GMOs in the market. The discount would be applied to all GMOs and not just to the proportion of GMOs that are in surplus. (Miranowski et al. 1999) Scenario 2: The demand for GMO-free products is relatively small compared to the available supply. The marketing of the GM crop would not be affected by the relative surplus of GMO-free crops. Any GMO-free crop would be accepted by the conventional production chain. In this case, the purchasers will not pay a premium or discount for GMO-free products and producers of GM-products will not have to take a discount. However, farmers have to invested in producing GMO-free crops and - at least for some of them - the additional costs will not be covered by the conventional marketing. It would be those farmers and the consumers of GMO-free products who are very likely to bear the costs under scenario 2. |
Agronomic traits address the producer and the crops are marketed similar to conventional crops. Thus any consumer without particular preference for GMO-free food should be indifferent when comparing GM and GMO-free products. A high degree of substitutability can be supposed, because the consumer could easily switch completely to the conventional product if additional cost for IP would increase the price of a product. This would mean that IP costs would be passed back to primary producers and processors of GM crops. The producers of conventional crops would not be affected and the additional IP costs at the farm level would reduce the profitability of GM crops.
The relative position of GM and conventional crops could be altered, if the agronomic trait is sufficiently advantageous at the farm level. As soon as the GM crop accounts for a significant proportion of all traded crops, it becomes the norm and will set the baseline for the commodity price of this crop (Buckwell et al. 1999, p.21). This would reduce the competitiveness of conventional crops and increase the incentive to adapt the production programme.
5.5.1. EU markets for soybeans and corn
| Soybeans |
|
|
|
|
| EU Production |
|
|
|
|
| Imports |
|
|
|
|
| Exports |
|
|
|
|
| Availabilities |
|
|
|
|
| Self-sufficiency (%) |
|
|
|
|
| Cake
and
cake equivalent (meal) |
|
|
|
|
| EU
Production
- from Community seed - from imported seed |
11 865 |
11 164 |
11 067 |
10 880 |
| Imports |
|
|
|
|
| Exports |
|
|
|
|
| Availabilities |
|
|
|
|
| Self-sufficiency (%) |
|
|
|
|
| Oil and oil equivalent |
|
|
|
|
| EU
Production
- from Community seed - from imported seed |
2 738 |
2 576 |
2 554 |
2 511 |
| Imports |
|
|
|
|
| Exports |
|
|
|
|
| Availabilities |
|
|
|
|
| Self-sufficiency (%) |
|
|
|
|
| Source: European Commission 2000 | ||||
Most soya-bean/meal production and imports are used for animal feed, but a small share (less than 1 Mio tonnes) is used for food. The EU main - and nearly exclusive - trading partners for soya beans and meal imports are Brazil, Argentina and the US (table 5.8). During the last years, soybean imports from the USA have been reduced, while imports from Brazil increased. On the other hand, soymeal imports from Brazil decreased and imports from USA and Argentina increased.The European market is of particular importance for Brazil and Argentina. 40 to 50% of their soya production is sold to the EU. The USA as the world's leading soybeans exporter, are sending 10 to 15% of their production towards the EU, which, is equal to around 30% of USA soya exports. Thus, for soya bean and meal trade, there is a mutual dependency between the three main exporters and the EU as the main importer.
(in soymeal equivalents - soybeans = 79% meal)
|
|
|
|
|
|
||
|
|
|
|
|
|
|
|
|
USA Brazil
Argentina
others |
% of total Mio t
Mio t
Mio t
|
33.1% 10.0
5.8
1.2
|
32.1% 8.9
5.2
0.9
|
34.7% 8.6
4.0
1.0
|
28.2% 10.2
6.1
1.5
|
20.9% 9.8
8.0
0.9
|
| Source: European Commission 2000 | ||||||
- more than 50% of the US soybean area and almost three quarter of the Argentinean soybean area are under GM crops,
- segregation of GM and non-GM crops is still limited in the US and there is no evidence on segregation in Argentina,
Corn: In corn production, the EU has reached a degree of self-sufficiency which is around 100% (table 5.9). Imports contribute 4 - 8% to total availability on the internal market. Feed use absorbs about 75 - 80% of the EU market volume, industrial use accounts for 4.2 Mio t each year (11-12%), and human consumption for 2.6 Mio tonnes (7%).
|
|
|
|
|
|
| EU Production |
|
|
|
|
| Imports |
|
|
|
|
| Exports **) |
|
|
|
|
| Availabilities |
|
|
|
|
| Self-sufficiency (%) |
|
|
|
|
| *)
estimation **) includes 85-95% processed products and animal feed
Source: European Commission, Grains Outlook March 2000 |
||||
However, imports of corn by-products, in particular corn gluten feed, surmount the imports of corn grains. In 1999, around 4.7 Mio tonnes of corn gluten feed was imported by the EU. The value of EU corn gluten feed imports from the US (1998: 500 Mio €) for instance is higher than the value of corn imports (1998: 240 Mio €).
| EU imports |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
USA Argentina
others |
% of total Mio t
Mio t
|
86% 0.5
0.02
|
77% 0.6
0.04
|
64% 0.9
0.03
|
12% 1.4
0.4
|
1.1% 2.0
0.53
|
| Source: European Commission, Comext, 2000 | ||||||
|
|
|
|
|
|
||
|
|
|
|
|
|
|
|
|
Corn Gluten Feed Brewers grains Corn germ cake |
Mio t Mio t |
0.9 |
0.5 0.53 |
0.6 0.39 |
0.7 0.10 |
0.628 0.129 |
| Source: DG Agriculture / Member States | ||||||
The main producers, in particular the US have already reacted to the EU and the Japanese demand. The Iowa State University has estimated that the US market should handle the situation quite easily, if about of 7 to 10% of EU demand would switch to non-GMO soya products. However, if EU food retailers and consumers should decide to reject meat from animals fed with GM soymeal, a significant price difference between GM and conventional soya would emerge. Therefore, the consumer attitude on meat from animals fed with GMO feed-stuff will be a crucial factor for the price development.
Furthermore, other factors
may influence EU import demand for non-GM soybeans:
- there is certain scope to substitute soya by other products,
- EU soymeal import demand has proven to be quite price elastic.
- Sourcing non-GM soybean suppliers often implies establishing new trade partnership, including contracts governing identity preservation, which has a cost (e.g. transaction) and requires time. When the number of significant exporters is limited as is the case for soybeans, it is even more difficult to find alternative suppliers.
| US reaction
to non-GM demand
In the US, segregation initiatives are mainly export driven, or they concern specific clusters like baby food. According to a recent survey of nearly 1200 US elevators about a quarter of the respondents will segregate GM and non-GM corn and 20% will segregate soya in autumn 2000. One out of ten elevators has declared to offer a price premium for conventional corn and 14.3% are planning to offer a premium for conventional soya. The resistance to buy GM crops also differs among the two crops. Only 12% of the elevators are planing to refuse biotech soybeans in fall 2000 and 18.4% of the elevators will refuse to buy biotech corn (Pioneer Hi-Bred International, 2000). According to a Reuters' survey of 400 US farmers, 15% of them have made or are planning to make investments to handle or segregate GM crops. (Reuters Business Brief 13 Jan 2000). |
For the USA, some estimations of possible market share have been made: If the entire US food processing industry switched to non-GM corn, the market for non-GM corn would constitute 8% of the 1998 US corn market. If the sweetener and the ethanol (by-product of corn) industries joined, non-GM corn would constitute 20% of the US corn market. Finally, 17% of the US 1998 production was exported of which 80 to 90% is fed to livestock and only a small percentage is directly processed into food products. This implies that an upper limit of the market share for non-GM corn in the US is 37% (Miranowski et al., 1999).A French research team (Valceschini 1999) is assessing the economic relevance and the technical feasibility of non-GM supply chains. Preliminary results on consumer reaction with regard to GM food were presented in December 1999. The researchers observed the buying decisions of consumers when choosing between GM and conventional products, the GM ones being properly labelled. Based on the observed sample, one third of consumers reject GM-labelled products, another third would buy them if they were cheaper than presently and a last third does not care and buys them. On this provisional basis, the authors assumed that appropriate labelling of the GM nature/origin of foodstuffs will have a significant impact on consumer demand. However it is difficult to quantify this impact.
It is interesting to note that three consumer groups of the same size are identified. This could echo the three-tiers market previously identified: GM, non-GM and GM-free. The "middle" group of consumers shows a very price-elastic behaviour. If GM foodstuffs are cheaper than presently, which means cheaper than conventional food, these consumers could adopt them. This is another factor suggesting a possible decline in the market share of the non-GM tiers.
5.5.3. Different stance on food and feed uses
The EU balance sheets for soya and corn have shown that the main use of soya and corn is in the feed sector, which will have a significant effect on the breakdown of demand between the GM, conventional and GMO-free segments. The EU Commission has announced to table a proposal dealing with novel feed, including GM feed in the second half of 2000. The labelling rules and in particular the level of the tolerance threshold will be key elements influencing market behaviour.
In Europe, some operators are already organising non-GM soybean supply chains for animal feed (see box on Soya de Pays). Depending on the quantities needed, the origin is mainly domestic (French and Italian soybean production) or foreign, in particular imports from Brazil. However, these initiatives concern a limited share of the feed market. Most initiatives are taken in the poultry sector. This might echo the attempts to restore market confidence after the dioxin crisis. In addition the market for poultry is a segmented one, there are already price premia for identified quality (example red label chicken).
| Soya de Pays,
France
Feed producer Glon Sanders and poultry producer Bourgoin have established a production chain for non-GM eggs and poultry meat production based on French non-GM soya. Participating farmers are not allowed to plant imported US soybean seed, have to enable traceablity back to the producer, respect distance from pollution sources and other requirements. The costs of IP are entirely borne by soymeals, as non-GM soybean oil cannot be easily valued because of substitution with rapeseed oil. French non-GM soybeans cost 30% more than imported ones. The first chickens fed with non-GM soya ("soja de pays") have been on shelves in April 2000. The first eggs were already introduced in February and their price is 15% higher than standard eggs. Farmers producing chicken said that thanks to the "soja de pays" initiative, they could get a premium of 15 €/t. Based on increasing demand from processing industries, areas under "soja de pays" are forecast to raise from 20,000 ha in 1999 to 60,000 ha in 2000, which represents 60% of the French soybean area. |
Soymeals, corn and its co-products account for key elements in animal feed. Three groups can be distinguished among key marketable feedstuffs:
- cereals (54% of marketable feeding stuffs in the EU);
- energy rich elements (27%),
- protein rich elements (19%).
In short term, segregation of the feed market into GMO and non-GMO stuff would increase feed production costs and thus animal production costs within the EU. Depending on the market development, imports in soybean meal, corn gluten products and other ingredients might be reduced and demand for locally produced feedstuffs, particularly rapeseed meal, barley and wheat could increase. (Gill, 1999)
As long as there are significant origins for non-GM crops, the need to set up IP systems would be limited. Trade flows would just adapt to this new demand. Secondly, if a product can easily be substituted, then IP is also unlikely to occur, because it will be far easier to switch to the substitute. Thirdly, if the commodity in question has many outlets around the world, the reaction on other markets will be relevant to the EU market. For instance, if Japan is paying a premium for non-GM soya then any IP system set up is going to supply this market first.
Non-food/feed uses of GM crop are expected to provide market opportunities in the medium or long term. There are possibly good prospects for renewable resources used in energy production and in the chemical industry. In general, the societal and ethical acceptance of these applications is higher than that of GM food products (Menrad, 1999 and Eurobarometer, 2000).
However, according to Menrad (1999), non-food applications of biotechnology would need a concerted effort involving science, industry and politics, also taking into account the interests of other groups (eg farmers) to speed up.
Not surprisingly, this situation has become a trade issue. However, it is difficult to isolate the possible effect of biotechnology on developments in trade, as many other factors play a role, like changes in competitiveness, transportation costs and the transaction costs of giving up of long-established trade links.
The issues at stake are of a different order of magnitude for soybeans and for corn. Between 1995 and 1997, EU imports from the US were worth, on average, 2 billion € for soybeans and soymeals and 0.03 billion € for corn. In addition, EU imports of Corn Gluten Feed are estimated to be worth around 500 Mio €.
US soybean exports declined from 26 to 20 Mio tonnes between 1997 and 1998, while world soybean trade held fairly steady. EU soya imports from the US have been partially replaced by imports from Argentina. The USDA has concluded that "traditional competitive forces (primarily prices) appear to be the main driving factors behind the changes in observed bilateral trade patterns". As the share of GM soybeans is much higher in Argentina than in the US, this shift in trading pattern cannot be attributed to reluctance to import GM soybeans.
The drop is even sharper for corn than for soybeans. US corn exports fell from 60 Mio tonnes in 1995 to 41 Mio in 1998. Most of the drop occurred on South-East Asia markets (with the exception of Japan) and is explained by the situation of China, which became again a net exporter of corn. On the EU market for corn, the share of US has steadily fallen while the share of other partners, in particular Argentina and Hungary, has significantly increased. The USDA considers that the loss of shares on the EU market results from issues related to biotechnology, in particular the differences in regulatory approaches.
While 11 types of GM corn have been approved in the US, only 4 have been cleared at EU level (table 5.12), and some Member States have decided to suspend authorisations for growing. Non-authorised GM crops cannot be placed on the EU market. In the absence of tolerance thresholds, if traces of such crops are found in a given consignment, it cannot be cleared for importing into the EU. According to the USDA, this situation has created uncertainties.
However, the type of GM soybeans which is mostly grown in the US (herbicide tolerant) is authorised in the EU for imports and processing (but not for growing purposes). According to the USDA, only a small part of US areas have been sown to non-EU approved corn varieties and the EU only accounts for 1% of US corn exports.
Trade issues have been addressed in the Biosafety Protocol, which aims at ensuring an adequate level of protection for transfer, handling and use of GMOs which might have an adverse effect on biodiversity. Reference is made to the precautionary principle in this respect. It is hoped that procedures foreseen under this Protocol, in particular information sharing and accompanying documentation, will help improving the predictability of transboundary movements of GMOs.
| Biosafety
Protocol
The Biosafety Protocol provides a framework for addressing environmental impacts of bioengeneered products that cross international borders. It was concluded in Montreal in January 2000 by delegates from 138 countries. "In accordance with the precautionary approach (.), the objective of this [Biosafety] Protocol is to contribute to ensuring an adequate level of protection in the field of safe transfer, handling and use of Living Modified Organisms resulting from modern biotechnology that may have an adverse effect on the conservation and sustainable use of biological diversity, taking also into account risks to human health, and specifically focussing on transboundary movements" (Article 1). The procedures foreseen under the Protocol are different for Living Modified (LM) seeds and commodities.
|
Finally, it is worthwhile noting that Identity Preserved markets are expected to increase in number and market share, with or without GMOs entering the markets. Trade experts have estimated a 25% market share for IP corn and IP soybeans by 2005 (Clarkson, 1999). Identity preservation systems in the US currently account for 8-10% of US agricultural production, and in ten years' time would be accounting for 25-30% (Young, 1999).
[ Agriculture
] - [ Publications
]
