abare.gov.au - research report 08.04

2	Costs and benefits of GM crop adoption

This chapter gives an overview of the potential on-farm benefits and costs arising from the adoption of GM crops in Australia. It highlights the potential impact on yield, use of pesticides and herbicides, and labour inputs, using the Australian and international experiences where available. Potential effects on prices and environmental impacts are also presented.

Crops considered

This report considers the adoption of first generation GM crops — that is, crops with modified traits that provide on-farm production benefits such as protection against insect pest infestation (Bt crops), crops that have been modified to be tolerant to herbicide treatments (HT crops) or crops with a combination of these modifications (stacked trait crops) (see box 1). These GM crops currently account for the bulk of commercial plantings of GM crops globally and in Australia.

Second generation crops are expected to give enhanced value to consumers by incorporating traits that lead to enhanced quality attributes in farm products. Third generation crops are expected to produce pharmaceutical or industrial products. Second and third generation crops are still in their early stages of development and are not considered in this report.

box 1

First generation GM crop traits

Within first generation GM crops there are currently two dominant traits: herbicide tolerance (the HT trait) and insect resistance (the Bt trait). Most of the global first generation GM crop area in 2007 was planted to crops with the HT trait (63 per cent), followed by the Bt trait (18 per cent), with stacked traits accounting for the remainder (19 per cent) (James 2007). Crops with stacked traits contain more than one GM trait — for example, two HT traits or a combination of HT and Bt traits.

Most HT crops have been genetically modified to be resistant to glyphosate. Glyphosate is a non-selective, broad spectrum herbicide that has been used extensively during the past two decades (Reddy 2001). The introduction of HT crops into crop production systems has enabled farmers to better manage weed infestations. GM crop varieties resistant to other herbicides such as sulphonyl urea, imazethapyr and glufosinate-ammonium have also been adopted by farmers to improve weed control (Norton 2003; Sanogo and Yang 1998).

For insect resistance, crops have been modified to include a gene found in soil bacteria Bacillus thuringiensis (Bt). This gene produces substances toxic to certain insects. The introduction of Bt crops has led to reductions in insect damage and/or pest control costs.

On-farm benefits

The adoption of GM crops is aimed at delivery of on-farm benefits such as higher yields and/or reduced operating costs. GM crop adoption may also lead to indirect impacts on upstream industries (industries that produce agricultural inputs) and downstream industries (ones that use agricultural products). Cumulatively, these benefits can result in increased trade flows and higher national incomes. These results are evaluated in a number of illustrative scenarios presented later in this report.

Yield effects

Cultivation in Australia of GM crops could reduce yield losses to insect pest and weed infestations. The potential yield of a crop depends on a range of factors, including the genetic potential of a crop variety, farm size, scale, geographical location, water and nutrient availability, soil condition, climatic factors and farmers’ management skills. However, in general, the higher the pest and weed incidence is, the greater will be the gain from growing GM crops (Fulton and Keyowski 1999; Marra et al. 2002; Qaim et al. 2006) (table 1 and box 2). To the extent that GM crops reduce yield losses from insect and weed pressures compared with non-GM crops, their adoption will increase farm output.

Pesticide and herbicide cost savings

GM technology has led to a reduction in on-farm costs associated with pest and weed management. For example, Bt crops provide their own protection against pests and therefore reduce or eliminate the need for pesticide. Also, current HT crops allow the use of relatively inexpensive broad spectrum herbicides, such as glyphosate or glufosinate-ammonium, which effectively control most weeds commonly found in agricultural fields. This eliminates the need for complex herbicide application regimes that require a mix of different, expensive and more toxic weed-specific herbicide sprays.

There is significant variation among countries in the extent to which GM crops lead to cost reductions for pesticide and herbicide inputs. This variation arises from differences in pest/weed pressure, economic trends and conditions, input costs and farm management skills and techniques.

The evidence presented in Box 3 is taken largely from on-farm experiences around the world. Experiences from countries that have similar production systems to Australia, such as Canada, are likely to provide the best indication of possible advantages to Australian producers.

box 2

Yield effects by crop

Cotton: Australian farmers have not received significant yield benefits from Bt cotton adoption as they have been traditionally very effective in controlling helicoverpa caterpillar infestations (Brookes and Barfoot 2006). The main advantage of Bt cotton adoption to Australian farmers has been reduced operational and pesticide costs.

Existing literature indicates that Bt cotton has increased yields in many other countries, although with significant variations among them. Yield increases for cotton producers range between 9–11 per cent in the United States to 43–87 per cent in India (see table 1).

Canola: Australian field trials have demonstrated that GM canola has the potential to increase yields in Australia. The results of Monsanto field trials, conducted side by side with HT canola and alternative non-GM canola varieties and weed management systems, indicate that glyphosate-tolerant canola provides substantial yield gains (8–24 per cent) compared with non-GM varieties. Glyphosate-tolerant canola was found to have similar or higher oil content than currently available varieties. Bayer Crop Science trials of InVigor Canola also indicate yield gains of 9–38 per cent (ACIL Tasman 2007). Pratley and Stanton (2007) also conducted a five-year field trial, in Wagga Wagga, NSW, comparing HT canola to non-GM canola production systems.

They reported that HT canola resulted in generally higher yields. Also, Norton (2003) estimated that HT canola has the potential to raise average Australian yields from 1.27 tonnes a hectare to 1.38 tonnes a hectare, an increase of 8 per cent.

In North America, GM canola yield advantages have averaged 6–10 per cent (Carew and Smith 2006; Mayer and Furtan 1999; Serecon Management Consulting Inc and Koch Paul Associates 2001).

Soy bean: On the basis of international experience, yield gains from cultivation of GM soy bean are likely to be limited. Some studies have reported marginal yield gains (Bernard et al. 2004; Fernandez-Cornejo and McBride 2002) and yield losses (Benbrook 2001); however, the small differences in yields reported in the literature are likely to be caused by differences in agronomic conditions and farm management skills. Although significant increases in yield have been reported by Romanian farmers, this could be because a tradition of poor weed control has meant a significant yield advantage with the introduction of HT soy bean (table 1).

Maize: Insect pests damage maize crops, resulting in losses in yields and in farm revenues. For example, in countries affected by the corn borer, such as the United States, or the helicoverpa caterpillar, such as Australia, there is only a limited time during crop rotations in which to apply a pesticide before the pest larva burrows into the stem of the plant. Bt maize has been engineered to provide protection against such pests, thus reducing yield losses in countries where it has been adopted. Bt maize varieties in the United States, Spain and South Africa have achieved yield advantages of between 5 and 11 per cent compared with non-GM varieties (table 1).

Prospective GM crops: GM wheat and GM rice are crops that may be adopted in the future. GM wheat field trials conducted in northern America show a yield advantage of 9 per cent for GM wheat resistant to glyphosate and 1 to 3 per cent yield advantage for fusarium-resistant wheat (Berwald et al. 2006). Chinese field trials of GM rice have indicated yield advantages of 3–7 per cent (Huang et al. 2005).

Farm management and labour cost savings

Managing crops with GM traits is generally easier and less time consuming. For example, GM crops reduce the number of annual sprays required, therefore reducing labour, machinery and fuel costs. In addition, many HT crops enable farmers to switch to broad spectrum herbicides which allow more flexibility in cropping decisions. Because these herbicides bind to soil particles, they are insoluble and therefore non-persistent. They allow crop decisions to be made on a season-by-season basis without locking fields into particular crop rotations (Norton 2003).

The labour savings and convenience of GM production systems have been widely discussed in the literature and a number of studies have quantified the benefits.
In cotton production the labour savings arising from adoption of GM technology will vary according to seasonal conditions and, in particular, the pest and weed pressures. In Australia, Bt cotton provides savings to farmers through reduced labour and fuel costs and reductions in the time spent in the field applying pesticide. Fitt (2003) estimated spray applications were reduced 66 per cent on average over the first six years of Bt cotton production in Australia. In South Africa, Morse et al. (2005) estimated that, because of the reduced number of sprays, labour required for spraying fell by about 50 per cent. In India, available estimates indicate that Bt cotton can lead to a 75 per cent reduction in the number of sprays each year (Bennett et al. 2004; Marra et al. 2002; Qaim and Janvry 2003; Traxler and Godoy-Avila 2004).

Adoption of HT canola and soy bean has also resulted in fewer herbicide sprays and a movement toward no tillage or minimum tillage cropping — where seeds are sown into the ground with minimal disturbance to top soil — reducing costs through lower machinery, fuel and labour requirements. For example, Gómez-Barbero and Rodríguez-Cerezo (2007) reported 80 per cent of Argentine farmers growing GM soy bean had adopted minimum tillage practices compared with 42 per cent of those growing conventional soy bean. A report published by the Canola Council of Canada in 2001 quantifies fuel savings from GM canola adoption at 5.1 to 6.3 litres per acre. However, no attempt was made to quantify labour cost savings (Serecon Management Consulting Inc and Koch Paul Associates 2001).

Planting HT canola also enabled farmers to practise minimum tillage, saving C$7.50 an acre in operating costs. Qaim and Traxler (2005) have reported labour savings of 8 per cent (US$3.60 a hectare) and fuel and maintenance savings of 28 per cent (US$6.82 a hectare) for Argentine soy bean farmers.

HT technology may allow farmers greater flexibility in planning future crop rotations. Some herbicides used for weed control in crops remain in the soil for as long as 34 months and therefore restrict what can be cropped in future rotations. Herbicides used with HT crops, such as glyphosate or glufosinate-ammonium, typically do not persist in the soil following a crop rotation, thus increasing farmer flexibility in planning future crop rotations (Norton 2003). Early field trials also suggest crops following GM crops have higher yields than those following non-GM crops, because control of weeds reduces weed seed levels and possible hosts for pests in the subsequent crops (Pratley and Stanton 2007).

From the literature survey, the impact of Bt maize adoption on labour and fuel input costs is uncertain. Where Bt maize adoption leads to a substantial reduction in the number of pesticide sprays, costs associated with spraying are likely to be reduced. In contrast, Gómez-Barbero and Rodríguez-Cerezo (2007) studied the impact of Bt maize on farm labour in Spain and concluded Bt maize does not affect the extent of paid or non-paid labour used.

box 3

Pesticide and herbicide cost reductions from GM crops

Cotton: In many countries, the introduction of GM cotton has resulted in a reduction of pesticide use of between 50 and 80 per cent. In Australia, for example, it has been estimated Australian farmers have achieved an average reduction in pesticide use of 56 and 75 per cent from adoption of Bt cotton varieties (Fitt 2003 and Knox et al. 2006). In the United States pesticide cost savings of between US$63 and US$74 a hectare have been achieved since commercialisation in 1996 (Brookes and Barfoot 2006). In South Africa, Bt cotton has led to a 53–63 per cent reduction in pesticide cost, with greater savings achieved in years of high pest incidence and in irrigated areas (Morse et al. 2005). In India, pests have traditionally had a devastating impact on cotton crops and so the use of GM cotton has enabled a reduction in pesticide use of 71–83 per cent (Bennett et al. 2006). Qaim and Janvry (2003) and Huang et al. (2003) have reported cost savings of around 50 per cent and 87 per cent for Argentine and Chinese cotton farmers respectively.

Canola: A survey commissioned by the Canola Council of Canada in 2001 found that 80 per cent of farmers reported a 40 per cent (C$9 an acre) reduction in herbicide use when using Liberty Link® or Roundup Ready® GM canola (Serecon Management Consulting Inc and Koch Paul Associates 2001). Similar results are presented in Fulton and Keyowski (1999), Mayer and Furtan (1999) and Gianessi (2005).

Soy bean: The major benefit of HT soy bean is easier and cheaper weed control. Gianessi (2005) compared the cost of weed control in the United States under HT soy bean to the most effective alternative and concluded HT soy bean deliver a saving of US$20 an acre. Brookes and Barfoot (2006) and Fernandez-Cornejo and McBride (2000) reported similar cost savings for United States soy bean farmers of US$22-34 a hectare. Qaim and Traxler (2005) have quantified the herbicide cost savings to Argentine farmers at US$24–30 a hectare. Cost savings of 29 per cent have also been achieved in Romania (Brookes 2005).

Maize: South African farmers have achieved pesticide cost reductions of US$7–21 a hectare, with larger cost savings occurring in irrigated GM maize crops where pest infestations are more prevalent (Gouse et al. 2005). Spanish farmers have reported cost savings of €4.50 to €20 a hectare (Gómez-Barbero and Rodríguez-Cerezo 2007).

Off-farm income

Where adopting GM crops requires farmers to spend less time in the field, they may be able to work off-farm to increase their household income. This was supported by Fernandez-Cornejo et al. (2005), who found a positive relationship between off-farm income and HT soy bean adoption in the United States.

On-farm costs

Seed prices, technology fees and user agreements

Farmers opting to grow GM crops are likely to face additional costs in terms of higher seed prices, technology fees and restrictive user agreements.

GM seed providers’ commercial practices in setting GM seed prices and charging technology fees to users — largely based on the area of land planted to GM crops — has led to significant variation in the costs faced by GM crop growers for different crops and in different countries (table 1). Gómez-Barbero and Rodríguez-Cerezo (2007) note that seed prices are correlated to pesticide cost savings for Bt maize in different regions in Spain, with premiums being highest where these cost savings are greatest. They also suggest the actual seed price appears to be influenced by farmers’ bargaining power. However, where there is competition in the GM seed supply market, as has been the case with HT soy bean seeds in Argentina, seed prices appear to be lower.

The use of GM seed is normally accompanied by a technology user agreement, imposed either by the technology provider or user industry groups. The agreements are often designed to protect intellectual property rights, reduce or delay insect/weed resistance and protect the environment. These agreements could entail additional costs for growers in adhering to regulations such as mandatory buffer zones.

An example of such an agreement is a mandatory refuge area where non-GM crops must be grown. For example, when Bt cotton was first introduced in Australia, the cotton industry restricted adoption to 30 per cent of total farm area because of uncertainties about heliothis pest developing resistance to the Bt protein (Fitt 2003). Now, with improved technology, these restrictions have been removed. For Bt cotton in India, regulations require a minimum of 20 per cent to be planted to conventional cotton, with a five-row buffer zone around all Bt cotton plantings.

Segregation cost

On-farm segregation arrangements can mean higher costs because of the need for certified planting seed; various crop management techniques (including appropriate separation distances between crops and control of ‘volunteer’ growth); and cleaning after harvesting, handling, storing and transporting GM grain types (Foster 2006).
Additional costs in the central receival system include extra grain testing requirements and more labour because of a longer receival period. The additional costs are likely to be small relative to on-farm costs and benefits, reflecting the economies of scale with bulk handling of grain (Foster 2006).

Price premium/discount

If a GM crop is differentiated from its non-GM counterpart in the market, the possibility of either a price premium or price discount exists. A price premium occurs when a product is perceived to have enhanced quality attributes. Quality here refers to the quality of the grains and to the extent that it is free of weeds and other impurities.

As HT crops allow more effective weed control, it is possible they will result in a cleaner harvest and therefore attract a farm gate price premium. This has been observed in Canada where cleaner GM canola has received a price premium of 1.27 per cent over non-GM canola (Serecon Management Consulting Inc and Koch Paul Associates 2001).

GM canola also enables earlier sowing, favouring production of crops with higher oil content. Generally, a price premium is paid for canola with higher oil content. It has been estimated that, if grown in Australia, HT canola would attract a price premium of 1.5–3 per cent, based on the increase in oil content (Foster 2003; Norton 2003).

Environmental aspects

Bt crops reduce the need for farmers to handle toxic pesticides, and hence improve the occupational health and safety environment for farmers (Marra et al. 2002). Reductions in pesticide applications also result in a reduction in environmental costs associated with spraying, such as spray drift and leaching to groundwater.

HT crops may also deliver environmental, occupational health and safety benefits to farmers. For some crops, such as canola and cotton, HT technology allows replacing more toxic herbicides (such as atrazine) with safer herbicides such as glyphosate or glufosinate-ammonium. It has been estimated that glyphosate is three times less toxic than most herbicides it replaces (Reddy 2001). HT technology also allows farmers to reduce tillage operations before sowing and rely on post-emergence weed control (Norton 2003). Minimum tillage helps to preserve soil structure and protects it from excess wind and water erosion.

Environmental and health issues associated with GM crops have been widely discussed in the literature. Morse et al. (2005) noted there had been fewer hospital admissions resulting from pesticide illness after introducing Bt cotton in South Africa. Pray et al. (2002) reported only 5–8 per cent of Bt cotton farmers in China suffered from pesticide illness compared with 22–29 per cent of other cotton farmers.

Knox et al. (2006) looked into the specific environmental benefits of Bt cotton in Australia. They reported Bt cotton has contributed considerable environmental benefits. Fitt (2003) also found evidence for environmental benefits associated with a reduction in pesticide use in cotton growing areas in Australia of 1.75 million litres. Qaim and Traxler (2005) reported the wider adoption of less toxic herbicides after introducing HT soy bean in Argentina.

Regarding biodiversity, a study conducted in Spain found there had been no detrimental effect of Bt maize on predatory insects (Gómez-Barbero and Rodríguez-Cerezo 2007). Morse et al. (2005) also found adoption of Bt cotton in South Africa led to an increase in on-farm biodiversity and a return of beneficial predatory insects owing to the reduced use of pesticides. These findings agree with a comprehensive review of the literature on the environmental impact of GM crops conducted by Romeis et al. (2006).

Minimising the impact of GM cropping technologies on the environment will require best practice farming techniques to be adopted and sustained. For example, the potential development of resistance in the weed spectrum from HT crop production systems arising from regular use of a single herbicide will require monitoring and changes to farming practices where appropriate.

Evidence for increased resistance in the pest/weed spectrum as a result of GM crop production is scant. These issues are considered on a case-by-case basis during the OGTR’s approval process for the growing of GM crops and measures are put in place to manage resistance based on the level and nature of the risk.

Economic costs and benefits to other sectors

Adoption of GM crops can have a number of indirect impacts on interrelated upstream industries such as seed, fertiliser and pesticide industries and downstream industries such as transport, storage, feed processing, livestock, food processing and textile industries. This section examines how interrelated industries may be affected by GM crop adoption.

Technology providers

When a new technology enters the market, the total benefit of that technology is usually shared among the technology provider, technology users and consumers. The degree of competition in the market will determine how the total benefits of GM technologies are shared among the three groups (see box 4). In some instances GM seed supply companies hold a patent over their technology and have therefore been successful in retaining a share of the total benefits from their innovation. Farmers and consumers have also received benefits.

Herbicide and pesticide industries

The introduction of GM crops may also affect herbicide and pesticide producers. Adoption of GM crops reduces the use of herbicide and pesticide and, as a consequence, the demand for these products decreases.

The decrease in demand may put downward pressure on prices of conventional herbicides and pesticides. Bullock and Nitsi (2001) found herbicide prices in the United States fell by 50 per cent between 1996 and 1999. Part of the decline might be attributed to GM technology adoption during that period. Desquilbet and Lemarie (2002) also argued there would be substantial losses for conventional herbicide producers if HT rapeseed were to be introduced in France, as a result of lower sales and prices. In the case of a price fall, farmers who continue to produce non-GM crops will benefit.

As farmers adopt GM technology the demand for herbicides used on HT GM crops will increase, possibly leading to price increases for these herbicides. For example, Bullock and Nitsi (2001) reported the price of glyphosate increased between 1995 and 1998 as a result of the adoption of Roundup Ready® soy bean.

box 4

Empirical studies on the benefits to seed suppliers from the sale of GM seeds

Empirical studies investigating the benefits to upstream GM seed suppliers have found benefits vary. Flack-Zepeda et al. (2000) studied data from the introduction of Bt cotton in the United States from 1996–98 and found seed companies were extracting 36 per cent (US$77.4 million) of the total benefit (US$215 million). The remainder of the benefit was shared between farmers (45 per cent of total) through increased productivity and consumers (19 per cent) through lower prices. In Argentina, Trigo and Cap (2003) estimated that, from 1998 to 2003, Bt cotton technology suppliers received 83 per cent (US$35 million) of the total benefits (US$42 million).

Other studies have estimated the gain to GM soy bean seed suppliers. Traxler (2004) reported that although lower prices have led US consumers to benefit the most from GM soy bean adoption, receiving 53 per cent of the total benefits, biotechnology companies are extracting a substantial proportion (34 per cent). The case is different for GM soy bean technology suppliers in Argentina. Weak protection of intellectual property rights for GM soy bean seed has enabled competition in the GM seed market, and benefits have largely been passed on to farmers. Trigo and Cap (2003) and Traxler (2004) have estimated seed companies are receiving only 4–8 per cent of total benefits.

Downstream industry impacts

As a cost-reducing technology, GM crop adoption is likely to lead to falls in crop prices. Assuming no change in the demand structure, downstream industries and consumers will all benefit from lower prices arising from the technology.

Economy-wide models can be used to conduct economic analysis of benefits from GM technology adoption. Huang et al. (2003) assessed the potential economic benefits of Bt cotton and Bt rice production in China. They estimated a 10.9 per cent decline in cotton prices as a result of the introducing Bt cotton. This resulted in a 4.8 per cent increase in supplies to the domestic textile sector. In the case of rice, which is currently in the pre-production trial stage in China, economic modelling indicates a 12 per cent reduction in price resulting from higher yield and lower production costs.

Anderson et al. (2006) studied the global impact of Bt cotton adoption using a global general equilibrium model. The results show a fall in the world price of cotton. As a result, cotton importing countries would receive benefits for their textile industries.

Regional impacts

The impact of GM crop adoption on a regional economy will be determined by the importance of the agricultural and local processing sectors for that economy. For economies where agriculture contributes a larger share of gross regional product (GRP) the impact of GM crop adoption will be more pronounced.