Introduction

Food security exists when all people, at all times, have physical and economic access to sufficient amounts of safe and nutritious food meeting their dietary needs and preferences (World Food Summit 1996; Farré et al. 2010, 2011). Everyone should therefore have access to food which is available on a sustainable basis (FAO 2006). The nutritional quality of food is just as important as the calorific value (Pérez-Massot et al. 2013). Food insecurity therefore not only affects the 870 million hungry people in the world (FAO 2012a) but also the additional 3 billion people who achieve the minimum dietary energy requirement (MDER) but nevertheless suffer from diseases caused by inadequate nutrition (Christou and Twyman 2004; Gómez-Galera et al. 2010; Farré et al. 2010). Food insecurity is prevalent in developing countries, particularly in sub-Saharan Africa and South Asia, which account for 98 % of the world’s hungry and the largest proportion of undernourished people (FAO 2012a).

Despite isolated episodes of sudden food insecurity caused by unpredictable production deficits (FAO 2011a), global food production at present is generally sufficient to feed everyone in the world with an average 2,790 calories per person per day (FAO 2012a). Therefore today’s food insecurity has more to do with limited access to food in poverty-stricken regions than limited availability (Smith et al. 2000).

Many factors prevent access to food in developing countries, including natural disasters (caused by biotic and abiotic stress), conflict, civil strife, lack of infrastructure, land ownership disputes, unsafe water and poorly-developed health and education systems (POST 2006). However, poverty is considered the main underlying cause of chronic food insecurity in the developing world (Van Wijk 2002; Christou and Twyman 2004; Yuan et al. 2011). More than a quarter of the population in developing countries subsists on less than $US 1 per day, increasing to at least half the population in sub-Saharan Africa and in the least developed countries on each continent (UN 2011). Therefore, poverty must be tackled to address food insecurity in the long term, through increasing rural employment-based income and boosting the agricultural productivity of subsistence farmers (Christou and Twyman 2004). Currently, most global initiatives addressing global food insecurity and malnutrition embrace short- and middle-term strategies. For example, the Micronutrient Initiative (MI 2010) along with the Global Alliance for Improved Nutrition (GAIN 2012) and the Sprinkles Global Health Initiative (SGHI 2009) strive to reduce malnutrition through supplementation and food fortification programs.

The most obvious consequences of food insecurity are undernourishment and malnutrition, leading to illness, disability, impaired cognitive development and premature death (FAO 2012a). In addition, children can suffer behavioral and psychosocial problems as well as impaired learning (CHP 2002). The lack of access to food often means that the poor are unable to work, or even excluded from income-generating opportunities, thus perpetuating this status. Poverty, illness and food insecurity thus form a self-reinforcing negative cycle from which many people (sometimes entire communities) find it impossible to escape (Farré et al. 2011). On a larger scale, this translates into productivity losses that can account for 2–4 % of gross domestic product (GDP) as demonstrated for several countries in South Asia (FAO 2012a).

This review considers the economic consequences of food insecurity in developing countries by measuring direct and indirect costs and dissecting them into their main components. We also examine the cost of today’s unsustainable food insecurity solutions, as well as the potential cost benefits of more sustainable solutions, such as the development of nutritionally-enhanced genetically engineered (GE) crops.

Food insecurity in developing countries: consequences for industrialized countries

Food insecurity and poverty have led to mass immigration from developing to industrialized countries, which has resulted in social problems caused by the mismanagement of immigrant populations (de Haan and Yaqub 2009). Immigrants are often blamed for displacing the native population from the employment market and thus increasing unemployment rates, and for overburdening public services such as healthcare and education (IPC 2009; MAC 2012). Despite the inaccuracy of such claims, they are often used as propaganda by extremist parties to gain public support (van Spanje 2010).

Food insecurity and poverty also perpetuate unacceptable labor conditions in developing countries by forcing people to carry out menial work for low wages as an alternative to starvation (Meyers 2004). Large companies in industrialized countries often take advantage of this situation to save labor costs and avoid more rigorous (and costly) regulatory scrutiny (Hippert 2010). Therefore, tens of thousands of jobs have been lost through relocation, and working conditions in industrial countries have deteriorated as workers are forced to compete with a less expensive labor force in developing countries (Levine 2011; Pedersini 2006).

Most of the rural population in developing countries is made up of subsistence farmers, aiming to grow enough food to feed their families (Christou and Twyman 2004). They sometimes cultivate export-oriented crops for additional income, and in some cases these can be illicit cash crops such as opium (Afghanistan) or coca (Colombia) with local chiefs and warlords taking a cut of the profits (Díaz and Sánchez 2004; Goodhand 2005). Drug trafficking has become an extraordinary income-generating activity for many criminal groups worldwide, which benefit from demand in the industrial world while exploiting developing country farmers (UNODC 2012). Although governments in the industrialized world have tried with limited success to block the import of drugs (GCDP 2011), a more sensible approach would be to simultaneously improve living standards in developing countries by tackling poverty and ensuring food security, so farmers are less likely to turn to drug production to supplement their incomes (GCDP 2011).

Food insecurity and poverty are also perpetuated by poor governance and corruption, exacerbated by the exploitation of valuable natural resources such as oil, minerals and timber by some governments, criminal organizations and some transnational companies (TNCs; Ascher 1999). Far from benefiting the industrialized countries where TNCs are located, such arrangements are harmful both to the developing country (where land is depleted and becomes unsuitable for agriculture) and the industrialized country (because the natural resources are exploited in a non-sustainable manner; Giljum et al. 2008). Addressing the basic needs of the population would contribute to a more equitable society with the ability to control its own resources in a sustainable manner (Baland and Platteau 1996).

The economic cost of food insecurity

The current FAO estimate of 870 million hungry people in the world (FAO 2010, 2012a) is 150 million higher than 10 years ago, reflecting the consequences of two crises that were different in nature and origins but had a similar impact on food security (FAO 2011b). The first was the food price crisis that peaked in 2008, reflecting the slow increase in food prices between 2003 and 2006 followed by a surge between 2006 and 2008 before declining in the second half of that year (Mittal 2009). These increases took many by surprise, increasing concerns that the world food economy was unable to adequately feed billions of people (FAO 2011b) (Fig. 1). Although opinions varied as to the relative importance of different contributory factors, there is a strong consensus that multiple factors sparked the price increases that began in 2003 (Mittal 2009; Wiebe et al. 2011), including the slowing of agricultural production reflecting lower investment and adverse weather conditions (Zeigler and Mohanty 2010; Zhao and Running 2010), declining global grain stocks (FAO 2012a) (Fig. 2), higher energy prices which increased production costs and thus the export prices of major food commodities (Mitchell 2008), the increased food demand from emerging economies such as India and China (Mittal 2009), speculation in financial markets causing the hyperinflation of basic food staples (Mittal 2009), and the increased use of land for biofuel production. However, further investigation revealed that the record grain prices in 2008 were not caused by higher biofuel production, but were based on a speculative bubble concerning high petroleum prices, a weak US dollar, and increased volatility due to commodity index fund investments (Mueller et al. 2011) (Fig. 3).

Fig. 1
figure 1

Global monthly grains price index (blue) and total food price index (red), January 1990–November 2011. The values for 2002 to 2004 are set at 100. Source: FAO (2012b)

Full size image
Fig. 2
figure 2

World grain stocks shown as days of consumption between 1980 and 2012. This shows strong decline since 2000 reflecting policy shifts and greater dependence on trade. Source: EPI from USDA (2012a)

Full size image
Fig. 3
figure 3

Corn used for fuel ethanol in the US between 1980 and 2010 in millions of tons. The corn used for fuel is derived from the previous year’s harvest, thus the 119 million tons of corn used for fuel ethanol in 2010 represents 28.7 % of the 2009 grain crop (416 million tons). Source: EPI from USDA (2012b)

Full size image

The second crisis is the ongoing severe global financial and economic decline which in 2011 saw international food prices rise to levels not seen in decades, coinciding with the global population reaching 7 billion people. The export-weighted FAO food price index climbed to a record 238 points in February 2011 (Fig. 1), and the price of wheat and corn doubled reflecting the drought and subsequent wheat export ban in the Russian Federation and the poor yield of corn in United States as well as the weakening dollar (FAO 2011b). The global food system is becoming more vulnerable to episodes of high prices and volatility (The Economist 2011). Specialization in a few export commodities such as coffee or cocoa has increased the dependence of developing countries on food imports converting them from net exporters, with an overall agricultural surplus of US$7 billion in the 1960s, to predominantly net importers in the 1990s and 2000s, with a deficit of US$11 billion in 2001 (FAO 2004; Action Aid International 2008).

In light of these crises, earlier progress towards food security based on the 1996 World Food Summit goal (reducing the number of hungry people in the world) and Millennium Development Goal 1 (halving the hungry population by 2015) has been reversed (Wiebe et al. 2011; Yuan et al. 2011). Multilateral investments in developing country agriculture by industrialized governments and global institutions such as the World Bank have steadily declined (Jomo 2008). USAID, the United States International Development Agency, has cut agricultural aid by 75 % over the last 20 years (USAID 2004). Only 4 % of current development aid to Africa is spent on agriculture, and agricultural research grants were cut by more than 80 % between 1980 and 2006, with the United States alone reducing its contribution from US$ 2.3 billion to US$ 624 million (Jomo 2008).

In many parts of the world, agricultural growth is needed to address the current world food crisis by contributing to overall economic growth and helping to achieve MDG1 (Yuan et al. 2011). There have been numerous attempts to estimate the cost of achieving MDG1, mostly at the global or regional levels, including the United Nations Zedillo Report, studies by the World Bank and the United Nations Development Program, and the International Food Policy Research Institute (IFPRI). These estimates have varied widely, mostly because of different methodologies, assumptions, coverage, measures and interpretations. The Zedillo report contains some rough estimates of the additional aid required to achieve the MDGs, with US$ 20 billion of the US$ 50 billion total required to halve poverty and hunger (UN 2001). Using two different approaches, the World Bank estimated that the additional foreign aid required to achieve MDG1 by 2015 is US$ 40–60 billion per year (Devarajan et al. 2002). IFPRI estimates that a total global annual investment in agriculture of US$ 14.3 billion per year is necessary, although under a high-investment scenario, these requirements would double to US$ 28.5 billion per year (Fan and Rosegrant 2008).

Direct costs of the causes and consequences of food insecurity

The direct costs of food insecurity have been estimated by the US Department of Agriculture (USDA) based on food assistance and nutrient supplementation programs. Such programs promote food security in developing countries by providing food aid to save lives and help low-income families (World Bank 2012). Additional direct costs reflecting the burden on healthcare systems dealing with hunger and malnutrition are included, although these are expressed as disability adjusted life years (DALYs) and are considered indirect costs in the discussion below.

Food assistance and nutrient supplementation programs

Malnutrition has been addressed directly by supplementation (short-term micronutrient delivery), industrial fortification, biofortication, dietary diversification and the support of public health measures (Stein et al. 2006). The United States has led international efforts to combat malnutrition and hunger for more than 60 years. Through food aid and assistance programs, the USDA provides support to the agricultural development sector, as well as food security and humanitarian help following natural or manmade disasters in developing countries, with average annual donations of $US 2.2 billion. In 2012, USDA food assistance benefited more than 9.7 million people through the Food for Progress, Food for Peace, Local and Regional Procurement Pilot Project, and McGovern-Dole International Food for Education and Child Nutrition Program initiatives (Table 1). The major participants are nonprofit charitable organizations, governments, intergovernmental organizations and academic institutions (Ho and Hanrahan 2010).

Table 1 The social and economic benefits of the Food for Progress Program and the McGovern-Dole International Food for Education and Child Nutrition Program in 2012
Full size table

Food for Peace (also known as Public Law 480) encompasses four sections: Trade and Economic Development Assistance; Emergency and Development Assistance; Food for Development; and Farmer-to-Farmer Assistance. This program aims to improve global food security and nutrition, promote agricultural development, expand international trade and foster private enterprise. Food for Progress donates US commodities to developing countries in order to initiate and expand free enterprise in the agricultural sector. The objective is to improve agricultural productivity and commercialization by training farmers, developing agricultural infrastructure and improving farming methods (e.g. irrigation systems, cooperatives and microcredit support). The Local Regional Food Aid Procurement Pilot Project provides food aid following price fluctuations and thus promotes food security in developing countries. The McGovern-Dole International Food for Education and Child Nutrition Program contributes to educational programs, as well as maternal and child nutrition in developing countries, by providing technical and financial assistance and agricultural commodities such as dairy, cotton, fruits/vegetables, poultry and livestock. This promotes primary school attendance, maternal health during pregnancy and breastfeeding, and children’s health and hygiene at school. Under this program, the Micronutrient-Fortified Food Aid Products Pilot (MFFAPP) explores the potential to fight micronutrient deficiencies through the distribution of micronutrient-fortified food aid (Ho and Hanrahan 2010). USDA supports the development of micronutrient-fortified foods by investing more than $US 8.5 million in nutrient-enhanced food to address micronutrient deficiencies in women and children. Furthermore, private companies such as Heinz offer support for the distribution of micronutrient powders to children in developing countries, e.g. Heinz has provided $US 5 million thus far to support the Micronutrient Campaign.

Burden on healthcare systems dealing with hunger and malnutrition

The global impact of hunger and malnutrition on healthcare systems includes the costs of mortality, morbidity and disability, but also longer-term consequences on physical and mental health (Black et al. 2008). Cost-effectiveness analysis (CEA) has been used to investigate the efficiency of healthcare resources by comparing the relative costs and health gains of different interventions. Current data are predominantly derived from high-income countries although the results can be extrapolated to developing countries (Hutubessy et al. 2003). For example, the additional costs of managing malnutrition in Dutch nursing homes are up to US$ 366 million per year (US$ 10,494 per patient at risk of malnutrition and US$ 13,117 per malnourished patient) based on the extra costs of nutritional screening, monitoring and treatment (Meijers et al. 2011). In Brazilian hospitals, malnourishment results in an average daily cost of US$ 228.00/patient, compared to US$ 138.00/patient for well-nourished individuals, an increase of 60.5 % (Correia and Waitzberg 2003). But the actual costs are even higher because malnutrition increases the length of hospital stays by an average of 43 % (Pirlich et al. 2006). More recent data from the UK suggest that the healthcare costs of malnourished patients over a 6-month timeframe (US$ 2,829) are more than twice those of well-nourished patients (US$ 1,210) (Guest et al. 2012). Nutrient supplements in hospital can result in substantial savings because well-nourished patients recover better and faster and have fewer complications (Russell 2007) (Table 2).

Table 2 The economic benefits of commercial oral nutritional supplements for hospital patients (adapted from Russell 2007)
Full size table

Indirect costs

Among the 26 major global health burdens, iron deficiency anemia, zinc deficiency and vitamin A deficiency (VAD) rank 9, 11 and 13, respectively (Fig. 4). This means that 30 % of the global population suffers from one or more of these diseases (WHO 2009). Furthermore, large numbers of people also suffer from diseases caused by a lack of selenium, folate, calcium and iodine (Stein et al. 2007; Stein 2010).

Fig. 4
figure 4

The top 20 major disease burdens in the developing world. The x-axis represents the attributable DALYs (% of global DALY). Source: Stein (2010)

Full size image

Women and children are the most vulnerable groups because pregnancy, breast-feeding and menstruation, as well as rapid body growth in children, increase micronutrient requirements and make it even more difficult to achieve adequate intakes (Benoist et al. 2008). More than one third of child deaths are attributed to malnutrition. If a child is undernourished or malnourished during pregnancy and/or the first 2 years of life, this affects physical and mental health for life. For example, children suffering from iron deficiency show poor cognitive development, psychomotor development and socio-emotional activity (Lozoff et al. 2006, Beard 2008). Iron deficiency anemia also affects reproductive performance and increases the risk of death during pregnancy (Hunt 2002). Malnutrition directly affects school enrolment and class performance, reducing the likelihood of a complete education (Khanam et al. 2011; Liu and Raine 2006). Micronutrient malnutrition also reduces the aggregate productivity and economic development of communities and countries (World Bank 1994; Qaim et al. 2007).

The accurate measurement of losses caused by micronutrient deficiency is difficult but the usual approach is to calculate DALYs representing the sum of years of life lost (YLL) plus the years lived with disability (YLD). One DALY is equivalent to 1 year of healthy life lost. Although human life cannot be measured in monetary terms, in this case a value for human illness and loss of life can be calculated based on the annual average per capita income of people in a particular country (Stein 2010). The WHO has reported that 28 million DALYs were lost due to zinc deficiency in 2002 and 35.1 million DALYs were lost to iron deficiency anemia. All mineral deficiencies combined result in 65.6 million DALYs lost and this figure increased to 92 million DALYs when vitamin A deficiency was included (WHO 2002, 2004). In an average year India loses up to 4 million DALYs to iron deficiency anemia, and 2.8 million to zinc deficiency (Stein 2010). In 1994, the World Bank reported that DALY losses due to protein energy malnutrition, vitamin A, iodine and iron deficiency was 5 % of GDP in Sub-saharan Africa, China, India, Latin American and Middle Eastern countries (Stein 2010), which can be easily overcome by investing 0.3 % of the GDP into malnutrition alleviation programs in these countries (Adamson 2004). In South Asia, where iron deficiency anemia is most prevalent, annual economic losses of US$ 5 billion are estimated, with productivity reduced by 1.5 % for every 1 % loss of hemoglobin (Dickinson et al. 2009).

The cost of unsustainable solutions (10-year survey)

The most effective solution to food insecurity is a varied diet including fresh fruits, vegetables, fish and meat. This is impractical in many countries resulting in persistent malnutrition at the population level (Gómez-Galera et al. 2010). To deal with the widespread iron, zinc and vitamin A deficiency in developing countries, nutritional planners have developed three solutions: short-term supplementation, mid-term fortification and long-term dietary modification. The short-term solution involves the provision of high-dose micronutrient capsules (e.g. 200,000 IU of vitamin A to all young children at 6-month intervals) and is the most widely implemented but also the least sustainable intervention (Greiner 2012). The mid-term fortification of staples or condiments and the long-term modification of diets for vulnerable groups are more sustainable but also more difficult to establish in developing countries.

The total costs of short-term micronutrient intervention have been estimated at US$ 1.5 billion per year, followed by an additional US$ 2.9 billion over 10 years to implement behavioral interventions and a further US$ 1.0 billion to establish more complex and targeted programs so that nutrition can be improved on a sustainable basis. Another US$ 100 million would be needed to monitor and evaluate these large-scale programs, conduct follow-up research and provide technical support (Horton et al. 2010). The initial US$ 1.5 billion annual investment would provide therapeutic supplements of iron, zinc and vitamin A, as well as universal salt iodization and iron-folate fortification during pregnancy, but even this intervention is not yet included in the WHO recommendations (Table 3) (Horton et al. 2010). These costs are estimated for the 36 countries identified in the 2008 Lancet series on maternal and child malnutrition, which are home to 90 % of moderately or severely stunted children worldwide (Horton et al. 2010). Additional costs for scaling up these interventions to include 32 smaller countries where 20 % or more of all children under the age of five are stunted or underweight (mainly in Sub-Saharan Africa) were estimated, showing that this expansion of coverage would increase the target population by 6 % and increase overall costs by a comparable amount (Horton et al. 2010) (Table 4).

Table 3 Estimated investment costs and annual operational costs for micronutrient interventions (Horton et al. 2010)
Full size table
Table 4 The 36 countries with 90 % of the global burden of stunting and an additional 32 high-burden countries with underweight or stunting rates greater than 20 % (Horton et al. 2010)
Full size table

There are at least four potential sources of funding for such interventions: private-sector households, private-sector corporations, public/government funding and innovative financing models such as the High Level Taskforce on Innovative Financing for Health Systems (Horton et al. 2010). Households are accustomed to bearing most of the recurrent costs of market-based strategies, such as salt iodization (estimated at US$ 400 million) and fortification (Horton et al. 2010). Developing country governments already contribute substantial amounts to nutritional programs, e.g. India allocated nearly US$ 1.3 billion for its Integrated Children Development Service Program in 2007–2008 (Horton et al. 2010).

Private charitable foundations and philanthropists such as the Bill and Melinda Gates Foundation and the Children’s Investment Fund are also emerging as a significant financing source for nutritional programs (Horton et al. 2010). Unlike the US donor agencies, which must demonstrate the worth and impact of their work to taxpayers, European donor agencies tend to work quietly, knowing they have the support of most of the public, although this makes the process less transparent (Greiner 2012).

Cost savings with nutritionally-enhanced crops

Vitamin A deficiency poses a significant public health challenge in 118 countries, especially in Africa and Southeast Asia (Van Wijk 2002). Genetic engineering is an emerging strategy for the control of VAD in the developing world, typically the development of β-carotene enriched crops, such as canola (Brassica napus) (Shewmaker et al. 1999) and mustard (Brassica juncea) (Agricultural Biotechnology Support Project 2003). Golden Rice (Ye et al. 2000; Paine et al. 2005), Multivitamin Corn (Naqvi et al. 2009), and high lutein/zeaxanthin corn (Naqvi et al. 2011) are further examples of crops that have been engineered to increase their content of β-carotene and other nutritionally important carotenoids.

Several recent studies have examined the potential economic impact of Golden Rice and Golden Mustard on VAD by calculating the avertable health burden in terms of DALYs in the Philippines and India (Zimmerman and Qaim 2004; Stein et al. 2006; Chow et al. 2010). The widely-used DALY metric allows comparisons of alternative health strategies using a single index that combines information about mortality and morbidity (Zimmerman and Qaim 2004; Stein et al. 2006; Chow et al. 2010).

Golden Rice

There are no conventional rice varieties with enough β-carotene in the grain to use in conventional breeding strategies to increase β-carotene levels and combat VAD. Golden Rice has been engineered to produce β-carotene (pro-vitamin A) in the grain endosperm, so that polished rice grains can be used to prevent VAD in the developing world. Golden Rice grains contain up to 23-fold more total carotenoids than conventional rice (37 μg/g dry weight) including β-carotene levels of up to 31 μg/g dry weight (Paine et al. 2005). An ex ante analysis of Golden Rice in India was carried out considering the entire sequence of events between cultivation and consumption to calculate its overall impact on health (Stein et al. 2006). In a high-impact scenario, India’s annual burden of VAD (2.3 million DALYs lost) could be reduced by up to 59.4 % by the consumption of Golden Rice, saving 1.4 million healthy life years. In a low-impact scenario, where Golden Rice is consumed less frequently and delivers less β-carotene, the burden of VAD would be reduced by 8.8 %. However, in both scenarios, the cost per DALY saved by using Golden Rice (US$ 3.06–19.40) is lower than the cost of supplementation, and Golden Rice outperforms international cost-effectiveness thresholds. The total annual cost of vitamin A supplementation is approximately $US 21 million, and the total annual cost of a flour fortification program has been estimated at $US 4–6 million (Fiedler et al. 2000). Golden Rice should therefore be introduced immediately as a complementary intervention to fight VAD in rice-eating populations (Stein et al. 2006).

Another ex ante analysis of Golden Rice was carried out in the Philippines, suggesting a gain of 15,000–85,000 DALYs per annum and a reduction in the health burden of 5.7–31.5 %, with the lower figures representing a pessimistic scenario and the higher figures representing an optimistic one (Zimmerman and Qaim 2004). The greatest overall benefits were predicted in children, reflecting the lower initial prevalence of corneal xerophthalmia among children in the Philippines. Golden Rice was predicted to avert 798 child deaths per year in the optimistic scenario (Zimmerman and Qaim 2004).

Bioavailability studies demonstrated that the total amount of β-carotene in Golden Rice (35 μg/g dry weight) was the same before and after cooking, i.e. boiling for 30 min (Tang et al. 2009). Therefore, eating 348 g of Golden Rice per day would achieve the dietary reference intake for vitamin A.

Golden Mustard

Golden Mustard has been engineered to accumulate up to 600 μg/g β-carotene (Agricultural Biotechnology Support Project 2003) and is particularly suitable for deployment in India, which consumes large amounts of mustard oil (Chow et al. 2010). India has the greatest number of clinical VAD cases in the world (more than 35 million) and the greatest percentage of subclinical VAD in children under six (31–57 % of the population) (West 2002). A cost analysis of Golden Mustard was carried out based on a conservative efficacy rate of 4 % and an optimistic efficacy rate of 23 % averted mortality. The number of DALYs averted over a 20-year time frame was estimated at 18–34 million, and the number of lives saved was 113,000–654,000, with the lower figures representing the conservative scenario and the higher figures representing the optimistic one (Chow et al. 2010). Golden Mustard was also estimated to avert 5–6 million more DALYs and 8,000–46,000 more deaths than supplementation, mainly because it would benefit the entire population and not only children and women (Chow et al. 2010).

The amount of β-carotene in mustard oil derived from Golden Mustard containing 600 μg/g β-carotene has been estimated at 185 μg/g (Chow et al. 2010), although only 71 % remained after baking, seasoning, deep-frying and shallow frying (Manorama and Rukmini 1991). The mustard variety used to produce Golden Mustard accounts for 70–80 % of the mustard grown in India, therefore only 75 % of the mustard seed pressed into oil would be the fortified variety. Based on these assumptions, the effective concentration of β-carotene would be 49.3 μg/g of consumed oil and the fixed costs would amount to approximately US$ 0.01 per person. In conclusion, only a few drops of the fortified oil would satisfy the dietary reference intake for vitamin A.

The current cost of technology

GE crops offer a number of potential solutions to tackle food insecurity in developing countries but the adoption rate for such crops is low at present (Qaim 2009; Ramasamy et al. 2007). This is often because developing countries have a limited capacity to carry out research and development (R&D), coupled with the high regulatory burden of GE technology, market barriers and the inadequate protection of intellectual property (Cohen 2004; Pray et al. 2005; Ramessar et al. 2009; Paarlberg 2001). In this context, it is important to establish a cost-effective approach for the development of nutritionally-enhanced GE crops. Currently, GE technology requires substantial upfront R&D investment plus additional funding to overcome the immense regulatory burden (Ramessar et al. 2010; Twyman et al. 2009; Ramessar et al. 2007). Even when a GE crop has been authorized for cultivation, a breeding program is required to commercialize the novel trait into locally adapted varieties. Additional marketing costs are necessary to promote the public acceptance of GE crops.

Following the lead of Golden Rice (Ye et al. 2000), several other nutritionally-enhanced crops have been developed, including Multivitamin Corn which accumulates β–carotene, ascorbate and folate (Naqvi et al. 2009). However, none of these enhanced varieties have been commercialized. In the case of Golden Rice, the R&D costs (involving international projects) reached US$ 3 million (Zimmerman and Qaim 2004) but this may increase to $7.5 million for other crops depending on the circumstances (Stein et al. 2006).

GE crops must go through a risk assessment procedure where they are evaluated in laboratory tests and field trials, and must undergo safety analysis (Gómez-Galera et al. 2012; Arjó et al. 2012). The costs of regulatory compliance include the direct costs of testing (to provide information for the regulators) and also the costs of the administrative structure to ensure compliance (Pray et al. 2005). The necessary tests include molecular characterization, compositional assessment and 90-day rat toxicity assays which can cost $US 4.2–7.7 million (Kalaitzandonakes et al. 2007). Agronomic, phenotypic, environmental and allergenicity testing may also be required (EFSA 2010). After field trials, regional breeding programs must be carried out to introduce the desirable characteristics into high-yielding varieties and/or hybrids grown in those areas, resulting in additional costs. For example, the cost of the first breeding program for Golden Rice in India was approximately $US 1 million (Stein et al. 2006).

The costs associated with regulatory compliance represent a significant portion of the total costs of bringing a GE product to market, erecting a significant barrier to adoption particularly in developing countries (Jaffe 2006; Kalaitzandonakes et al. 2007; Pray et al. 2005). However, assuming that GE regulatory mechanisms for licensing such products are already in place and the R&D program is complete, one-time fixed costs for the adoption of GE crops in India have been estimated at $US 5.6 million for Golden Mustard (Chow et al. 2010) and at least $US 2 million for Golden Rice (Stein et al. 2006).

Once GE crops are approved for cultivation, promotion campaigns and social marketing is necessary to ensure consumer acceptance. In this context, marketing costs for Golden Rice could exceed $US 15 million in India (Stein et al. 2006) although these costs may be lower for second-generation, quality-enhanced crops combining agronomic and quality traits (Qaim 2009). The new phenotype of GE crops (such as the different color of rice or corn seeds containing high levels of β-carotene) can also affect public acceptance. However, a study in Mozambique looking at the consumer acceptance of an orange corn variety with high levels of β-carotene showed that existing preferences for white corn do not prevent the acceptance of orange biofortified corn, and that the colored kernels may act as a self-targeting nutritional intervention (Stevens and Winter-Nelson 2008). Taking previous data from Golden Rice together with the acceptance of orange corn, it is expected that promotional campaigns and social marketing for Multivitamin Corn will cost less than $US 15 million. Altogether, these data suggest that the total technology costs for GE crops include $US 3–7.7 million (32 %) for R&D, $US 1 million (4 %) for breeding, $US 2–5.6 million (23 %) for regulatory compliance and up to $US 15 million (41 %) for marketing, making a total of US$ 20–29 million (Fig. 5).

Fig. 5
figure 5

The proportion of R&D, breeding, regulation and marketing costs required for the development of a genetically engineered crop (adapted from Stein et al. 2006)

Full size image

In addition to technology costs, the adoption and implementation of a regulatory framework for GE crops in developing countries may have a direct impact on the economy, particularly on current and future agricultural exports to countries with stringent regulations, such as those in the EU (Zarrilli 2005; Sabalza et al. 2011). For example, the approval of Bt eggplant (Chong 2005) led to the exclusion of India from international markets and blocked financial support for biotechnology research from EU countries (Ramessar et al. 2008, 2009). The Indian Government therefore banned cultivation of this eggplant variety due to political pressure rather than scientific evidence.

Intellectual property (IP) issues also need to be taken into consideration. Even in the case of the more recent biotechnology patents, IP is built at least in part on earlier innovations which have been protected through patents, including vectors, selectable markers, transcription regulatory elements, sub-cellular targeting systems, etc. (Krattiger and Mahoney 2007). Consequently, IP issues influence decision making in transgenic crop development (Sechley and Schroeder 2002). Humanitarian applications such as Golden Rice (GR), was constrained by up to 70 patents. Thus the creation of GR required the use of many technologies that were patented by different companies, including Bayer AG, Monsanto, Novartis AG, Orynova BV and Zeneca Mogen BV (Krattiger and Potrykus 2007). Following discussions and negotiations with IP holders, the developers of GR were able to secure free access to all constraining IP gratis for defined humanitarian research purposes only, allowing the use of GR in developing countries by resource-poor farmers (Krattiger and Potrykus 2007). Although GR demonstrated that IP and Freedom to Operate (FTO) issues can be resolved, biotechnology companies are generally uneasy about such humanitarian ventures because most developing countries lack enforceable IP policies that would ensure that their IP rights are protected adequately (Wendt and Izquierdo 2001). In order to establish and maintain international technology transfer agreements, industrialized and developing countries need to cooperate in the development of a manageable system for IP protection. This can be achieved in a number of ways, e.g. local ownership and the involvement of scientists from developing countries or by the implementation of appropriate IP policies and effective enforcement procedures in developing countries (Kowalski 2002). Several organizations have been set up to promote biotechnology in developing countries, particularly in the area of subsistence agriculture, focusing on IP access and management They offer advice on a complete set of biotechnology technologies under IP by third parties and a road map for FTO. Some examples are CAMBIA (Center for the Application of Molecular Biology to International Agriculture), PIPRA (Public Intellectual Property Resource for Agriculture), AATF (African Agricultural Technology Foundation) and SIPPI (Science and Intellectual Property in the Public Interest).

Conclusions and recommendations

Numerous studies have shown the social and economic benefits of GE technology, including its ability to address global food insecurity. GE crops can tackle food insecurity in a number of ways. First-generation crops, with modified input traits, can address food insecurity by increasing the yields of food crops grown by subsistence farmers to avoid hunger, and can increase the profit margins of smallholder farmers growing cash crops e.g. by reducing labor and pesticides, thereby reducing poverty and empowering a greater proportion of the population (Qaim 2010; Sanahuja et al. 2011). For example, the major benefits of the herbicide tolerant Roundup Ready soybean in Bolivia include a 30 % increase in yield, a 22 % savings in labor and other variable costs (ISAAA 2012). Herbicide tolerant soybean adopters in general cultivate larger areas, are more educated and are more likely to own their farm and farm machinery (Smale et al. 2012). Second-generation crops, with modified output traits, can address food insecurity directly by increasing the nutritional value of food, e.g. the examples of Golden Rice, Golden Mustard and Multivitamin Corn, all of which have higher levels of key vitamins compared to conventional varieties (Naqvi et al. 2009).

A number of studies have considered the relative costs and benefits of GE technology in terms of the overall development costs compared to conventional intervention strategies and the benefits on the ground to farmers. These studies have clearly demonstrated the benefits of first-generation crops to farmers in many developing countries and the consequential positive effects on national GDP and GNP values thus increasing the ability of governments to invest in infrastructure and improve the health and well-being of their populations. For example, the adoption of Bt cotton in India has generated a profit of US$ 51 billion during the period 2002–2008 (Devasahayam et al. 2011) and herbicide tolerant soybean in Bolivia provided a net return of US$196 per hectare, resulting in US$175 million benefits at the national level (ISAAA 2012).

Furthermore, we have reviewed several cost–benefit studies relating to nutritionally enhanced GE crops. For example, the expected cost of developing Golden Rice, including R&D costs, was estimated to be $US 10.7 million, with a continuing cost of $US 0.5 million per year (Zimmerman and Qaim 2004). Balanced against this, Stein et al. (2006) showed that introducing Golden Rice 2 which accumulates substantially more β-carotene in the polished grain than the original variety, could recover between 204,000 and 1.4 million DALYs per annum in India at a cost of only $US 21.4–27.9 million over 30 years, which is an average annual cost of only $US 713,000–931,000. Providing vitamin A supplements was an inexpensive intervention, at $US 23–50 per DALY and US$ 1000–6000 per death averted, although this would target the most vulnerable groups such as children and pregnant women (Chow et al. 2010). GE crops could avert 5–6 million more DALYs and 8,000–46,000 more deaths by covering the entire population rather than just the most vulnerable groups. Although the cost of GE crops was up to five times higher than their non-engineered counterparts, this predominantly reflected the one-off cost of regulatory compliance and approval ($US 5.6 million) which could be reduced if the regulatory burden on GE technology was lowered (Chow et al. 2010). The deployment of Golden Rice and other nutritionally-enhanced crops should also be considered in a geographical context – for example, rice is a staple in Asia but not in Sub-Saharan Africa, where nutritionally-enhanced corn varieties would be preferable (Zhu et al. 2008; Naqvi et al. 2009).

The global costs of food security need to be considered when developing efficient strategies to address hunger and malnutrition. It would cost $US 130 million to provide supplements for the 17.3 million acutely-affected children (6–59 months of age) in the 36 countries with the highest burden of malnutrition (Horton et al. 2010) but it is important to consider that these are recurrent costs, necessary to address the symptoms but not the causes of malnutrition. In contrast, the costs of developing GE crops have to be borne upfront in order to overcome technological and regulatory barriers as well as sociopolitical factors. However, once crops have been cultivated they are largely sustainable without further investment and the running costs are mainly associated with distribution and formulation to ensure adequate doses. Rice is a staple in South Asia, East Asia and the Pacific so nearly 2 billion people could benefit from Golden Rice, whereas corn is a staple in sub-Saharan Africa, South America and the Caribbean, where an additional 700 million people could benefit from Multivitamin Corn. It is also clear that GE crops enhanced with multiple vitamins and minerals are desperately needed because this would allow a single crop to treat multiple deficiency diseases. Stacking nutritionally-enhanced GE crops with agronomic traits will be the next logical step in addressing food insecurity in developing countries in a more meaningful way. This will assure that subsistent farmers will maximize the benefits of the new products and technologies in the most optimal way.