that over the long-term, the growth of output, employment and productivity progress in the same direction (ILO, 2005). However, social costs in the short-term can be high
.
4.4.6.3 Gender perspectives in agricultural labor
As agriculture and food systems evolve over the next dec­ades, gender issues and concerns are highly likely to con­tinue to be central to AKST development, at least in the de­veloping countries where women have played a significant role in traditional agricultural production. Over the years improvements  in  agricultural  technologies  have  seldom been targeted as recipients of improved technologies. Yet there are more women working in agriculture than men, e.g., women in rural Africa produce, process and store up to 80% of foodstuffs, while in South and South East Asia they undertake 60% of cultivation work and other food produc­tion (UNIFEM, 2000).
     Employment of female vis-à-vis male workers is likely to decline in the future as women obtain more employment in other sectors. Historically, there has been a global decline in the world over the last 10 years (47% in 1994 to 43% in 2004). While this may imply further declines in the future, it may also be true that female employment in agriculture will increase as a result of changing production patterns.
     The increasing participation of women in subsistence production in agriculture is highly likely to continue facili­tating male out-migration to urban areas and to other sec­tors such as mining and commercial farming (at lower costs to society than would otherwise be possible). The greater number of men moving out of the agricultural sector is highly likely to continue. A slightly increasing feminization of the agricultural labor force in most developing countries may reflect the fact that women are entering into high value production and processing and thus less likely to abandon their agricultural ventures (Mehra and Gammage, 1999).

4.5 Existing Assessments of Future Food
Systems, Agricultural Products and Services

4.5.1 Assessments relevant for changes in food systems
Existing assessments provide information how agricultural and food systems might change in response to the changes in the direct and indirect drivers discussed in the previous subchapters (note that the outcomes of these assessments may be compared to the reference scenario presented in Chapter 5). Over the past 50 years, there have been at least 30 quantitative projections of global food prospects (sup­ply and demand balances). We have reviewed several recent global assessments (see 4.2) that provide information rel­evant for future agriculture and food systems, either directly (i.e., assessments with an agricultural focus) or indirectly (other assessments that include agriculture). Important or­ganizations that provide specific agricultural outlooks at the global scale include the Food and Agriculture Organi­zation of the United Nations (FAO), the Food and Agri­culture Policy Research Institute (FAPRI), some of the re­search centers of the Consultative Group on International Agricultural Research (CGIAR) such as the International Food Policy Research Institute (IFPRI), the OECD, and the

United States Department of Agriculture (USDA). Other food projection exercises focus on particular regions, such as the European Union. Finally, many individual analyses and projections are implemented at the national level by agri­culture departments and national level agricultural research institutions.
     In subchapter 4.2, we introduced a selection of global assessments and discussed their objectives and use of sce­narios. None of these—IPCC's Assessment Reports (IPCC 2001, 2007abc), UNEP's Global Environment Outlooks (UNEP 2002, 2007; RIVM/UNEP, 2004), the Millennium Ecosystem Assessment (MA, 2005a), IFPRI's World Food Outlook (Rosegrant et al., 2001), FAO's World Agriculture AT 2015/2030 (Bruinsma, 2003) and IWMI's Comprehen­sive Assessment of Water Management for Agriculture (CA, 2007)—address the full spectrum of the food system and AKST from the perspective of a range of different plausible futures (Table 4-14). This is not surprising given the different objectives of these assessments (see 4.2), but it does imply that an assessment that meets development objectives can­not be met solely through analyzing earlier assessments. Al­though the projections provided by FAO and IFPRI address agricultural production and services to some degree, the attention paid to AKST elements is relatively limited. This highlights a need for new work to integrate plausible futures with regard to the interactions between driving forces and food systems while addressing AKST in more detail. Analy­sis of recent scenarios exercises indicates that while some elements related to the future of food systems are touched upon, the focus of these exercises is more on production and consumption than on the distribution component of food systems. Most studies addressed qualitative and quan­titative production indicators, and provide assumptions on yields for various crops, area under certain crops, input use or exchange mechanisms. Consumption as well as access to food (including affordability, allocation and preference), has often been addressed through modeling food demand in different scenarios. For example, assumptions regarding allocation of food through markets are made indirectly un­der different scenarios by assuming whether and how well markets and governance systems function. Food preferences are usually covered in a more qualitative manner through assumptions made about changes due to various cultural and economic factors (Zurek 2006).      The area least covered by the reviewed scenario exer­cises is food utilization (Zurek 2006). The IFPRI and MA exercises calculated the number of malnourished children under each scenario (a very basic indicator of hunger and whether nutritional standards are met), but nutritional out­comes under different diets and their possible changes are seldom addressed. Little, if anything, is said in any of the exercises concerning food safety issues or the social value of food, both of which can have important consequences for food preferences. The MA does quantitatively assess certain health indicators; these could be used to give a further in­dication on human nutritional status in different scenarios. Further in-depth research is needed on some of the specific food systems variables and their changes in the future, spe­cifically for those related to food utilization, as well as a number related to food accessibility.