Over the last decade agricultural and rural populations in the developing world have become more extensively and directly affected by several new processes in a rapidly changing global context. Globalization and liberalization of trade have brought about a greater integration of rural economies in both national and world markets. Rural populations are being confronted with more dynamic and, therefore, less predictable market-dominated conditions of production. The responses of rural populations to uncertainties and risks are also becoming less predictable but increasingly more important as an economic factor. In keeping with the rapid transformation occurring in global markets, markets for agricultural labour are likewise experiencing changes and becoming more dynamic. Because of production constraints, many farmers are now migrating to the cities to work in an attempt to maximize their access to scarce resources and improve the lives of their families. For those who stay behind, access to cash and a wage income is becoming as critical as access to land (ECOSOC, 1999a).
Demographic projections by the United Nations (UN), supported by findings of the International Food Policy Research Institute (IFPRI), suggest that, over the next two decades, rural populations will continue to exceed urban populations in the developing countries. Moreover, food needs in the developing world could double over the next three decades (World Bank, 1999a). Since economic growth in the developing countries is correlated with agricultural growth (FAO, 1998), the pressure on agriculture to increase its output, and hence its productivity, will be enormous in the years to come. Production on existing land will almost have to double in order to provide the required food supply sustainably (World Bank, 1999a).
Contributory factors in the intensification and increase of agricultural productivity are thus likely to be research and the organization of research-supporting databases; new technology; the development of rural human resources, including the agricultural labour force (to be achieved through the targeting of small- and medium-scale farming operations); and the use of public environments as a forum for dialogue on policy issues. Clearly, all these developments will destabilize the existing division of labour in rural areas and will have important consequences for gender relations at the subnational and national levels.
Recent studies have shown that more varied employment becomes available as workers move out of agriculture and subsistence production and into paid employment in the expanding manufacturing and service sectors (Mehra and Gammage, 1999). However, this trend does not necessarily reflect healthy growth in the agricultural sector. Sooner or later, planners will be faced with the question of how best to invest in and maximize the productivity of the remaining human resources.
There seem to be important gender deviations in the general trend of labour moving out of agriculture. A slightly increasing feminization of the agricultural labour force in most developing countries may reflect the fact that women are lagging behind men and abandoning agriculture at a slower rate (Mehra and Gammage, 1999). Furthermore, women tend to work in low-productivity jobs more often than men, especially those who remain in the agricultural sector.
Viewing workers as an important resource for agricultural development is economically advantageous. Efficiency losses related to sex segregation in agricultural activities may result in a reduction of total output as a consequence of labour force misallocation and the underutilization of existing gender-specific skills (such as women's traditional expertise in plant genetic resource management), or even loss of skills. Experts in estimating labour force efficiency have suggested that reducing labour force segregation by sex is not only an issue of redistributive justice, but can also have beneficial effects for both men and women in terms of welfare improvement (see Tzannatos, 1999, who claims that "the `size of the pie' increases with women claiming a bigger share"). From this perspective, policy interventions to improve women's productivity may involve more investment in women as agricultural producers and a reorganization of the gender-based division of labour.
In the developing world as a whole, agriculture accounted for about 63 percent of total female employment in 1997 (see Table 4, p. 14) and is still the most important sector for female employment in sub-Saharan Africa and Asia. At the same time, the share of women's work in the informal sector, both urban and rural, remains high. Informal occupations provide the livelihood (paid or otherwise) of more than 80 percent of women in low-income countries and 40 percent of those in middle-income countries. These countries together account for 85 percent of the world's population (Chen, 1999; Mehra and Gammage, 1999). However, this work remains largely unrecognized because the informal sector (both within and outside agriculture) is often considered to be a "residual" category that is supposedly short-lived and does not contribute substantially to national economies in terms of output. Research in recent years has suggested that women's labour force in agriculture is often substantially underestimated. For example, in 1994 the International Labour Organisation (ILO) reported that the difference between the employment rates registered through a narrow versus a broad definition of "work" in India amounted to 75 percent of the female labour force. The 1981 census in the Dominican Republic estimated a rural female labour rate of 21 percent, a figure corrected up to 84 percent three years later in a special survey that included activities such as gardening and animal care in the definition of "work" (Tzannatos, 1999).
With globalization, rural planners face the challenge of how to account for (and rely on) the indirect effects of social factors on economic processes. For example, how can low-income countries afford to consider cash crop growth (involving expensive infrastructure) as a developmental priority if large rural populations remain constrained by low labour productivity in subsistence farming and thus are candidates for urban (un)employment in the near future? Indeed, they (primarily women) cultivate the food crops necessary for food security but, with no additional income, they cannot participate directly in development or derive benefits from it.
How can the two development processes - industrial cash crop expansion and subsistence farming - be combined without hindering one another? This may require the targeting of subsistence farming populations with specific programmes for sustainability and growth (such as microenterprise development) or learning how to assess the hidden work involved in coping strategies and the subsequent returns that are used informally by farmers.
As far as women are concerned, structural adjustment programmes have brought additional "pull-back" forces. A recent report by the Secretary-General of the United Nations on the advancement of rural women stresses that the globalization and liberalization of trade may bring more difficulties than returns for rural women because of the reduction of international aid and, correspondingly, diminishing national social budgets in the poorest countries of the world (ECOSOC, 1999a). Capital disinvestment from the social sphere may have affected the working lives of women by adding additional constraints deriving from their household roles. For example, lower health budgeting is likely to increase the care-giving functions falling on women, while disinvestment from community infrastructure, such as water supply systems, is likely to increase women's household work. In development planning, it is of crucial importance to define the target population (by sex, age, sector of agricultural production and type of occupation/work) in view of a specific development objective. However, this will be impossible if rural planners continue to base their decisions mainly on data relating to physical inputs and outputs, while ignoring the perspective of human resources: the human labour input and the social dimensions of agriculture.