Peter Rogers 1 and Jorge Ramirez-Vallejo 2
Division of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138, USA
1 Gordon McKay Professor of Environmental Engineering, and Professor of City Planning,
2 Visiting Research Scholar, Division of Engineering and Applied Sciences
Agriculture is one of the most important sectors in many developing countries, and thus is a key determinant of growth. Increased trade provides economic development, as many developing countries have a comparative advantage in some form of agricultural production. For the least developed countries it represents more than 36% of GDP and 35% of total trade (IFPRI, 2003). Yet, in these countries the ability of agriculture to contribute to growth and development has been heavily circumscribed by global trade protection, much of it
originating from OECD countries.
Direct agricultural subsidies in OECD countries amount to over $300 billion every year with the EU leading followed by the US and Japan (EU, 2003). Farmers in the OECD countries typically receive more than one third of their income from government programs. The value of total agricultural support in OECD countries is more than five times higher than total spending on overseas development assistance and twice the value of agricultural exports from developing countries. Payments per cow per day in the EU exceed the generally accepted definition of poverty pegged at $2 per day for hundreds of millions of humans in the developing world. As a result, the reduction of subsidies and liberalization of agriculture will continue to be at the top of the negotiating agenda of future international trade talks. As this paper was being drafted the WTO meeting in Cancun, Mexico, collapsed because of the failure of the OECD countries to address some of these contentious issues. Therefore, the linkages between agricultural trade and water resources need to be identified and analyzed to better understand the potential impacts that a full liberalization, or lack thereof, will have on water resource exploitation under future trade regimes.
Water is one of the most important inputs to agricultural production. Water is embodied in all products to a greater or lesser degree, but the amounts used in production generally exceed the embodied water by orders of magnitude. This is particularly true with agricultural products. For example Hoekstra and Hung (2002) claim that 16,000 cubic meters of water are required to produce one ton of beef, compared with 1,200 per ton of wheat. The supply of water to agriculture can be from natural sources of rain or snow or from human controlled sources by irrigation. The total amounts of water from all sources used to produce a crop and including the water embodied in the crop are referred to as virtual water. The concept of virtual water has been widely discussed since its first introduction by Allan (1993). Figure 1 gives a schematic definition of the concept. Recently there has been a resurgence of interest in operationalizing the concept (see Hoekstra and Hung, 2002, Oki, 2002, and Hoekstra et al., 2003). So far the literature has focused mainly on the amounts of virtual water traded under the present trade regimes. The main purpose of this study, however, is to examine the impact of trade liberalization on virtual-water trade in the future.