We use the global LINKAGE model to assess the impact of trade and support policies in agriculture on income, trade, and output patterns. We provide order-of-magnitude estimates of the impacts of policy changes rather than point estimates. Two sets of simulations are used to identify key drivers in the results. One set decomposes the aggregate results by looking at the impacts of partial reforms, regionally and across instruments, to identify the relative contribution to global gains of reforms in industrialized and developing countries and of border protection versus domestic support. The second set responds to critics of trade reform (inflated gains for developing countries, no transition costs for industrial country farmers, uncertain supply response in developing countries).

Reform of agriculture and food provides 70 percent of the global gains from merchandise trade reform of $385 billion. The global gains are shared equally among industrial and developing countries. Developing countries gain more as a share of initial income, and income gains occur in developing country agriculture, reducing poverty. Both groups of countries gain more from their own reforms than from the other group's reforms. Productivity and supply assumptions affect impact assessment, but their influence is small and does not alter the main aggregate findings. Trade elasticities, however, are key in determining the overall level of the income gains. Higher elasticities dampen terms-of-trade effects and increase trade and real income gains more than proportionally and the converse is true for smaller elasticities. These effects can be very large for individual countries.