Although empirical evidence and theoretical analyses both overwhelmingly support the view that trade liberalization yields efficiency gains, agricultural liberalization has made slow progress to date. In Discussion Paper No. 624, Santiago Levy and Research Fellow Sweder van Wijnbergen assess the effects of including agriculture in the Free Trade Agreement (FTA) now under negotiation between Mexico and the US. They show that the efficiency gains of liberalization tend to accrue to richer groups in both rural and urban areas, which may explain farmers' resistance to easing agricultural protection. They assume that there are no instruments available to effect lump-sum transfers, so the result of standard trade theory that winners can compensate losers and still be better off does not apply. Indirect taxes and subsidies involve major administrative difficulties and two incentive problems. First, compensating maize farmers in proportion to their current production encourages them to continue the production that the reform is intended to reduce; second, maize liberalization has major effects on rural labour markets and migration, which can be mitigated by rural employment programmes.

Levy and van Wijnbergen argue that these require the design of specific adjustment programmes to accompany any major trade reform. In particular, improvements to the productivity of assets owned by the groups harmed by reform may break the linkage between transfer payments and recipients' past activities and thus remove their incentives to continue them. For Mexico, this entails investment to increase the productivity and hence the value of rain-fed land through irrigation and other improvements. By improving farmers' access to credit when they most need it and increasing rural labour demand, such a programme can credibly establish its transitory nature by providing workers with alternatives once it ends.

Levy and van Wijnbergen advocate gradual reform, since those most affected by reform are among the poorest social groups and are least able to borrow to smooth consumption. Spreading the reform over five years reduces the efficiency gains by only 5-6%, and careful timing of the liberalization and adjustment programmes can ensure that the rural poor always have higher utility under adjustment than under protectionism. The commitment technology of embedding trade liberalization in an FTA weakens the case for `cold turkey', but it also requires assurance to potential beneficiaries that the programmes will begin once the negotiations over trade liberalization are complete. This can be provided by making support from external organizations contingent on the completion of adjustment programmes.

Transition Problems in Economic Reform: Agriculture in the Mexico-US Free Trade Agreement
Santiago Levy and Sweder van Wijnbergen