Rapid and comprehensive reduction in barriers to international trade has often been followed by a sharp deterioration in the current account, as demonstrated by Mexico's experience following the accelaration of trade reform in 1987, and by a decline in savings. Theoretical predictions of the impact of trade reform on the current account tend to be counterfactual, since a permanent reduction in tariffs will affect current and future goods equally, leaving intertemporal relative prices and private savings unchanged while gradual tariff reduction raises the price of current vis-à-vis future goods and improves private savings. Permanent reduction of tariffs will therefore affect intratemporal not intertemporal prices and should therefore have no impact on savings.

In Discussion Paper No. 441, Research Fellow Sweder van Wijnbergen analyses how these conclusions are affected by the possibility of a policy reversal. First, a positive probability of future reimposition of tariffs reduces the price of current vis-à-vis future goods, so the anticipation of a future tariff increase will raise current consumption if the intertemporal elasticity of substitution is greater than one. Further, the anticipation of the impact of future tariff revenues on consumers' after-tax income, will always have a negative impact on savings, whatever the intertemporal elasticity of substitution. Second, quite apart from the impact of the impact of shifts in expected intertemporal relative prices, policy uncertainty will in itself further reduce private savings provided that there is positive risk aversion and that the intertemporal elasticity of substitution elasticity exceeds unity.

These results have important policy implications. If the trade reform will not be reversed, but the government cannot credibly communicate this to the private sector, consumers effectively use the wrong intertemporal prices, so private savings will be sub- optimal. Policy intervention to increase private savings is therefore justified, preferably through a temporary increase in consumption taxes, or through temporary tariffs as a more feasible second best.

Mistaken beliefs concerning future policy regimes act as distortions and will be self-fulfilling if the decline in private savings following reform leads to a large enough current account deficit for the reform to be reversed. This possibility further strengthens the case for policy intervention to increase private savings, and arguably for external support in the early periods of trade reform through institutions such as the World Bank or the IMF.

Trade Reform, Policy Uncertainty and the Current Account: A Non- Expected Utility Approach
Sweder van Wijnbergen