Trade and Optimal Monetary Policy

Using a stochastic business cycle model of a small open economy we ask how the problem of the optimizing policy-maker changes endogenously as the international trade structure is altered. From input-output data for OECD and emerging market economies, we document that differences in trade across countries are substantial, and that the composition of trade varies as much as openness to trade. The trade and production structures examined in our theoretical model easily map into measurable macroeconomic characteristics of countries. Our results are as follows. First, the traditional imports to GDP ratio turns out to be close to irrelevant for the ranking of monetary policies. Second, the composition of imports has a very substantial impact on welfare. Given the same degree of openness, the exchange rate peg is more inefficient if the bias towards non-tradable goods is high, and less inefficient if the share of imported intermediates is low. That is, open countries engaging in much vertical trade find it more costly to peg the exchange rate. Third, while a stylized regularity of emerging market economies is the very high share of imported capital goods, compared with industrial countries, this turns out to have no impact on the welfare result. Fourth, local currency pricing has a muted impact . This is the consequence of two assumptions: the existence of a long chain of substitutability in demanded goods, and the mobility of labor across sectors, so that imported inputs affect factor prices in all sectors directly, rather than through general equilibrium effects.