The analysis of Chapters 2 and 3 was conducted in terms of an economy which was too small to affect the prices at which it bought and sold goods in world markets. Such an approach offers both a simple means of obtaining insights into the workings of various policies and a reasonably accurate representation of most real world economies. However, it is an inadequate tool for analysing the trade relations among such major trading countries or blocs as the United States, Japan and the EEC. In this chapter we temporarily leave the small-country case and consider economies which are large enough to use their buying and selling power to influence world prices. For the time being, we retain the assumption that producers are atomistic price takers. The logical starting point for any analysis of protection in a large-country context is the observation that a country which can, by its own actions, affect the price at which it buys or sells a traded good may gain by using a tariff or an export tax to restrict trade. The resulting gains from an improved terms of trade must be balanced at the margin against the increased domestic deadweight losses from a higher trade tax. The outcome of this exercise is the so-called optimal tariff. In the first section we consider how the optimum tariff is derived. We then employ a two-country model to consider the outcome when one country's optimal tariff induces the other country to impose a retaliatory tariff.