Trade liberalisation - lifting import restrictions and promoting exports - has been a central feature of economic policy reform in Uganda. Yet, does a rise in exports automatically follow liberalisation? There are two reasons why this is not necessarily the case: a) the ability to increase production is curtailed by a shortage of appropriate inputs, especially in the primary sector; and b) reducing trade policy barriers does not eliminate obstacles to trade. Poor infrastructure and institutional inefficiencies can significantly increase trade or transactions costs, notably by increasing transport costs. If such non-policy barriers remain high they will constrain the ability of producers to increase exports.

Uganda has made significant progress in reducing anti-export bias in its trade policy. Taxes on exports have been abolished and import tariffs reduced. Import tariffs often result in a net subsidy for domestic producers of import-competing goods, but a net tax on producers of exports. Producers of exports have no protection on the price of their final outputs yet they still may have to pay a higher price on production inputs if they are imported. Protection against imports from other parts of Africa has all but disappeared in most sectors. Tariff protection (so-called nominal protection) against imports from outside Africa was less than 20 percent on average in 1997, resulting in real (so-called effective) levels of protection being less than 30 percent.

Non-policy barriers remain high, however. Transport costs do not appear to be any lower than in 1994. For food products and manufactured goods transport costs provide far greater protection than do tariffs. Natural protection (the excess costs of trade due to remote location and/or inefficient transport networks) on domestic sales arising from transport costs was equivalent to an average 48 percent tax in 1994 - around a quarter higher than the 38 percent effective protection due to trade policy. Implicit export tax-cum-transport costs reached 100 percent for manufactured foods, almost 40 percent for food products, almost 25 percent for coffee, cotton and tea, and about 20 percent for fish. The tax on manufactures is greater both because unit transport costs tend to be higher and more imported inputs are used (and subject to tariffs plus transport costs).

Transport costs, it is evident, are often inordinately high, in many cases representing a greater cost to exporters than trade policy. One reason is simply that Uganda is land-locked and far from the sea ports through which most goods pass. This is exacerbated by the poor quality of road and rail networks and long delays at customs and ports. Such non-policy barriers partly explain why Ugandan manufacturing has been so slow in responding to trade liberalisation and throws up a number of policy implications: