Throughout the second half of the 20th century, Indonesia's trade policies favoured the modern manufacturing sector relative to agriculture. Within agriculture, they favoured import-competing sectors at the expense of export-competing sectors. Rice is the most important import-competing agricultural sector and, until 2000, assistance to rice growers came mainly from input subsidies. The extent of subsidies to rice and other import-competing agriculture varied with foreign exchange availability and the government's budgetary circumstances: 'good' times produced more interventionist policies and 'bad' times produced less government intervention. Since the mid-1980s, the aggregate bias in favour of manufacturing has declined greatly and agriculture is now only lightly taxed relative to manufacturing. This change resulted from the reduction in manufacturing protection that has characterised this period and from the increased protection of rice and sugar occurring after 2000. We attempt to explain the political forces underlying these changes in trade policy.