The liberalization of food marketing has been implemented as part of structural adjustment programmes in Sub-Saharan Africa. In this study we assess i) the aims of reform policy, ii) the implementation of specific reform measures, iii) the politics of reform, and iv) the impacts of reform. The study highlights the diversity in the initial circumstances: during the late 1970s, state-governed marketing was firmly established in many, but not all countries. Where it existed, it varied further in terms of the variety of crops collected and the efficiency of actual collection. In general, pre-reform marketing policies were extensive, serving simultaneously the political aims of nation-building and food security, and for this purpose, provided subsidies to both producers and consumers. These subsidies, together with the high costs of marketing operations, caused huge financial problems for the governments. Marketing reforms have been implemented through reforming pricing policies, institutional set-up and macro-economic environment. Several governments embarked with doubts on liberalization policies, but during recent years even the most hesitant countries have also implemented reforms. In certain countries, the liberalization of key food crops was a delicate political issue because of complex vested interests. Behind the debate over the supply of cheap food for urban consumers existed a whole range of factors like the patrimonial linkages of marketing boards, regional politics, and the interests of large-scale millers and estate producers. In the debate on economic liberalization, political issues are largely simplified and marketing boards are crudely evaluated in terms of economic efficiency. The study compares countries which always relied on private food marketing to countries which liberalized food marketing in 1992 and those still retaining state interventions. The data show that the best growth rates for the production of key food crops are in the countries with more liberal food marketing regimes. Differences within country groups are significant. Variations can partly be explained by the achieved level of self-sufficiency and by the nature of key crops. Differences within country groups are also very high. Maize production in eastern and southern Africa has been the focus of a detailed case-study because of the complexity of its extreme politization. Marketing reform has had relatively little impact on food production, which is still growing more slowly than is the population in Sub-Saharan Africa. Major gains were expected from decreased fiscal costs of active (and extensive-coverage) food policies. These gains have materialized to a lesser extent than expected. If liberalization releases government resources for other uses, these should be directed to measures to increase agricultural production: land reform, input subsidies and the construction of feeder roads. The marketing, milling and consumer support of food crops should be targeted at crops mainly consumed by the poor.