Existing studies of trade reform in Sub-Saharan Africa (SSA) -- and elsewhere -- are mainly qualitative in character, partly because of the difficulty of measuring the intensity of trade liberalisation. This paper uses a quantitative measure of trade liberalisation to provide statistical evidence on the conditions accounting for Africa's timid reforms. This measure captures the change in the tariff equivalent of all trade barriers and is therefore a summary indicator of the intensity of reform. The measure provides a fairly accurate description of the trade liberalisation experience of SSA countries, indicating that the most significant reforms occurred during the period 1986--90. The empirical model includes a set of hypotheses suggested both by theory and the anecdotal evidence of Africa's liberalisation experience. The empirical methodology distinguishes between the conditions that affect the likelihood of reform and the factors that determine the intensity of trade liberalisation. The empirical results justify this distinction. The findings suggest that larger aid flows, higher levels of urbanisation, a strong current account position and a relatively large manufacturing sector enhance the probability that trade reform is adopted. There is also some evidence that economic crises facilitate reform. Conversely, heavier fiscal dependence on trade taxes, greater import competition and a large government make trade reform less likely. However, only five of these factors -- current account balance, size of manufacturing sector, size of government, fiscal dependence on trade taxes and aid -- are found to exert a statistically significant effect on the intensity of trade reform. Copyright 2003, Oxford University Press.