Terms of trade shocks in Africa: are they short-lived or long-lived?

This paper examines the persistence of shocks to the terms of trade using annual data on 42 sub-Saharan African countries for the period 1960-96. Instead of using unit root tests to distinguish between stationary and random walk processes for the terms of trade, this paper characterizes the persistence of shocks using point and interval estimates of the half-lives of terms of trade shocks. Using median-unbiased (MU) estimation techniques that remove the downward bias of standard [least squares (LS)] estimators, we find that the cross-country average of the half-lives of terms of trade shocks is about 6 years. It is also found that the estimated persistence of African terms of trade shocks varies widely-for about one-half of the countries, their half-lives are short-lived (less than 4 years), while for one-third of the countries, their half-lives are long-lived (permanent). The majority of African countries are found to have terms of trade that typically experience finitely persistent (transitory) shocks, which is consistent with the reversion of terms of trade to their long-run trends. (C) 2003 Elsevier B.V. All rights reserved.