¹ The statistics reported in this paragraph are taken from World Bank, Africa: development indicators (Washington, DC, 1992),Google Scholar and United Nations Conference on Trade and Development,The Least Developed Countries Report(Geneva,1992.)Google Scholar

² World Bank, Review of Adjustment Lending II: adjustment lending policies for sustainable growth (Washington, DC, 1989).Google Scholar

³ These alternative analyses of the causes of sub-Saharan Africa's economic crisis, and the associated policy recommendations, are exemplified in the competing views of the ‘Washington consensus’ and the UN Economic Commission for Africa. See, for example, World Bank and UNDP, Africa's Adjustment and Growth in the 1980s (Washington, DC, and New York, 1989),Google Scholar and UN ECA, Africa's Alternative Framework to Structural Adjustment Programmes for Socio-Economic Recovery and Transformation (Addis Ababa, 1989).Google Scholar

⁴ Elbadawi, I. A., Ghura, D., and Uwyaren, G., World Bank Adjustment Lending and Economic Performance in Sub-Saharan Africa in the 1980s: a comparison with other low income countries (Washington, DC, 1992), World Bank Policy Research Working Paper No. 100, p. 5.Google Scholar

⁵ The comparative performance figures for 1986–1990 are as follows: (i) real growth of GDP (%): 3.5 for adjustment and 3.9 for non-adjustment loan countries; (ii) exports/GDP (%): 28.0 for adjustment and 31.7 for non-adjustment loan countries. World Bank, Adjustment Lending and Mobilization of Private and Public Resources for Growth (Washington, DC, 1992), Policy and Research Series No. 22.Google Scholar

⁶ Mosley, Paul and Weeks, John, ‘Has Recovery Begun? Africa's Adjustment in the 1980s Revised’, in World Development (Oxford), 21, 10, 1993, p. 1589.Google Scholar

⁷ The approach adopted is broadly similar to that used in World Bank, World Development Report 1983 (Oxford and New York, 1983), for an index of price distortions.Google Scholar

⁸ The question of weighting to apply to various criteria is a fundamental problem in this type of analytical work. The expedient of using equal weights is common: see, for example, World Bank, World Development Report 1983 (Washington, DC, 1983).Google Scholar

⁹ The relation tested across countries is EX (export growth) = a+b INDEX (liberalisation). Using average export growth for the 1980s the result found is: Ex = -2.86+0.15 INDEX, where R² = 0.09, n = 25, and 1.47 is the t-ratio. The coefficient b is significant only at the 10% level.

¹⁰ UNCTAD, op. cit. 1993.

¹¹ Exchange-rate data come from World Bank, op. cit. 1992. The real exchange-rates given are where nominal exchange-rate movements are adjusted for relative price changes in a country and its trading partners. Real indices are defined so that a decline gives a depreciation. Ibid. Table 3.9.

¹² Of the countries covered by the liberalisation index the above-average liberalisers pursued a variety of exchange-rate policies during the 1980s. Four were in la zone franc (Benin, Burkina Faso, the CAR, and Chad) and experienced a nominal appreciation against the US dollar. Six others (The Gambia, Guinea, Malawi, Botswana, Lesotho, and Burundi) adjusted their exchange rates over the decade and experienced nominal devaluations of varying magnitudes. However, there is no tendency for the liberalising economics to experience larger than average real exchange-rate depreciations. In fact the reverse holds with the below-average liberalisers showing on average the largest real depreciation of their exchange rates.

¹³ The relation tested is EX (export growth) = a+b RER (change in real exchange rate). The results for period I are: EX = 2.65+0.04 RER where R² = 0.11, n = 22, and 1.61 is the t-ratio. The positive coefficient b is significant at the 10 % level. For period 2, however, using World Bank instead of UNCTAD export data, there is no significant relationship.

¹⁴ The relation is MAN (change in share of manufacturing in GDP) = a+b RER (change in the real exchange-rate). The result for the 1980s is: MAN =1.88+0.04 RER, where R² = 0.40, n = 18, and 3.51 is the t-ratio. The coefficient b is significant at the 1% level.

¹⁵ For a seminal statement of the ‘dependent economy’ two-sector model that lies behind this reasoning, see Dornbusch, Rudiger, Open Economy Macro-Economics (New York, 1980).Google Scholar

¹⁶ Mosleys and Weeks, loc. cit. p. 1591.

¹⁷ Diakosavvas, Dimitrius and Kirkpatrick, Colin, ‘Exchange Rate Policy and Agricultural Exports in Sub-Saharan Africa’, in Development Policy Review (London), 8, 1, 1990, pp. 29–42,Google Scholar and Gylafson, T. and Radetzki, M., ‘Does Development Make Sense in the Least Developed Countries?’, in Economic Development and Cultural Change (Chicago), 40, 1992.Google Scholar

¹⁸ The export growth for a group of developing countries over the 1960–1983 period is negatively correlated with the instability and misalignment of the real exchange-rate. See Cottani, J. A., Cavallo, D. F., and Khan, M. S., ‘Real Exchange Rate Behaviour and Economic Performance in LDCs’,Google Scholar in ibid. 38, 1990.

¹⁹ The difference between the average for the two groups is significant at the 1% level. Data on terms of trade movements refer to barter terms of trade and come from World Bank, op. cit. 1992, Table 5.15.

²⁰ Faini, Riccardo and de Melo, Jaime, ‘Adjustment, Investment and the Real Exchange Rate in Developing Countries’, in Economic Policy (Oxford), 10, 10 1990, pp. 492–519.Google Scholar