Trade liberalisation in small open economies : the case of Kenya
The object of this thesis is to determine the consequences of trade liberalisation on the Kenyan economy. This is done by simulating the effects of tariff reduction, devaluation of domestic currency and export subsidies. In addition, the effects of quantitative controls and markup pricing are simulated. The structure of the economy is modelled through the specification of alternative closure rules. Policy changes are simulated using a computable general equilibrium model (CGE). A nine sector model based on a Social Accounting Matrix is constructed using the TV-approach to modelling introduced by Drud, Grais and Pyatt (1986). We depart from neoclassical models, and therefore other CGE models of Kenya, by assuming product differentiation between domestic goods and imports and between gross output sales to domestic and export markets. Our model is essentially Keynesian but for comparative purposes, neoclassical closures are specified in some simulations. In general, the basic argument for or against trade liberalisation concerns its contribution to economic growth. The neoclassicals argue that by improving efficient allocation of resources, liberalisation stimulates higher economic growth. The structuralists, on the other hand, argue that because of structural rigidities in LDC economies and because of unfavourable international conditions, liberalisation will have minimal effect on economic growth. CGE models are useful in sorting out these arguments. It should be noted however that the assumptions underlying these models often reflect the modeller's view about the structure of the economy. The usefulness of CGE models for policy purposes will therefore depend on how realistic they reflect the structure of the economy being modelled. The results of our model show that the gains from trade liberalisation, in terms of the growth of real GDP, are low. This applies to both neoclassical and Keynesian closures. However, it is shown that changes in returns to factors, consumption levels and aggregate price levels, depending on the closure adopted, are significant. This is also true for the policy effects on exports, imports and on the prices and quantities at the sectoral level. These results reinforce the view that for policy purposes it is important that the model being used reflects the structure of the economy under consideration. It also means that it will not make sense to have tailor made policy recommendations for all LDCs.