Co-integration and error-correction modelling of agricultural export supply in Cameroon

Estimates of factors influencing Cameroon's exports of cocoa, coffee and cotton are derived in a system of equations using the Engle-Granger and Johansen co-integration and error-correction representation procedures. Two co-integrating vectors involving cocoa and coffee exports as endogenous variables are identified in the system while tests for exogeneity of cotton exports are consistent with the independence of cotton from the other two commodities. These findings are corroborated by estimates of a restricted error-correction model which lead to acceptance of the hypothesis that cocoa and coffee exports are indeed determined endogenously to the system and not linked to cotton exports. Statistical significance of the error-correction terms for cocoa and coffee validates the existence of an equilibrium relationship among the variables in each of these cointegrating vectors. The combined short-run dynamic effect of lagged quantities of cocoa and coffee, export/domestic price ratio and GDP jointly explain changes in exports of cocoa whereas lagged quantities exported do not seem to have a significant short-run dynamic effect on changes in coffee exports. © 1999 Elsevier Science B.V. All rights reserved.