This study investigates whether the export-led growth or growth-led export business strategy is suitable in the Tanzanian economy from 1970 to 2013. The research aims to check if the export-led growth (ELG) and growth-led export (GLE) doctrines are appropriate in the Tanzanian economy following the trade liberalisation in mid 1980s. Augmented Dicker Fuller test was employed to examine the unit root and the empirical results reveal that all variables are non- stationary at level and stationary at the first difference. Engle-Granger residuals and Johansen co integration test were utilised and results confirmed that variables are co integrated. Furthermore, co integration and error correction model investigates the long run and short run coefficients. In long run coefficients, the results conclude that both doctrines (ELG and GLE) are important in Tanzanian economy whereas in short run coefficients were not significant. Error terms confirm the existence of long run relationship amongst the variables. Granger causality test results indicate that in long run coefficients, there are bidirectional relationship between the Gross Domestic Product (GDP) and exports in Tanzanian economy while in short run coefficients there is unidirectional moving from exports to GDP. Structural break results show a stable contribution from GDP to exports while there is an unstable contribution from exports to GDP. In this context, the government should improve the production and export sector to boost the economic development in the long run in Tanzania.
Economic growth, export-led growth, growth-led export, structural break, Tanzanian economy