Abstract
This study examines the empirical relationship between manufactured exports, primary commodity exports and output per worker in Malawi over the period 1980-2019. The Solow model, with physical and human capital as factor inputs, is augmented with manufactured exports and primary commodity exports as potential determinants of total factor productivity (TFP). This study contributes to the existing literature on Malawi by disaggregating exports into manufactured and primary commodities and expressing these export components as ratios to total merchandise exports to avoid spurious regression results seen in other studies. The empirical analysis uses the Autoregressive distributed lag (ARDL) modelling procedure developed by Pesaran et al. (2001) to examine if manufactured and primary commodity exports determine output per worker in the long run. The main finding is that manufactured exports have a positive and significant long-run effect on output per worker. Primary commodity exports show a positive and significant long-run effect, which is larger relative to manufactured exports. These differential results imply that the economy remains resource-dependent and vulnerable to volatile primary commodity prices. The long-run effect of physical capital is positive and statistically significant. The large magnitude of the physical capital accumulation estimate suggests some positive externalities associated with capital accumulation. The long-run effect of human capital, on the other hand, is positive but statistically insignificant. All the variables show positive and significant short-run effects. To sustain growth, the government should continue diversifying the economy into manufacturing, increase the total fixed investment rate to realise positive externality effects, improve educational quality, and correct demand and supply mismatches for educated labour.
Keywords: Manufactured exports, primary commodity exports, physical capital, human capital, output per worker, augmented Solow model.