External shocks and macroeconomic variability in a small developing economy

The main aim of this thesis is to examine external shocks faced by small developing countries and the impact of these shocks on the economy, using Botswana as a case study. We present three papers which explore the external shocks within a variety of macroeconometric frameworks, and make policy recommendations based on our findings. In the first instance, we utilize a sign restricted vector autoregressive (VAR) approach to identify four generalized terms of trade shocks relevant for small commodity exporting countries. The findings of this paper show that the impact of shocks to Botswana's terms of trade is mainly determined by the global context in which the shocks occur. The response of macroeconomic policy therefore differs depending on the underlying determinants of each shock. Secondly, we employ a principal component analysis (PCA) to consider the common factors driving variations in a panel of Botswana's bilateral exchange rates. Here we find that the unobservable component most responsible for variations in these bilaterals is not a Botswana factor, but rather a risk-related factor that responds to conditions in the US and South African economies. Further, models incorporating this latent component improve predictive accuracy at high frequencies for some of the exchange rates. In the last paper, we examine the incidence of the resource curse and utilize a structural VAR framework to explore the response of key sectors in the Botswana economy to positive natural resource shocks. We find that while Botswana has successfully avoided the political dimensions of the resource curse, it cannot be said with certainty that the country escaped the economic dimensions of the phenomenon. Our results show that the economy exhibits some mild Dutch disease effects, however, these are minimized and brief due to successful policy efforts to maintain a competitive exchange rate.