Export volatility is a serious problem in the economic growth of any developing country. This study uses quarterly data on Export price, Export Volume, Labor, Capital, Imported capital good and Real Gross Domestic Product (RGDP) over the period 1991-2014. To model Export price and Export volume volatility as time varying process, an extended form of GARCH (Generalized Autoregressive Conditional Heterosedasticity) model i.e. GARCH (1,1) is used and Extended Cobb-Douglas production growth model is also used to see the long-run relationship between export volatility and economic growth in Ethiopia. According to the estimation result of GARCH (1, 1) model, previous day's volatility of export price and volume can influence the current day's volatility of both export price and volume. And from the extended Cobb-Douglas production model result, both export price and export volume volatility indexes negatively affect long run economic growth. As remedial measure the study suggested that the country should transform and shift towards production of high quality goods from low quality goods along with diversifying its export products where it has comparative advantage.