This study is concerned with time series empirical analysis on sources of GDP growth in Ethiopia for the period 1981 to 2009. An aggregate Cobb-Douglas production function for Ethiopia is estimated by expressing the production function in intensive form. Growth equation is estimated with a time trend to capture the rate of technological progress within a cointegrational framework. Assuming constant returns to scale, the intensive production function is estimated with OLS and the regression result showed that input elasticities for capital and labor were 0.43 and 0.57 respectively. And average rate of technological progress was .001. These parameters are used to compute the growth contributions of capital, labor and technical progress. Capital, labor and technological progress contributed about 56%, 42%, and 2% respectively to GDP growth in Ethiopia the period 1981 to 2009. From 1981 to 1991 GDP, capital and labor annually grew on average by 1.2%, -0.03% and 3.1% respectively. And the contribution of capital, labor and technology to growth were -1.06%, 95.74% and 5.32% respectively in the same period. The dismal GDP growth performance and negative growth of capital formation during 1981 to 1991 occurred because of the socialist government’s inappropriate economic policy and fall in agricultural output due to drought. From 1992 to 2009 GDP, capital and labor grew on average by 5.9%, 8.5% and 3.2% respectively. Capital, labor and technology contributions to growth were 66%, 32% and 2% respectively. The increasing capital formation and GDP growth attributed to EPRDF regime’s liberalization policy and creation of relatively favorable conditions for private sectors and increase in agricultural out put. Capital labor ratio had positive effect on economic growth in short run as well as long run in Ethiopia during 1981-2009.