This paper uses time series data from 1972 to 2012 on Ghana to test the hypothesis that foreign aid can promote growth in developing countries. The ARDL approach to cointegration (bounds test) was employed to examine both the long run and short run relationships between aid and economic growth. The results of the bounds test showed that there is cointegration between foreign aid and economic growth in Ghana. This was further confirmed by the error correction term which was very significant and correctly signed. The error correction term showed that the speed of convergence to long run equilibrium is moderate. From the results, labour, capital and government expenditure have positive impact on economic growth in Ghana in both the long run and the short run whereas foreign aid and interest payment on external debt have negative impact on growth. In order to derive a positive benefit of foreign aid, we recommend the provision of economic aid which is geared towards capital formation and skills development of labour through education and training rather than political aid since the results show that capital and labour have positive impact on economic growth. We also recommend the provision of more grants and less loans as aid to Ghana because interest payment on external debt has been found in the study to have negative effect on economic growth because most foreign aid are not invested in projects with direct future cash flows.