Aims: The objective of this paper is to examine whether public debt enhances economic growth in Sri Lanka by separating debt into domestic and external debt. Study Design: The paper adopts the Keynesian demand-side model to estimate GDP on demand-side and supply-side factors such as domestic debt, external debt, gross fixed capital formation (investment in short), labour force, general price level, money supply, government spending and trade openness. In a model where debt causes investment, public debt effects are netted out from investment by removing the estimated investment effects of domestic and external debt. Duration of Study: quarterly data from Q1 2010 to Q1 2023 are utilized to trace the long-term and the short-term effects of domestic and external debt on GDP. Methodology: The ARDL method of estimation is used and the model is estimated in double log form so that estimated parameters reflect output elasticities. A general to specific modelling approach is adopted to find the best suited model. Results: Variables such as government spending, money supply, general price level and labour force are not relevant while domestic debt, external debt, investment (debt effects netted out) and trade openness are significant in describing GDP. The long-term domestic debt elasticity of GDP is 0.588 while short-term domestic debt elasticity (sum of contemporaneous and lag effects) is about 0.63. The cumulative effects of domestic debt on GDP tappers off to 0.60 in the long-run. The result suggests negative impacts (-0.272 and -0.254) of external debt on the economic growth in short-term and long-term respectively. The negative effect of external debt can be justified if recent external borrowings have been used to repay loans taken to finance war and unproductive projects. Conclusion: The paper finds that the domestic public debt creates relatively strong positive effects on GDP in the short-term and the long-term. Contrary to the expectations, the effects of external debt on GDP have become negative both in the short-term and the long-term. The paper also finds that debt effects on GDP have been significantly reduced when debt effect-ridden investment variable is included in the regression in place of debt-effect netted out investment. Thus, this study recommends researchers to carry out further research on why the effects of external debt on GDP become negative. Further, it recommends policy makers to use both domestic and external debt to finance public investment and productivity enhancement projects rather than borrowing merely for the purpose of consumption and repaying of existing debt.