The paper examines the causal effect of public debt on economic growth in developing Asia over the last two decades, using an instrumental variable approach. The author equipped the debt-to-GDP ratio with multiple instruments, one of which is a new index to address the potential endogeneity in statistical modeling of debt-growth nexus. The multiple instruments hold the strength of identification. The consistent estimations provide robust evidence for the detrimental impact of debt on growth. The estimation of a kinked constrained IV-GMM estimator finds a statistically significant debt threshold effect at around 50-60 per cent of GDP, where the growth pays off significantly.