Dynamic effects of trade and output volatility on the trade‐growth nexus

Purpose - This paper seeks to revisit the highly debated trade‐growth hypothesis by considering the effects of trade and output volatility on the relationship between trade and economic growth. Design/methodology/approach - The relationship is modeled by testing for the existence of output and trade (export and imports separately) using the conditional variances of the variables and then specifying an autoregressive conditional heteroskedastic (ARCH) process in a vector error correction model. Findings - Using Singapore as a case study, the paper finds the two‐way relationship between export growth and trade‐adjusted GDP growth is robust even after controlling for the effects of income and export volatility. In addition, neither trade nor GDP volatility bears any impact on the bi‐directional causality between imports and unadjusted GDP growth thereby highlighting the crucial role of imports as intermediate inputs and embodying foreign technology in promoting economic growth. There is also evidence that output volatility impedes output and trade growth, while trade volatility exerts a negative influence on the trade‐adjusted income growth. Practical implications - Ignoring the presence of trade and output volatility in modeling the trade‐growth relationship provides biased empirical results which have serious implications for trade‐oriented growth strategies that policy makers cannot afford to ignore. Originality/value - This is the first attempt to explicitly model output, export and import volatility in empirically testing the trade‐growth hypothesis. Second, the robustness of the hypothesis is also tested by considering GDP and non‐trade GDP as it has been argued that use of GDP may lead to the problems of simultaneity and specification bias since exports and imports are themselves a component of GDP.