Following Sadorsky (Energy Economics, 2011, 33, 739), we examine the nexus between energy and trade in South Asia in a panel framework using annual data from 1990 to 2015. Given the cross-sectional dependency issue, we use robust second-generation panel econometric methods for the analysis. We find that variables are stationary at the first difference and cointegrated in the long run. The Dumitrescu-Hurlin panel causality test shows a long-run unidirectional causality from trade openness to energy consumption but not vice versa. Moreover, the panel symmetric and asymmetric estimation techniques reveal that energy consumption rises by 0.47 per cent on average and 0.83 per cent, with a 1 per cent rise in trade openness in the long run. However, a negative shock of similar magnitude reduces energy consumption by 0.37 per cent. We further fortify the panel results in the country-level data with robust methods. We recommend energy efficiency and conservation policies and renewable energy technology augmentation policies through bilateral or multilateral trade strategies for sustainable development in the selected South Asian countries.