This paper examines the saving-investment nexus and the level of capital mobility for the BRICS and Fragile Five countries within the scope of the Feldstein-Horioka (1980) puzzle that asserts a substantial correlation between domestic investment and domestic savings in spite of the increasing capital mobility in the world. It is used the Autoregressive Distributed Lag (ARDL) bounds test approach for the period 1980-2018 on a country-by-country to identify the nature of the saving retention coefficient. Findings of the paper reveal the Feldstein–Horioka (1980) puzzle holds for China, South Africa and Turkey period examined while there is not any long-run relationship between savings and investment in Brazil, India and Indonesia. Based on the savingretention coefficients range from 0.46 to 0.74 for the four countries which there is a cointegration between saving and investment, there is moderate to low capital mobility in these countries.