The empirical studies in the context of Bangladesh provide contradictory evidence regarding the relationship between domestic saving and foreign aid. The present article examines the domestic saving-foreign aid relationship using the co-integration and error-correction modelling strategies to tackle the problem of non-stationary data, which the earlier studies overlooked. Three different co-integration techniques have been used and in every case a valid long-run inverse relationship between saving and aid is observed. The short-rum error-correction models have also confirmed the negative relationship between these two variables. The Granger Causality test identifies a long-run causal relationship running from foreign aid to saving while in the short-run a bi-directional causality is found. This paper finds that the negative relationship between saving and aid is quite ‘robust’ and not sensitive to the choice of variables in the saving function.