Studying the stability of a vertically suspended, fully submerged pipe conveying water (water riser), researchers have found a contradiction between theoretical predictions and experiments. Theory predicts instability at small velocities of convection, while experiments did not show such instability. The only attempt to explain this contradiction was made by Païdoussis, suggesting that the theoretical prediction is wrong because of an improper description of the negative pressurisation of water at the inlet of the pipe. In this paper, it is shown theoretically that the negative pressurisation influences the stability only slightly and cannot explain the contradiction. What can explain it, is the hydrodynamic drag caused by surrounding water, which is shown to be an essential stabilising factor.