Reinforced concrete (RC) shear walls in tall buildings are found to have a short shear span, particularly in high-degree coupled walls supported on transfer structures in low-to-moderate seismic regions. These non-seismically detailed walls in existing buildings are exposed to a high risk of failing in shear or compression before plastic hinges are formed at their base. Whilst previous research have focused on squat walls used in low-rise structures tested with zero or low axial loads, the structural response of these walls with a short shear span and limited ductility under high axial load is rarely discussed. Therefore, an experimental study that investigates the influence of the axial load ratio (ALR) on RC walls with a short shear span is presented in this paper. The specimens are designed with a low shear span-to-length ratio (SLR) and detailed with a characteristic 2% vertical and longitudinal reinforcement to represent a wall sub-structure above the transfer structure of tall buildings. Four walls are tested under reverse cyclic loading and subjected to target ALRs that range from 0.1 to 0.4 to investigate the seismic performance until gravity collapse. The ALR is found to have significant effects on crack patterns, failure modes and deformability. Two modified empirical prediction models are proposed to estimate the shear strength capacity and ultimate drift ratio of rectangular RC shear walls with a short shear span under the effects of the ALR. A unique model of the drift limit of collapse under axial load as a function of the reinforcement ratio is put forward for performance based design and assessment.