ABSTRACT This paper compares the axial shortening behaviors of reinforced concrete (RC) and concrete-filled tube (CFT) columns in high-rise building structures and proposes a relationship between the differential axial shortening and the dead load portion. This relationship can be used to determine the initial column sections in the preliminary design of building structures. Factors that can influence the differential axial shortening in columns were considered, and numerical analyses were conducted to compare the relative effectiveness of CFT columns. The time-dependent deformation of concrete was found to be the most significant influencing factor when compared with the flattening process of each floor level and construction sequence analysis. However, parametric studies demonstrate that axial shortening behavior can be ignored in buildings of fewer than 20 floors, and that it has a negligible effect even in taller buildings. Notably, the differential axial shortening effect can be excluded in the preliminary design stage when the dead load portion is less than 30% of the ultimate axial resistance. In the absence of design standards recommendations for reference values to minimize serviceability or damage problems in building structures, the proposed relation between the differential axial shortening and the dead load portion can be effectively used to determine initial column sections in the preliminary design of building structures.