A scalable video encoding method includes splitting a lower-layer image into a maximum coding unit, and encoding the lower-layer image based on coding units having a tree structure obtained by hierarchically splitting the maximum coding unit; splitting a maximum coding unit of an upper-layer image corresponding to the maximum coding unit of the lower-layer image into coding units having the tree structure, based on a split structure of the maximum coding unit of the lower-layer image; determining a structure of a prediction unit included in the coding units of the upper-layer image corresponding to the coding units of the lower-layer image, based on structure information indicating a structure of the prediction unit included in the coding units of the lower-layer image; and determining whether to split the prediction unit included in the coding units of the upper-layer image by comparing a first cost to a second cost.