Although the 2011 Tohoku‐Oki earthquake (Mw 9.0, 11 March 2011) occurred more than 1,000 km from South Korea, it significantly changed the magnitudes and orientations of Global Positioning System (GPS) velocity vectors in the country, which took two years to return to preearthquake values. Then, areas with relatively thick crust show contractional surface deformation, whereas areas with thinner crust show dilatation, suggesting that the deformation pattern is long‐term. Although there is no one‐to‐one correlation between earthquakes and surface deformation, earthquakes tend to be concentrated in areas with higher shear strain rates. In South Korea, the stress field inferred from earthquake data is inconsistent with surface deformation calculated from GPS data. This may arise because stress orientations at earthquake focal depths are different from those at the surface, or surface deformation measured by GPS may represent interseismic elastic deformation leading to future large earthquakes, while recent earthquakes without surface ruptures might not contribute to the surface deformation. Some effects of crustal thickness on surface deformation and inconsistency between surface deformation from GPS data and kinematics from earthquake data may be a common feature of intraplate regions.