A 3-year field study was undertaken to investigate the effect of row spacing on vegetative growth, grain yield and water-use efficiency of wheat. All 3 years of the study experienced 21–51% below-average rainfall for the growing season. Widening row spacing led to reduced biomass and tillers on per plant basis which could be related to the reduction in light interception by the wheat canopy in the wide rows which in turn could have reduced assimilate production. Reduction in vegetative growth in 54-cm rows translated into a significant reduction in grain yield which was strongly associated (r2 = 0.71) with the loss of spike density. The pattern of crop water use (evapotranspiration, ET) during the growing season was very similar for the three row-spacing treatments. However, there was some evidence for slightly lower ET (~5%) in 54-cm rows in two growing seasons. More importantly, there was no evidence for increased ET during the post-anthesis phase in wide rows as has been speculated by some researchers. Over the 3 years of the study, grain yield declined by 5–8% as row spacing increased from 18 to 36 cm and by a further 12–20% as row spacing increased from 36 to 54 cm. There was a consistent decline in water-use efficiency for grain (WUEG) with increasing row spacing over the 3 years. WUEG declined by 6–11% as crop spacing increased from 18 to 36 cm and declined further by 12–15% as row spacing increased to 54 cm. Lower light interception at wider row spacing could have reduced assimilate production by wheat as well as increased soil evaporation due to lower shading of the soil surface in more open canopies. Growers adopting wider row spacing on these relatively heavy textured soils are likely to experience some reduction in grain yield and WUEG. However, some growers may be prepared to accept a small yield penalty from intermediate row spacing as a trade-off for increased stubble retention and soil health.