Teeth resection is a method of controlling the injurious effects of the aggression displayed when newborn piglets fight to establish a teat order. Recent European legislation discourages the practice. The objective of this study was to assess the effect of clipping and grinding piglets’ needle teeth, compared to leaving them intact, on the welfare of piglets in farrowing crates. Six days pre-partum, 60 sows were assigned to one of three treatments. Litters had their teeth clipped (C), ground (G) or left intact (I) at birth. The time taken to carry out each procedure was recorded. Piglet weights and facial lesions, which were scored according to severity, were recorded on days 1, 4, 11, 18 and 27. Piglet weights were also recorded at birth. Mouth lesions were recorded on days 1, 4 and 27. Instantaneous scan samples of piglet behaviour were carried out for 30 min post-teeth resection procedure (1 min intervals), and for 6 h on days 1, 4, 8, 14, 21 and 26 (5 min intervals). One male and one female piglet per litter were chosen as focal animals and observed for 5 min each post-procedure and for 10 min each twice per day on days 1, 5, 12, 20 and 26. Mortalities were recorded throughout lactation. Grinding took significantly longer than clipping the teeth or leaving them intact ( F = 638.87, P < 0.001). I piglets had higher facial lesion scores than C and G piglets ( F = 10.58, P < 0.001). A smaller proportion of piglets in I litters than C and G litters and a smaller proportion of piglets in G litters than C litters had at least one mouth lesion ( F = 4.74, P < 0.001). During 30 min post-procedure, I piglets were active on the heatpad in more observations than C and G piglets ( F = 3.49, P < 0.05). During 5 min post-procedure C piglets spent longer chomping than I piglets ( F = 5.92, P = 0.05). On day 21, I piglets were active in more observations than G piglets ( F = 2.11, P < 0.05). On day 26, G piglets were inactive in more observations than C and I piglets ( F = 5.02, P < 0.05). On days 14 and 26, C piglets were sleeping in more observations than G piglets ( F = 2.87, P = 0.05). There was a tendency for a larger proportion of I than C piglets to die due to overlying ( F = 2.68, P = 0.08). In conclusion, although all three options were associated with welfare problems, grinding can be recommended in preference to clipping or leaving the teeth intact.