The objective of the present study was to compare the performance of Hereford bulls in an insulated tie-stall, an uninsulated barn and a forest paddock in Northern Finland (Ruukki: 64°44'N, 25°15'E). In November 1999, thirty Hereford bulls (age 6.8±0.5 (mean±SD) months; live weight (LW) 285±35 kg) were divided into six groups of five animals according to their LW and the groups were randomly allotted to one of three treatments: tie-stall in an insulated barn (IB bulls, ten animals in individual stalls), pen in an uninsulated barn (UB bulls, 5 animals per pen, two pens) and forest paddock (PAD bulls, 5 animals per paddock, two paddocks). The experiment ended in October 2000, when the bulls were 18.3±0.5 months of age and 772±51 kg LW. During the entire experiment the live weight gain (LWG) of the UB bulls was 7% higher than that of the IB bulls (1436 vs. 1339 g d -1 , p<0.05) and there was a tendency that the LWG of the PAD bulls was 6% higher than that of the IB bulls (1414 vs. 1339 g d -1 , p<0.1). Carcass gain of the UB bulls was 8% higher than that of the IB bulls (812 vs. 751 g d -1 , p<0.05). The carcass conformation score of the PAD bulls was 23% higher than that of the IB bulls (7.6 vs. 6.2 EUROP conformation, p<0.05), but there were no significant effects of treatments on the dressing proportion or carcass fat score. The dry matter and energy intakes (g/kg W 0.75 and MJ/kg W 0.75) of the IB bulls were significantly lower than those of the UB and PAD bulls during both feeding periods (winter: from 2 November 1999 to 25 April 2000, summer: from 26 April 2000 to 16 October 2000) and during the entire experiment. According to this study, finishing beef bulls can be overwintered outdoors in Northern Finland without warm housing facilities. Compared with the tie-stall system, the energy expenditure of walking and other exercise increases in outdoor housing systems, which also means increasing energy intake.