Residual feed intake (RFI) is the difference between an animal’s actual and expected feed intake. Two experiments were conducted comparing energy and nitrogen partitioning in mid-lactation, in Holstein–Friesian cows selected for high or low RFI measured previously as growing calves. Each experiment used 16 cows (8 high-RFI and 8 low-RFI); the first used primiparous (PP) cows and the second used multiparous (MP) cows. Cows were housed individually for 4 days in metabolism stalls, then open-circuit respiration chambers for 3 days. Each cow was offered ad libitum lucerne hay cubes plus 6 kg DM per day of crushed wheat grain. Individual feed intake, milk yield, milk composition and faecal and urine output were measured. Methane and carbon dioxide output and oxygen consumption were measured in the chambers. In MP cows, a greater proportion of energy intake was partitioned to milk and less to heat in low-RFI than high-RFI cows. The proportion of gross-energy intake per kilogram metabolic bodyweight partitioned to milk production was greater and the proportion partitioned to methane and heat production was lower in MP than in PP cows. Energy from tissue mobilisation was not affected by RFI or parity. The amount of nitrogen consumed from feed was greater in MP than PP cows. As a percentage of N intake, N partitioned to milk was greater in PP than in MP cows, but there were no overall effects of RFI on N partitioning. However, there was a trend towards a positive association between N excreted in the urine and RFI, which could have environmental implications. Both RFI and parity were associated with variation in energy and nitrogen partitioning and should be examined in a larger subset of animals in future.