Rodents with mutations in the leptin, or leptin receptor, genes have been extensively used to investigate the regulation of energy balance and the factors that underlie the development of obesity. The excess energy gain of these mutants has long been considered as being due in part to increased metabolic efficiency, consequent to reduced energy expenditure, but this view has recently been challenged. We argue, particularly though not exclusively, from data on ob/ob mice, that three lines of evidence support the proposition that reduced expenditure is important in the aetiology of obesity in leptin pathway mutants (irrespective of the genetic background): (i) milk intake is similar in suckling ob/ob and +/? mice; (ii) ob/ob mice deposit excess energy when pair-fed to the ad libitum food intake of lean siblings; (iii) in several studies mutant mice have been shown to exhibit a lower RMR ‘per animal’ at temperatures below thermoneutrality. When metabolic rate is expressed ‘per unit body weight’ (inappropriately, because of body composition differences), then it is invariably lower in the obese than the lean. It is important to differentiate the causes from the consequences of obesity. Hyperphagic, mature obese animals weighing 2–3 times their lean siblings may well have higher expenditure ‘per animal’, reflecting the costs of being larger and of enhanced obligatory diet-induced thermogenesis resulting from the increased food intake. This cannot, however, be used to inform the aetiology of their obesity.