It has been assumed that at the whole organismal level, the mitochondrial reactive oxygen species (ROS) production is proportional to the oxygen consumption. Recently, a number of researchers have challenged this assumption, based on the observation that the ROS production per unit oxygen consumed in the resting state of mitochondrial respiration is much higher than that in the active state. Here, we develop a simple model to investigate the validity of the assumption and the challenge of it. The model highlights the significance of the time budget that mitochondria operate in the different respiration states. The model suggests that under three physiologically possible conditions, the difference in ROS production per unit oxygen consumed between the respiration states does not upset the proportionality between the whole animal ROS production and oxygen consumption. The model also shows that mitochondrial uncoupling generally enhances the proportionality.