In this review we attempt to link the efficiency by which animals behave (economy of animal behaviour) to a neuronal substrate and subjective states to arrive at a definition of animal welfare which broadens the scope of its study. Welfare is defined as the balance between positive (reward, satisfaction) and negative (stress) experiences or affective states. The state of this balance may range from positive (good welfare) to negative (poor welfare). These affective states are momentary or transient states which occur against the background of and are integrated with the state of this balancing system. As will be argued the efficiency in behaviour requires that, for instance, satisfaction is like a moving target: reward provides the necessary feedback to guide behaviour; it is a not steady-state which can be maintained for long. This balancing system is reflected in the brain by the concerted action of opioid and mesolimbic dopaminergic systems. The state of this system reflects the coping capacity of the animal and is determined by previous events. In other words, this integrative approach of behavioural biology and neurobiology aims at understanding how the coping capacity of animals may be affected and measured. We argue that this balancing system underlies the economy of behaviour. Furthermore we argue that among other techniques anticipation in Pavlovian conditioning is an easy and useful tool to assess the state of this balancing system: for estimating the state of an animal in terms of welfare we focus on the conditions when an animal is facing a challenge.