If there is just a single household (individual) in an economy, there is no problem in evaluating policy changes. We can use this household's utility function (or ask it) to rank different social states, for example, corresponding to different prices. If the household prefers state A to state B (corresponding to, say, other prices than state A), ‘society’ makes the same choice. However, real world economies have millions of households with different preferences (tastes) and incomes. Some households may gain while others may lose from a proposed policy change, such as a move from state A to state B. Therefore, in order to be able to say something about the desirability for society as a whole, one must aggregate preferences so as to arrive at meaningful and operational welfare criteria. This will be the focus of attention for the next few chapters.

The Pareto criterion

The Italian economist Vilfredo Pareto has specified a condition of optimal or efficient allocation referred to as the Pareto condition. By this criterion, a policy change is socially desirable if everyone is made better off (the weak Pareto criterion) or at least some are made better off while no one is made worse off (the strong Pareto criterion). Obviously, when the possibilities of making such policy changes are exhausted, we are left with an allocation of commodities that cannot be altered without someone being made worse off. Such an allocation is called Pareto-optimal or efficient.

The Pareto criterion is often considered to be the common core of welfare economics, but it is indeed a weak core.