The Pareto Criterion and Ethics

22 September 2006 at 7:31 pm David Gordon 1 comment

| David Gordon |

Some economists defend use of the Pareto criterion in welfare economics in this way: Value judgments are subjective, so it would be unscientific for an economist to use them in recommending policy. But the Pareto criterion is a value-free statement. All it says is that if one person in society is made better off by a change, and no one is made worse off, then social welfare has increased. Of course, there is a problem with exclusive reliance on the criterion. Very few changes count as Pareto improvements, and thus situations that intuitively are unjust, such as a regime of slavery, count as Pareto optimal. Nevertheless, it is alleged, in the few cases where a Pareto superior change is possible, we have a value-free reason to support such changes.

This contention seems to me incorrect. The criterion is neutral about the preferences of people in society: it doesn’t say that only certain preferences, and not others, count as increases in social welfare. But preference-neutrality does not make the criterion value-free. The claim that people’s preferences, other things being equal, should be satisfied, is itself a value judgment. Someone could consistently deny it; suppose, e.g., that one thinks it bad that people get what they want, or bad that certain classes of people get what they want. Some people might think it obvious that these opinions are mistaken, but their truth is not here at issue. The point rather is that they, and their denials, are value judgments. If so, the Pareto criterion is a value judgment as well.