CHAPTER 2 THEORETICAL TOOLS OF PUBLIC FINANCE 53

individuals and letting them trade freely. Indeed, this is the Second Funda-
Second Fundamental Theo-
mental Theorem of Welfare Economics: society can attain any efficient rem of Welfare Economics
outcome by a suitable redistribution of resources and free trade. Society can attain any efficient
In practice, however, society doesnt typically have this nice choice. Rather, outcome by suitably redistribut-
ing resources among individuals
as discussed in Chapter 1, society most often faces an equityefficiency and then allowing them to freely
trade-off, the choice between having a bigger economic pie and having a trade.
more fairly distributed pie. Resolving this trade-off is harder than determining equityefficiency trade-off
efficiency-enhancing government interventions. It raises the tricky issue of The choice society must make
making interpersonal comparisons, or deciding who should have more and between the total size of the
economic pie and its distribu-
who should have less in society. tion among individuals.
Typically, we model the governments equityefficiency decisions in the
social welfare function (SWF)
context of a social welfare function (SWF). This function maps the set of A function that combines the
individual utilities in society into an overall social utility function. In this way, utility functions of all individuals
the government can incorporate the equityefficiency trade-off into its deci- into an overall social utility
sion making. If a government policy impedes efficiency and shrinks the eco- function.

nomic pie, then citizens as a whole are worse off. If, however, that shrinkage in
the size of the pie is associated with a redistribution that is valued by society,
then this redistribution might compensate for the decrease in efficiency and
lead to an overall increase in social welfare.
The social welfare function can take one of a number of forms, and which
form a society chooses is central to how it resolves the equityefficiency
trade-off. If the social welfare function is such that the government cares sole-
ly about efficiency, then the competitive market outcome will not only be the
most efficient outcome, it will also be the welfare-maximizing outcome. In
other cases where the government cares about the distribution of resources,
then the most efficient outcome may not be the one that makes society best
off. Two of the most common specifications of the social welfare function are
the utilitarian and Rawlsian specifications.
Utilitarian SWF With a utilitarian social welfare function, societys goal is to
maximize the sum of individual utilities:
SWF U1 U 2 . . . UN
The utilities of all individuals are given equal weight, and summed to get
total social welfare. This formulation implies that we should transfer from
person 1 to person 2 as long as the utility gain to person 1 is greater than the
utility loss to person 2. In other words, this implies that society is indifferent
between one util (a unit of well-being) for a poor person and one for a rich
person.
Is this outcome unfair? No, because the social welfare function is defined in
terms of utility, not dollars. With a utilitarian SWF, society is not indifferent
between giving one dollar to the poor person and giving one dollar to the rich
person; society is indifferent between giving one util to the poor person and
one util to the rich person.This distinction between dollars and utility is impor-
tant because of the diminishing marginal utility of income; richer people
gain a much smaller marginal utility from an extra dollar than poorer people.
With a utilitarian SWF, society is not indifferent between a dollar to the rich