In contrast to conventional analyses of monetary union between two par
ticular countries or sets of countries, this paper treats the possible
expansion of a given currency area as a continuous variable ranging f
rom zero to one; zero if there is no expansion and one if all sources
of imports and competition in trade are included in the union. The opt
imal order in which new members are admitted to the union then becomes
a central aspect of the problem. Along with other advantages, this ap
proach makes it easier to defend the argument that many nations are to
o small for form optimal currency areas. The demands upon the origin o
f the shocks, factor mobility, and political organization are smaller.