See, for example, Hardin (1971), Taylor (1976), Axelrod (1980, 1981, and 1984), and
14
Lichbach (1992)
See Samuelson (1954, 350-51).
15
20
Countless articles and books have discussed the unique qualities of the Prisoner's Dilemma.
Many authors have shown that some level of supply of the public good is a likely (or at least
possible) outcome of N-player, repeated play, or negotiated models of Prisoner's Dilemma.
14
The large latent group in Olson's analysis is least likely to be successful at providing a
public good. The group is composed of individuals which "by definition, cannot make a
noticeable contribution to any group effort, and since no one in the group will react if he makes
no contribution, he has no incentive to contribute." (Olson, 1965: 50) Olson contends that such
latent groups are large relative to their possible contribution to the supply of the good.
Individuals in the group must also have preference orderings for the good which are unlikely to
be affected by strategic interaction.
Permutation #17 (DC>DD>CC>CD) represents the essential qualities of the latent group
member's preference ordering. Like the intermediate preference ordering, DC>CC implies that
the individual's costs of providing the good exceed the benefits derived from its own supply.
Unlike the intermediate preference ordering, DD>CC implies through Corollary 3.1 that costs
also exceed the utility of the good supplied by collective provision. Thus, the latent individual
prefers no good being provided to either sole provision or collective provision of the public
good. Figure 6 illustrates the flat indifference curves that are necessary to obtain this
preference function.
15
Convention