Neil Komesar

Imperfect Alternatives

In describing political behavior, the most important factors are the per capita stakes and their distribution across various interest groups. Interest groups with small numbers but high per capita stakes have significant advantages in political action over interest groups with larger numbers and smaller per capita stakes.

Higher per capita stakes make it more likely that the members of the interest group will know and understand the issues. In the extreme but not uncommon case, the members of the losing majority (often consumers or taxpayers) do not even have the incentive to recognize that they are being harmed, in some instances, they may even be convinced that they are being aided. Legislation that effectively excludes competition is often cast in terms of consumer health and safety. The majority is not stupid or innately passive. In these instances, the per capita impact on each member of the majority is so low that it does not even justify the expenditure of resources necessary to recognize the issue involved.

But, even if a member of an affected group recognizes the impact of the legislation and his or her per capita benefit exceeds the allocate share of costs of political participation (or even if the per capita benefit exceeds the total costs), one may observe no willingness to contribute from this member and, more importantly, no collective action from this group. Whatever benefits an individual might gain by producing the collective action through his or her contribution, the net benefits to that individual would be even greater if he or she did not have to contribute. There is, therefore, an incentive to refuse to contribute and allow others to hear the costs of political participation; in other words, there is an incentive to free ride.

The severity of the shortfall in the representation of a group depends on the degree or extent to which members of the group free ride. At one extreme, if only a few free ride and the efforts of others take up the slack, there is no underrepresentation. At the other extreme, if all free ride, they will have no political representation and everyone in the group will lose.

The extent or probability of free riding can be related to group size and the characteristics of the distribution of impacts. Small groups have an advantage in collective action because the costs of overcoming free riding and inactivity are lower for them- Small groups can more easily make and police an agreement among the members. Because groups with small numbers and higher per capita stakes are less likely to face free riding problems, they are more likely to produce collective political action and be better represented in the political process.

The same analysis of free riding also suggests that results might differ among large groups depending on the unevenness of the distribution of per capita stakes in the large group. In this connection, George Stigler suggests that a large group with a very uneven distribution of stakes can be analyzed in much the same fashion as the small number group:

The small number solution has a wider scope than a literal count of numbers would suggest. The size distribution of individuals is highly skewed when these individuals have a size dimension (sales of firm, property of family). The large individuals In a group may therefore properly view themselves as members of a small number industry if their aggregate share of the group resources is large.

The analysis points to two aspects of the distribution of per capita stakes that can affect the likelihood of political action by larger groups. First, the higher the average per capita interest of the larger group (albeit still low relative to the average per capita interest of the smaller group), the more likely that members of the larger group will incur such basic expenses as acquiring an understanding of their position. Second, the greater the heterogeneity of the distribution, the greater the likelihood of collective action on behalf of the larger group because of the existence of small, high per capita stakes subgroups.

Thus, the degree and form of political action for both large and small groups depends in good part on cerlain characteristics of the benefits of political action—in particular, the distribution of the per capita benefits of political aclion, including the mean, variance, and skewness of that distribution. The greater Ihe mean per capita stakes the more likely a particular group member will benefit from political activity. Greater skewness and variance indicate pockets of high-benefit group members.

In addition to the level and distribution of benefits, the degree and form of political action also depend on the costs of political participation and, in particular, the costs of overcoming the free-rider problem. Foremost is the cost of information. It shows up in several forms. As we have already seen, one important form of information is the basic recognition of the existence of an interest. The most dormant groups are those whose members do not even recognize a need for political action. The more complex the social issue the more difficult or expensive it is to recognize one's position.