Free Riders and Collective Action Revisited.

• Author: STROUP, RICHARD L.

The free-rider problem associated with public goods was recognized by David Hume, even before the time of Adam Smith's writings. Each citizen who can enjoy the benefit of a public good has an incentive to try to lay the whole burden of provision on others, whenever the exclusion of nonpayers is very costly or impossible. Hume recommended in 1739 that government provide the goods in question, such as bridges (Musgrave 1985). Two and a half centuries later, economists typically recommend a similar solution (Arrow 1970; Atkinson and Stiglitz 1980; Auerbach and Feldstein 1985; Cornes and Sandier 1986; Nicholson 1989; Samuelson 1954).

The public-provision prescription is seldom questioned, although today's economists and policy analysts, having been exposed to public-choice logic and empirical analysis, do recognize that government is an imperfect institution. Government provision of public goods, it is conceded, will not be free of problems (Shleifer 1998). For example, the rational ignorance of voters is widely recognized, and so too is the disproportionate influence of organized special-interest groups. Lobbyists and their campaign contributions are the facets of the problem that receive the most attention.

Even though problems associated with the imperfection of government are commonly recognized, it is seldom noted explicitly that the root of those problems is precisely the same as that of the free-rider problem associated with private production of public goods. The formation and successful control of a government program in the public interest, for any reasonable definition of that nebulous term, are themselves public goods. Who will pay the price in time, effort, and other lobbying costs to originate a program and to control it in the interest of the general public? Does efficiency in serving the public have a constituency? Adam Smith pointed out long ago that no individual can be expected to seek the public interest. Markets work to exhaust the gains from trade and cooperation because each individual has an interest in finding and capturing any and all such gains. Of course, when free riders can enjoy a public good without payment or trade, production and the potential gains from it may never occur. Efforts to originate government programs and to control them in the public interest are no different. As Gordon Tullock (1971) put it, "The public decision-making process is a procedure for generating a public good; and the persons involved in it, whether they are the voters, judges, legislators, or civil servants, all can be expected to treat it as any other public good" (917).

Tullock recognized the point very clearly.(1) He perceived the likelihood of shirking, in the form of spending too little time and effort in researching the issue subject to public decision and in the form of utilizing the decision maker's own preferences rather than the interest of the public in general. For these reasons, public decisions will not necessarily promote the well-being of the general public. Indeed, a program justified in the name of producing a public good may in fact be utilized by special interests to help only themselves, harming the public in the process. Analyses of programs gone awry are common, but the free-rider problem that surely causes many of these problems is seldom mentioned. Tullock's observations, though published in a prominent economics journal, seem not to have made a large impression on policy-relevant discussions by economists since that time. The problem of free riding seems to this day to be discussed almost entirely in the context of market failure.

One constructive use of Tullock's basic insight can be in systematic side-by-side comparisons of the incentive problems built into private provision of public goods, on the one hand, and those built into every case of public provision of any good (and of public regulation), on the other. This is the sort of comparison called for by James Buchanan (1987), by Kenneth Shepsle and Barry Weingast (1984), and by Neil Komesar (1994). A careful and realistic evaluation of the incentives facing participants in the public policy process--that is, of the free-rider problem inherent in all politically directed public activity--would be useful in comparing alternative institutions whether the output of the public policy were a public good or not, and in determining whether a change in policy might provide a superior result.

Public Provision of Public Goods: Solving the Free-rider Problem or Expanding It?

Economists who discuss public goods and the free-rider problem use many examples. Among the most common are national defense, public health measures such as mosquito abatement, and roads and bridges (Varian 1984, 253; Atkinson and Stiglitz 1980, 486-87; Nicholson 1989, 727). Each is subject to the most common problem cited: the lack of any producer's ability to exclude beneficiaries in a low-cost fashion, a condition that generates the free-rider problem, resulting in an expectation that the good will be underprovided. The standard solutions offered are government provision of the good, through purchase or production, or government subsidization of its private provision. Each, however, introduces many free-rider problems of its own.

The production of goods and services, whether in the private sector or by the government, is a complex undertaking. In meeting the demand for a good, a starting point is to define specifically the quantity and the qualities of the good to be provided.

Decisions on What to Produce and How to Produce It

Consider national defense. What is the proper type and level of national defense? What sort of fleet should the navy build and support, and how large should it be? Where should the ships and their support be based in order to provide the best defense for a given expenditure? Similar questions must be answered about the air force and its airplanes and about the army and its forces.

Each of these decisions has intensely important ramifications for military suppliers and for departments within the military bureaucracy. The interest of each of these groups is likely to be well represented in the decision process, both directly and in lobbying. Members of Congress and relevant members of the current administration will be strongly lobbied on behalf of each supplier group and probably by each bureaucratic department.(2)

Who, on the other hand, will persistently lobby for the diffuse interest of the general public by, for example, identifying and then lobbying for the most cost-effective set of resources, or deployment of those resources, to deter potential foreign aggressors? Citizens who are employed by defense contractors or who live in an area where a defense contract is locally important may become active when their specific issue is being debated. For them, economic benefits for themselves and their localities will loom large, whereas the search for cost-effectiveness will surely be secondary. Certainly the firms, the chambers of commerce, and other organized interests will lobby intensively, often with large budgets to do so, as the narrow issues of specific interest to them are being considered.

Most citizens and most groups, however, are not apt to be involved when a specific defense procurement or deployment issue is settled. Each may well recognize that the many decisions on defense procurement and deployment are important; but for each person, the cost of learning about the issue and becoming involved is borne privately by the citizen, whereas the payoff for making better decisions in the service of cost-effectiveness is spread among the general public. The classic free-rider problem presents itself very strongly, even with respect to decisions about the goals of public provision of national defense.

The same problem appears in complex decisions on government provision of mosquito abatement. Which wetlands should be treated? Should aerial spraying of mosquitos be utilized? Which chemical pesticides are acceptable to reduce mosquito populations? Which is best in each situation? Again, those with income directly at stake or with strong views on these questions will probably be heard. But by whom will the general public's diffuse interests be strongly and persistently represented? Each ordinary citizen is likely to act as a free rider when the level and description of government provision of a public good is being decided. And such decisions are only the beginning of the government provision process.

Control of the Production Process

A large literature in economics shows that in the private sector, where minimizing the cost of producing anything (of a chosen quantity and description) is necessary to maximize the profits of a firm, organizing for cost-minimizing production nonetheless remains a complicated process. Thrainn Eggertsson, in his review and extension of this literature, lists the following activities that typically must be undertaken in the production of a good in a modern economy; each is applicable to production or regulation by the government as well:

1. The search for information about the distribution of price and quality of commodities and labor inputs, and the search for potential buyers and sellers and for relevant information about their behavior and circumstances [Who might be the least-cost provider of each good and service needed?]

2. The bargaining that is needed to find the true position of buyers and sellers when prices are endogenous [What prices can actually be negotiated on behalf of the public?]

3. The making of contracts [What specifications and stipulations should be included to...