Law and Economics Handout #1
Public Goods

1. The Basic Concept

a. Definition: See text.

b. The Four-Good Rectangle: Exclusion and rivalry. See Figure 1.

c. Role of Property Rights: When property rights are created they provide a means of exclusion and therefore move items in the left-hand side of the diagram to the right-hand side of the diagram. Normally, goods are non-exclusive because the transactions costs of exclusion are too high (i.e. the number of parties involved is typically quite high). But ingenious entrepreneurs who discover ways to lower the transactions costs of exclusion will be rewarded with profits.

d. Classifying Goods: Goods are not classified forever in one place; their classification depends on the institutional arrangements and how these develop over time.

e. Is Information a Public Good?
(1) Cooter and Ulen argue that information for producers is a public good because it is non-rival (one person's consumption of a unit of information does not reduce any one else's consumption of that same unit of information) and non-exclusive (once a unit of information is produced and transmitted to another person the recipient cannot be prevented from reselling the information again at relatively low cost). The inability to prevent resale means that producers cannot appropriate the full value of what they produce and will under-produce information. Additionally, potential consumers cannot estimate the value of the information before receiving it. As a result:
(a) Government should either produce all of the information itself (weather forecasting); subsidize private parties to get more of it (basic research); establish property rights in the information (the patent or copyright systems); or regulate the private production of information (such as the FDA or SEC)

(b) All of these actions should produce more information, solving the free-rider problem.

(2) Information is largely a toll good: non-rival and exclusive. Information is an experience good (a good whose value will become apparent only after it has been purchased) instead of a search good (whose value can be ascertained before purchase). As a result, producers/sellers depend on repeat sales and reputation which disciplines them and gives consumers some notion of the value of the information before purchase. Also if information can be transmitted in a relatively excludable manner then there is an incentive to produce information. Note that:
(a) The transmission of information requires that information first be collected, then organized, stored, and then retrieved for transmission; thus significant resources must be devoted to these activities which tends to prevent most free-riders from retransmitting the information. That is, technical constraints deter most free-riding (but not all).

(b) Sellers can claim that some information is proprietary and therefore excludable (see Klein, 1997, p, 108).

(c) Klein (1997) observes that Consumers Union, Credit Bureaus, Underwriters' Laboratories, and the Better Business Bureau all produce information privately without government intervention.

(d) Even if information is a public good in some situations that does not imply that government should intervene and impose some policy. The institutional context within which information is produced also matters. See below.

f. There are Few Public Goods. The following are not public goods (can you explain why?):
(1) Roads

(2) Schools (Education)

(3) Parks

(4) Police services

(5) Courts

(6) Airspace (see text p. 107)

(7) The ocean (see text p. 107)

2. Example of the Public Goods Argument
a. Cooperative solution: 1,000 farmers want to build a dam upstream-i.e., they want to privately produce a public good. (A dam produces flood control which is both non-exclusive-once it is in place, anyone who lives along the river and does not pay for flood control services can still obtain them-and non-rival-one farmer's consumption of flood control services does not prevent other farmers from consuming the same service at the same time.) Notice that 500 farmers have land next to the river on both sides and 500 farmers have land next to the first 500 farmers (meaning they are further from the river and therefore suffer lower damages when the river floods). If all farmers voluntarily contribute then we get:
                    Cost                 Benefits                 Net Benefits
Farmer       of Dam               of Dam                      of Dam
                        1              $600                $1,600                       $1,000
                        2              $600                $1,600                       $1,000
                        3              $600                $1,600                       $1,000
                        .                   .                         .                                 .
                        .                   .                         .                                 .
                        .                   .                         .                                 .
                     500            $600                 $1,600                       $1,000
                     501            $300                    $800                          $500
                     502            $300                    $800                          $500
                     503            $300                    $800                          $500
                        .                  .                           .                                 .
                        .                  .                           .                                 .
                        .                  .                           .                                 .
                  1,000            $300                    $800                          $500

                  Totals      $450,000            $1,200,000                  $750,000

Results: The dam size generating the maximum net benefits is built. Public goods are efficiently produced by private parties.

b. Free-rider solution: Only 100 of the 1,000 farmers actually contributes to the building of the dam with the other 900 being free-riders.

                          Cost                 Benefits                 Net Benefits
      Farmer       of Dam               of Dam                      of Dam

 1             $500                  $700                         $200
  .                 .                          .                               .
  .                 .                          .                               .
  .                 .                          .                               .
100          $500                  $700                         $200
101              $0                  $700                         $700
  .                   .                       .                                .
  .                   .                       .                                .
  .                   .                       .                                .
500             $0                   $700                         $700
501             $0                   $350                         $350
502             $0                   $350                         $350
  .                  .                        .                                .
  .                  .                        .                                .
  .                  .                        .                                .
1,000         $0                   $350                           $350

Totals    $50,000           $525,000                      $475,000

Results: A smaller dam is built with free-riders (a $50,000 size dam which generates $475,000 in net benefits versus $450,000 size dam). It is also possible that the dam won't be built at all. Therefore, flood protection from the dam is under-produced or not produced at all (i.e., produced inefficiently) when public goods are privately produced.

c. Summary: In small numbers situations, it is highly likely that public goods can be efficiently provided by private parties because the transactions costs of achieving agreement are relatively low and the ability to identify free-riders is relatively easy (the cost of being a free-rider in a small group is high). So the small numbers situation should result in the cooperative solution. But in large numbers situations, as above, it is more likely that public goods cannot be efficiently provided by private parties because transactions costs of achieving agreement are high and the ability to identify free-riders is more difficult (the cost of being a free-rider in a large group is relatively low).

3. Solutions to the public goods problem:
a. The Mainstream View-The Free Rider Problem and the Justification of Government Provision: The example of the dam with large numbers of farmers.
(1) High transactions costs prevent the farmers from negotiating among themselves.

(2) Therefore the government uses coercion to lower transactions costs: Government taxes some farmers $600 and other farmers $300 (for a total of $450,000) and eliminates the problem of free-riders.

(3) In general, private producers cannot solve the free-rider problem because they must rely on voluntary exchanges. In contrast, government can solve the free-rider problem by coercing free riders into paying for their share of the public good (through taxes). Therefore, when government finances the provision of public goods, the inefficiency of private provision is significantly reduced or eliminated.

b. A Critique of the Mainstream View: Critics of this view argue that government merely transforms the free-rider problem into a forced-rider problem. The mainstream view assumes that government levies taxes according to the benefits received. That is, if farmers 1 through 500 receive relatively large benefits from flood control but farmers 501 through 1000 receive smaller benefits from flood control, then farmers 1 through 500 should pay higher taxes ($600) while farmers 501 through 1000 should pay lower taxes ($300). But there is no accurate way for government to determine how much each farmer benefits from the provision of flood control. This means that taxes must be levied in a manner that will not match the benefits derived from the public good (such as each farmer paying $450): some farmers will pay too little (easy-riders) while others will pay too much (forced-riders).
(1) The government solution is redistributive: it eliminates free-riders but creates forced riders (those who value public goods less than their tax price) and easy riders (those who value public goods more than their tax price). This is a cross-subsidy from forced riders to easy riders.

(2) Private solutions: Based on entrepreneurial search for least cost solution to lowering transactions costs voluntarily.

(a) General methods used by entrepreneurs: Pre-commitment contracts and tying contracts

(b) Examples:

[1] Private gated communities (See Foldvary, 1994)

[2] Private streets in East St, Louis (See Gage, 1981)

[3] Private roads: Dulles Tollway in Virginia; a private road in Riverside, CA; private roads in 19th century America (See Gunderson, 1989; Klein 1990, 1992, 1993, 1994, and 1996; Roth, 1990 and 1996; Samuel, 1995; Semmens, 1988)

[4] Private police: Private beats in SF, CA; private security in malls; RR police; Starrett Protective Service, etc. (See Blackstone and Hakim, 1996; Benson, 1990, 1994, 1998; Carlson, 1995; Dorffi, 1979; Fixler and Poole, 1992; Gage, 1982; Poole, 1980; Walsh, Donovan, and McNicholas, 1992; Wooldridge, 1970)

[5] Private Courts: Private arbitration services such as AAA; Rent-A-Judge in CA (See Benson, 1990, 1992, and 1995; Fitzgerald, 1988; Gordon, 1985; Poole, 1980; Pruitt, 1982; Wooldridge, 1970)

[6] National Defense: Privateers (Anderson and Gifford, 1991; Sechrest, 2000) and private military companies (Isenberg, 2000).

4. The Importance of the Institutional Setting
a. The Public Interest Approach: This approach assumes that government legislators and bureaucrats will act in the public interest to provide only those public goods which produce net benefits for their principals, the voters. Thus the institutional context for the provision of public goods is irrelevant; all that matters is the intentions of the providers.

b. The Public Choice/Institutional Approach: Political institutions do not have direct, transferable ownership shares. But voters (principals), as a group, elect agents who represent them in the government. Most voters have widely dispersed interests and find it difficult to organize and lobby their agents because the transactions costs of organizing and coming to an agreement are quite high (the large numbers problem). But some voters have a more concentrated interest in a single issue find it easy to organize and lobby their agents (the small numbers situation). This asymmetry in organizing costs gives organized groups a comparative advantage in gathering and interpreting complex information.

(1) Well organized groups will be able to monitor their agents (legislators and bureaucrats) very well because they can discover and understand information concerning a single issue with relative ease (these groups have the resources to hire experts such as lawyers, accountants, etc. who help them understand the impact of complex laws on their actions) and influence the actions of government agents accordingly. Government agents must heed the pressures brought to bear by these organized groups or pay the price and so are very responsive to them.

(2) Unorganized voters face very high transactions costs of discovering and understanding the effects of complex laws and of influencing legislators. As a result, government agents have significantly more slack to pursue their own interests and the interests of organized groups while ignoring the interests of unorganized voters. This also means that elections do not discipline government agents who ignore the wishes of unorganized voters.

(3) Results: Government agents act on behalf of special interests who obtain benefits which are paid for largely by unorganized voters (special interests can externalize costs onto unorganized voters) and who have strong incentives to increase the complexity of rules so that the special interests can achieve a comparative advantage in information discovery (i.e., special interests have a strong incentive to increase asymmetric information problems). Government agents also tend to have a short time horizon because of their lack of property rights-legislators must run for election periodically and tend to support legislation that promises significant short-run benefits that appear before an election (or oppose legislation that imposes significant short-run costs) and bureaucrats will support actions that have short-run benefits (and postpone actions that have short-run costs) because their tenure in office is relatively short (primarily because most bureaucrats will move to the private sector where their government contacts allow them to earn a salary many times what they were earning in government service).

(4) Summary: The previous results (the inefficiencies that arise from the incentive to externalize costs, to widen and exploit information asymmetries, and the incentive to be short-sighted) are all characteristic of the institutional arrangement known as the open access commons.

(5) Political Entrepreneurs have incentives to increase transactions costs for unorganized voters (not lower them) while simultaneously lowering transactions costs for organized groups (so that the latter groups may more readily exploit the former ones).

(6) The inefficiency generated by rent-seeking is quite large. Rent-seeking occurs when organized groups compete for the right to transfer resources to themselves from unorganized voters. Since government agents have control over these transfer rights, organized groups are willing to spend large amounts of resources to convince government agents to transfer wealth to themselves instead of their competitors. Rent-seeking is thus highly inefficient because the resources used to transfer wealth (a zero-sum game) could have been used to create it (a positive-sum game) instead.

(7) Inefficiency of a Monopoly Provider: The lack of competition means that government providers will typically supply a given service at greater cost than private providers who face competition would.

(8) Government agencies do not have a good record when it comes to regulating information. See Calfee, 1992, Rubin, 1995; and Tabarrok, 1999 for the effects of the Food and Drug Administration's regulation of advertising; see Benston, 1983 for the effects of the Securities and Exchange Commission's regulation of information disclosure. Alternatively, private parties have a good record. See Campbell, 2000.

(9) When all of the inefficiencies of government are also considered it is not clear that the private provision of public goods is less efficient than government provision of them.

c. Examples of Private Production of Information
(1) Private Basic Research: Discovery of DNA, penicillin, vaccines for smallpox (Jenner), polio, and pneumonia. (See Kealey, 1996, Kealey and Al-Ubaydli, 2000; Steelman, 1998.)

(2) Private Methods of Protecting Information without Patents: Use of contracts and technological fences. (See Bessen and Maskin, 1997; Desrochers, 2000; William W. Fisher, 1997; Holcombe, 1983; Hunt, 1999; Karn, 1994; Krummenacker, 2000; Merges, 1999; Palmer, 1989; Yu, 1981.)

(3) On the Napster Controversy see Boldrin and Levine; DeLong; Kinsella, 2000; Matlick, 2000a, 2000b, 2000c. 2000d; Steinrich, 2000, Walker, 2000.

(4) On the Problem of Copyright see Cohen, 1999; Friedman, 1999; McElroy, 1985.

NOTE: See Public Goods Bibliography for the references listed above.