Nonrival Goods—Exclusion Undesirable or Inefficient

Private goods also share another important characteristic, namely, that the marginal cost of consumption is nonnegligible. In the case of nonrival public goods, however, the marginal cost of consumption is zero or very low. Once a bridge is built, for example, the marginal cost of letting another car use it is virtually zero, up to the point of congestion. Likewise, the cost of informing 1,000 consumers over the air waves is the same as the cost of informing 2,000. The information available to 1,000 additional consumers does not reduce the amount available to others—the marginal cost of consumption is zero. Although private production of nonrival goods is possible, the private sector will produce suboptimal quantities. Socially optimal pricing requires that the price of goods or services be equal to the marginal cost of consumption. If the price equals marginal cost, private provision may be unprofitable. For an uncongested bridge, for example, optimal pricing would require a very low toll, too low to recover the initial investment and, hence, too low to interest the private sector. If the toll were set high enough to interest the private sector, too few cars would use the bridge. Low marginal cost of consumption is often used as an argument for public provision of research and extension, utility services, and public information services such as agricultural prices and weather patterns. The argument for public involvement in the provision of nonrival public goods is strong, but the nature of the involvement need not be provision of the good, as public funding of private provision may be optimal in many cases. For example, a government may achieve the optimal quantity of research and extension services with public funding of private provision

Complementary Markets

In some cases the production of a good requires the production of a complementary good: computers and computer programs, for example. Software companies flourished only after the advent of personal computers. This example of complementary markets involves only two goods. In some cases, many markets—and large-scale coordination—must be involved. Public intervention in urban renewal programs and rural development have been justified on the grounds of this market failure. The renewal of a large section of a city or the development of rural areas requires extensive coordination among many actors, including factories, retailers, landlords, transport, and so on. Similarly, the development of rural areas requires extensive coordination among various actors. If markets were complete, the coordination would take place through the price system. Incomplete markets require that someone act as coordinator.

Risk Aversion

The public sector, as representative of a country’s entire population, can spread risk over every citizen in the country and is, therefore, in a unique position as an investor. For this reason, Arrow and Lind  argued that when governments act as investors, they should be risk-neutral. They should neither prefer nor avoid risk. Governments should normally choose projects based on their expected net present value and disregard the variance around the mean of the net present value. For private investors, who are normally risk-averse, a tradeoff always exists between risk and return, often expressed as a tradeoff between the variance and the mean. If problems of moral hazard did not exist and insurance markets were complete, private investors would be able to buy insurance against commercial failure and undertake riskier projects. But investors cannot insure against commercial failure and normally shy away from excessively risky projects. The absence of an insurance market against commercial failure and government risk neutrality implies that some risky projects may be attractive to the public sector but not to the private sector If a project is not attractive to the private sector because of a high risk factor, public provision may be justified, even if the project produces a private good.