Contrary to common misconceptions, public goods are not "goods provided by the public" (read: by the government). Public goods are sometimes supplied by the private sector and private goods - by the public sector. It is the contention of this essay that technology is blurring the distinction between these two types of goods and rendering it obsolete.
Pure public goods are characterized by:
I. Nonrivalry - the cost of extending the service or providing the good to another person is (close to) zero.
Most products are rivalrous (scarce) - zero sum games. Having been consumed, they are gone and are not available to others. Public goods, in contrast, are accessible to growing numbers of people without any additional marginal cost. This wide dispersion of benefits renders them unsuitable for private entrepreneurship. It is impossible to recapture the full returns they engender. As Samuelson observed, they are extreme forms of positive externalities (spillover effects).
II. Nonexcludability - it is impossible to exclude anyone from enjoying the benefits of a public good, or from defraying its costs (positive and negative externalities). Neither can anyone willingly exclude himself from their remit.
III. Externalities - public goods impose costs or benefits on others - individuals or firms - outside the marketplace and their effects are only partially reflected in prices and the market transactions. As Musgrave pointed out (1969), externalities are the other face of nonrivalry.
It is evident that public goods are not necessarily provided or financed by public institutions. But governments frequently intervene to reverse market failures (i.e., when the markets fail to provide goods and services) or to reduce transaction costs so as to enhance consumption or supply and, thus, positive externalities. Governments, for instance, provide preventive care - a non-profitable healthcare niche - and subsidize education because they have an overall positive social effect.
Moreover, pure public goods do not exist, with the possible exception of national defense. Samuelson himself suggested [Samuelson, P.A - Diagrammatic Exposition of a Theory of Public Expenditure - Review of Economics and Statistics, 37 (1955), 350-56]: