beyond a certain point of minuteness, the interest would be so small as to elude perception, and would obtain no hold whatever on the human mind.
(Lloyd, 1832, p. 18)
Modern economists have attempted to express the general challenges related to how large groups consider and share public goods. Paul A. Samuelson is one of the pioneers in this field. A question posed by Samuelson (1954) was whether it is possible to calculate the optimal public expenditure by applying an econometric model. He started out with two categories of goods: âordinary private consumption goodsâ and âcollective consumption goodsâ. In a formalized mathematical model, he demonstrates that an optimal public expenditure, defined as a Pareto efficient equilibrium, is not calculable in a decentralized market, or by voting mechanisms. According to Oakland (1969), Samuelson demonstrates how over- or under-production may lead to problems when we deal with both pure and impure public goods. It could be shown that pareto optimal conditions demand that public policy addresses the distribution of these goods (Oakland, 1969, p. 268). The distribution of public goods was addressed by Charles M. Tiebout in 1956. If consumers are mobile, we may see a large number of communities offering different variants of public goods.
The consumer-voter may be viewed as picking the community which best satisfies his preference pattern for public goods.
(Tiebout, 1956, p. 418)
It was assumed that there is an optimally sized community at each location. Consumers will sort themselves into homogenous communities offering the efficient amount of public goods. In these communities a Pareto efficient equilibrium, as described in Samuelson (1954), could be calculated in a decentralized market. It has been left to others to provide more detailed calculations and more realistic assumptions (Batina & Ihori, 2005, p. 311). According to Stiglitz (1982), Tieboutâs conclusion only holds under very special and unreasonable assumptions.
Head (1962) elaborates on the results of Samuelson (1954), when he shows that the discussions on âjointnessâ and âexcludabilityâ of goods are unrelated. Richard Musgrave has since made major contributions in particular by synthesizing earlier contributions in the field (see Sturn, 2010), and Musgrave and Musgrave (1973) and Cornes and Sandler (1996) popularized these insights.
Buchanan (1965) focused on âclub goodsâ which refer to collective goods where exclusion is possible. Later these were labelled âtoll goodsâ in Ostrom and Ostrom (1977). Buchanan found that an optimal relationship between the good and the number of members in a club is calculable given there are certain rules for participation. A key implementation of this is to allow for more flexible property arrangements and to develop exclusion mechanisms. Musgrave and Musgrave (1973, pp. 53â54) summarized the reasons causing market failures of public goods in a table where four categories of goods were determined by two dichotomies âexcludabilityâ/ânonexcludabilityâ and ârival consumptionâ/ânon-rival consumptionâ.
Ostrom and Ostrom (1977) explored the question of how we ought to organize and manage collective goods in general. They referred to the nature of public goods by categorizing public goods in four quadrants determined by two dimensions, first presented in Musgrave and Musgrave (1973): Excludability/non-excludability and joint/individual (rival/non-rival). According to Ostrom and Ostrom (1977) empirical evidence show that small groups and individuals are more successful in providing public goods than large groups.2 This is why people seek governmental institutions. However, government distribution (bureaucracy) could lead to tyranny and difficulties in measuring performance, and a lack of communication between production units and consumers. Ostrom and Ostrom (1977) suggested that public goods are manageable when collective consumption is organized apart from production, and when we apply a market-like arrangement among producers and collective consumption units. This has been viewed as a âthird wayâ apart from private and government property and has received much criticism (Block & Jankovic, 2016).
In the popular text by Hardin (1968), the prerequisites for a sustainable management of commons is discussed. A âcommonsâ is exemplified by a plot of land used for grazing by a herd of livestock. According to Hardin, the rational individual will add animals to the herd to expand their activity, and thereby to increase their revenue. However, this will lead to the soil being depleted, and threaten the livelihood of all. Hardin (1968) claims that the history of human collaboration shows that commons are being abandoned in area after area. He claims that collective management of a commons only succeeds under conditions of low population-density. This view was challenged by scholars (e.g. Ostrom, 1990; Solstad & Brekke, 2011). A common argument is that the capacity of individuals to extricate themselves from various types of dilemmas related to common pool goods is contingent on the institutional environment and that there are many examples of this. Referring to Hardin (1968), Ostrom described a game in which the livestock owners can succeed in making a binding contract to commit themselves to a cooperative strategy that they themselves work out. Based on a number of case studies, Ostrom (1990, p. 90) suggested there are eight design principles that characterize a robust management of a common pool resource. These governance principles ensure broad participation in the governance of these resources and predictability and risk mitigation for the participants.3 Solstad and Brekke (2011) show that a pareto-optimal Nash equilibrium4 is achieved in a context where rational individuals manage a common pool resource, modelled as a two-stage sequential game. First, the harvesting of renewable natural resources takes place. Then the surplus from this stage is used for buying private goods and contributing to public goods. In this setting it is shown that the individuals share the objective of maximizing the total surplus.
In experiment studies, a public good is typically regarded as a collective asset managed by voluntary contributions. These studies include different setups involving participants contributing to the asset in the experiment. Central questions are whether people are more or less cooperative or selfish, and what the factors are that influence their willingness to pay for a good. Experiment studies show how participation and attitudes depend on different design factors surrounding the experiment (Ledyard, 1994), e.g. how the participantsâ cooperation is improved when certain social norms are internalized (e.g. Rege & Telle, 2004). Other studies show that the willingness to pay for a public good cannot always be calculated by aggregating units belonging to a more general category (e.g. linked to territory or time). If a private purchase of the good is conceivable, as in the case of access to fishing resources for example, we tend to see aggregated values more frequently than if private purchasing of the good is unconceivable, as in the case of air traffic controls (Kahneman & Knetsch, 1992).
In general, most of the economic literature on public goods focus on the optimal distribution and the optimal price, and on problems related to collective governance. There is some agreement among economists on the difficulties involved in calculating the optimal expenditure of a public good. And several social scientists are critical to the mere notion that we are able to calculate or govern public goods by referring only to properties of the good itself. According to Cowen (1985) we need to include the institutional framework surrounding the good, and how the good is produced and provided to determine the category of the public good. Stretton and Orchard (1994) argue that the amount and kind of public goods in mixed economies must be ordered and allocated by a mixture of political, administrative, and market choices. Majority rule â directly at the political level, or indirectly at the administrative level â should influence the amount and kind of public goods, not only markets. Malkin and Wildavsky (1991) also believe the institutional and political framework should influence the public good category, but they underline that efforts to categorize public goods are, and must be, a normative exercise; public goods are socially constructed.
Neither of these critiques present a model which incorporates characteristics of both the good itself, and the relevant institutional framework, though they take the position that institutional or political contexts should be recognized when the type of public good is determined. Cowen (1985) points to the lack of institutional variables in such analysis, Stretton and Orchard (1994) argue that elected entities should have a say, while Malkin and Wildavsky (1991) dismiss the notion that we should look for discrete criteria determining public goods categories.
The aim of the multi-layered approach to public goods is to meet the critique of the institutionalists, the levellers, and the social constructivists by suggesting a model that incorporates the main features of the good itself and the relevant institutional framework.
Notes