Printer Friendly

Public goods: an ideal concept.

1. Introduction

In this paper I want, first, to present a number of problems encountered with the concept of public good. Economists point to many characteristics for public goods. Some authors take this as an argument that the concept of public good has no validity. Furthermore, several economists hesitate as to whether the concept of public good is an exclusive tag for an economic event or whether one economic event can on the contrary be called at the same time a public and something else too (a private good, or a merit good). Second, I want to offer a solution for the two kinds of problems encountered in the first part of the paper. I will argue that the three economic concepts of public, private and merit good are ideal concepts that can be more or less present in an economic event, that an economic event can embody characteristics of more than one ideal concept, and that the ideal concept of public good consists of two basic characteristics to which the economic profession has pointed by many different descriptions (I counted thirteen of them). I hope my paper will show that the concept of public good has validity because it points to an opportunity for gain if collective action can be taken. However, as no general method exists for financing the collective action in a Pareto efficient way, different societies might choose different approaches to such opportunities for gain. The fact that there exist many culturally different approaches to realize the potential gain embodied in public goods does not make the concept itself culturally relative.

2. Problems with the concept

2.1. Objections from the outside: Malkin and Wildavsky

One of the most radical objections against the concept of a public good was undertaken in a joint article by Jesse Malkin and Aaron Wildavsky entitled "Why the Traditional Distinction Between Public and Private Goods Should be Abandoned" (Malkin and Wildavsky, 1991). The article, which sympathetically quotes libertarian writers, was published in a political science journal. Since libertarians object to most governmental functions, the concept of a public good must for them be placed under suspicion, because its very existence presumes a legitimation of governmental activity. Thus, the fundamental objection of Malkin and Wildavsky seems to me to be that the concept of a public good gives rise to the possibility of justifying governmental functions which they regard as excessive or even illegitimate.

It is my view that Malkin and Wildavsky are right in pointing out that the concept of a public good is sometimes, maybe most of the time, abused in day-to-day political reality. However, they are mistaken when they try to locate this political abuse in the idea that the concept 'public good' itself has no true content but is actually nothing but a social construct (Id.,).(1) For my part, I will use the arguments of Malkin and Wildavsky to articulate the strongest possible objection to the concept of a public good, but then, in opposition to their position, I will argue that the political abuse they observe is caused by a misapplication of that concept.

The part of Malkin and Wildavsky's thesis that is relevant for our study here is the claim that the concept of a public good is a social construct which can therefore have many meanings.

The first observation made by Malkin and Wildavsky is that the specific signifier 'public good' is used for expressing the particular idea under discussion. The choice of this signifier has consequences, they argue, because our language has automatic associations connected with the signifier 'public' (as demonstrated, for example, by the definition of this word in Webster's Third New International Dictionary) (Id., p. 357). Malkin and Wildavsky then quote two economists (Suits and G. L. Bach) who explicitly espouse what is but a connotation of the signifier 'public;' they take for granted that use of this word must indicate that the good or the service has to be provided collectively or by the government if it is to be provided at all. Malkin and Wildavsky point out that only a few economists defend the idea that it is not self-evident that a good must be provided by the government just because it is a "public good."

Now, I agree that the choice of signifier is important for the meaning of a word. This observation, however, is more relevant for words in everyday language than for scientifically defined words. On the other hand, alternative signifiers for the idea under consideration, such as collective good (used by Olson) or social wants (used by Musgrave), are at first sight no improvement. For these reasons, I would prefer to concentrate on the task of looking at the scientific definition of the idea (i.e., the signified) behind the term 'public good.'

Malkin and Wildavsky claim that, speaking descriptively, economists use different definitions for the concept 'public good' (Id., p. 358). A first candidate for a definition of public goods is those goods that have the characteristic of nonrivalness in consumption (Ibid.). Such goods can be simultaneously enjoyed by many people (e.g., clean air), in contrast to private goods, which are marked by rivalness in consumption (e.g., bread). A second candidate for a definition of public goods is those goods that have nonexclusion possibility (Ibid.). This term means that we are confronted with goods that anyone, whether he pays or not, can enjoy (clean air), in contrast to private goods, where property right enforcements prevent consumption if one does not pay (bread). A third strategy for defining public goods uses the two previous characteristics simultaneously (Id., pp. 358-9). Economists who use this strategy differ as to which of the two characteristics they consider the more crucial one. Thus, some take nonrivalness in consumption as crucial and nonexclusion possibility as secondary. On this view, television and radio signals would be public goods, even though technology exists for excluding nonpaying citizens. Other economists opt for nonexcludability as the crucial characteristic. Thus, clean air, which is nonexclusive but rival in that a firm cannot pollute while at the same time others have clean air, would be a public good for these economists. Still other economists argue that both characteristics are essential. They give as examples of public goods lighthouses and national defense.

Malkin and Wildavsky next point to a fourth definition of public goods. That definition uses three essential characteristics; in addition to nonrivalness and nonexcludability it includes impossibility of rejection (Id., p. 360). This term points to the fact that whether consumers like it or not, they must consume the good - for instance, breathe the air, be it polluted or clean.

Malkin and Wildavsky find a fifth definition in Samuelson's strong polar case of a public good, sometimes called the concept of a pure public good. Besides nonrivalness and nonexcludability, Samuelson seems to add equal consumption for all consumers (Ibid.).

At this point Malkin and Wildavsky shift their argument. They ask which goods would be classified as public goods under the different definitions. Clearly, what concrete goods count as public goods will differ from definition to definition. Malkin and Wildavsky make their strongest theoretical claim by building on a mistake made by Samuelson (Id., p. 361). Samuelson had concluded that his concept of a pure public good made it imperative to say that in concrete cases we have many goods which more or less embody the idea of a pure public good with a "knife-edge pole of private-good case, and with all the rest in the public-good domain" (Samuelson, 1969, p. 108). Malkin and Wildavsky now point out that bread certainly is a candidate for the knife-edge pole of the private good case. But, they argue, if we assume that some people enjoy seeing others consume private goods (e.g., seeing poor children eat bread), then these private goods automatically become public goods. Thus, under Samuelson's definition of public goods the distinction between private and public good cannot be maintained. For Malkin and Wildavsky, maintaining the distinction requires drawing a line at some point arbitrarily specified by economists or by the public, say at 90%, 80% or 70% publicness (Malkin and Wildavsky, 1991, p. 364).

Malkin and Wildavsky therefore conclude that a good is not definable as a public good by any objective criteria, or better, by any criteria inherent in the goods themselves. On the contrary, they argue, a good becomes public by the social decision to treat it that way.

2.2. Mistakes made by insiders

2.2.1. Samuelson

Samuelson wrote three very influential articles on public goods (Samuelson, 1954, 1955, 1958). In those articles he introduced the idea of a pure public good, i.e., a good that if provided to one person needs to be made available in equal amounts to all (Samuelson, 1955, p. 350). By introducing the idea of a pure concept he was able to develop two conclusions with logical necessity: that there is an opportunity for collective gain and that individuals are motivated neither to pay for it nor to reveal truthfully their interest in the public good to an agency (the government) if that agency intends to use taxation power to force people to pay according to their benefits (Samuelson, 1955, p. 350; Samuelson, 1954, p. 389). When Samuelson reflected on the applicability of his theory for real policy matters, he correctly understood the nature of his theoretical effort and thus maintained the ideal nature of the concept 'public good.' However, he did not give the concept 'private good' the same status. He was therefore forced into maintaining counter-intuitive conclusions.

Let me quote the passage where Samuelson makes the above reflections.

In my papers I often spoke of 'polar' cases: e.g. the polar case of a 'pure private good' . . . . At the other pole was what I called a 'pure public good' . . . . I did not demur when critics claimed that most of reality fell between these extreme poles or stools, but instead suggested that these realistic cases could probably be analyzed fruitfully as a 'blend' of the two polar cases.

I now wonder whether this was optimal semantics. . . .

Thus, consider what I have given in this paper as the definition of a public good . . . : 'A public good is one that enters two or more persons' utility'. What are we left with? Two poles and a continuum in between? No. With a knife-edge pole of the private-good case, and with all the rest of the world in the public-good domain by virtue of involving some 'consumption externality'. . . .

So I now think the useful terminology in this field should be: pure private goods in which the market mechanism works optimally, and possibly close approximations to them, versus the whole field of consumption-externalities or public goods.

This does, however, lead to an uncomfortable situation. If the experts remain nihilistic about algorithms to allocate public goods, and if all but a knife-edge of reality falls in that domain, nihilism about most of economics, rather than merely public finance, seems to be implied (Samuelson, 1969, pp. 108-9).

As Samuelson himself connects the free market with the concept of private good it seems to me legitimate to clarify the ideal concept of private good by quoting Adam Smith's understanding of the free market.

All systems either of preference or of restraint, therefore, being thus completely taken away, the obvious and simple system of natural liberty establishes itself of its own accord. Every man, as long as he does not violate the laws of justice, is left perfectly free to pursue his own interest his own way, and to bring both his industry and capital into competition with those of any other man, or order of man. The sovereign is completely discharged from a duty, in the attempting to perform which he must always be exposed to innumerable delusions, and for the proper performance of which no human wisdom or knowledge could ever be sufficient; the duty of superintending the industry of private people, and of directing it towards the employments most suitable to the interest of the society (Smith, 1937, p. 687).

If one accepts the notion that the ideal concept 'private good' requires allowing the market to work fully, then the application of that ideal concept to the real world requires allowing the market to work as much as possible. Thus, even though national defense is considered a public good, there is no a priori demand that the government produce that good. Cannons and rifles can very well be produced by the free market and bought by the government through a bidding system. They would thereby remain in some sense also private goods.(2) Similarly, if pollution units are determined, then nothing prevents society from letting the free market find the most effective way of preventing pollution. In my view, it therefore follows that if one accepts the thought of ideal concepts it is wrong to argue that one ideal concept can be applied to a greater or lesser extent (Samuelson's pure public good) and another not (Samuelson's private good).(3)

Furthermore, truth is established first of all at the level of the ideal concept. The problem of seeing correctly how ideal concepts apply to reality is a totally different matter from evaluating an ideal theory or an ideal concept. Samuelson saw correctly the consequences of calling his definition of public goods an ideal concept. Unfortunately, he either did not see that the concept of the free market (or the concept of a private good) is an ideal concept as well or he did not correctly see the consequences of calling it an ideal concept.

2.2.2. Musgrave

Musgrave introduced the concept 'merit good' as a label for a number of economic activities of the government which he did not consider to be public goods proper. His definition of the concept 'merit good' is goods or wants "considered so meritorious that their satisfaction is provided for through the public budget, over and above what is provided for through the market and paid for by private consumers" (Musgrave, 1959, p. 13). He also writes that a merit good "by its very nature, involves interference with consumer preferences" (Ibid.). He gives as examples "free hospitals for the poor or public subsidies to low cost housing" (Musgrave, 1956, p. 341). Musgrave also allows for demerit goods. "[T]heir satisfaction may be discouraged through penalty taxation, as in the case of liquor" (Musgrave, 1959, p. 13), he writes.

The difficulties with Musgrave's position become obvious when one looks at what he sees to be the relation between concrete economic events and the theoretical concepts of private, public, and merit good. In his first writings Musgrave classifies merit goods as public goods ("social wants" in his terminology) because, as he says, "the benefits derived from such services extend beyond the specific beneficiary" (Musgrave, 1957, p. 111). Later on he concedes that merit goods could also be private goods. Thus with reference to merit goods he writes, "Wants are satisfied that could be serviced through the market," and "Separate amounts of individual consumption are possible" (Musgrave, 1959, p. 9). Nevertheless, Musgrave continues to write that some merit goods have many public good aspects. In the German translation and in the English original from the third edition on, Musgrave introduces hesitantly the idea that there could be polar cases of private and public goods (Musgrave, 1959, p. 89).(4) In order to relate the concepts of private, public, and merit goods he proposes two vectors, each with three possibilities. The first vector is the degree of externality or percent of benefit which is social. This first vector is allowed three possibilities: all, part or none of the benefits of the good might be social. The second vector is the degree to which consumer sovereignty applies. This vector is also allowed three possibilities: consumer sovereignty applies either fully, partially or not at all. The case of a good for which vector one indicates that all benefits are social or are externalities, and for which vector two indicates that there is full consumer sovereignty, he calls the case of a 100 percent public good (his category "social want"). The case of full consumer sovereignty in vector two and no degree of externality in vector one is then the case of a 100% private want. The case of full consumer sovereignty in vector two and a partial degree of externality in vector one Musgrave calls a mixed situation with both public (social) and private benefits. All the other cases are merit wants. Musgrave adds without explanation that "the case [merit want situation] at closer inspection frequently proves to be one of social [i.e. public] want" (Musgrave, 1959[Harvard], p. 89. My emphasis). Clearly, Musgrave seems to understand the concepts of private, public, and merit good as tags that can be attached to concrete cases. The polar case nature is introduced only as a surprising result, however, not as a permanent feature. In his publication Fiscal Systems he considers merit goods to be mainly private goods. He writes, "they are quite capable of being subjected to the exclusion principle" (Musgrave, 1969, p. 12).

Clearly, Musgrave is shifting his opinion back and forth on whether merit goods are private or public goods. Yet, it is not that Musgrave does not know the definitions of the concepts of private, public and merit good. Let me suggest that the weakness of Musgrave's view lies in his conception of the nature of the concepts of private, public and merit good. Musgrave seems to work with the idea that these three concepts are tags or boxes which are able to identify concrete economic goods. The concepts are supposed to separate economic activities which have mutually exclusive characteristics. The difficulty for such a conception of the three economic concepts starts when Musgrave notices that economic activities which have to be tagged as public goods (because they are provided for by the government) have a secondary characteristic that does not agree with the definition of public goods (violation of consumer wishes). A new difficulty arises for Musgrave when he sees that goods provided by the government on meritorious grounds can also easily be provided efficiently through the market (medical services and housing). Musgrave ends up accepting the position that merit goods are not a subcategory of either the tag public good or the tag private good, because a merit good can be both a private and a public good.(5)

Now, Musgrave has no difficulty holding on to a fixed definition of the three concepts. I propose that the three concepts are unambiguously different, but that they can apply to the same economic activities. For this to hold the concepts must be considered tags not for real characteristics but for ideal aspects. The three concepts are not like tags that answer the question of whether an object is made from wood, iron or aluminum. These are mutually exclusive. The three concepts are more accurately tags for answering questions like: is this person just, or friendly, or generous? These concepts are not mutually exclusive and apply in degrees. If the three concepts of private, public, and merit good are similar to the concepts of just, friendly, or generous, then a concrete economic activity can be said to possess all three characteristics and to possess them in a greater or lesser degree.

Let us take the example of milk. Milk is a prime example of a private good since it is excludable and is marked by rivalness in consumption. But one could also rightly argue that in some countries milk is a merit good because the state subsidizes it in order to increase its consumption by the poor. Finally, one could argue further that milk is a public good because many people enjoy the idea that children in their country have milk available to them. In this sense, consumption of milk represents a psychic externality and to that extent is a public good. These three arguments can be true simultaneously only if the three concepts of private, public and merit good point to aspects of goods rather than being tags for concrete goods.

The three concepts are in a second way similar to tags like just, friendly, or generous in that they point to less tangible aspects of economic activities or economic goods. For less tangible characteristics we can expect many different descriptions. And indeed, economists have used many words to describe the characteristics of economic events possessing public good aspects. They have also used many words to describe the special relations human beings have with such economic events. I will assemble both the words pointing to characteristics of economic events and the words describing human relations with such economic events. In what follows, I will then argue that all the characteristics of economic events with public good aspects can be classified by means of two characteristics which form the ideal concept 'public good.'

3. Towards a defensible solution

3.1. An enumeration of the characteristics used to classify an economic activity or an economic good as a public good

Malkin and Wildavsky object that according to their reading economists do not give multiple characteristics of economic events having public good aspects, but rather present five different definitions of the concept. If this were the case, then Malkin and Wildavsky would be right in objecting to the economic profession.

I am convinced that a better approach is to look at the many characteristics that economists have observed in goods with public good aspects and then see if one definition can be developed.

In his Public Goods and Public Welfare, (1974)(6) Head enumerates the following characteristics as features used to define public goods: (1) decreasing costs in production, (2) externalities, (3) joint supply, (4) nonexclusion, (5) nonrejectability, (6) benefit spillovers, (7) unenforceability of compensation, (8) indivisibility, (9) nonappropriability, and (10) nonrivalness. Other authors add: (11) free rider possibility (Buchanan, 1975, p. 207), (12) multiple user good (Shoup, quoted in Head, 1974, p. 79, n. 17.), (13) lumpiness (Head, 1974, p. 168). There are, therefore, at least thirteen terms referring to characteristics of economic events with public good aspects. However, some of the thirteen characteristics are obviously related to each other in as much as they point to the same aspect from somewhat different angles. I will therefore group the above-enumerated characteristics with the purpose of arriving at a single definition of the concept 'public good,' a result which, if achieved, would undermine one of Malkin and Wildavsky's attacks on the concept.

3.1.1. Group I. Not internalizing the price of an aspect into the price of the total good

(2) Externalities are costs and/or benefits from consumption or production which are not reflected in market prices (Penguin Dictionary, 1972, pp. 158-9).

(6) Benefit spillovers, according to Head, refer to positive externalities resulting from the provision of services by one jurisdiction which are enjoyed by residents of another jurisdiction (Head, 1974, pp. 270-8). Clearly, this concept is a subcategory of the concept 'externalities.' It restricts the beneficiaries to lower level governmental jurisdictions.

(7) Unenforceability of compensation is understood by Head as the central characteristic of externalities (Head, 1974, pp. 185-6).

(5) Impossibility of rejection is defined by Head as an extreme form of external diseconomy (Head, 1974, p. 83).

3.1.2. Group II. Violation of infinite divisibility theoretically required by the concept 'private good'

(8) Indivisibility: This concept means that certain goods are not available in all desirable quantities, but only in specific sizes (The Penguin Dictionary, 1972, p. 135).(7)

(13) Lumpiness is a synonym for indivisibility (Head, 1974, p. 168).

3.1.3. Group III. One good, many users. Decreasing cost possibilities

(3) Joint supply: This is a confusing term. In order to avoid confusion, one should distinguish between Samuelsonian and Marshallian joint supply. The term Marshallian joint supply refers to a situation where two or more products are necessarily produced by one process, such as meat and wool from sheep (The Penguin Dictionary, 1972, p. 239; Head, 1974, pp. 78-9). This concept does not belong per se in the discussion of public goods. The term Samuelsonian joint supply refers to a situation where because one product can be enjoyed by many, it becomes efficient for consumers to join together in the production process. Samuelsonian joint supply is thus a production reaction to a characteristic of the consumption condition (Head, 1974, p. 77).

(12) Multiple user good: Sharp introduced this term to avoid the confusion that is possible with the term 'Samuelsonian joint supply' (Head, 1974, p. 168; Bird and Head, 1972, p. 4).(8)

(10) The term nonrivalness in consumption means to convey the same characteristic as Samuelson's concept 'joint supply'. The only difference is that this term describes the characteristic from the point of view of consumption and not of the solution in production. This concept also means to convey the same characteristic as the one referred to by Sharp's term 'multiple user good'. The difference is that Sharp describes the characteristic from the point of view of the economic good under consideration, and not from the point of view of consumption of that good.

(1) Decreasing production costs simply refers to the fact that there are economies of scale (The Penguin Dictionary, 1972, pp. 135-7). Head takes decreasing production costs to be only one case of decreasing costs for a consumption unit. Indeed, he argues that the important point is the fact that cost per consumption unit can be diminished because production costs decrease or because more individuals consume the same good, as in the case of nonrivalness in consumption (Head, 1974, pp. 28, 176-9). Thus, if there are more consumers for the same good, or the same kind of good, then in both cases consumers will receive the same good at a lower price.

3.1.4. Group IV. Payment problems: the inability to prevent enjoyment without pay

(7) Nonexclusion is a term used to describe the enviable position of a consumer who can enjoy a product without having to pay for it. This situation arises when a producer or another consumer has no economically sensible method of excluding another consumer from enjoying the good or service without the latter paying his/her share in the good or service that s/he co-consumes.

(11) A free rider is a person who makes use of the advantages of the nonexclusion situation (Buchanan, 1975, pp. 37, 148).

(7) Nonappropriability is a term used to describe this situation from the point of view of an economic good. Head defines it as "that property of a good which makes it impossible for private economic units, through ordinary private pricing, to appropriate the full social benefits (or be charged the full social costs) arising from their production or consumption of that good." (Head, 1974, p. 28)(9)

3.2. Reduction of the many characteristics to a few crucial ones

Let us now reflect somewhat more formally on each of the four groups of characteristics of public goods.

Group I includes externalities, benefit spillovers, unenforceability of compensation, and impossibility of rejection. The relationships among these characteristics can be presented as follows. Externalities can be either positive or negative. In the former case they are often referred to as 'external economies' while in the latter they are often called 'external diseconomies.' An extreme form of external diseconomy is the impossibility of rejection. A particular form of external economy is the benefit spillover of local government actions onto nonresidents or people outside the locality. Externalities are a problem because the price of an aspect of a good cannot be included in the price of the good itself. There is thus a price internalization problem. Until the price internalization problem is solved, there is the problem of unenforceability of compensation. These complex relations are represented in Table 1.

Group II includes indivisibility and lumpiness. These two terms are essentially synonyms.

Group III includes Samuelsonian joint supply, multiple user good, nonrivalness in consumption, and decreasing costs. The first three concepts are essentially synonymous, describing a single characteristic from three points of view. The term 'multiple user good' describes the characteristic under consideration from the point of view of a good which has special features in its consumption possibilities: it can be used by many consumers and thus possesses the characteristic of nonrivalness in consumption. Samuelsonian joint supply is an efficient production strategy for goods with the special consumption feature of nonrivalness in competition.(10) Head proposes the fourth term 'decreasing cost' as the most general term. Decreasing costs can be obtained from the production side and from the consumption side. The former is in the economic literature called 'economies of scale'. The latter is labeled with one of the three essentially synonymous terms mentioned above, i.e., Samuelsonian joint supply, multiple user good, or nonrivalness in consumption. The relationships among the items in Group III are represented in Table 2.

Group IV includes nonexclusion, free rider possibility, and nonappropriability. These terms emphasize the same characteristic from different angles. The term 'nonappropriability' refers to objects which have qualities that cannot be appropriated or internalized because consumers cannot be excluded from their consumption (nonexclusion characteristic), and this allows for the emergence of free riders.

These four groups can now be further combined. Combination A relates Groups II and III. Combination B relates Groups I and IV.

Combination A: Group II and Group III are related as cause to effect. Indivisibility or lumpiness is one of the reasons for economies of scale or for the availability of decreasing costs, i.e., of an opportunity for gain.

Combination B: Group I and Group IV are related to each other because the problem with externalities is at bottom the unenforceability of compensation. This concept is closely related to the concept of nonexclusion or nonappropriability.

Head, however, points to two differences between these two seemingly similar concepts. First, unenforceability of compensation in the case of externalities takes place concerning services which are often of a different nature than the ones that are paid for, as when production of apples co-produces flowers, giving free nectar for the honey industry (Head, 1974, pp. 185-6). Nonappropriability because of nonexclusion possibilities in the enjoyment of pure public goods, on the other hand, refers by definition to the same service. Second, externalities may extend to only one or a few persons; pure public goods extend by definition to all members of the relevant group (Id., p. 186).

In my view, the second difference discussed by Head concerns a feature of public goods which does not touch its essence. Indeed, nothing prevents us from transforming the concept of the relevant group, used by Head for the concept of a pure public good, to the domain of economic events where there are externalities. The number of persons who enjoy the externalities could then be called the relevant group for the externalities.(11)

In my view, the first distinction is also unessential from the point of view of economic optimalization. I see a pure public good as a good that necessarily produces externalities, even if it is privately consumed or provided. Thus, the theorem about the difficulties occurring in the search for a social optimum with pure public goods has dramatic generality precisely because these same difficulties arise with all goods having externalities.

As a consequence, we are left with the idea that the thirteen characteristics can be reduced to two combinations: Combination A and Combination B. Combination A stresses the opportunity for gain resulting from the existence of goods which can be used by many. This characteristic can then be elevated to an ideal level. Instead of stressing that a good can be used by many, we now can say that for that good there is nonrivalness in consumption. Combination B stresses the problem related to the realization of the opportunity for gain: unenforceability of compensation because of nonexclusion possibility. This nonexclusion possibility can be treated as a technical problem, namely, the problem of finding barriers for nonpaid consumption (toll-booths, TV signals that are usable only with a descrambler, taxation schemes). The nonexclusion possibility can, however, also be elevated into an absolute problem, because barriers can not be found or because implementing barriers is too expensive. Thinking of nonexclusion possibility as being without a solution is equivalent to elevating it into an absolute and thus ideal level. These connections are illustrated in Table 3.

Thus the thirteen characteristics by which public goods aspects or problems are described can be reduced to a related pair:(12)

1) Decreasing costs from multiple users thereby offering an opportunity for gain.

2) Unenforceability of compensation because of nonexclusion possibility. This makes financing the opportunity for gain difficult if not impossible.(13)

Samuelson chooses as the primary characteristic for his analysis decreasing cost from multiple users.(14) From this he is able to derive an optimal level of provision.(15) Assuming a positive economic good, the optimal level of provision is bigger than the sum of quantities which the consumers would buy individually. Left to the free market, public goods will in that model be under provided. There thus exists an opportunity for economic gain. According to Head, and I agree with him, Samuelson's statement provides a benchmark for the assessment of market performance.

Samuelson includes the second characteristic (unenforceability of compensation) in his analysis when he looks for a method to realize the opportunity for gain. Samuelson's proposal consists of two steps.(16) First, he asks that the government inquire how much each citizen is willing to pay for a particular public good (e.g., a bridge). If an entrepreneur is willing to provide the public good at a price that is less than the citizens are willing to pay, then there is an opportunity for gain for all in the provision of that good.

Second, the government must use its tax power to force people to pay what they declared they were willing to pay. Samuelson thus uses the government to overcome the unenforceability of compensation. He then points out that citizens will realize that their declared willingness to pay for a public good will be used twice by the government: once to decide whether or not to provide the good and once to decide how much to tax citizens for a particular good. Thus, claims Samuelson, citizens will have a selfish incentive not to reveal their true preferences. Consequently, he concludes, an ideal solution exists, but it cannot be realized by the government.(17)

Samuelson thus argues that the government is capable of dealing with the unenforceability of compensation for public goods. However, says Samuelson, one can not hope, even at the level of the ideal concept, that the government will succeed in using its tax power in such a way that the opportunity for gain is fully realized.(18)

Even at the ideal level, the concept of public good represents a difficult-to-realize opportunity for gain. It is difficult to get the information about the desirability of the public good.

A further difficulty arises when the government decides to separate taxation for the provision of a public good from the benefits received. Ideally, citizens should never be taxed more for a public good than what they receive in benefits.(19) Paying for education by real estate taxes, for example, violates the ideal of public good provision, because it taxes according to ability to pay, not according to benefits received. Failure to apply benefit taxation for payment of public goods violates the ideal provision of public goods. Such a failure is not, I would argue, a failure of the ideal concept. It is a failure of the application.(20)

Olson is an author who does not look for the government to realize the opportunity for gain. His suggestion, though, is that public goods enjoyed by a large group will not be provided if that large group can not organize itself.(21) Such organization is costly and requires payment of dues. And self-interest-motivated economic actors will not pay dues unless the share of the group benefit derived from their dues exceeds the dues themselves.

For a group of N taxpayers that would mean that, on the assumption of equal interest, they would only join a taxpayers lobby group if their dues resulted in group benefits that were at least N times bigger than the actual dues. (For taxpayers that might be 50 million times bigger). As this is unlikely, Olson assumes that such groups will not be successful in organizing themselves.

Other large groups, such as workers or farmers, might only be able to organize themselves if they could provide selective incentives whose benefits outweighed the cost of the dues. If unions are able to get union shop legislation passed, then the private incentives are big enough to motivate the payment of dues.

Olson now argues that the public goods pursued by interest groups are financed by incentives not directly related to these goods.(22) Successful lobbying groups are thus able to over provide public goods. If the beneficiaries of a public good cannot organize themselves, that good will be under provided or not provided at all.(23) Olson is thus able to show that a voluntary approach to public goods provision leads to either feast or famine. Neither, of course, is optimal.

4. Conclusions

1. Conceiving the idea of a public good as an ideal concept allows us to show that there is an opportunity for gain. Samuelson's analysis showed us that the government is not capable of fully realizing that opportunity. Olson's study demonstrates that voluntary provision through group formation also does not guarantee an ideal provision of the good. In short, the ideal concept 'public good' points to the presence of an opportunity for collective gain, but at this level of analysis the realization of that opportunity remains a problem.(24)

2. Samuelson's approach can be used to justify partially the claim by Malkin and Wildavsky that the government cannot be counted on to provide a public good optimally. Olson's approach can be used to justify further the claim by Malkin and Wildavsky that private initiatives can provide some public goods.

3. It is wrong, however, to hope that private initiatives will provide public goods at optimal levels. That is the insight provided by Olson. It is again wrong to suggest that demonstrated difficulties in realizing the opportunities present in public goods justify saying that the concept 'public good' is nothing but a social construct. There are clearly opportunities for gain beyond atomistic economic activities.(25)

4. It is a legitimate question whether private or governmental initiatives are better at realizing those opportunities. Most likely, private initiatives will be better in some cases, while government initiatives will be better in other cases. But to accept that conclusion is to agree that the concept 'public good' is a valid, if problematic, concept for analyzing certain economic problems. Even if the solutions for realizing the opportunities demonstrated by the concept 'public good' have culturally and socially determined components, it is wrong to claim that the very concept 'public good' is merely a cultural construct.


Editorial help was given by Thane Naberhaus, along with the bibliography by Bridget Johnson. Funding was provided by the Georgetown University Center for the Advanced Study of Ethics and the Center for Business-Government Relations of the Georgetown University Business School.


1. Adams and McCormick come to the same two conclusions when they write: "both valid and invalid concepts have been abused in debates about the proper role of government. This, however, is irrelevant in determining the validity of the concepts per se" (1993, p. 114).

2. This observation is often dealt with in public good's discussion by arguing that a public good can either be provided or produced by the government (Ostrom et al, 1991, p. 144).

3. The idea that the concept of public good is an ideal concept is also defended, although in other terminology, by Adams and McCormick (1993, p. 109). Thus they write there: "The concepts of rivalry in consumption and excludability on non-payers are central to discussions of public goods. . . Usually, goods and services are categorized as either rival or non-rival in consumption, but these are only poles on a continuum, with many goods falling in between."

4. See also Andel, 1984, p. 634. What Andel refers to as the "third edition" is actually the edition in which the title page refers to Musgrave as being at Harvard. The earlier edition identifies him as a professor at the University of Michigan.

5. Even though I present here a critical evaluation of Musgrave's hesitation as to whether a merit good is a public or a private good, elsewhere I present a positive appreciation of his insights in the concept of merit good (W. Ver Eecke, 1998).

6. I use this publication because the author has done extensive conceptual work in the area, and has collected his most important papers about public goods in one publication.

7. Head accuses Buchanan of using 'indivisibility' confusingly as a portmanteau tenn for two characteristics: joint supply and impossibility of exclusion (Head, 1974, pp. 78-79).

8. Head also claims that upon occasion some unfortunate terms are used for this characteristic such as: jointness of demand, joint consumption, consumption externality. He even mentions nonrivalness in consumption, which I discuss as a legitimate term (Head, 1974, 78 n. 15).

9. Head also explicitly mentions that it is meant to convey the same problem as Musgrave's "impossibility of exclusion" (1974, p. 28 n. 55, 180).

10. It is very important to see the difference in point of view taken when using these synonyms. Failing to do so easily leads to confusion. See Olson's attempt to relate his concept of exclusive collective good to jointness of supply (Olson, 1968, p. 38 n. 58).

11. This move is in fact made by Samuelson, when he changes his verbal definition of public goods without changing his mathematical model. Thus, in the Biarritz conference he writes: "A public good is one that enters two or more persons' utility" (Samuelson, 1969, p. 108).

12. Ostrom and Ostrom too define the same two characteristics as essential of public goods (1991, p. 165). In attacking the article by Malkin and Wildavsky, Comes and Sandler demonstrate that the same two characteristics can be used as valid criteria by the government for deciding whether or not to provide a good publicly (1994, p. 384).

13. Political scientists are sensitive to the many dimensions involved in implementing a program for capturing the potential gain of public goods. Thus Adams and McCormick write: "Economic efficiency is only one of several considerations which are correctly considered in political decisions. Others include effects on distribution of income and well-being, on freedom and individual liberties and on regional autonomies" (1993, p. 114). Cornes and Sandler point to a more economic problem of implementing public goods provisions when they write: "When a policy maker decides whether to provide a good publicly, the nonrivalry and nonexcludability characteristics may prove to be an important consideration even though other influences, such as transaction costs, come into play" (1994, p. 384). Still, in many cases the difficulty is connected with the need to come to a collective decision about the financing of the public good.

14. "Collective consumption of goods . . . which all enjoy in common in the sense that each individual's contribution of such a good leads to no subtraction from any other individual's consumption of that good" (Samuelson, 1969, p. 387).

15. The mathematical proof is given in Samuelson, 1969, pp. 98-123; the geometric proof is given in Samuelson, 1955, pp. 350-56.

16. Samuelson, 1954. First he looks for 'optimal conditions' (pp. 387-88). Second, he looks for implementation in Section 3: 'Impossibility of decentralized spontaneous solution' (pp. 388-89).

17. "Although the optimum is definable, rational people will not, if left to themselves, be led by an invisible hand to the bliss point. On the contrary, it will pay for each rational man to dissemble, trying to mask his preference for the public goods and to engage in other game strategy maneuvers which, when all do them, will necessarily involve deadweight loss to society" (Samuelson, 1958, p. 334).

18. "[My theory] is in fact an attempt to demonstrate how right Wicksell was to worry about the inherent political difficulty of ever getting men to reveal their tastes so as to attain the definable optimum" (Samuelson, 1955, p. 355).

19. David Schmidtz uses this aspect of public good's implementation to make a philosophical objection to most governmentally provided public goods. He argues that the use of governmental coercion (payment forced upon an unwilling minority) requires additional arguments beyond the public good's argument of efficiency (Schmidtz, 1991: XVI). For Schmidtz the fact that coercive production of public goods would involve the survival of society would be an acceptable argument (Ibid., 159). He mentions that others might use the argument of equality (Ibid., XVI). He also demands that the government use "sufficiently delicate way[s]" to provide public goods (Ibid.,). One such delicate way might be the proposal by Mackscheidt to deal with epidemics. Rather than make inoculation obligatory in order to provide the public good of avoiding the epidemic, he suggests using a merit good technique: i.e., the government should provide inoculation so much below cost that enough previous free-riders consider it in their own self-interest to buy inoculation at the reduced cost in order to prevent an epidemic (Mackscheidt 1997).

20. Such failures, according to Priddat are not an exception, they are the rule since implementation of public goods programs more often than not relies on a majority vote opposed by a minority (Priddat, 1992, p. 246).

21. "[I]n a large group . . . it is certain that a collective good will not be provided unless there is coercion or some outside inducements that will lead the members of the large group to act in their common interest" (Olson, 1968, p. 44).

22. "The common characteristic which distinguishes all of the large economic groups with significant lobbying organizations is that these groups are also organized for some other purpose. The large and powerful economic lobbies are in fact the by-products of organizations" (Olson, 1968, p. 132).

23. "The existence of larger unorganized groups with common interests is therefore quite consistent with the basic argument of this study . . . they also suffer if it is true" (Olson, 1968, p. 167).

24. Further distinctions can be useful for deciding how to implement a program to capture the opportunity for gain present in public goods such as: is exclusion technically feasible in the future and is it economically justifiable. I opted not to analyze that problem. For a study that does just that, see Ostrom and Ostrom 1991, p. 168.

25. Other authors came to a similar conclusion. Thus Comes and Sandler write: "public characteristics may not determine the provision choice. But . . . these properties are a crucial consideration in evaluating allocation" (1994, p. 384). And Adams and McCormick write: "What the government does is ultimately a political decision involving choices about the trade-offs among conflicting objectives. One (but only one) of the considerations is economic efficiency. Public goods theory, carefully stated, can help identify potential economic efficiency gains" (1993, p. 115).


Adams, R. D., and McCormick, K. (1993). The traditional distinction between public and private goods needs to be expanded, not abandoned. American Journal of Political Science, 5(1), 109-116.

Andel, N. (1984). Zum Konzept der meritorischen Guter. Finanzarchiv, 42, 630-648.

Bird, R. M., and Head, J. G. (Ed). (1972). Modern fiscal issues: Essays in honor of Carl S. Shoup. Toronto: University of Toronto Press.

Buchanan, J. M. (1975). The limits of liberty: Between anarchy and Leviathan. Chicago: University of Chicago Press.

Cornes, R., and Sandler, T. (1994). Are public goods myths. Journal of Theoretical Politics, 6(3), 369-385.

Head, J. G. (1974). Public Goods and Public Welfare. Durham: Duke University Press.

Mackscheidt, K. (1997). Letter with course notes. Koln.

Malkin, J., and Wildavsky, A. (1991). Why the traditional distinction between public and private goods should be abandoned. Journal of Theoretical Politics, 3(4), 355-78.

Musgrave, R. A. (1956). A multiple theory of budget determination. Finanzarchiv, pp. 333-343.

Musgrave, R. A. (1957). Principles of budget determination. In Joint Economic Committee (Ed.), Federal Expenditure Policy for Economic Growth and Stability (pp. 108-115). Washington, D.C.: Government Printing Office.

Musgrave, R. A. (1959a). The Theory of Public Finance. New York: McGraw-Hill Book Company ([Musgrave at Michigan]).

Musgrave, R. A. (1959b). The Theory of Public Finance. New York: McGraw-Hill Book Company ([Musgrave at Harvard]).

Musgrave, R. A. (1969). Fiscal Systems. New Haven: Yale University Press.

Olson, M., Jr. (1968). The Logic of Collective Action. New York: Schocken Books.

Ostrom, V., and Ostrom, E. (1991). Public goods and public choices: The emergence of public economies and industry structures. In V. Ostrom (Ed.), The Meanind of American Federalism (pp. 163-197, 276). San Francisco: Institute for Contemporary Studies.

Ostrom, V., Tiebout, C. M., and Warren, R. (1991). The organization of government in metropolitan areas: A theoretical inquiry. In V. Ostrom (Ed.), The Meaning of American Federalism (pp. 137-161, 275). San Francisco: Institute for Contemporary Studies.

Priddat, B. P. (1992). Zur Okonomie der Gemeinschaftsbedurfnise: Neuere Versuche einer ethischen Begrundung der Theorie meritorischer Guter. Zeitschrift fur Wirtschafts-u. Sozialwissenschaften, 112, 239-59.

(G. Bannock, R. Baxter, and R. Rees, Eds.). (1972). The Penguin Dictionary of Economics. Middlesex: Penguin Books Ltd.

Samuelson, P. A. (1954). The pure theory of public expenditure. Review of Economics and Statistics, pp. 387-389.

Samuelson, P. A. (1955, Nov.). Diagrammatic exposition of a theory of public expenditure. Review of Economics and Statistics, pp. 350-356.

Samuelson, P. A. (1958, Nov.). Aspects of public expenditure theories. Review of Economics and Statistics, pp. 332-338.

Samuelson, P. A. (1969). Pure theory of public expenditure and taxation. In J. Margolis and H. Guitton (Eds.), Public Economics (pp. 98-123). London: Macmillan Press Ltd.

Schmidtz, D. (1991). The Limits of Government. An Essay on the Public Goods Argument. Boulder: Westview Press.

Smith, A. (1937). The Wealth of Nations. New York: The Modem Library.

Ver Eecke, W. (1998). The concept of a 'merit good.' The ethical dimension in economic theory and the history of economic thought or the transformation of economics into socio-economics. Journal of Socio-Economics, 27(1), 133-53.
COPYRIGHT 1999 JAI Press, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Eecke, W. Ver
Publication:The Journal of Socio-Economics
Date:Mar 1, 1999
Previous Article:Reactance to taxation: employers' attitudes towards taxes.
Next Article:Two kinds of order: thoughts on the theory of the firm.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters