11. Are Roads Public Goods, Club Goods, Private Goods, or Common Pools?

Minasian (1964) explains, in his criticism of public goods theory, that different institutional arrangements create different incentives for the allocation of resources. He discusses television signals, but consideration of other allegedly public goods has lead to similar criticism (e.g., Coase 1974; Benson 1994, 1998, 2010).

Alternatives may exist (e.g., public roads fully financed by tolls, toll roads provided by private entities that are allowed to set tolls and retain profits), but these two dominated in the United Kingdom.

Non-rivalrous consumption means that even though one person consumes the benefits of the good, everyone else can consume the same undiminished benefits. Non-excludability means that, not only unlimited numbers of people consume the benefits, but no one can be prevented from consuming them even if they do not pay their share of the costs. Free access to a non-rivalrous good creates “free-rider” incentives: individuals recognize that they can consume the benefits without paying, so they will not voluntarily pay for the good, and this means that private producers will not produce the good because they cannot collect revenues to cover costs (or at least, that they will not produce enough of the good because, while everyone can free ride, some may not).

Buchanan (1965) contends that the dichotomy between pure private goods and pure public goods suggested by Samuelson and others is inappropriate. Instead, he suggests that all goods should be considered as club goods where the size of the club depends on the good. In theory, clubs can, in this sense, be as small as one person, or as large as infinity. A large theoretical literature on clubs has developed since the seminal work of Buchanan (1965). See McNutt (2000) for a review. Here, the issue is not how many people can consume a good, but rather, it is how many share ownership and the right to exclude others. This is more in line with the analysis of Ostrom (1990, 2005).

A club good can be rivalrous, in that use by one individual reduces the value somewhat, but not completely, for other users (e.g., congestion). If such costs are born by club members, and therefore, internal to the collective decision-making process of the club, then the club may still provide the good.

They can be related because initially a good or resource can have the characteristics of a public good but given the inevitable congestion that arises with free access, it will become a common pool.

Key theoretical conclusions from the two person game can hold for an N person model (Milleron 1972; Bergstrom et al. 1986; Bernheim 1986; Shitovitz and Spiegel 1998, 2001, 2002). Specifically, both a unique Cournot–Nash and a unique trading equilibrium can exist, and furthermore: (1) a trading equilibrium must be a Pareto optimum for the N individuals involved in the game (the summation of the equilibrium Marginal Rate of Substitution for all N individuals equals 1); (2) the Cournot–Nash solution is not a Pareto Optimum (the Marginal Rate of Substitution \(=\) 1 for some individuals, so the sum of Marginal Rates of Substitution over all N consumers cannot equal one); (3) the total amount of the club good produced in the trading equilibrium is greater than the amount produced in the Cournot–Nash equilibrium; (4) each individual contributes more to club good production in the trading equilibrium than in the Cournot–Nash equilibrium; and most importantly, (5) the trading equilibrium is strongly preferred by all N individuals over the Cournot–Nash equilibrium. Therefore, in theory at least, a non-rivalrous club good can be efficiently produced as a result of the voluntary decisions by the members of a club who then have free access to use the good. This does not mean that the trading equilibrium arises in any institutional setting, however. Institutions that facilitate achieving this are discussed below.

Whether there are net social benefits from expansion depends in part on the consequences of expansion on characteristics of club good itself. At some point, as club membership expands, congestion sets in, so further increases in membership has offsetting marginal effects on members’ utility: the quantity of the club good can increase as membership increases, while its “quality” decreases due to congestion costs, and higher maintenance costs. Thus, the optimal membership is likely to be finite, and for at least some club goods, quite small, although optimal membership for other club goods may be very large Buchanan (1965).

These institutional developments clearly do not exhaust the possibilities, however, and others may be equally or even more important for other times (e.g., where technologies differ), places (e.g., where different social, political or economic conditions apply) or club goods.

Buchanan (1994, p. 132) refers to the resulting arrangement as “ordered anarchy” and explains that “Much of human activity takes place in a setting described as ‘ordered anarchy,’ by which I refer to the simultaneous presence of apparent order and the absence of formal law governing behavior. How is such ordered anarchy possible? ... The answer suggested by my argument here is that interacting parties choose to constrain their separate choices in such fashion as to create non-intersecting and therefore non-conflicting outcomes.” Others use the term, “customary law” to characterize such arrangements (e.g., Pospisil 1971; Fuller 1981; Benson 1988, 1989, 1999a, 2011).

Vanberg and Congleton (1992, p. 421) suggest that another strategy is unconditional cooperation until or unless non-cooperative behavior is confronted, and explicit punishment of the non-cooperative player as exit occurs. They label this strategy “retributive morality,” but such violence is risky, so with competitive options and the ability to spread information, prudent morality tends to be a superior strategy.

Indeed, people often live and interact in many different communities (clubs) with voluntary governance from many different overlapping arrangements (Benson 1999b; Ostrom 2007, 2005).

Various levels of custom can have different content and procedure (Fuller 1981, pp. 241–241):

That the family cannot easily organize itself by a process of explicit bargaining does not mean there will not grow up within it reciprocal expectancies... Indeed the family could not function without these tacit guidelines to interaction... At the midrange, it should be observed that the most active and conspicuous development of [custom]... in modern times lies precisely in the field of commercial dealings. Finally, while enemies may have difficulty in bargaining with words, they can, and often do, profitably half bargain with deeds... That [customary norms are]... at home across the entire spectrum of social contexts does not mean that [they retain]... the same qualities.... At the terminal point of intimacy [custom]... has to do, not primarily with prescribed acts and performances, but with roles and functions.... In the middle area, [custom] ... abstracts from qualities and disposition of the person and concentrates its attention on ascribing appropriate and clearly defined consequences to outward conduct. Finally, as we enter the area of hostile relations.... the prime desideratum is to achieve - through acts, of course, not words - the clear communication of messages of rather limited and negative import; accordingly there is a heavy concentration on symbolism and ritual.

Transactions costs also imply that there are limits to how extensive an inter-group network of cooperation can be, but there are other reasons to expect that these limits can be broken down if it is desirable. After all, as Mises (1957, p. 257) explains, “Man is not the member of one group only and does not appear on the scene of human affairs solely in the role of a member of one definite group. In speaking of social groups it must be remembered that the members of one group are at the same time members of other groups. The conflict of groups is not a conflict between neatly integrated herds of men. It is a conflict between various concerns in the minds of individuals.” For example, a medieval merchant generally was simultaneously a member of the merchant community, a religious organization, and perhaps an urbanized community or neighborhood association, and the geographic dimensions of each varied. Thus, he was in fact familiar with the behavioral rules of several different groups and was in a position to facilitate the development of inter-group ties.

The good could be rivalrous in the sense that congestion occurs with increased use, thereby reducing the benefits for all consumers, as explained below, but for now, assume that it is non-rivalrous, given the size of the community.

The knowledge problem facing a benevolent authority is actually much broader than suggested here (Hayek 1973; Mises 1949). The authority also does not have the knowledge required for efficient production, for instance, as such knowledge is generally dispersed through large numbers of individuals.

Note the analogy to Coase’s theory of the firm (1937) wherein allocations can occur through markets or through firms if the transactions costs of market allocation are higher than the costs of making allocations within a firm hierarchy. This discussion suggests a third choice - cooperation within a club.

Groups relying on hunting tended to develop improvements in technologies for hunting which enhanced their wealth in the short run, but the long-run effect was often quite different. Many migratory animals were hunted into extinction by primitive groups (Ridley 1996, pp. 227–247), for instance, because ownership could not be established until an animal was killed. However, because the members of the group relying on hunting developed new weapons and other inputs to hunting (e.g., domesticated horses, ships), and became skilled in the use of those inputs, they developed a comparative advantage in violence. Carneiro (1970) agrees but adds that successful creation of relatively permanent states of this type occurred where exit by those being subjugated was very difficult due to the surrounding hostile environment (e.g., oceans, desserts, mountains, other hostile communities).

Also see Levi (1988, pp. 10–14).

The resulting incentives for the use of tenuously owned property are similar to those with common pools. The incentives are to consume the property quickly before the government takes it away. Furthermore, incentives to invest in maintenance and improvement are very weak for the same reason - the current owner cannot be confident that he/she will be able to consume the benefits of these investments because the property may be transferred before the benefits are fully realized.

This section draws from and expand on Benson (2006).

There is evidence of a network of roads in Britain prior to the Roman conquest, although much of it is indirect (Jackman 1966, p. 3; Gregory 1931, pp. 45–55).

Also see Gregory (1931, p. 94). These other roads were not funded or maintained by the state (Roman roads that survived into the Middle Ages were not maintained by the state either).

Both can probably be characterized as club goods since cattle were generally held in community pastures to capture scale economies in herding (Dahlman et al. 1980), and policing presumably produced community wide deterrence effects.

See Benson (1998, pp. 198–203) for more detailed discussion of these organizations.

The hundreds were described in some of the king’s early codes, so Lyon (1980, pp. 67, 84) argues that as kingdoms grew kings needed a way to organize local government; thus, they presumably established the tithings and hundreds as local judicial administrative units. However, as Blair (1956, p. 235) points out, such an interpretation is erroneous because it “mistake[s] the nature of Anglo-Saxon legal codes which were not so much concerned with promulgation of new law as with codification of established custom. There is little doubt that the hundred [and tithing] was functioning as a unit” before it appeared in any code.

Above the shire court there was, apparently, a third level of courts “which were, so to speak, hundreds in themselves” (Stephen 1883, p. 67). Note, however, that the higher level courts were not anything like modern courts of appeal. They were simply increasingly inclusive with jurisdictional rules requiring that a dispute be handled by the least inclusive group that encompassed the parties in the dispute. Also note that the rules applied in these courts were customary rather than royal in origin. A panel of jurors with equal numbers of representatives from the separate tithing supervised the trial, which generally involved oath taking (reflecting the important role of reputation in these communities) or ordeal (reflecting the strong religious beliefs of the time).

The Normans did mandate that local freemen travel to royal courts, although there was considerable resistance to such requirements (Benson 1998, pp. 210–212).

Note that the “commercial community” of this period can also be characterized as a club ruled by customary norms, as merchants established their own participatory dispute resolution forums at each market and fair (Benson 1989). The earliest merchant guilds also arose spontaneously, both to provide protection for foreign merchants who were away from their homes, and to protect against unknown foreign merchants who might take advantage of a local merchant and then never return (Milgrom et al. 1990, p. 4).

See also Pawson (1977, p. 68).

This section also draws from and expands on Benson (2006).

The JP office was created in 1326 with a mandate “to keep the peace” (Stephen 1883, p. 190). Appointed by royal commission for each county, JPs were to pursue their duties without monetary compensation. Over thirty statutes instituted between the late fourteenth and the middle of the sixteenth centuries establishing additional functions for JPs, including those dealing with the road maintenance.

See Webb and Webb (1913, pp. 14–26) for more details on this statute and others which followed.

See Albert (1972, p. 8), Darby (1973, pp. 290, 372), and Pawson (1977, pp. 68–69).

Indeed, these cost were often made even higher because the best time of the year for road repairs was also the busiest time of the year for most parishioners since they were engaged in agricultural production (Parkes 1925, p. 9).

Also see Jackman (1966, pp. 48–49).

For detailed discussions of Turnpike Trusts, see Pawson (1977), Webb and Webb (1913), and Albert (1972).

The correspondence between the timing of the turnpike era and the beginnings of the industrial revolution is more than accidental. As Webb and Webb (1913, pp. 143–144) explain,

With the coming the Industrial Revolution, with a rapidly increasing population, with manufactures ready to leap from the ground, with unprecedented opportunities for home and foreign trade, improvement of communication between different parts of the kingdom became, from the standpoint of material property, the most urgent requirement. Today, the railway and the tramway, the telegraph and the telephone, have largely superseded roads as the arteries of national circulation. But, barring a few lengths of canal in the making, and a few miles of navigable river estuaries, it was, throughout the eighteenth century, on the King’s Highway alone that depended the manufacturer and the wholesale dealer, the hawker and the shopkeeper, the farmer, the postal contractor, the lawyer, the government official, the traveller, the miner, the craftsman and the farm servant, for the transport of themselves, and the distribution of their products and their purchases, their services and their ideas.... And all contemporary evidence indicates that, what with the surface-making and embanking, widening and straightening, leveling and bridging, the mileage of usable roads was, by the eighteenth-century Turnpike trusts, very greatly extended.

Indeed, the tremendous increase in economic activity that began during the mid to late 1700s could not have occurred without the simultaneous improvements in transportation. Furthermore, the development of the British railroad system did not really begin until the 1820s as the turnpike system was nearing its peak (Pawson 1977, p. 8), and well after the beginnings of the industrial revolution.

Fletcher (1891) provides a very detailed discussion of the development and technological advances in steam powered road vehicles, including information about both the successful and unsuccessful entrepreneurs and inventors involved.

See Gurney (1831, p. 12), Dance (1831, p. 45), and Dalgleish (1980, p. 117).

See Gurney (1831, p. 18) and Dalgleish (1980, p. 122).

Gurney (1831, p. 12) and Dance (1831, p. 45).

See, for example, Gurney (1831, p. 22) and Dalgleish (1980, p. 117).

Dance (1831, p. 46) also notes that coach proprietors, coachmen, and postboys were in the opposition.

The steam carriage industry did not give up. For instance, at least one group including Thomas Telford initiated an effort to run steam-carriage services on their own improved road between London and Birmingham, with intentions of extending the services beyond this route (Dalgleish 1980, pp. 125–128). This group organized the “Steam Company,” surveyed the route, and gained support from innkeepers and canal operators (who hoped to compete with railroads by connecting with the steam carriages). The railway serving the route objected strongly, but the group apparently was relying of Telford’s prestige to carry them through parliamentary approval. Telford died in September, 1834 however, and the project was abandoned. Yet another initiative by the advocates of steam-powered road travel was the formation of the “Institute of Locomotion for Steam Transport and Agriculture” for the purpose of pursuing the application of steam power to transportation, agriculture and other economic purposes through both economic and political means (Gordon 1833, p. 1). Their political efforts to alleviate the restrictions on steam carriages clearly continued after the 1831 Select Committee report. See for instance, the report of the Select Committee on Mr. Goldsworthy Gurney’s Case in 1834; Gurney was an active advocate and promoter of steam carriage transportation (Gurney 1831), but to no avail. There were also numerous additional efforts to develop steam-powered transportation (Fletcher 1891) but politically imposed limitations also continued to be established, even as tolls were eliminated, as discussed below. A significant blow to the development of horseless road transportation in England came with passage of the Locomotive on Highways Act of 1865, for instance, often referred to as the “Red Flag Law”. The Act stipulated that all self-propelled vehicles on public highways in country areas be limited to a maximum speed of four mph (two mph in towns) and that they be preceded by a man on foot carrying a red flag or lantern. However, this was just one of the many actions taken to prevent the introduction of steam carriages in England. See Fletcher (1891, pp. 279–288) for details. Indeed, he laments that “All the high-speed engines of recent times have been built for service in foreign countries - our foolish and meddlesome laws prohibiting sensible speeds in this country - hence Russia, Greece, Turkey, India, Ceylon, France, New Zealand and Germany are all ahead of Great Britain in this matter” (Fletcher 1891, p. 257). England was far ahead of the rest of the world in the development and improvement of road vehicles (using steam power) at the beginning of the 18th century and the advantage continued for some time, but the political resistance to horseless transport on roads undermined these advantages, leading to a shift in innovative activity to other countries where the internal combustion engine was favored over steam. The early 1900s saw around 125 different manufacturers producing steam cars in the U.S., however, and one, the Stanley Motor Carriage Company, remained very competitive with internal combustion cars in the U.S. until Henry Ford’s development of mass production methods (StanleyMotorCarriage.com 2016, p. 1):

While their car looked similar to most others, their real advantage was simplistic automation. Their early steam engine boasted 13 moving parts with the count for the entire car at 37. It was light, quiet, and perhaps the most powerful vehicle of its time; definitely the fastest. Once lit, the car automatically generated steam to meet demand with little additional attention required except perhaps watching the water level. All that was required of the driver was to set the throttle to a comfortable speed and to move the tiller for steering.

Steam engines are the only engines (or motors) that generate maximum power from rest. Through the simple movement of a lever the power was precisely controlled. With their finicky ignition systems, balky carburetors, and gear-grinding transmissions the “internal explosion engines” as the Stanley’s called them were no match for the simplicity, reliability, and power of steam. The thrashing, banging, clattering, and smell of the internal combustion car was no match for the discernable hiss of a Stanley burner and the sound of tires rolling over stones on the dirt roads of the era. The smell of raw gasoline, partially burned hydrocarbons, along with a mechanical complexity of hundreds of parts for the engine and transmission not to mention the car’s audible noise further tarnished the early image of the gasoline-powered automobile. In America the steam car gained popularity, especially with the rich, and Stanleys became the premier steam cars to own.

Stanleys were the most popular car in the U.S. from 1900 to 1904, and they set numerous land-speed records in January 1906, reaching 127.6 miles/hour on January 26. 86 major models were produced with six different body styles. About 11,000 Stanleys were sold over roughly 25 years, with peak production of 750 in 1907. The company consciously chose not to compete with Ford, however, and in 1914 twice as many Model Ts were produced in a day than the annual output of Stanley Steamers. Model Ts also sold at about 25 % of a Steamers price. The company was sold in 1817 and production ended in 1924.

The Local Government Act of 1858 authorized any parish to become an “Urban Sanitary District,” and these districts could not be included in any larger Highway District (Gregory 1931, p. 195). Parishes whose local officials wanted to maintain control of their roads used this process to do so, and as a consequence, a number of small districts avoided political consolidation.

Pounding of the road surfaces when dry created an unanticipated problem of “waviness” (Gregory 1931, p. 257), the use of “armoured tires” with iron studs on automobiles to prevent side slipping further damaged road services, and ruts were created during rainy weather.

For example, the British experience with toll roads has an American counterpart. The first toll road company in the U.S. was chartered by Pennsylvania in 1792, in order to provide a highway between Philadelphia and Lancaster. 1,562 turnpike companies established over 10,000 miles of roads in the Eastern United States between 1792 and 1845 (Klein and Fielding 1992). As Gunderson (1989, p. 192) notes, “Relative to the economy at the time, this effort exceeded the post -World War II interstate highway system that present day Americans assume had to be primarily planned and financed by the federal government.” Similarly, Klein and Yin (1996) point out that about 150 private toll roads were opened in California between 1850 and 1902. Klein (1990) explains, however, that numerous government mandated toll exemptions for powerful interest groups tended to undermine the incentives to build private toll roads (government regulations often explicitly prevented profit taking, just as in Great Britain).

Many developing countries are franchising roads to private firms which construct the roads and them operate them, charging tolls to earn the costs of construction and operation, and to cover franchising fees paid to the government (Pereyra 2002). Indeed, providing such roads are so attractive, in part because of their impact on real estate values, that it is becoming increasingly common for governments to auction franchises (Engel et al. 2002).

The Intermodal Surface Transportation Efficiency Act of 1991 actually attempted to stimulate privately provided toll roads, bridges and tunnels in the U.S. (as long as they are not part of the interstate highway system) by making them eligible for a 50 % grant from the Highway Trust Fund, and in an effort to take advantage of these available funds a number of states have passed their own legislation to allow private provision of roads. However, some of the largest and most important private highway projects have refused federal funds in order to avoid the added complications that accepting such funding entails.