Introduction
In the last section of this essay the concept of a congestion toll was introduced. It was argued that the optimal toll was given by the size of the congestion externality where the MSC curve intersected the demand curve. It was also suggested that one of the complications in determining the optimal congestion toll was the presence of positive externalities. This issue will now be addressed.
The Positive Externality
The concept of the positive network effect was introduced earlier in this essay. The positive relationship between network size and marginal private benefit formed the basis of the relationship between network size and demand. It was argued previously that whilst ever the marginal private benefit of increased network size was positive, demand would increase. However existing network users also receive a benefit from increased network size, giving rise to a positive externality. If we now assume that there is a corresponding positive externality associated with internet usage1, how does this impact on our optimal congestion toll?
Figure 12 represents a typical internet connection with a given demand curve and a congestion toll in place. The MSB curve has been drawn in above and to the right of the demand curve to represent the external benefits of usage. In the presence of positive externalities the optimal quantity of usage occurs where MSC = MSB. That is, at quantity Q1 of traffic. This was the original private market outcome and thus the positive externality associated with usage has cancelled out the negative externality of usage. This need not always be the case. Figure 13a represents the case where the positive externality is greater than the negative externality whilst Figure 13b represents the reverse situation.
In Figure 13a, the positive externality of the usage outweighs the negative. The socially optimal amount of traffic, given by MSC = MSB, occurs at quantity Q3 of traffic. In order to achieve this level of usage it is necessary to offer a subsidy of P1 - P2.
In Figure 13b, the positive externality is less than the negative externality. The optimal level of traffic in Figure 13b occurs at quantity Q1 of traffic. In order to achieve this level of traffic a congestion toll needs to be imposed. However this toll will be lower than the original toll since it incorporates the positive benefits of usage. Thus the toll need only be P1 - P2, giving the new curve MPC +(toll-benefit).
Conclusion
The determination of the optimal congestion toll is thus complicated by the possible presence of positive externalities to usage. Whether a toll of subsidy is the appropriate policy thus becomes a matter of empirical observation. If the congestion costs are observed to outweigh the positive externalities of usage then a dynamic congestion toll is the appropriate response. If the positive externality of usage is found to outweigh the costs of congestion then the appropriate response is to subsidise usage. It should be remembered that the presence of the positive externality of usage has been assumed in this essay and may or may not exist.2 By contrast, congestion is a recognised problem. (MacKie-Mason & Varian (1997:41); Bohn et. el. (1993:2-3) In the next section of this essay, possible areas where this model may be applied will be discussed, as will areas requiring further research.
Endnotes
1. Usage might be argued to bring about positive externalities through increased communication and dissemination of information. The positive network associated with network size arises because there are more users or web sights to view. The positive benefits of increased usage may arise from greater volumes of content on web sites or users engaging in communication with a greater number of people.
2. Note, the assumption used is that there is a positive externality generated by usage. This is distinct to the positive externalities associated increased network size, explained earlier in this essay. See (
Economides (1996:6);Villasis (1996:60)) .
Last update 12-Jun-98. To visit my personal web site click here.