Welfare Economics And Congestion Pricing

Akiva Leeder Law & Economics

Welfare Economics and Congestion Pricing

Introduction Since at least the time of the industrial revolution economists have tried to account for the existence and ideally solve the problem of externalities. Much of this focus has been on the negative externalities associated with pollution and how the free market has or has not been able to adequately address the problem. Despite the general agreement that a problem exists, there is no shortage of opinions regarding why the problem exists in the first place and how to deal with it. This paper will discuss the externalities associated with traffic congestion and analyze one potential solution, congestion pricing, within the framework of welfare economics. Of the many economists who addressed the free market’s failure to compensate for externalities, few have been more influential than Arthur Cecil Pigou and Ronald Coase. While their approaches to the problem of externalities may seem different at first glance, some have argued that many of Coase’s insights are borrowed from and built

 

1 

upon Pigou’s work.1 These two economists’ views have been instrumental in shaping free market policies with regard to externalities. After the problems of externalities are recognized, a further question of how to deal these problems emerges. While both Pigou and Coase offer some guidance, an offshoot of welfare economics known as cost-benefit analysis has developed, providing policy makers with a practical decision-making framework. Though imperfect, costbenefit analysis seeks to solve the problem of externalities by asking “whether society as a whole will become better off by undertaking [a] project rather than not undertaking it, or by undertaking instead any of a number of alterative projects.”2 Policy makers can use welfare economics to identify cases of externalities and explore different methods of addressing them. Solutions can range from relying on the free market to correct itself to government intervention in the form of taxes and/or subsidies. Although it may be argued that policy decisions are ultimately political in nature, welfare economics does provide an important framework for solving the problem of externalities. Even when no active solution is undertaken, perhaps due to political inertia, it should be recognized that this inaction is still a form of policy. Leaving problems unsolved however is only sensible if “it would cost more to remove them than we would gain.”3

                                                         1

See Herbert Hovenkamp, The Coase Theorem and Arthur Cecil Pigou, University of Iowa Legal Studies Research Paper 08-44, April 2009 (arguing that Coase built upon Pigou’s work more than Coase ever acknowledged). 2 E.J. Mishan, Cost-Benefit Analysis, at xxix (4th ed. 1988). 3 RH Coase, ‘Social Cost and Public Policy’, in GA Edwards, ed Exploring the Frontiers of Administration, Toronto (Canada), Bureau of Research, Faculty of Administrative Studies, York University, 1970 at 40.

 

2 

The problem of congestion and the congestion pricing solution

Most commuters in urban environments, and many in suburban environments, have encountered the phenomenon of congestion. Congestion, and more specifically traffic congestion, occurs when traffic demand exceeds roadway capacity.4

Traffic

congestion affects people beyond those immediately involved in any particular manifestation, as it tends to increase transportation costs, which in turn hurts the economy in the form of lost jobs and wasted productive time.5 In addition, traffic congestion adds to costs that are more difficult to measure, such as pollution,

6

public

health (including the hindrance of emergency vehicles), and quality of life.7 The idea to alleviate congestion in urban areas is not new, though the typical approach has been to simply increase public transportation.8 Adding an underground railway, for example, gives commuters an alternative to driving, thus tending to reduce traffic congestion. But, any reduction in congestion solely as a result of adding an underground railroad fails to directly account for the social costs imposed by the driving commuters whose choice to drive contributes to the congestion. The driving commuters would enjoy the benefits in disproportionately high amounts as compared to the rail commuters since the drivers’ costs do not increase.

                                                         4

See http://www.fhwa.dot.gov/congestion/describing_problem.htm (defining congestion in terms of the United States highway system). 5 See Partnership for New York City, Growth or Gridlock? The Economic Case for Traffic Relief and Transit Improvement for a Greater New York (2006). 6 Id at 41. 7 Id at 5. 8 See E.J. Mishan, Cost-Benefit Analysis, 3-5 (4th ed., 1988) (discussing a cost-benefit analysis to the implementation of an underground railway).

 

3 

William Vickrey emerged as an early proponent of congestion pricing to ease traffic congestion in urban areas so as to more accurately account for costs imposed on the transit system by drivers.9 While congestion charges can take many forms, the basic idea is to charge drivers a fee to enter congested urban areas. How much to charge and the details of implementing such a plan as well as how to allocate the proceeds (if any) present additional issues that must be resolved before any such plan should take effect. While there are several examples around the world of successful congestion pricing schemes, none has yet been implemented in a major US city. The mayor of New York City, Michael Bloomberg, proposed such a plan in 2007,10 but the proposal failed to achieve enough support in the New York State Assembly.11 Ultimately it was a political as opposed to economic failure as congestion pricing schemes can be justified by welfare economic theory.

A Pigovian Approach to Externalities Though Pigou did not use the word “externality” in his writings, it is clear that he understood the concept in discussing the “divergences between marginal social net product and marginal private net product.”12 For Pigou, a negative externality exists

                                                         9

See William Vickrey, Pricing in Urban and Suburban Transport, The American Economic Review, (May, 1963) at 460 (discussing the impact of a congestion charge). 10 See Mayor Bloomberg Presents PLANYC: A Greener, Greater New York, http://www.nyc.gov/ (Search “PR- 119-07”). 11 See http://cityroom.blogs.nytimes.com/2008/04/07/congestion-pricing-plan-is-dead-assembly-speakersays/?hp. 12 AC Pigou, Economics of Welfare, 172 (Transaction Publishers 2009)(1920).

 

4 

when marginal private net product exceeds marginal social net product – that is, situations in which “incidental uncharged disservices are rendered to third parties.”13 One example Pigou gives of this phenomenon is that of an “owner of a site in a residential quarter of a city … when he invests resources in erecting buildings in a crowded centre, which, by contracting the air space and the playing-room of the neighborhood, tend to injure the health and efficiency of the families living there.”14 In this example, the costs to the families, in the form of health and efficiency loss are not borne by the builder and thus represent a classic example of a negative externality. In discussing how to deal with these externalities, Pigou notes that simply modifying the contractual relationship between any two contracting parties to a transaction which causes externalities will not work “because the [externality] arises out of a service or disservice rendered to persons other than the contracting parties.”15 However, Pigou notes that the State can remove these externalities by imposing “‘extraordinary encouragements’ or ‘extraordinary restraints.’”16

Specifically, Pigou

notes that the most obvious “encouragements and restraints…[are] those of bounties and taxes.”17 As an example of the implementation of tax to combat negative externalities, Pigou cites the special taxes imposed on businesses producing and distributing alcoholic drinks.18 In this case, the “extra costs in policemen and prisons which [alcohol] indirectly

                                                         13

Id at 185. Id at 185-86. 15 Id at 192. 16 Id. 17 Id. 18 Id. 14

 

5 

makes necessary”19 is paid for by the tax on alcoholic drinks. By using the proceeds to combat the negative externalities produced by the taxed item, the tax works to eliminate the negative externalities altogether. In pigovian terms, this tax would equate marginal private net product with marginal social net product. Such a taxing scheme, often referred to as a pigovian tax, presents the ideal situation where those who cause the negative externalities in the first place are the ones who must pay to eliminate them. The pigovian tax scheme also works in the case of positive externalities where, instead of a tax, bounties are paid to the producers of positive externalities so as to encourage their production. Pigou cites agriculture as an example of an industry that produces a positive externality - “the indirect service of developing citizens suitable for military training.”20 Regardless of whether that benefit holds true today, the idea for Pigou is that since “the private net product of any unit of investment is unduly small,”21 a bounty can be given to encourage investment in those industries, such as agriculture, to increase the amount of the resulting positive externalities. Expanding on his agriculture example, Pigou notes that bounties can take the form of “spreading information about improved processes of production.”22

Since

farmers “lack the appreciation on the part of potential beneficiaries, it would be difficult to collect a fee for undertaking that task.”23 Assuming the agricultural industry does produce this positive externality, giving farmers a bounty in the form free information to increase their production accrues a net benefit to society at large.                                                          19

Id at 186. Id at 193. 21 Id. 22 Id at 194. 23 Id. 20

 

6 

In addition to any tax or bounty scheme, Pigou acknowledges “the Government may find it necessary to exercise some means of authoritative control.”24 He gives the example of city planning where “power must be held by some authority to limit the quantity of building permitted to a given area, to restrict the height to which houses may be carried, for the erection of barrack dwellings may cause great overcrowding.”25 He goes on to argue for the necessity of a government authority to “tackle the collective problems of beauty, of air and of light, as those other collective problems of gas and water have been tackled.”26 Even though Pigou’s “name remains attached to the notion that negative externalities are ubiquitous and that government is systematically superior to markets in reducing the associated social costs,”27 he was weary of the government’s ability to successfully regulate. Pigou was cognizant that the government is subject to “sectional pressure and to personal corruption by private interest,” and that companies “may employ corruption, not only in the getting of their franchise, but also in the execution of it.”28 Furthermore, the fluctuating nature of a representative democracy “may lead to action based on short views – views bounded by the next election, not extending to the permanent interests of the community.”29 It is clear that although Pigou advocates for tax or bounty schemes in order to deal with externalities, he understands that it is the government that must implement them and therefore they either may not happen all, as

                                                         24

Id. Id. 26 Id at 195. 27 A.H. Barnet and Bruce Yandle, The End of the Externality Revolution (20060 at13, available at http://mises.org/journals/scholar/barnett.pdf. 28 Pigou at 332. 29 Id at 334. 25

 

7 

was the fate for New York City’s congestion pricing plan, or it may not be implemented in a way providing the maximum benefit to society.

Pigovian Approach to Congestion Pricing When Pigou was writing about externalities in the early 20th century he did not address the specific problem of urban traffic congestion, but he did address the concept of congestion pricing in his first edition of The Economics of Welfare.30 As laid out by Pigou: Suppose there are two roads ABD and ACD both leading from A to D. If left to itself, traffic would be so distributed that the trouble involved in driving a ‘representative’ cart along each of the two roads would be equal. But, in some circumstances, it would be possible, by shifting a few carts from route B to route C, greatly to lessen the trouble of driving those [sic] still left on B, while only slightly increasing the trouble of driving along C. In these circumstances a rightly chosen measure of differential taxation against road B would create an ‘artificial’ situation superior to the ‘natural’ one.31 In this example, ABD represents a shorter route than ACD such that ‘natural’ situation is for drivers to choose take the shorter route until, due to resulting congestion, the travel time for both roads is equal. Pigou offers a solution to this problem in the form of a tax on the shorter route which would cause some drivers to shift to the longer route so as “to

                                                         30

See Barnet at 7. Robin Lindsey, Do Economists Reach a Conclusion, Econ Journal Watch, at 300 (quoting from the first edition of Pigou’s The Economics of Welfare). 31

 

8 

lessen the trouble of those still left [on the shorter route], while only slightly increasing the trouble of driving along [the longer route].”32 A critic of Pigou, F. H. Knight, responded to the “two roads” problem by proposing a private as opposed to public solution.33 Knight claimed that “if the roads are assumed to be subject to private appropriation and exploitation, precisely the ideal situation which would be established by the imaginary tax will be brought about through the operation of ordinary economic motives.”34 The idea is that if the shorter road was privately owned and the owner could impose tolls, the owner would “charge for its use a toll representing its ‘superiority’ over the free road.”35 The owner will set this toll “up to the point where congestion and diminishing returns set in,”36 and it is this point that “maximize[s] the total product of both roads.”37 Thus Knight was able to propose a solution that did not involve direct government intervention to solve the congestion externality, but rather utilized the free market mechanism. Instead of responding to Knights argument, Pigou simply removed the example from later editions of his book and arguably conceded the point by doing so.38 Despite this, Pigou continued to advocate for government imposed taxes (or bounties) as solutions for externalities that the private market failed to correct. If presented with the problem of urban traffic congestion, Pigou might very well refer to his solution to the "two roads" problem. Since the roads in an urban environment                                                          32

Id. See F.H. Knight, Some Fallacies in the Interpretation of Social Cost, The Quarterly Journal of Economics, Vo. 38, No. 4 (Aug, 1924). 34 Id at 587. 35 Id. 36 Id. 37 Id at 588. 38 See Barnett at 7. 33

 

9 

are a public good and it would be hard to envision Knight’s idea of their private ownership, a properly administered pigovian tax could solve the problem. Pigou would describe the problem of traffic congestion in terms of marginal product - private net product exceeding social net product. That is, the private benefit accruing to the individual drivers exceeds the societal benefit when traffic demand exceeds capacity. Drivers travelling into urban centers under these traffic conditions impose costs on society beyond what they bear themselves. A pigovian solution to the problem would entail a form of tax on the drivers, the proceeds of which would go towards abating the problems caused by the congestion. It must be noted that a congestion problem only occurs when traffic demand exceeds capacity, and this only occurs during certain times of day and only on certain days of the week. Since the problem is that of congestion rather than any one driver’s choice to travel into an urban center, a proper pigovian tax should only target those drivers whose choice adds to the congestion problem. Pigou addressed this notion of variable pricing when writing about railroad rates.39

In addressing services generally, he notes that “though physically similar,

[services] are not necessarily similar in respect of cost when they are rendered at different times or seasons of the year.”40 For example, consumers of electricity supplied during peak hours are charged higher rates than during off-peak hours. In order to supply the electricity required during peak hours, “a large quantity of equipment must be erected

                                                         39 40

 

See Pigou at 290-317. Id at 294.

10 

additional to what would be required if there were no hours or seasons of exceptional demand.”41 Pigou refers to this pricing scheme as the “cost of service principle.”42 Pigou extends this “cost of service principle” to transportation by noting that it “would seem to warrant higher fares for travel at busy seasons and at busy hours of the day than are charged at other times.”43 Further, even if rail ticket prices are not adjusted to reflect the application of this principle, differential charges exist in “a concealed form.”44 Specifically: When a man traveling as a straphanger in the London Tube at 5 o’clock in the evening pays the same absolute price as he does when traveling in comfort at 3 o’clock, he is paying that price for a different and much inferior service. There is just as real a differentiation as there would be if he traveled in equal comfort on both journeys and paid a considerably higher fare at the crowded time.45 This same logic can be applied to a traffic congestion scenario. A driver who travels into an urban center during congested times may pay the same absolute cost (i.e. fuel, wear and tear, etc.) as the driver making the same trip during an un-congested time, except the former pays more in the form of increased travel time and frustration. The application of Pigou’s “cost of service principle” to traffic congestion does not capture the costs imposed by drivers during congested times on others, particularly other commuters.

For example, drivers during congested times not only pay costs

themselves in the form of increased travel time and frustration, but also increase both of these costs on commuters utilizing public transportation in the form of busses.                                                         

In

41

Id (Note Pigou’s analysis assumes that electricity must be produced at the time it is supplied). Id. 43 Id at 295. 44 Id. 45 Id. 42

 

11 

pigovian terms, the marginal private net product of the drivers during congested times exceeds marginal social net product of all commuters. A pigovian response to this divergence between marginal private net product and marginal social net product would be to impose a tax on drivers who travel into urban centers during congested times and use the proceeds to combat congestion. For example, a toll could be imposed on drivers who travel into urban centers during periods of congestion and the proceeds of the toll could be used to expand bus service. Assuming such a toll could be properly administered, it would tend to reduce congestion by preventing some drivers from traveling into urban centers during congested times while offering an alternative to driving in the form of expanded bus services which reduces congestion.

A Coasean approach to Externalities Coase was able to take a different approach to the problem of externalities from that of Pigou by framing the problem in the first instance as a reciprocal one. Coase notes that the traditional approach to dealing with externalities involves the scenario where “A inflicts harm on B” and the question to be decided is “how should we restrain A?”46 Instead, Coase argues that the “real question that has to be decided is, Should A be allowed to harm B or should B be allowed to harm A? The problem is to avoid the more serious harm.”47 By framing the question in this way Coase seeks to maximize social

                                                         46 47

 

RH Coase, The Firm, the Market and the Law, 96 (University of Chicago Press, 1990)(1988). Id.

12 

welfare through an efficient allocation of resources. In this regard, Coase’s objective is similar to Pigou’s goal of maximizing marginal social net product. Unlike Pigou’s system of government imposed taxes and bounties to address externalities, Coase had faith in the free market to deal with these problems on its own so long as an initial determination of rights has been made.48 Moreover, Coase argues “the ultimate result (which maximizes the value of production) is independent of the legal position if the pricing system is assumed to work without cost.”49 Essentially Coase argues that as long as rights with respect to liability are clear (regardless of how they are assigned) and there are no transaction costs involved in the market, the ultimate result of maximization of the value of production will occur naturally through the free market mechanism, eliminating the need for government intervention. While this prospect is certainly intriguing, it relies on the highly theoretical assumption of a market free of transaction costs. Understanding that the assumption of zero transaction costs is “very unrealistic,”50 Coase acknowledges that a “rearrangement of rights will only be undertaken when the increase in the value of production consequent upon the rearrangement is greater than the costs which would be involved in bringing it about.”51 Furthermore, once transaction costs are taken into account, “the initial delimitation of legal rights does have an effect on the efficiency with which the economic system operates.”52 That is, after transaction costs are taken into consideration there is no                                                          48

Id at 104 (arguing that without the establishment of initial delimitation of rights there can be no market transactions to transfer and recombine them). 49 Id. 50 Id at 114. 51 Id at 115. 52 Id.

 

13 

guarantee that the market will be able to bring about the “optimal arrangement of rights, and the greater value of production which it would bring.”53 In certain situations where transaction costs are sufficiently high, Coase accepts that government intervention could “lead to an improvement in economic efficiency.”54 He then offers the example of the “smoke nuisance,” where “a large number of people is involved and when therefore the costs of handling the problem though the market or the firm may be high.”55 While it seems here Coase would be in favor of government regulation, he is quick to point out that a “further alternative” exists; “to do nothing about the problem at all.”56 The idea for Coase is that government regulation may bring about the optimal solution, but it is far from certain. It may be the case where, despite the existence of negative externalities, the best course of action is simply inaction. It is at this point in “The Problem of Social Cost” that Coase is essentially calling for a type of cost-benefit analysis. He argues “all solutions have costs, and there is no reason to suppose that governmental regulation is called for simply because the problem is not well handled by the market.”57 Instead of defaulting to government action, Coase would have economists and policy makers undergo “a detailed investigation of the actual results of handling the problem in different ways.”58 It is in this regard that Coase’s solution to a problem requiring governmental action is similar to that of the cost-benefit analysis championed by E. J. Mishan.59

                                                         53

Id. Id at 118. 55 Id. 56 Id. 57 Id. 58 Id at 199. 59 See E.J. Mishan, Cost-Benefit Analysis, at xxix (4th ed. 1988). 54

 

14 

A Coasean approach to Congestion Pricing The ideal Coasean solution to the problem of traffic congestion would be market based. This would entail a clear delineation of rights and a transaction costless market through which all participants could bargain to reach an optimal solution.

If the

congestion problem were neatly confined to two rural roads, like that of the “two roads” problem, Coase would likely support Knight’s solution. Coase would explain that by clearly defining property rights in the roads and assuming low costs in extracting and enforcing tolls, the free market mechanism would achieve the optimal level of congestion thus eliminating the need for government intervention. This example breaks down, however, when applied to urban traffic congestion. It is hard to conceive of private ownership of city roads, but even this would fail to account for other negative aspects of urban traffic congestion such as pollution. Furthermore, even if property rights could be established so as to account for all the externalities, transaction costs would be prohibitively high considering the large number of people involved. Because of these issues, Coase might treat urban traffic congestion in a similar manner to the problem of smoke nuisance.60 If so, urban traffic congestion may represent an “occasion” where, due to the “large number of people involved,” government regulation may “lead to an improvement in economic efficiency.”61 However, Coase would only support such intervention if it can be shown that the costs associated with such intervention do not exceed the benefits. If the costs were to exceed the potential benefits, Coase would likely advocate for the alternative of doing nothing at all.                                                          60 61

 

Coase at 118. Id.

15 

Coase is generally critical of the pigovian tax scheme as a method of solving externalities. In the context of smoke pollution, Coase argues that the “aim of such regulations should not be to eliminate smoke pollution but rather to secure the optimum amount of smoke pollution, this being the amount which will maximize the value of production.”62 In the context of traffic congestion, Coase would argue that the aim of any congestion charge should not be to rid urban areas of congestion completely but rather to achieve the optimum level of congestion.

It may be the case that some level of

congestion actually produces higher net benefits than a charge sufficiently high that it eliminates congestion altogether. It is clear that Coase would only support a pigovian congestion charge if the empirical data supports the conclusion that the benefits exceed the costs. In Coase’s own words; “it is all a question of weighing up the gains that would accrue from eliminating these harmful effects against the gains that accrue from allowing them to continue.”63 Essentially Coase is calling for a type of cost-benefit analysis before undergoing a program intending to reduce a negative externality.

Cost-benefit analysis As stated by Aaron Wildavksy, “the purpose of cost-benefit analysis is to secure an efficient allocation of resources produced by the governmental system in its interaction with the private economy.”64

The idea behind the analysis is that only

                                                         62

Id at 153. Id at 131. 64 Aaron Wildavsky, The Political Economy of Efficiency: Cost-Benefit Analysis, System Analysis, and Program Budgeting, Public Administration Review (December 1966) at 293. 63

 

16 

projects whose benefits exceed its costs should be undertaken.

To make this

determination, “streams of costs and benefits are discounted so as to obtain the present value of costs and benefits.”65

While seemingly straightforward, assumptions and

judgments must be made in conducting the analysis, many of which amount to educated guesses. A classic example of a cost-benefit analysis involving a major public project is that of an underground railway.66 Mishan frames the analysis as “estimating the initial capital outlays which will be spread over five years against … the stream of future net benefits (gross benefits less operating costs) spread over fifty years.”67

Curiously,

Mishan claims that “there is no difficulty in estimating” these initial costs for such a project despite the fact that cost estimates for construction projects, particularly for public transportation projects, are routinely understated, while the benefits are routinely overestimated.68 Even if construction and operating costs on an underground railway can be accurately estimated, one still has to estimate the benefits, for both expected users of the service and others. For example, “ for some portion of the expected users … the benefits would be calculated as a cost saving as compared with their existing form of travel” and may be “augmented by some estimate of the value of the greater comfort” which

                                                         65

Id. See Mishan 3-5. 67 Id at 3. 68 See Bren Flyvbjerg, How (In)accurate Are Demand Forecasts in Public Works Projects?, Journal of the American Planning Association (Vol.71, No.2 Spring 2005) (arguing that routinely inaccurate cost and benefit estimates for major transportation projects lead to flawed decisions). 66

 

17 

traveling by underground rail might provide.69 Also, there may be some people “that might not travel at all if there were no provision for underground railway services.”70 For those who will not be users of the underground railway but will continue to travel on roads, benefits may include “some saving of time,” as well as “an increase in safety and an increase in comfort.”71 Mishan goes further to point out that even those who never plan on using the new form of transportation may still benefit from “a partial form of insurance” – the railway will be “an alternative that is always open to them should their customary means of travel fail.”72 Mishan cautions, however, not to include among the benefits “any rise in land values resulting from the new underground service,”73 for doing so would be double counting. Mishan argues that the advantages leading to any increase in property value, such as increase “job opportunities, shopping opportunities, or outings,” should have “been entered into the cost-benefit analysis of the railroad on an annual basis.”74 To add the “capital gains,” in the form of property value increases, “would amount to a clear case of double-counting.”75 A Cost-Benefit Analysis of Congestion Pricing To justify the imposition of congestion pricing based on a cost benefit analysis, the present value of the initial and operating costs should be less than the present value of

                                                         69

Mishan at 4. Id. 71 Id. 72 Id. 73 Id. 74 Id at 78. 75 Id. 70

 

18 

the benefits. In his comprehensive study of the London Congestion Charge,76 Jonathan Leape outlined the costs and benefits of the scheme the London Congestion Charge Research Program (LCCRP) took into account in conducting a cost-benefit analysis.77 Costs for the congestion charge were divided up into five categories; the initial set-up costs, the operation costs, the supervisory costs associated with managing the scheme, traffic management costs, and charge-payer compliance costs.78 The benefits included the “time savings to drivers and passengers of vehicles that continue to use the road system after charging is introduced,” “improved journey time reliability,” as well as “reduced accidents and lower carbon dioxide emissions.”79 The net proceeds of the scheme were not included as a benefit, but “given the UK government requirement that all charge revenues be spent on transport improvements, the charge payments are likely to be generating significant additional benefits in reduced travel times and accidents and in other savings.”80 Costs involved with the set-up and enforcement of a congestion pricing plan has long been an attack against the implementation of such schemes. As Vickrey notes in his 1963 essay, “talk of direct and specific charges for roadway use conjures up visions of a clutter of toll booths, an army of toll collectors, and traffic endlessly tangles up in queues.”81 Vickrey accepts that “conventional methods of toll collection” would be costly in terms of “manpower, space, and interference with the smooth flow of traffic,”82                                                          76

See Jonathan Leape, The London Congestion Charge, Journal of Economic Perspectives, (Vol. 20, No.4, Fall 2006). 77 See Id at 171. 78 Id. 79 Id. 80 Id. 81 Vickrey at 457. 82 Id.

 

19 

but he had the foresight to envision technological solutions to these problems. His vision of cars being scanned by roadside equipments and bills being sent out to drivers83 is not so different from London’s enforcement scheme of “video cameras at every entry point… using automatic number plate recognition technology… to identify vehicle registration number[s].”84 The main benefits of a congestion pricing scheme take the form of reducing traffic congestion that would tend to decrease travel time, accidents and pollution while increasing journey time reliability. These benefits are harder to estimate than the set-up and enforcement costs largely because they depend on “assumptions made about the value of time, the value of lower pollution and reduced accident rates.”85 Making difficult assumptions like these is all part of conducting a cost-benefit analysis as noted by Wildavsky; “if certain costs or benefits are deemed important but cannot be quantified, it is always possible to guess.”86 Though seemingly unwieldy in application, computer modeling techniques allow for the easy adjustment of assumptions to create reliable models – this was how the cost-benefit analysis was conducted prior to the implementation of the Congestion Charge in London. Various Congestion Charge Models One of the basic theories underlining a congestion charge is that “a journey should be undertaken only if its value is at least as great as the costs which it imposes on                                                          83

See Id at 458. Leape at 163. 85 Laura Blow, Andrew Leicester & Zoë Smith, London’s Congestion Charge, Briefing Note 31 at 8, The Institute For Fiscal Studies (2003). 86 Wildavsky at 296. 84

 

20 

the community.”87 Since “the congestion problem arises because drivers are not faced with the full costs of their actions, … an obvious solution is to make them pay these external costs.”88 The charge works by making sure “that only those drivers with a valuation of their journey (above private costs already incurred) greater than or equal to the charge continue to travel.”89 This raises the question of what level to set the charge so as to adequately reflect the cost on the community. License Plate Rationing While not a charge per se, license plate rationing programs seek to reduce traffic congestion by only allowing certain vehicles to travel into congested urban areas on certain days of the week. A typical plan “restricts a set of vehicles from entering a specified area on certain days based on the last digit of the vehicle’s license plate.”90 The theory behind such plans is that the restrictions will compel drivers to shift their mode of transportation, change the time of day they travel, carpool, or simply not make the trip at all. If successful, such plans would result in less congestion, thereby deriving similar benefits to those of a congestion charge outlined above. Such plans are usually accompanied by the imposition of fines for violators set at high levels. While one could theoretically avoid the plan simply by paying the fine, the price is usually set sufficiently high and may be coupled with vehicle impounding91 so that drivers seeking to avoid the plan resort to other means. One common avoidance                                                          87

David G. Tipping, Time Savings in Transport Studies, The Economic Journal at 852 (Vol. 78, No. 312)(Dec. 1968). 88 Blow et al at 3. 89 Id. 90 Cambridge Systematics, License Plate Rationing Evaluation, Technical Memorandum ES-1 (Dec. 2007). 91 See Id at 2-5 (noting that Mexico City’s plan penalty includes impounding the violating vehicles for a period of 48 hours).

 

21 

method for those who can afford it is to simply purchase a second vehicle. If the particular plan only distinguishes between odd and even numbered license plates (like that of Athens),92 this method of avoidance is particularly effective. If the second car purchased is of lesser quality and therefore emits more pollution, then the plan may actually end up causing more pollution while not reducing congestion. Furthermore, since the acquisition of a second vehicle to avoid the restrictions may only be available for the wealthy, such plans may disproportionately impact those with lower incomes.93 Though several of these programs are in effect in cities around the world, the long-term effects are questionable and the programs require strict enforcement.94 License plate rationing schemes cause drivers to expend resources avoiding the plan while requiring funds from other sources for the plan’s implementation and thus may be inferior to congestion pricing schemes. London London’s congestion charge imposes a £8 daily fee95 to drive into central London. The charge is a form of “‘area license’ in that the fee effectively buys for the purchaser the right to drive into and out of the charging zone as many times as desired throughout the day, with charges applicable between 7:00am and 6:30pm.96 The charge is the same for all eligible vehicles, except that disable persons, emergency vehicles, and certain alternative fuel vehicles are exempt.97 The charge is enforced through vehicle plate                                                          92

See Id at 2-23. See Id at 4-1. 94 See Id. 95 The charge is £8 if you pay on the day of travel or £10 if you pay the first charging day after travel. 96 Blow et al at 5. 97 See Id at 6. 93

 

22 

recognition technology that captures each vehicle’s plate number, which is then stored in a database and compared at the end of each day with those numbers for which the owners have paid the charge.98 Those who fail to pay receive a fine of £120, which is reduced by a 50 percent discount if payment is made within two weeks or increased to £180 if no payment is received after 28 days.99 It can be argued that a flat blanket charge is not the optimal solution since it does not “internalize the externalities imposed by road users slowing down everyone else.”100 The ideal charge would differentiate “by both the specific route taken and by the time of day, to reflect different levels of congestion in both a geographic and time-specific sense.”101 However, this would require more sophisticated systems and likely increase monitoring costs, though such a scheme is currently place in Singapore. In any event, London’s Congestion Charge has been a success by several measures. The congestion charge has successfully been able to influence “the decisions of road users on various margins: whether to take a particular trip, which mode of transport to use, and when to travel.”102 Since implementing the charge, average travel speeds have increased almost 17 percent and congestion has “dropped an average of 30 percent from the start in February 2003 to mid-2005, at the top end of expectations.”103 Another result was an increase in journey time reliability; surveys showed that the “standard

                                                         98

Id. See Penalties and enforcement, http://www.tfl.gov.uk/roadusers/congestioncharging/6714.aspx. 100 Blow et al at 7. 101 Id. 102 Leape at 165. 103 Id at 166. 99

 

23 

deviation of travel times decreased 27 percent during the morning peak and 34 percent for return journeys.”104 The congestion charge also had a significant effect on public transportation. Though there was no increase in rail trips,105 there was a large increase in bus ridership. From autumn 2002 to autumn 2003 there was an increase in bus passengers of 38 percent, half of which it is estimated is from an improved bus service and half to the congestion charge.106 One explanation for the better than expected increases in bus ridership is due to the so called ‘virtuous circle’: The higher price of rush-hour car travel induces some to switch to public transport, increasing revenues to transport providers. At the same time, reduced congestion leads to increased travel speeds for buses which, in turn, further encourage patronage while also reducing average costs per passenger to transport providers. Increased passenger numbers and reduced average costs enable providers to offer some combination of improved service levels (more routes, higher frequencies) and lower fares. Improved services and reduced fares stimulate further shifts from car travel to public transport, resulting in additional reductions in congestion and gains to public transport.107 Additionally, this ‘virtuous circle’ leads to more reliable bus service, as evidenced by the 30 percent drop experienced in excess waiting times for buses in the charging zone in the first year after the introduction of the charge.108 In a certain sense the congestion charge had been too successful since revenues from the charge have been far lower than estimates. The shortfall in revenues, at about                                                          104

Id at 167. Id at 168 (Leape cites other factors for this including the prolonged closure of the Central Line, downturn in the local economy, the war in Iraq, and the drop off in tourism). 106 Id. 107 Id. 108 Id. 105

 

24 

half the level originally predicted, was “due to the scheme’s greater-than-expected impact on car traffic – a drop of 30 percent in potentially chargeable vehicles entering the zone compared to the mid-point prediction of 20 percent used to project revenues.”109 The drop in revenue was also affected by a higher than expected number of discounted or exempt vehicles, “reflecting, in part, the success of the scheme’s incentives for using low-emission vehicles.”110 The shortfall in revenue was also attributed to higher than expected implementation and enforcement costs. Overall, London’s Congestion Charge has been “both a practical success in reducing congestion and a popular success.”111 Though the benefits “appear to be largely in line with expectations,” the resource costs of running the scheme “have been twice as high as expected.”112 Despite this, the “net benefits of congestion pricing seem to be positive, but less than commonly anticipated.”113 While it can be called a success, London’s experiment shows the pitfalls associated with a cost-benefit analysis – the costs are routinely under-estimated while the benefits are routinely over-estimated. However, London’s experience provides a valuable guide for other major cities considering such a plan by using the results to conduct a more accurate cost-benefit analysis. Singapore Singapore first implemented a paper-based congestion charge scheme (Area License Scheme) in 1975, which was converted to an electronic system (Electronic Road                                                          109

Id at 169-170. Id at 170. 111 Id at 173. 112 Id. 113 Id. 110

 

25 

Pricing) in 1998.114 Singapore’s scheme differs from London’s in a few major respects; each car is fitted with an in-vehicle unit which requires drives to use pre-paid cards, the amount charged varies throughout the day, charges are paid each time a vehicle enters the charging zone, and busses and taxis are not exempt from payments.115 Singapore’s system of fitting each car with an in-vehicle unit is in line with Vickrey’s vision116 and allows the system to adjust the price based on prevailing traffic conditions. Since a driver is charged each time they enter the charging zone, it forces a driver to make the conscious decision whether to incur the payment for each trip, as compared to London where a driver may make unlimited trips in a single day and only make one payment. In this respect, it could be argued that Singapore’s system is more “optimal” than London’s in that is seeks to more accurately “internalize the externalities imposed by road uses slowing down everyone else.”117 Singapore’s original Area Licensing Scheme (ALS) was more like that of London in that it offered unlimited number of entries into the charging zone and only charged one flat rate.118 While the ALS was effective in initially reducing by 44 percent, by 1988 the drop was only 31 percent.119 Since the charge was either $3 or nothing, the system led to “sharp and short peaks of entering traffic volume.”120 A variable pricing system could

                                                         114

Blow et al at 16. Id. 116 See Vickrey at 459. 117 Blow et al at 7. 118 See Kian Keong Chin, Road Pricing Singapore’s Experience at 3 (2002), http://www.imprinteu.org/public/Papers/IMPRINT3_chin.pdf. 119 Id at 5. 120 Id at 6. 115

 

26 

reduce these peaks in traffic, but such a pricing scheme was thought to be too difficult to administer with the manual ALS.121 With the introduction Electronic Road Pricing Scheme (ERP), it was possible to charge variable rates and easier to charge drivers each time they entered the charging zone. As a result, the ERP reduced traffic volume by about 10-15 percent as compared to the manual ALS.122 Much of this reduction is attributed to ERP charging for multiple entries, forcing multiple trip-makers to reduce the number of trips and/or utilize public transportation.123 The ERP system also allows for optimal road use; “when too few vehicles are deemed to be using the roads…the road pricing charge can be reduced to allow more vehicles to use the roads,” and conversely if “too many vehicles are on the roads… the road pricing can be increased.”124 While the effect on drivers’ decision at the margin to make a trip based on the variable price is unclear, in theory such pricing should provide for more optimal road use. While it can be argued that Singapore’s ERP provides a more ‘optimal’ use of roads than that of London’s Congestion Charge, it must be noted that the ERP requirement of each car having its own device adds a considerable amount of cost to the initial set-up costs of a congestion charge scheme. Though these costs may be justified in the long term due to increased efficiency, this would require a separate cost-benefit analysis.

                                                         121

Id. Id at 8. 123 Id. 124 Id at 9. 122

 

27 

New York City

After studying the successes of cities around the world, New York City recently sought to implement its own congestion pricing scheme. The proposed plan would be modeled after London and impose an $8 daily charge for cars and $21 for trucks who enter Manhattan below 86th Street between 6 a.m. and 6 p.m., Monday through Friday. Like the London program, all the revenue raised from the charge would be entirely dedicated to transportation investments.125 Despite various studies126 indicating such a scheme would be successful in reducing congestion and political support from most of the city, the plan failed to win the requisite support among city representatives from outer boroughs and among State Representatives. Since the implementation of the plan required the support of New York State, the lack of support among State Representatives prevented the plan from being implemented. Common among the complaints was the belief that congestion pricing tends to be regressive by imposing a disproportionately higher burden on low-income drivers. Such beliefs have strong political implications even if economic theory does not support them. Indeed some economists, including David Tipping, would claim this is a “silly argument,” since “only the rich can afford to do anything which happens to be expensive. There is no public insistence on subsidizing villas in the South of France.”127 These economists would argue that the congestion charge more accurately reflects the true cost                                                          125

See Transportation Report (2006), http://www.nyc.gov/html/planyc2030/downloads/pdf/report_transportation.pdf. 126 See Id (incorporating figures from various studies). 127 Tipping at 853.

 

28 

of driving into congested areas and therefore if the charge makes a trip too expensive then this is no different from any other good that is too expensive. As New York City’s ill-fated plan makes clear, ultimately the imposition of a congestion charge is not simply an economic decision. As Tipping notes, “it is a matter for political decision, influenced by but not determined by the sums done by economists and engineers.”128 Recognizing this, the London Congestion Charge program made a point to ‘earmark’ revenues of the charge for public transport.129

Specifically,

“earmarking was seen as important in offsetting the potentially regressive effects of the congestion charge which, as a flat-rate charge, imposes a heavier burden on low-income drivers. Thus … earmarking played a crucial role in securing political support for the congestion charge.”130 Though the New York City plan called for such earmarking, it was not enough to convince a majority of the State Representatives that the plan would not be regressive. Conclusion Welfare economics can be used to explain and justify the imposition of a congestion charge in order to reduce traffic congestion.

While Pigou and Coase’s

theories diverged in many respects, both economists’ theories can be used to support the implementation of a congestion pricing scheme. Pigou’s concept of taxing those who cause negative externalities and using the proceeds of the tax to combat the effects of the negative externalities is a rough                                                          128

Id at 854. Leape at 170. 130 Id. 129

 

29 

approximation of what a congestion charge seeks to accomplish. If the charge is set up so that the proceeds are used to provide enhanced bus service, like it is in London, the congestion charge begins to look like a classic pigovian tax. The optimal solution envisioned by Coase, the assignment of rights and letting the free market work itself out, is not feasible in the case of urban traffic congestion. However, this would not stop Coase from blessing a congestion charge as a solution to the problem. Ultimately Coase advocated for a change of approach in dealing with problems of welfare economics; start with a “situation approximating that which actually exists, to examine the effects of a proposed policy change, and to attempt to decide whether the new situation would be, in total, better or worse than the original one.”131 In the end, a public project like that of a congestion charge should past muster under an actual cost-benefit analysis and not be based solely on “blackboard economics.”132 While tending to be complex and requiring assumptions - even outright guesses at times - such analysis is effective in determining whether a project should be undertaken. Before implementing the program in London, the government conducted an extensive cost-benefit analysis that demonstrated the likely benefit of the program. Though it turned out that costs were underestimated and benefits overestimated, the positive benefits predicted in the analysis did materialize and the program is considered a success. It may be argued that, economic theory and cost-benefit analysis aside, the decision to implement a project like a congestion charge is ultimately a political one.                                                          131

Coase at 154. See Coase, Social Cost and Public Policy, at 41-42 (arguing that economic policy conclusions derived from calculations which fail to consider how social institutions work in practice have little relevance for economic policy in practice). 132

 

30 

New York City’s experience is an example of a policy that ultimately failed not because of unsound theory or a negative cost-benefit analysis, but rather due to politics. It must be realized, however, that this failure to implement the congestion charge in New York City was itself a form of policy – one of inaction. Coase would support a policy of inaction if it indeed resulted in the least cost to society. But, if economic theory and cost-benefit analyses show that action, a congestion charge in this case, would result in lower costs to society than inaction, then even Coase would support it. In this regard, a congestion charge represents a welfare economics based solution to the negative externalities imposed by urban traffic congestion that policy makers should take seriously.

 

 

31