A Companion to Urban Economics - Arnott and McMillen - Part 4

URBAN TRANSPORT ECONOMIC THEORY 241 A Companion to Urban Economics Edited by Richard J. Arnott, Daniel P. McMillen Copyright © 2006 by Blackwell Publishing Ltd P A R T I V Urban Transportation 242 Y. KANEMOTO A Companion to Urban Economics Edited by Richard J. Arnott, Daniel P. McMillen Copyright © 2006 by Blackwell Publishing Ltd Urban Transportation Transportation economics is a well-defined field of economics. Its subfields can be categorized along three dimensions: transport mode, passenger/freight, and supply/demand/policy; maritime freight supply is therefore a subfield accord-ing to this categorization. Supply deals with both technology and industrial organization, and policy subsumes regulation. Urban transportation economics by definition deals with transportation in cities. But, as with urban economics, its core material has been determined by the intellectual history of the subject. Two bodies of urban transport economic theory, the theory of congestion and applied discrete choice theory, are especially import-ant. Both have been extensively adapted to problems in other areas of economics, and the principal developer of each is a Nobel Prize winner – William Vickrey and Daniel McFadden, respectively. The theory of marginal cost pricing has a long and distinguished history. The basic idea is that individuals will make socially efficient decisions if they face the social cost of their decisions and enjoy the social benefits from them. When they do not, there is an uninternalized externality. Arthur Pigou proposed taxation as a way to internalize externalities. The best-known example is that of pollution. When a firm emits pollution that it does not pay for, the marginal social cost of its production exceeds the marginal private cost. Since the firm does not face the full cost of its production, it overproduces. This externality can be corrected by imposing a tax on the firm for each unit it produces, equal to the difference between its marginal social and marginal private costs. Imposition of such a tax will cause the firm to face the full marginal social cost of its production, and hence to make efficient output decisions. William Vickrey was among the first to recognize that this argument applies to congestion in transportation, since an individual’s travels impose a cost on others, by slowing them down, and the individual is not charged for this cost. Pigouvian pricing of the externality in this context is called congestion pricing. Vickrey championed congestion pricing of transportation throughout his life, and most urban transport economists have followed his lead. URBANUTRRBAANNSPORRATNSPCOORNTOAMTIOICN HEORY 243 The economic theory of congestion has three principal components. The pricing of congestible facilities is one. The choice of the capacity or size of congestible facilities is another. The optimal choice of capacity is in principle simple: capa-city should be expanded to the point at which the marginal social benefit from increasing capacity equals the marginal social cost. Combining efficient pricing and efficient capacity yields the “self-financing result” that when a congestible facility exhibits constant long-run average costs, the revenue from the optimal conges-tion toll just covers the cost of constructing optimal capacity. More generally, the efficient subsidy rate is simply related to the degree of returns to scale of the facility. The current wisdom is that freeway travel exhibits more or less constant long-run average cost, while mass transit exhibits decreasing long-run average cost due to economies of density and service frequency. The economic theory of congestion therefore provides a rationale for mass transit receiving a limited sub-sidy from the public coffers, but not freeway travel. These are first-best principles that apply when full efficiency can be achieved. The third principal component of the economic theory of congestion deals with how these principles need to be modified when full efficiency cannot be achieved. A sample question asked in the second-best theory of congestible facilities is how capacity should be chosen when congestion tolls cannot be applied. Yoshitsugu Kanemoto’s essay, “Urban Transport Economic Theory,” provides an exposition of this body of theory. In the early 1970s, Daniel McFadden was asked to forecast the demand for a proposed subway line in San Francisco, which at that time had no subway. Note that modal choice is a discrete or qualitative choice, a choice of which rather than how much. How to proceed? He could have studied the modal split in other cities that had subways, and attempted to apply what that revealed about the demand for subway travel to San Francisco. But he adopted a much more ingeni-ous approach. He reasoned that, on a particular journey, individuals choose the mode that gives them the highest utility, and base their choices on the attributes of each potential mode – travel time, schedule frequency, accessibility, comfort, and so on. He posited that the utility an individual obtains from a mode on a journey is the sum of two components, a systematic component, which is com-mon across individuals in a particular income-demographic group, and an idiosyncratic component capturing unobserved individual attributes and per-sonal tastes. Positing too that the systematic component of utility for individuals in a particular group for a particular journey on a particular mode is a linear function of that mode’s attributes for that journey, and treating the idiosyncratic component as an error term, on the basis of microdata that he had on indi-viduals’ modal choices in the San Francisco Bay area, he was able to estimate the parameters of the systematic component of utility for the various groups in the population. Then, knowing what the characteristics of the subway would be for the various journeys, he was able to estimate the systematic component of utility for subway travel for individuals in a particular group on a particular journey. Then, applying the principle that each individual chooses the mode that provides the highest utility, he was able to forecast the proportion of individuals in each group on each journey who would choose to switch from their current mode to the subway. 244 URBA . TRAANNESMPOORTTOATION The procedure that McFadden developed forms the conceptual basis for all sophisticated travel demand forecasting done today. The details of McFadden’s procedure and how it has been adapted to forecast the effects of alternative congestion pricing schemes, in one study on a private toll road in Orange County and in another on the San Francisco Bay bridges, is the topic of “Urban Passenger Travel Demand,” by André de Palma, Robin Lindsey, and Nathalie Picard. It was mentioned in the Preface that one of the major developments that led to the creation of urban economics as a field was the transportation and land-use studies of the 1950s and 1960s. Interest in the relationship between urban trans-portation and land use has continued unabated. Transportation availability strongly affects land use, but land use also strongly affects transportation demand. Forecasting land use is the weakest link in urban traffic demand forecasting, and has been made more difficult by the suburbanization of employment and the increased importance of noncommuting trips. John McDonald’s essay, “Urban Transportation and Land Use,” examines how economists go about forecasting the effects of urban transportation improvements on land use. Urban transport policy generates heated debate. One reason is that it so directly affects people’s lives, another that winners and losers are usually clearly identifi-able, and yet another that professional cultures collide. Economists favor market-based, pricing solutions; planners, bureaucrats, and regulators instinctively want to regulate; and engineers incline toward technology- and infrastructure-based solutions. In “Urban Transport Policies: The Dutch Struggle with Market Failures and Policy Failures,” Piet Rietveld surveys the landscape of urban transport policy in the Netherlands, which has the highest density of population among devel-oped countries. Scarcity is central to economics. We – economists and students of economics – need to continually remind ourselves that a situation that is unpleasant or even morally offensive may, because of scarcity, nonetheless be efficient or even socially optimal. Traffic congestion is an example. The spatial concentration of economic activity that is the defining characteristic of cities generates agglomera-tion benefits but also traffic congestion. While our instincts may tell us that long delays in traffic are “outrageous,” as economists we should aim not to eliminate traffic congestion but to insure that its level is socially optimal. ... - tailieumienphi.vn