University of Southern California USC
Peter Gordon

Peter Gordon's Blog

A blog exploring the intersection of economic thinking and urban planning/real estate development and related big-think themes.

Recommended Books

Recommended Films

Giving the Plan a Bottom Line

Ward Elliott
Rose Institute of State and Local Government
Claremont Men’s College

Summary / Introduction
The central problems of California transportation are economic: smog, congestion, energy consumption, coat and quality of service to all sectors of society. Transportation is insulated from market disciplines which would otherwise control wasteful use of scarce resources. Our roads and air are chronically overused because we treat them as a commons, “free” to users. Public transit is expensive and unresponsive to demand because it is overregulated and oversubsidized. Jitneys, which in poorer countries do most of what we are trying to do unsuccessfully with subsidized buses, rapid rail, or carpools, are suppressed by law and neglected in transportation planning.

Planners focus on politically attractive alternatives which make little economic sense while ignoring economically attractive alternatives which are politically thorny. $70 million worth of transportation planning in California has avoided bottom-line cost comparison and kept from the public the kind of information about multi-billion dollar road and transit systems that the public demands, and gets, when it buys a box of Wheaties in the supermarket.

Proper cost comparison should consider hidden costs as well as conventional ones. In the Los Angeles Basin, conventional planning picks up $800 million in annual public transportation costs but misses $2,850 million in smog, congestion, and hidden budget costs, 2/3 of the smog costs. These cost the average household $700 a year and subsidize the average car user with $400 a year in hidden benefits. They are wasteful and regressive. The average peak-hour river gets about $2 a day in hidden subsidy in the smog season, $1 a day in the winter. Ten percent of daily traffic-upwind, morning, peak-hour drivers– is responsible for more than 40% of daily vehicular smog and congestion. Shaving that ten percent by a quarter would eliminate almost the full 40% of congestion and at least ten percent of vehicular smog, seven percent of all smog.

Only one strategy, smog and congestion charges, appears capable of making such large improvements without absurdly high costs, yet this strategy is all but ignored in transportation planning. The most prominent road and rail projects on which millions of planning dollars have been spent, offer negligible benefits by comparison, at much higher cost. The proposed Norwalk-El Segundo Freeway, for example, would cost about 7 cents per passenger mile but probably would save less than a penny in discounted congestion costs. The proposed SCRTD starter line, by official estimate, would cost about $2.50 -3.00 per passenger mile in capital costs and likewise save less than a penny in discounted congestion costs. Neither the freeway nor the rail line would have any effect on smog, and both would add to energy consumption.

Some small traffic diverters, such as diamond lines, ramp meters, bus subsidies, and carpool incentives, might justify their relatively low costs with small reductions in smog and congestion, but their combined effect, by generous reckoning, would be less than one or two year’s growth in demand. Legalizing jitneys would probably do more than any of these to cut transportation costs, and at less public expense. More powerful diverters, fuel rationing or surcharges, would cost more in added delay than they would save in smog and congestion costs.

Smog and congestion charges, being variable by time and place, could make the cut of 40% and 7% in daily congestion and smog with only a 2 -1/2% reduction in daily VMT. They would cost about a penny and a half per passenger mile in fuel system costs and return about 3¢ a mile in public benefits. Such charges appear to be more production and cost effective, and less regressive, than existing transportation financing and other proposed alternatives. They are technically feasible and have been used successfully in Singapore, where they cut morning peak-hour traffic by 40%.

The figures used are makeshift and need much more work, but they should be enough to show that planning emphasis has been poorly allocated and will probably remain so unless planners make more effort to compare the full costs and benefits of each major alternative. Such a comparison is possible and should be part of the state transportation plan. It could save millions, and perhaps billions of dollars of wasted investment by helping the public distinguish figs from thistles.


Aurelius Morgner
University of Southern California

There is little of which I am critical in Professor Ward Elliot’s paper. I agree, of course, wholeheartedly with his proposal to construct a full cost reference case from which new transit policy proposals may be evaluated. Within the framework of his full cost case I would like to comment on how we have come to this present situation in Los Angeles transportation and what it suggests for the future.

The present arrangement in the Los Angeles basis whereby people live at considerable distance from their work by virtue of its freeway system has produced for Los Angeles the lowest population density of any major city in the world. This pattern is only in part the result of the market system that permits people to decide where they will live and work and the means by which they will move between the two. Long ago the people of Los Angeles cast their dollar votes in the market place against the “big red cars” and for auto transport. More important is the fact that through the ballot box public officials further responded to the public demands for fast auto access to all areas by highway and freeway construction.

The new feature introduced by the ballot box was that in the market people no longer had to pay the full cost of transportation. In the past in most cities powerful, monopolistic privately owned transportation systems had even charged the consumer more than the full cost of his transportation. Furthermore such rail systems had not involved congestion and pollution costs which the automobile involved. Hence the economic incentives in the market toward less density of population, as in the case of a new city such as Los Angeles.

Urban transportation by automobile appears to be grossly underpriced. The basic full cost reference case suggests that probably only about half of the maintenance and capitol carrying costs are met. When smog and congestion costs are added it may be that auto users pay not half, but as little as a fifth or a sixth of their transportation costs. Thus the auto users fundamental decisions as to how much transportation to buy in connection with his workplace and residence decisions does not consider the full cost of his action.

By this process of the subsidization of automobile transportation to the consumer we in the Los Angeles basin have achieved a transportation system that is admired and envied by the people of the rest of the world. Auto transportation, as Ward mentions, reduces the time of the average trip in Los Angeles by 60% and offers transportation to many areas that no public system of rail or bus transportation could never economically serve. More important the automobile gives people a complete flexibility to come and go at the time they wish. Thus we have a system that would seem to befit the world’s most economically prosperous area, but which in truth is more that a reflection of economic well being, but of the over consumption of an underpriced good.

Ward Elliot’s cost reference case is an attempt to arrive at what the full cost this magnificent transportation system really entails. He has roughly pieced together a surprising amount of data. If the impreciseness of smog and congestion costs are disturbing, it may merely reflect the newness of our attempts to qualify such costs. Possibly the inexactness of the estimates will be less disturbing when we remember that such presumably solid matters as constructions costs have in recent years been substantially miscalculated in the case of the San Francisco and Washington, D.C. public transit systems. Elliot’s concern with the possible regressive nature of transportation costs relative to lower income groups does, however, to me seem misplaced. Our concern should be that non-users subsidize users. All public actions affect the distribution of income. It seems a needless restraint to make every policy fit an income distributional requirement when equity in income distribution can be treated as one single matter through taxes and transfer payments being designed to offset not only what may be deemed to be the undesirable income distribution effects of the market, but of governmental policies as well.

Lastly, what in general terms might we do in the future to think more rationally about the transportation problems of the Los Angeles basin? First, we can recognize that this is not a problem to be “solved” by drawing a public rapid transit grid. The variables in the picture keep changing and with their change there are further feedbacks. Second, it seems reasonable to believe that the increasing recognition of smog and congestion costs must lead to increased pressures to get the motorist out of his automobile. Incentives politically are likely to be more acceptable than disincentives. Offers of express buses, para transport systems, etc. meet this requirement, but probably will not be sufficient. An important objective may be not to get the autoist to give up his car, but to drive less or possibly give up his second car. To many economists the best disincentive is a high price for auto travel which will cause the motorist to voluntarily stay out of his car to at least some extent. Ultimately, the important matter may be the voluntary long run relocation of people so as to shorten their distance to work. Third, relocation of people and changed patterns of car use all suggest the desirability of marginal adjustments over time of the transportation system. Flexibility of adjustment is what seems to be important. Commitments to big plans should be avoided. To create a vast rail system might mean the underutilization of both it and the freeway system. The freeway system is already in place. The problem is to make the most effective use of it. Thus the appropriate strategy would appear to be to raise costs of freeway travel, change the full cost at time made of all modifications in the transit system as a whole and at every turn to evaluate the alternative use of funds in securing a more effective transportation system. This means that we should not turn back to the drafting board for more effective new systems of transit routes but first should move forward to the development of meaningful techniques of cost benefit analysis that would be applied to each measure proposed.


Robert C. Ellickson
University of Southern California

I wholeheartedly agree with the basic theme of Professor Elliott’s informative paper: that we need better cost-benefit analysis of transportation alternatives. I also support Elliott’s conclusion that a closer economic evaluation would demonstrate the foolishness of a substantial investment in a fixed-rail transportation system for Southern California.

However, to be fair to those who are charged with developing transportation policy, I must admit that cost-benefit analysis is an extremely difficult undertaking - - even in the hands of someone as skilled as Professor Elliot. For example, Professor Elliot treats the exemption of highway systems from local property taxes as a “cost” of highways, and implies that he would end this exemption. I disagree. The tax exemption for highways does not consume any resources, but is rather a pure wealth transfer to taxpayers of governments that won highways from local taxpayers. Thus, this “cost” should not properly have been included in Elliot’s cost-benefit analysis.

In addition, it may in fact be efficient not to levy property taxes on highways and other real property owned by government. When the major authority levying property taxes also owns the highway, exempting it from taxes saves the deadweight loss involved in assessing highway properties (no easy task given the absence of “like sales”) and having the government pay taxes to itself. The allocative case against the tax exemption is obviously stronger when the governmental unit that owns the highway is not a major taxing authority; Beverly Hills, for example, would have been more likely to accept the Beverly Hills Freeway had it been able to tax it. But my basic point is that the issue of property tax exemptions for highways is a rather complex one.

I also have some difficulties with Professor Elliott’s objections to using general local tax revenues to finance highway police and maintenance services. Professor Elliot considers this practice an inefficient and regressive subsidy. He would prefer that these services be finances with user charges on motorists. Although I generally am quite sympathetic to increased reliance on user charges, they are not always the most efficient financing mechanism. To be specific, if the administrative costs of levying user charges exceed the allocative gains they promise (by introducing the pricing mechanism into the provision of governmental services), user charges are not in fact efficient. For example, we will all be receiving inoculations against swine flu this coming fall. I hope the federal government will not impose user charges to finance that program. Since everyone will be receiving the benefits of those shots, it would be more efficient to use general federal tax revenues, and save the deadweight administrative costs of collecting flu-shot fees from beneficiaries. Although benefits from highways are not quite so equally distributed, it is still the case that even non-motorists needed highways for buses and trucks. At bottom, the efficiency of financing highway policing and maintenance services with user charges would depend on the type of charge used. Local gasoline taxes would be cheap to piggy-back onto state and federal gasoline taxes, but might be easy to evade in highly balkanized metropolitan areas. By declining to nominate any user fee system to substitute for general revenue financing, Elliott has again somewhat obscured the complexity of the problem.

My last criticism goes not to Elliott’s efficiency analysis, but rather to his assumptions about the incidence of the property tax. He adheres to the traditional view, popularized by Dick Netzer, that the property tax is in effect a regressive excise tax on the consumption of housing. The Netzer view is not under severe challenge. A growing body of literature argues that the common component of property taxes on improvements is in reality a tax on capital,* and thus progressive. If the modern theorists are correct, the use of property taxes to finance highway support services is not as regressive as Elliott implies.

I will focus the latter half of my remarks on the politics of transportation, a topic that may have been inadequately emphasized this afternoon. I certainly do not claim to have a full-fledged model of transportation politics; if I did, I might be able to explain the following paradoxes in current policy.

Paradox 1.
Like most economically-minded academicians, both Elliot and I would probably end up advocating higher fuel taxes, peak-hour surcharges on motorists, and the deregulation of transportation carriers as ways to improve transportation policy. These sorts or policies are not only supported by economic theory, but also by evidence from Singapore and elsewhere. The interesting puzzle is why politicians at all levels are ignoring our advice and rather are pursuing economically-suspect multi-billion dollar hardware systems. I can at least offer an hypothesis to explain their behavior. It is rather clear that academic experts on transportation are not very important politically. Ranged on the opposite side is a surprising alliance of two powerful groups that do favor fixed-rail systems - - the public lobby and the environmentalists.

To be provocative, I will venture a scenario to explain why we have been seeing such great interest in expensive fixed-rail systems in the last few years. A decade ago, the public works lobby (building trades unions, contractors, materials manufacturers, civil engineers, etc.) Filled their pork barrel in large measure with interstate highways and dams. As you know, public works are more lucrative for most of these interest groups than private projects; for example, under the Davis - Bacon Act, “prevailing wages” (i.e., union wages) must be paid on federally-funded public works. Under my scenario, the steady diet of highways and dams became increasingly threatened after the year 1970 with the passage of the National Environmental Protection Act, and the escalation of legal and political challenges by environmentalists to both highway and dam projects.

At that point, the public works lobby discovered two sorts of expensive construction projects that the environmentalists not only did not object to, but would in fact lobby to support: sewage treatment plants and fixed-rail transportation systems. Since 1970, the federal government has greatly increased its grants-in-aid for both. In short, my hypothesis is that the public works lobby found it could not beat the environmentalists, and therefore joined them.

Paradox 2.
As a commuter from West Los Angeles to U.S.C., I am currently being subjected to the indignities of the Diamond Lane Experiment. That program meters traffic at freeway entrances through lengthily queues, not through monetary toils that might rather easily be collected by automatic toll-collecting machinery. Elementary economics indicates that time spent waiting in line is a deadweight loss; it is thus highly doubtful that the current system is the most efficient. Why then was it adopted? Perhaps the explanation lies with the perceived distributive injustice of charging commuters tolls. My point is that the answer primarily requires a better understanding of politics, not economics.

Paradox 3.
It is ironic that governments are heavily subsidizing construction and operating of bus and rail transportation systems during a period of “energy crisis.” Those subsidies are usually defended as counter-balancing the sorts of subsidies to auto transportation that Professor Elliot has identified. But if one is concerned about distortion in consumer choices between transportation modes, surely the appropriate solution in a time of expensive energy is to stop subsidizing automobiles, not to increase subsidies to its alternatives. Subsidizing all modes of transportation is certain to increase energy consumption. The explanation for the paradox again must lie in the political forces that shape transportation policies.

My basic theme is thus that those who wish to influence transportation policy must develop more sophisticated models of transportation politics. As I hope my three paradoxes have illustrated, the current transportation policy makers are clearly not listening to their academic critics. They obviously do not share our concern about efficient resource allocation. Rather, they are motivated to mollify the most organized interest groups. Discussions of transportation alternatives should not ignore this political reality.

Top | Home | Contents