Light Rail - The Facts; Critics refute Federal Reserve arguments

• The Federal Reserve Bank of St. Louis releases a questionable report about light rail

Evaluating Public Transit Benefits in St. Louis
Critique of "Light Rail Boon or Boondoggle" By Todd Litman
Victoria Transport Policy Institute
27 July 2004

Summary

There has been considerable debate over the value of rail transit service. Critics argue that it is economically inefficient and more costly than alternatives, but their analysis often focuses on just one or two of the full economic, social and environmental benefits provided by high quality transit, and underestimates the full costs of accommodating additional automobile traffic on the same corridors. Rail transit can improve mobility for non-drivers, reduce automobile travel and associated costs, and support more efficient land use patterns. As a result, communities with major rail transit systems tend to have less per capita traffic congestion, lower per capita traffic fatalities, lower road and parking facility costs, and consumer cost savings. On congested urban corridors, automobile costs (road and parking facility capacity, congestion impacts, accident impacts and pollution emission damages) are higher than average, resulting in greater than average savings from shifts to alternative modes. Although it would not be cost effective to provide light rail transit service everywhere, when all costs and benefits are considered, rail transit is often the most cost effective way to improve transportation on major urban travel corridors. A recent paper by Molly D. Castelazo and Thomas A. Garrett ("Light Rail: Boon or Boondoggle" 2004) exhibits typical errors by rail critics. It ignores many benefits of rail transit and understates the costs of automobile travel on the same corridors. Their study fails to reflect current best practices for comparing highway and transit investment cost effectiveness.

To view a full copy of the report.

White Paper by John Roach
Development Programming Associates
721 Olive Street, Suite 1111
St. Louis, Mo. 63101
E-Mail: Troach1997@AOL.com Tel: 314-621-0800 Fax: 314-621-1970________

Introduction

The recent paper authored by Mr. Garrett and Ms Castelazo, St. Louis Fed economists, presented in the July issue of the Regional Economist titled "Light Rail Boon or Boondoggle?" is based upon a series of premises that neither reflect the real world nor lead to productive analysis of public investments. A later more extensive version has been posted that represents an extension of the comments in the article published in the Regional Economist. This paper represents comment on the expanded version The paper is incomplete in analysis and amounts to a tendentious recitation of standard Libertarian rhetoric.

Perhaps the most complete recent work on valuing public transit improvements was published July 23, 2004 by the Victoria Transport Policy Institute and is authored by Todd Litman. The authors of the instant study would do well to consult this material in order that they might at least start with a complete list of the potential beneficial effects of public transportation improvements as well as a more complete formulation of the externalities of an automobile based transportation system and its expansion.

Litman lists some five major categories of benefits: Mobility, Efficiency, Travel Time Savings, Land Use and Economic Development. He breaks these down into some twenty-two sub-categories and posits a series of mechanisms for their evaluation. The current paper treats but one of the four subcategories under Mobility, but two of seven under Efficiency, includes no discussion of Travel Time Impacts, a woefully inadequate discussion of Land Use Impacts and under Economic Development Impacts discusses few considerations beyond the possible impact on nearly residential values that it finds generally positive. It largely ignores all six of the sub-categories.

Assumption: value is solely determined by profit potential

However, the over-arching problem is the assumption that value in public investment is to be measured by the potential of the investment to produce a profit. This approach ignores the value of public goods and would support the proposition that almost every public investment is of questionable value. The fact is that public investment occurs when private persons, motivated by the profit motive, abjure from investments in the same or similar improvements. The decision-making with respect to public investments is inspired by a variety of motivations. They include all of Litman's five categories and twenty-two subcategories. They include mobility, education, reduction of health risk, economic development, cultural and recreational enhancement. To the degree that the value of these motivations is denigrated in favor of profit potential, public goods such as public transportation improvements (including construction of city streets and interstate highways), museums, zoos, public schools, public universities, subvention in the form of scholarships, public recreational facilities, public art, public communication, symphonies, opera companies and public health facilities would be absent from our society. The only substitute for such investment would be charitable giving.

Indeed, we would return to a feudal era in which the Lord of the Manor (the historical equivalent of the super rich of today) would be the sole arbiter of societal investment, a result that has no measurable public acceptance.

Assumption: taxes are bad when they confer a service on a group more narrow than the general population

The contention throughout the paper is that public transit and particularly rail transit confers benefit only on the daily commuter and is primarily to be justified by aid to the poor. Almost all public services save defense law enforcement and fire protection may be subject to same observation. From public education to art museums, parks to public playgrounds, public education to swimming pools the beneficiaries are few compared to the number of taxpayers. The societal conscensus has been, however, that the provision of public amenity is a worthy end. In summary, with taxes I buy civilization.

Assumption: sole value of public transit is access for the poor to employment

Just as public streets and highways are constructed to afford mobility to the larger society, public transit enjoys a constituency larger than those who are forced by economic necessity to patronize the service. If indeed other examples of public investment were measured solely on their value to the poor, none of these investments would measure well. There is a panoply of examples of public transportation investments oriented toward the larger purpose of affording mobility, light rail among them. They have included public subsidies for the development of air travel and space travel; for the construction of roads, railways and canals; land grants to afford access to developing areas and investment in regulatory authority to protect public safety in the use of such improvements.

The paper fails to value the benefit to the general public of the MetroLink in improvements in roadways foregone, air pollution not created to say nothing of fostering a more efficient land use that can be provided utility and general services at a lower cost. There is no value placed on the reduction of sprawl that otherwise promotes the costly relocation of public institutions and cultural infrastructure. No value is assigned to the enhanced availability of mobility to the elderly, the young and the physically handicapped nor to the speed and avoidance of congestion that light rail affords to its patrons. In short, the assumption that the value of the improvement is determined solely by its utility to the poor represents an incomplete and faulty analysis.

It is also worth noting as a parenthetical matter that services that are provided solely to the poor in American society are almost uniformly poorly run and inadequate. We need only recall the failures of public housing, public hospitals and services directed to poor neighborhoods. Services provided to a broader slice of the society reflect the expectations and demands of that portion of the constituency that expects and can demand quality.

Given the assumptions, the numbers are wrong

The numbers for the operating subsidy and cost of capital are just plain wrong. The most significant error appears to be that the capital cost figure for 2001 includes payment of construction cost taking place in that year as an element of the St. Clair County construction.

In 2001 the operating subsidy (Operating Expense less Operating Revenue) was $14,096, 428. The actual capital Amortization was $1,176,883 and the Opportunity Cost for $348,000,000 at 8% would be $27,840,367. Note that assigning opportunity for these investments at 8% seems grossly excessive, given the fact that local borrowing is tax exempt and the relatively low historical federal governmental borrowing rate. A 1998 financing for the St. Clair Extension was concluded at approximately 5.2%, a partial refinancing was concluded in the 3.5% range and the current federal government 30 year real cost of government borrowing according to OMB Circular C is 3.5% and in 2002 (an appropriate year for prediction based upon 2001 results) was 3.9$. A blended opportunity cost pegged at 4.2% would be $14,616,000. Thus depending upon the opportunity cost the annual subsidy based upon FY 2001 would range from would range from $43,113,311 to $29,889,311 or a subsidy of $2.43/trip to $2.09/trip (based upon 14,288,976 trips).

The tyranny of false choices

Perhaps the most glaring of the false premises that the paper indulges leads to its comparison of false choices. Any person who is familiar with the politics and public policies that animate American society knows that tolling all auto facilities or raising the gasoline tax to a level that would satisfy even the direct costs of the auto oriented society is a neither a probability nor a possibility. Perhaps the most dearly held value among Americans is the opportunity of personal mobility. What public transit and light rail represents is an attempt to leaven the effects of this over-riding motivation with a means of exercising some moderation in its pursuit as well as lend some modicom of choices to the larger society.

Assumption: subsidy for development or transit is a bad thing

First, much as the authors might wish, almost all development is the beneficiary of subsidy in one form or another. From FHA insured mortgages that built the suburbs after the Second World War to the TIF program responsible for the suburban commercial explosion to the fact that utility extensions are largely subsidized by existing rate payers there is obvious and not so obvious tinkering with the so-called free market. Who is to say that cities and inner suburbs out not receive some payback for the contribution they have made to sprawl development.

Value is a product of an investments contribution to larger purposes

If the numbers were correct (see above) perhaps the most trenchant criticism of the paper is that it is an example of analysis that it is aware of the cost of everything but the value of nothing. Rail transit is a concomitant to the success of every great city both in this country and elsewhere in the world. In this country New York, Boston, Washington D.C., Atlanta, San Francisco, Portland and San Diego come to mind. Cities that are only now developing rail have suffered from sprawl and are now attempting to redress the balance. Los Angeles, Salt Lake City and Denver are examples. Sole reliance on the auto is a hallmark of municipal failure. Detroit is perhaps the most notable example. Further, cities themselves are the repositories of the commerce, culture and learning of the society, their preservation and enhancement is a public value of the highest order.

A not inconsiderable benefit of MetroLink has been the economic development that it has already accompanied its development and is planned for the future. The new Cardinal Stadium was placed at its new Downtown site partially because of its MetroLink access, Webster University's enlarged Downtown presence has been stimulated in part by the expansion of MetroLink to serve a station near the main campus in Webster Groves. Downtown has seen a major residential revival, the area near the BJC complex is alive with new development both by the Medical Center and by independent development interests, The Delmar Loop Extension to the Delmar MetroLink Station was inspired by MetroLink, Emerson Park represents the first mixed income development in East St. Louis within the memory of man and a second phase is now under construction, there are several additional Transit Oriented Developments about ready to move forward at other St. Clair County Stations.

It is indeed impossible to engage in single factor analysis with respect to the motivating factor for these and other developments as well as the remarkable run-up in residential values along the alignment, particularly in St. Louis' Central West End and Skinker-DeBaliviere neighborhoods. It is also impossible to fully isolate the impact ofr transit improvements on residential values such as the instant study attempted.

What this City needs and is beginning to develop is a sustainable urban fabric. The City and the inner suburbs are beginning to prosper and the enhancement of our rail infrastructure is a key part of future success. All would be better off if the St. Louis Fed, long regarded as a distinguished example of economic scholarship, would avoid parroting the nostrums of a Libertarian fantasy world.

Light Rail Transit: A Bargain Compared to Toll Roads

By E. L. Tennyson, P.E. with editing and comments by Dave Dobbs, Texas Association for Public Transportation, publisher http://www.lightrailnow.org, and Lyndon Henry

Light Rail Transit (LRT) is far less costly than toll roads if they are both built on the same or similar right-of-way. Light Rail in a subway will be more expensive than a toll highway on terra firma, but LRT will have far more capacity.

Basically, Light Rail on a right-of-way (ROW) that has no buildings on it averages about $20 million per double-track mile excluding rolling stock and shops. It will typically carry 1,500 to 3,500 people in the peak hour one-way but could carry 14,000 people per hour if, as in Boston, there is a need to do so.

Where it is necessary to rebuild the streets, LRT costs jump to $30 million a mile and if it is elevated on structure, $60 million a mile is typical. LRT elevated on structures over streets may run $80 million a mile or more. In other words, project costs, highways or transit, vary with the amount of concrete, excavation, utility relocation, right-of-way acquisition, etc., required. To keep this simple, let us consider only ground-level LRT and freeways on undeveloped land.

Assuming 4.5 million passengers annually or about 1500 one-way peak-hour trips on a $20-million-per-mile Light Rail line, the annualized capital cost per passenger is approximately 11 cents. If, as is true for several U.S. LRT systems today, the one-way peak-hour travel exceeds 3000 riders, that cost drops to 5 cents per passenger or less (10 million riders annually or more). Operating costs will add 30 to 50 cents to that depending upon the skill of the management and the wage scales.

With a roadway, 1500 passenger trips per hour in one direction requires a four-lane freeway with interchanges* at $17 million a lane mile or $68 million per mile total compared with $20 million per mile for double-tracked LRT. The seemingly high cost of Light Rail Vehicles (LRVs) is often a target of criticism. However, with a 30-year economic life and high passenger capacity, LRVs compare very favorably to the cost of automobiles per passenger-mile carried. For example, an LRV will typically average about 1,000 weekday passengers and 290,000 per year, for a typical average passenger trip length of about 5 miles. At $ 100,000 per year to buy the LRV, this comes to 7 cents per passenger-mile.

Automobiles cost $20,000 to buy (some more, some less), and will cover approximately 10,000 miles a year in predominantly commuter service. If we assume a 10-year economic life, the automobile will cost about $2,000 per year, or about $0.20/mile for 10,000 miles/year. With an average occupancy of 1.2 persons, that becomes 17 cents per passenger-mile (p-m), which is more than twice the cost of the LRV at 7 cents per p-m.

In terms of total capital and operating costs, LRT - particularly surface LRT - is also significantly more economical. For example, San Diego's extremely cost-effective LRT has a total cost of about $0.52/p-m. Other surface-routed LRT systems have a total cost more in the neighborhood of $0.80/p-m. In contrast, the automobile has a total cost of about $0.53/p-m (APTA), to which must be added about $0.60 per p-m for commuter parking - for a total cost of about $1.10-1.15 per p-m. All in all, LRT by these measures is by far a more cost-effective mobility investment.

Now, if one compares freeway costs to LRT costs at higher passenger volumes (3500 people/hour), roadway capital costs go up 50% to $102 million for the six lanes required to move those same 3500 people per hour one-way in both AM and PM peak hours. Excluding the cost of the extra rail cars that may be required to handle higher volumes, LRT capital costs will either remain the same or rise only slightly if the transit agency has to purchase more cars and/or improve signaling. (See freeway capacity assumptions in Note 3.)

In terms of a "ballpark" estimate, this means that freeways capable of handling typical urban hourly peak passenger loads are five times as expensive per mile to build as equivalent LRT, systems while the costs per passenger mile are not that different. Since toll roads are usually built with borrowed money, a six-lane grade separated facility like the one discussed here could easily exceed $200 million per mile over the thirty-year life of the bonds due to interest costs, much more than an equivalent LRT line similarly built with debt (e.g., $20 million per mile plus $20 million debt service over 30 years).

While it can be argued that over a 24-hour day most six-lane urban freeways handle several times the total number of passengers annually than most LRT lines do, this does not address the peak-hour mobility capacity problem where rail delivers passengers on time for far less cost as traffic volumes increase. Additionally this discussion does not address the cost of parking, lost tax base, or pollution where the auto-roadway system has far higher negative externalities relative to LRT.

* NOTE: Without interchanges half the roadway's capacity is lost, as traffic signals at intersections will be required. Interchanges characteristically consume half a square mile or more and freeways typically need 300 to 500 feet of right of way. This represents an enormous loss of very valuable tax base, which is not included here. Although light rail can operate in very narrow rights of way (as little as 34 feet), most LRT ROWs are normally just 60 to 80 feet wide at most.

Sources: (1) Actual LRT construction costs are based upon Denver SW, Portland Airport, Sacramento east extension, Salt Lake City south (Sandy) and St. Louis Metrolink extension to Scott AFB. (2) Commuter automobile costs of 53 cents per passenger mile are based on AAA and Runzheimer data, plus parking. Because of high costs at VTA (Santa Clara), SEPTA, Pittsburgh, and Buffalo (the last three systems have subways, escalators, elevators, drains, pumps, lighting, etc., operating all day), the FTA national average for LRT is 52 cents per passenger mile, but most light rail systems carry passengers for 30 to 50 cents per mile. (3) Highway costs are based on the proposed InterCounty Connector in the outer Maryland suburbs. (4) AASHTO Highway Traffic Manual assumes 2,200 vehicles per lane per hour but illegal tailgating or perfect operation is implied in that number. At Level of Service "F", the normal state of urban freeways at peak periods, congestion drops capacity to 1,350 vehicles per hour or 1,500 people at 1.15 passengers per peak vehicle. With families traveling, vehicle loading is higher on Sundays. Even using the highest number (2,200) does not change the determination in this discussion enough to matter. E. L. Tennyson, P.E. o Vienna, Va o stennyson@webtv.com

Critique by Lyndon Henry

An article and report produced by personnel of the Federal Reserve Bank of St. Louis in the summer of 2004 vigorously attacking mass transit and light rail systems have far more of the character of political tracts than well-reasoned, fact-based analysis conducive to sound public policy decisionmaking. Identifying and understanding these fallacies and serious methodological weaknesses is critical to furthering sound transportation decisionmaking in US cities such as St. Louis.

University of Cincinnati Economist critiques Garrett paper