Alternatives To Rail: Rubber-Tire Transit

By Thomas A. Rubin and James E. Moore II

"Rubber Tire Transit" refers to a wide range of transit options that provide dedicated or priority lanes for buses, high occupancy vehicles, and (tolled) automobiles. Compared to rail transit, Rubber Tire Transit has advantages with respect to travel speed, capacity, number of passenger transfers and passenger access, speed of construction, emergency vehicle access, and cost. In many cases, however, Rubber Tire Transit has not been objectively considered as an alternative to new rail lines.

We perform some overdue comparisons. Our discussion of capital costs accounts for both local cost and external capital grants. Our discussion of operating costs accounts for operating revenues (mainly fares), external subsidies, local subsidies, and cost-effectiveness. The non-local share of operating costs is very small, and is ignored.

Consider Los Angeles. The El Monte Busway / High Occupancy Vehicle (HOV) facility consists of a dedicated lane in each direction on or near the I-10 Freeway from slightly east of El Monte to near Union Station in the northeastern corner of the Los Angeles Central Business District. Originally placed in service in the early 1970s as an exclusive bus lane, the facility subsequently opened to HOV traffic.

Measured per mile of guideway, transit ridership on the El Monte is over four times the ridership on the Long Beach-Los Angeles Blue Line train. Speed on the El Monte is well over twice the speed of the train. Buses on the El Monte provide approximately three times more passenger-miles per mile of guideway than does the light rail Blue Line. When HOV use of the El Monte is added, this factor jumps to 10.

The acknowledged capital cost of the Blue Line is $877 million for 21.6 miles of two- directional light rail guideway in (approximately) fiscal year 1988 dollars, or $41.6 million per two-directional mile. The original cost of the El Monte Busway project contracts was $36,191,300. 1/ Assuming a project spending midpoint of 1972, the capital cost of the El Monte in fiscal year 1988 dollars was approximately $85.3 million for 11.5 miles of busway/HOV lanes, or $7.4 million per two-directional mile. 2/ Thus, the El Monte produces approximately 10 times more transportation service per guideway mile than the Blue Line, at less than one-fifth the capital cost per mile.

Total Capital Cost And Local Tax Burden (1996 $Millions) Of Transit Options, Per Guideway Mile, by Mode Table 1 Transit Mode Total Capital Local Tax RUBBER TIRE GUIDEWAY     A. Pure Busway $ 5-35 a/ $ 2-25 B. Busway/HOV Lane 5-30 b/, 4/ 2-20 C. High Occupancy Toll (HOT) Lane 5-31 c/ 0-10 D. Bus Transit Mall 3-10 d/ 1-8 E. Dedicated Lane on General Use Street 1-5 e/ 1-3 F. Signal Preemption/Prioritization 1-2 f/ 1 RAIL GUIDEWAY     A. Heavy Rail $200-550 g/, 5/ $80-200 B. Light Rail 50-80 h/, 6/ 30-60 C. Automated Guideway 45-85 i/ 25-65 D. Downtown People Mover 20-50 j/ 15-40 E. Commuter Rail 2-6 k/, 7/ 1-4 &nbsp a/ The small increase over "Busway/HOV Lane" costs reflects costs of stations and more complex guideway entrances and exits. Costs reported for vehicles and supporting assets are our estimates. b/ The costs reported for vehicles and supporting assets are our estimates. c/ The small increase over cost of HOV lanes reflects the costs of electronic toll collection gear and other elements. "Local Tax" is lower due to the self-financing provided by tolls. This does not comprehend bus use, and bus costs are not included. In many corridors, HOT lanes could return a profit to investors. d/ This estimate is based on the costs of Portland and Denver busways. The "Local Tax" component can be lower if benefit assessment district or other non-general tax financing is used. the costs of buses and operating and maintenance facilities are not included. e/ This is our estimate based on existing streets and modifications limited to minor changes, such as curbs cuts for off-line bus stops, signage, and lane separation barriers. The costs of buses and operating and maintenance facilities are not included. f/ This is our estimate based on a system and bus fleet large enough to achieve economies of scale. The costs of buses and operating and maintenance facilities are not included. g/ The low end of the range is defined by the five-mile San Fernando Valley Red Line, which is $1,082 million in 2004-2013 dollars. The high end is defined by the three-mile Red Line Segment 1, which is $1,418 million in 1986-93 dollars. h/ The high end of the range is defined by the 13.5 mile Pasadena Blue Line, originally estimated at $998 million, and later reduced to slightly over $800 million in 1994-2003 dollars, plus the cost of rail cars and other items not included by MTA. The low end of the range is our attempt to construct a more likely estimate by discounting the extraordinarily high cost of the Pasadena line. i/ This is our estimate, constructed based on BC Transit's experience with Vancouver SkyTrain, and the additional costs of automation incurred by the MTA for the Green Line. We adjusted this initial range by subtracting of $5 million per mile from the low end of the interval, and adding $5 million per mile to the high end. It is difficult to properly classify the Green Line. At one point, the Green Line was intended to be an automated guideway system. Tens of millions of dollars were spent on automation, but the objective was eventually abandoned. The vehicles are designed to light rail standards, as is the electrical system and other components. But the method and speed of operation, combined with a 100 percent grade separated guideway, is more typical of heavy or rapid rail. Rapid rail generally operates with a minimum of four-car trains, and more commonly operates with six-to-ten car trains. The Green Line currently operates with one-car trains. j/ This is our estimate based on recent Los Angeles experience. These costs can vary significantly due to great differences in the cost of guideway construction, and who pays for additional costs. Guideway designs range from simple to tunnels to integrated with buildings. k/ We estimate the total capital cost of Metrolink at approximately $1.2 billion for approximately 300 miles of commuter rail right-of-way.

This result is representative. Table 1 shows a reasonable range of Los Angeles County local and total capital cost estimates for typical transit projects. Capital costs are shown in 1996 dollars as ranges on a per mile, two-directional basis. For pure busways and busway/HOV options, capital costs include vehicles and support equipment. If the purpose of the new facility is to allow existing buses and HOVs to operate faster or more cost-effectively, or merely to encourage greater use of transit and HOVs; then the costs shown account only for the physical guideway. Higher travel speeds reduce the number of buses needed to provide the same level of service.

"Local Tax" cost refers to the portion of the cost paid by local taxpayers, with emphasis on Los Angeles County experience. Unfortunately, the Los Angeles County Metropolitan Transportation Authority (MTA) has consistently excluded important rail components (such as rail cars) from rail line budgets. In past years, MTA has not allocated tens of millions of dollars in general and administrative overhead costs to the rail lines. Such costs should be shown under generally accepted accounting principles (GAAP), transit industry accounting practices, and PUC requirements. Also, MTA is required by both GAAP and the state legislation 3/ to capitalize interest costs during construction, but it has never done so. Properly recognizing capitalized interest would add from tens of millions to potentially over $100 million to the costs of individual rail lines.

Thomas A. Rubin, CPA, CMA, CMC, CIA, CGFM is a mass transit consultant in Oakland, California. He served as controller-treasurer of the Southern California Rapid Transit District from 1989 until the SCRTD/LACTC merger that formed the Los Angeles County Metropolitan Transportation Authority in 1993.

James E. Moore II is associate professor of civil and environmental engineering and of urban planning and development at the University of Southern California. He is associate director of USC's Center for Advanced Transportation Technologies, and director of the Transportation Engineering Program.

Following proper accounting practices would increase the reported costs of MTA rail lines by approximately 10 percent to 30 percent. We have increased MTA's reported costs by conservative estimates. Consistent with GAAP standards, we show interest on borrowings to build rail lines after the commencement of revenue operations as operating rather than capital costs.

Total Capital Cost And Local Tax Burden ($Millions) Of Transit Options, Per Passenger-Mile, by Mode Table 2 Transit Mode Total Capital Local Tax RUBBER TIRE TRANSIT     A. Pure Busway $.007-.046 $.003-.033 B. Busway/HOV Lane .002-.015 .001-.010 C. HOT Lane .002-.015 0 -.005 D. Bus Transit Mall .008-.027 .003-.027 RAIL GUIDEWAY     A. Heavy Rail .16-.43 .06-.16 B. Light Rail .20-.31 .12-.24 C. Automated Guideway .06-.10 .03-.08 D. Downtown People Mover .16-.41 .12-.33 E. Commuter Rail .008-.024 .004-.015         /a Costs are not computed for Dedicated Lane on General Use Street and Signal Preemption/Preference. The benefit of these technologies is increased speed for existing transit services. A comparison of costs per passenger-mile to other guideway modes is not meaningful.

We convert costs per unit of guideway into costs per passenger-mile, 8/ indexing the passenger-miles accruing per mile of guideway to one minute of operation. Our calculations include a number of implicit assumptions, e.g., a 40-year life and 254 working weekdays per year; but are based as fully as possible on National Transit Database (Section 15) data from the Federal Transit Administration, and on standard industry practices. Table 2 shows the total capital and local tax costs per passenger-mile for different transit modes, referenced against the midpoints of the cost intervals in Table 1.

Operating Costs And Subsidies Per Passenger-Mile a/ Table 3 Transit Mode Cost Per Passenger-Mile Subsidy Per Passenger-Mile RUBBER TIRE GUIDEWAY MTA Local Bus 11/ $447 $0.29 Los Angeles Privatized Bus b/, 12/ 0.249 0.137 New Jersey Long Haul Express Bus c/, 13/ 0.167 0.028 RAIL GUIDEWAY Red Line (heavy rail) 14/ 0.563 0.428 Blue Line (light rail) 15/ 0.829 0.773 Metrolink (commuter rail) 16/ 0.397 0.244       a/ All values are adjusted to 1996 dollars. b/ Los Angeles Privatized Bus is Foothill Transit Zone, fiscal year 1994. c/ New Jersey Long Haul Express Bus data total values for four long-haul commuter bus operators under contract to the New Jersey Transit Corporation in 1993.

Table 3 shows operating costs and taxpayer subsidies. The overwhelming majority of operating subsidies come from local taxes. The MTA Local Bus and Los Angeles Privatized Bus should be compared to the Red and Blue Line costs. Long Distance Busway costs should be compared to Metrolink commuter rail costs.

Figures 1 and 2 show that short- and medium-length rubber tire guideways provide bus transit modes with a substantial cost advantage relative to Los Angeles Red Line subway and Blue Line train. The subsidy advantage is even greater. Similarly, long-distance busways demonstrate a substantial financial advantage relative to Metrolink commuter trains. In all cases, the high end of the Rubber Tire Transit cost and subsidy ranges are lower than the low end of the corresponding rail guideway ranges. Our accounting deliberately overstates bus costs, because no costs are allocated to high occupancy vehicles.

In most cases, the capacity of Rubber Tire Transit is significantly higher than the capacity of comparable rail modes. 9/ Figure 3 shows the most common range of capacities. Rail capacities are slightly overstated.

The actual, utilized capacity of most rail guideways is less than that of a single freeway lane. This contradicts several years of frequent, continuing, and very incorrect statements from the MTA. 10/ Figure 4 compares the observed freeway lane equivalents for various Los Angeles modes against the very optimistic estimates provided by the MTA and its predecessor agency. The Blue Line, for example, has an actual utilized freeway lane equivalency of only approximately 0.6-0.7, and the Blue Line is by far the most heavily used light rail line in the nation. The individual Red Line subway segments will have usable capacities on the order of two to three freeway lanes. Commuter rail lines produce high theoretical capacities due to the possibility of longer trains, high operating speeds, and long trips; but in practice it is rare for an individual commuter rail line's utilized capacity to approach 0.5 freeway lane equivalents.

These results are clear. Transit agencies should consider more busway projects. High occupancy vehicle lanes should be given a higher priority for construction and implementation than rail guideways. HOV lanes are far more productive and cost-effective than rail. Indeed, we question why the existing and proposed Southern California rail lines should be constructed or operated at all.

Footnotes

1/ Southern California Rapid Transit District Engineering Department, "Summary-San Bernardino Busway Project," October 26, 1981.

2/ In 1972, the Department of Labor, Bureau of Labor Statistics Consumer Price Index (CPI-W) in the Los Angeles/Long Beach/Anaheim urbanized area was 41.4. In 1988, the index was 119.0, giving a multiplier of 2.87. The MTA reports that the Construction Cost Index increases at 82 percent of the rate of increase for the CPI (MTA, "A Plan for Los Angeles: Transportation for the 21st Century," March 1995, p. 114). This adjustment produces a multiplier of 2.36, and a comparable 1988 dollar cost of El Monte Busway contracts of $85.3 million.

3/ PUC 130510-130519, legislation originally passed as the "Los Angeles County Transportation Commission Revenue Bond Act."

4/ These values are based on data from "A Plan for Los Angeles County - Transportation for the 21st Century," MTA, March 1996, and supporting documents relating to six Los Angeles County HOV lanes.

5/ These values are based on data from "A Plan for Los Angeles County - Transportation for the 21st Century," MTA, March 1996, pp. 127-131, adjusted to account for costs not included by MTA, such as direct rail segment costs not shown as such, capitalized interest, overhead, and administration.

6/ This upper bound is based on data from "A Plan for Los Angeles County - Transportation for the 21st Century," MTA, March 1996.

7/ This is based on Southern California Regional Rail Authority budget documents.

8/ This methodology is adapted from Vukan R. Vuchic, "Urban Public Transportation Systems and Technology," 1981, p. 70.

9/ Thomas A. Rubin and James E. Moore II, "Ten Transit Myths: Misperceptions About Rail Transport in Los Angeles and the Nation," Reason Foundation Policy Study No. 218, November, 1996.

10/ Los Angeles County Transportation Commission 30-Year Integrated Transportation Plan, LACTC, April 1992, pp. 18-19.

11/ MTA FY95 National Transit Database (Section 15) Report, adjusted to account for MTA errors in allocation of expenses between modes and additional interest costs associated with borrowings to acquire capital assets.

12/ National Transit Data Base Data Tables for 1994.

13/ National Transit Data Base Tables for 1993.

14/ Red Line operating costs and revenues are from MTA "A Plan for Los Angeles County - Supporting Cash Flows" for 2003, a date by which the Red Line to North Hollywood will have been operating a sufficient period to experience mature ridership. Our estimate of interest costs and ridership is based on data from "A Plan for Los Angeles" and MTA financial statements.

15/ MTA FY95 National Transit Database (Section 15) Report, adjusted to account for MTA errors in allocation of expenses between modes and additional interest costs associated with borrowings to acquire capital.

16/ Metrolink data is from its fiscal year 1996 budget.