For Submission to the Transportation Research Boards 86th Annual Meeting, January 2007, Washington D.C.

PAPER 07-1538: APPLICABILITY OF BOGOTS TRANSMILENIO BRT SYSTEM TO THE UNITED STATES

By: Alasdair Cain Senior Research Associate National Bus Rapid Transit Institute (NBRTI) Center for Urban Transportation Research (CUTR) University of South Florida 4202 E. Fowler Ave, CUT 100 Tampa, FL 33620 - 5375 Phone: (813) 974 5036 Fax: (813) 974 5168 Email: cain@cutr.usf.edu Georges Darido, Senior Research Associate, NBRTI Michael R. Baltes, ITS Program Manager, Federal Transit Administration Pilar Rodriguez, Transportation Planner, TranSystems Corp Johan C. Barrios, Graduate Research Assistant, NBRTI

November 2006

WORD COUNT: TABLES AND FIGURES: TOTAL:

(10*250)

6,137 2,500 8,637

Cain, Darido, Baltes, Rodriguez, Barrios ABSTRACT

Bogots TransMilenio is the highest capacity BRT system in the world and one of the worlds best examples of a high level BRT system. As such, it demonstrates what can be achieved with BRT if certain high capacity design features and operating characteristics are adhered to. This paper has attempted to highlight the different capabilities of BRT that are demonstrated by the TransMilenio, taking into account the variety of differences between the city of Bogot and typical U.S cities. A series of observations are made in relation to the topics of passenger capacity, capital cost effectiveness, achieving modal shift objectives, urban renewal, business models, and politics. The paper concludes by discussing the various issues relating to the replication of the Bogot Model in the U.S. Overall, TransMilenios great strength is the fact that it is a true rapid transit network. Constructing such a network, in a relatively short time period, has given the system an economy of scale that has maximized operating efficiency and yielded city-wide mobility and urban renewal benefits. The relatively low capital costs associated with BRT also provides U.S cities with the opportunity to implement rapid transit networks capable of addressing city-wide mobility objectives.

Cain, Darido, Baltes, Rodriguez, Barrios

INTRODUCTION Serving the city of Bogot, Colombia, the TransMilenio is one of the worlds premier Bus Rapid Transit (BRT) systems. The system carries almost 1.3 million passengers on a 50 mile network of high capacity trunk corridors, supported by feeder services that extend system coverage to peripheral areas of the city (1). It is the highest capacity BRT system in the world, carrying up to 41,000 passengers per hour per direction on its busiest sections, comparable to the passenger volumes normally associated with heavy rail. TransMilenio is also the centerpiece of a long-term urban renewal and sustainable mobility strategy that has yielded large scale public space improvements, extensive walking and cycling networks, in parallel to a variety of disincentives to private vehicle use. Encouraged by the success of the TransMilenio, the Colombian government is now financing the construction of similar systems, on a smaller scale, in cities throughout the country. The great success of the TransMilenio has made it the poster child for BRT proponents across the world. However, some have rightly questioned the extent to which its achievements can be replicated in a developed world context. This paper aims to compare the context in which the TransMilenio operates with that in typical U.S cities, in order to assess its applicability to transit in the U.S. This paper is based on a report (of the same name) published by the National Bus Rapid Transit Institute (NBRTI) in May 2006. This report is available at www.nbrti.org.

A BREIF OVERVIEW OF THE TRANSMILENIO Construction of TransMilenio Phase I began in 1998, with initial sections opened in 2000. The full 41km (25.6 miles) were completed by early 2002, and the system was soon carrying approximately 800,000 passengers per day. Construction of Phase II, which also consists of three trunk corridors totaling 25.6 miles in length, began in 2000. Phase II has recently been completed and ridership is now just under 1.3 million riders per day (1). Figure 1 shows the alignment of the main trunk corridors and the supporting feeder routes, while Figure 2 shows a typical trunk-line station. The current system provides one of the best examples of what is referred to by Wright (2) as Full BRT, characterized by exclusive busways on trunk corridors, off-board fare collection and verification, fare free integration between feeder services and trunk services, and high quality rail like stations. Planning of Phase III is currently underway. Once completed, this phase will provide more extensive system coverage within the city, and extend the service to the western suburb of Soacha. The TransMilenio masterplan consists of 388km (241 miles) of dedicated trunk corridors to be constructed over a total of eight separate phases. This is expected to take several decades, and will depend on funding availability (3). The TransMilenio system is managed by TRANSMILENIO S.A, a public sector agency created in 1999 to manage system planning and daily service supervision. Private contractors are responsible for fare collection and operations (including vehicle costs), through concession contract agreements. Contractors are selected through a formal tendering process and paid per service kilometer. The TransMilenio service design is based on a closed system (also known as Feeder-Trunk service), where high-capacity articulated buses operate on the trunk corridors, and

Cain, Darido, Baltes, Rodriguez, Barrios lower capacity feeder buses are used to connect the trunk corridors with peripheral areas.

The TransMilenio has significantly improved mobility in Bogot, reducing transit user travel times by an average of 32 percent. The TransMilenio has also been successful in attracting choice riders, reducing the proportion of private vehicle trips from 18 percent to 11 percent. Localized improvements in environmental quality have also been observed. A much more detailed description of TransMilenio, the plans for its future expansion, and its impacts, are provided in the report on which this paper is based (3). ISSUES AFFECTING THE APPLICABILITY OF THE TRANSMILENIO TO THE U.S TransMilenios applicability to the U.S is a complex issue. The sheer size of Bogot, with a population of around eight million, makes it comparable to only a small number of U.S cities (only Los Angeles, New York and Chicago have comparable urbanized area populations, while no other urbanized area has a population significantly greater than five million (3)). Population density has a significant impact on the demand for transit within any given urban area. Bogot is characterized by an extremely high average population density of 210 people per hectare (3), high even by Latin American standards, compared to the low average population densities of around 10 people per hectare observed in modern U.S cities (4). U.S downtowns also tend to be relatively small with the majority of Americans now living and working in the suburbs, while Bogot still retains the traditional CBD focused activity center. Also, the geographical dispersion of income groups is different; in the U.S the lower income areas tend to exist in the urban cores, while in Bogot the lowest income areas are located in the city periphery. In the low-density urban environments observed in the U.S, environments trip origins and destinations tend to be highly dispersed, severely limiting transits ability to compete with the private auto. In summary, Bogot provides ideal circumstances for transit and non-motorized modes, while modern U.S cities provide the opposite. It is no surprise then that while 80 percent of Bogots population does not have access to a car (3), over 85 percent of all trips in the U.S are made by car (5). The TransMilenio replaced dozens of traditional microbus routes and consists of numerous service routes on the same right-of-way (e.g., local, express, super-express, etc.). Banning traditional buses from the TransMilenio corridors guaranteed high levels of system demand. TransMilenio can therefore be seen as a comprehensive reorganization and enhancement of road-based public transportation services from a multi-point to a feeder-trunk system. This highly integrated approach is different from that of most U.S New Start projects, which plan a new transit service to try to create new, or reinforce weak, travel patterns because of the availability of right-of-way or stations but do not necessarily serve existing travel patterns. In addition, New Start projects are usually corridor-focused instead of considering city-wide networks. The political contexts of the two countries are also very different. Like most cities in the developing world, Bogot is located in a political environment characterized by a strong executive. A powerful executive figure, such as a president at the national level or a mayor at the local level, wields much more power and influence than their counterparts in typical developed world democracies, and plays a much more important role in policy initiation. This type of political system, also featuring fewer environmental controls and much less stringent bureaucratic review, also permits much faster policy implementation.

Cain, Darido, Baltes, Rodriguez, Barrios

Although the public sector was largely responsible for TransMilenios infrastructure costs (similar to the way in which the federal government contributes to the capital cost of transit projects in the U.S), farebox revenue from the TransMilenio system covers operating costs and actually generates profit (reportedly as high as 30 percent according to TransMilenio managers), allowing the Colombian government to turn system operation over to the private sector, using a tendering process to maximize operating efficiency. In the U.S, transit systems operate at a loss, imposing greater limitations on private sector involvement. Although there are examples of private sector involvement in transit service provision, in most cases the public sector continues to be fiscally responsible for the system, either through providing subsidies to the private companies who run the system or running the system directly. Table 1 provides a summary of the demographics and mode use characteristics in Bogot versus typical modern American cities. Such differences limit the extent to which Bogots experience of BRT can be applied in the U.S. Indeed, most of the interest in TransMilenio has come from other developing countries in East Asia and South America, where more similar economic, socio-political and urban form conditions exist. Some express the view that what happened in Bogot cannot be replicated because it is the result of a unique set of circumstances related to the extraordinary administrations of Enrique Pealosa and Antanas Mockus (Mayor of Bogot before and after Pealosa). There are even doubts about whether the trend toward sustainable transport in Bogot can be maintained if the political agenda shifts towards other priorities (6). However, there are some similarities that should also be acknowledged. In both locations, the public sector is responsible for the capital cost of highway and transit infrastructure. In both locations, the over-utilization of transportation infrastructure by private vehicles has resulted in severe negative economic, environmental, and social impacts. Similarly, transit is looked upon in Bogot and in the U.S as a way to improve aggregate mobility through the more efficient use of road space, thus reducing traffic congestion. While limited in its direct applicability to the U.S, the TransMilenio does illustrate a number of important issues regarding the applicability of BRT treatments in the U.S, as well as in other developed world environments. The following sections discuss these issues, commenting on the different lessons that may be learned.

PASSENGER CAPACITY
Carrying up to 41,000 passengers per hour per direction (pphpd), actual system capacity is estimated at 45,000 pphpd. This makes the TransMilenio the highest capacity BRT system in the world. The accommodation of such high passenger volumes is made possible by a variety of different system design features: High capacity articulated vehicles (160 passengers) with multiple doors High average bus occupancies (occupancy levels on TransMilenio buses average 80 percent in peak periods and 70 percent in off-peak periods) (2) Exclusive running ways that guarantee high commercial speeds at all times of day and allow the co-existence of local, limited stop, and express services High capacity rail like stations that feature level boarding and off-board fare payment

Cain, Darido, Baltes, Rodriguez, Barrios

Centralized control of bus operations, which reduces bunching and improves reliability High service frequency (280 buses per hour per direction on busy trunk sections, resulting in a combined headway 13 seconds at busy stations) TransMilenios high capacity is also the result of high in-vehicle passenger loading. While the TransMilenio system was designed to carry seven standing passengers per square meter (7), capacity calculations in the U.S assume much lower standing passenger loads. This explains why the assumed capacities of different transit modes in the U.S context (10,000 to 12,000 for busbased transit, 26,000 for LRT and 50,000 for HRT) (8) are lower than the theoretical capacities of different transit modes presented by Vuchic (9) (15,000 pphpd for bus-based transit, 30,000 pphpd for LRT, and 72,000 for HRT). Applying standard U.S passenger loadings, as prescribed in the Transit Capacity and Quality of Service Manual (100 passengers per bus, 48 seated, 52 standing) (8), to the TransMilenio case, results in a busway capacity of approximately 28,000 pphpd. While significantly lower than the 41,000 pphpd carried by the TransMilenio, this figure is still much higher than the 10,000 to 12,000 pphpd maximum capacity figure assumed in the U.S. Why such a discrepancy? According to Samuel (10), calculation of BRT system capacity in the U.S is based on headways of 40 to 60 seconds. Such headways are relevant to the situation where no special bus terminals are provided and there is only limited passing availability at stops. If high capacity TransMilenio-style infrastructure is provided, Samuel states that it is possible to run buses at six to eight second headways (450 to 600 buses per hour per lane), equating to a seated capacity of 27,000 to 36,000 passengers per hour (10). The TransMilenio proves the validity of these calculations. As well as showing that BRT has a similar passenger capacity to Light Rail, the TransMilenio also demonstrates the extremely wide range of passenger volumes that can be accommodated by BRT. Wright states that BRT can cost-effectively accommodate volumes of 4,000 to 40,000 pphpd (2). In summary, BRTs passenger carrying capacity appears to be grossly underestimated in the U.S, often leading to its early rejection in favor of LRT in the alternatives analysis process. TransMilenio demonstrates that BRT systems are capable of carrying the passenger volumes commonly associated with rail-based transit, even assuming U.S passenger loading standards. Thus, BRT should not be ruled out of alternatives analysis in favor of LRT on the grounds of insufficient passenger capacity.

CAPITAL COST EFFECTIVENESS

The history of transit planning in Bogot provides a useful illustration of the comparative capital cost of BRT versus Heavy Rail Transit (HRT). For many years, heavy rail was the preferred transit improvement option in Bogot, and between 1947 and 1997 there were a total of 10 attempts to implement such a system. The primary reason for the failure of each attempt was the high capital expenditures involved. Table 2 compares the capital costs and predicted ridership associated with the 1997 heavy rail proposal with the first two phases of the TransMilenio, and

Cain, Darido, Baltes, Rodriguez, Barrios

the TransMilenio masterplan. The table shows that the heavy rail proposal would have provided one 18 miles metro line for a total cost of US$3,041M, equating to a cost of US$169M per mile. In return, it was predicted that the heavy rail line would have carried an estimated 795,000 passengers per day, equating to 16 percent of the citys total transit trips. In comparison, TransMilenio Phase I provided more trunk corridor (25.6 miles versus 18.0 miles) and similar ridership levels, for a total capital cost of US$340M, almost one tenth of the cost of the heavy rail option. Analysis of Phase II costs showed that only 37 percent of the costs are specifically for TransMilenio infrastructure (exclusive lanes, stations, terminals, pedestrian overpasses, and studies and designs). Of the remaining 63 percent, 40 percent is for general traffic lanes, nongrade intersections, sidewalks and utilities; 11 percent is for right of way acquisition; and 12 percent is for environmental and social mitigation, traffic management, detours, and maintenance. Thus, the TransMilenio portion of Phase II costed US$207M, equating to approximately $5M per km. Table 2 also shows the predicted capital cost of the complete eight phase TransMilenio masterplan. The capital cost of the system, featuring a 241 mile network of exclusive busways capable of accommodating 85 percent of the cities transit trips, is estimated at US$2,300M not including vehicle and fare collection costs (11), and approximately US$3,320M (USD) including vehicle and fare collection costs (12). Comparing this figure to the capital cost of the 1997 heavy rail proposal (US$3,041M), it can be concluded that TransMilenio offered Bogot a city-wide rapid transit solution for similar capital expenditure as one rail corridor. Table 3 compares the capital costs, trunk corridor lengths, and ridership associated with the TransMilenio in relation to a range of different rapid transit projects in the U.S. First of all, it can be seen that rapid transit capital costs are highly variable, ranging in this case from $3.2M for the San Pablo Rapid in Oakland, CA, versus over $16 billion for the complete Metrorail system in Washington D.C. Even accounting for system size by considering capital cost per mile, the variation in capital cost is huge (0.23M per mile for the San Pablo Rapid versus $340M per mile for the Red Line in Los Angeles, CA). Next, it can be seen that some of the variation appears to be related to mode, with BRT being the cheapest on average and heavy rail being the most expensive on average. However, even within each mode the variation is quite large. This is particularly true for BRT, where cost per mile for the small range of projects mentioned here ranges from $0.23M to $25M per mile. This can at least partly be explained by the fact that the term BRT has been used to refer to a wide range of different bus-based service improvements, from inexpensive improvements such as adding Traffic Signal Priority to express bus services that run in mixed traffic, through to high quality services featuring dedicated runningways and rail like stations. Such services, like the TransMilenio in Bogot and the Orange Line in Los Angeles, represent the high-end of the BRT capital cost range, which appears to be similar to the low-end of the range of LRT costs. Indeed, a recent paper examining BRT capital costs divided the BRT genre into three categories; arterial streets (average capital cost $0.22M), running in HOV lanes (average capital cost $5.83M), and dedicated busways (average capital cost $13.41M) (13). However, even consideration of both mode type and runningway type does not account for the full variation in the capital costs associated with different rapid transit modes. The issue of context, such as the land costs associated with Right Of Way acquisition and whether expensive tunneling is required, plays a crucial role in determining overall capital cost (13), making it difficult (and unwise) to attempt to make universal inferences by comparing the

Cain, Darido, Baltes, Rodriguez, Barrios

cost of rapid transit projects in different places. This is illustrated by the TransMilenio where, even within the same city, the capital cost per mile of Phase II was twice that of Phase I due to the more extensive public space improvements associated with the latter phase. This suggests that an accurate capital cost comparison across different rapid transit modes must be conducted in the same corridor during the same time period. Using average weekday ridership per mile as a measure of system performance, Table 3 shows that TransMilenio Phases I and II outperformed the U.S heavy rail systems by at least 3 times, U.S LRT by at least 12 times, and U.S BRT by at least 4 times. This result, however, is moderated by the fact that a transit trip in Bogot is not directly comparable to a transit trip in the U.S because of significant differences in the transit mode share, costs, levels and expectations of service, ride quality, and comfort, among other factors. In essence, one transit trip in the U.S. is equivalent to many trips in Bogot because transit does not have as many competitive advantages in terms of cost, utility, convenience, etc. Unfortunately, the demand function is not understood well enough to make such adjustments, but it can be said that the capital cost effectiveness of TransMilenio is likely to be superior to most transit investments of any mode in the U.S. Finally, it should be noted that overall system cost includes both capital and operating costs over the life of the project. While operating costs are less of an issue in Bogot, due to the fact that the TransMilenio is operated for profit by the private sector, the contribution of operating costs to overall system cost are much more of an issue in the U.S, where such costs require to be subsidized for the duration of the projects existence. Operating cost efficiency is an area where BRT has been criticized, due to the fact that high frequency service requires a large number of vehicle operators. Hidalgo (14) conducted an analysis comparing the total costs (capital plus operating) of different rapid transit modes over a 20 year period on a hypothetical 20km corridor carrying 35,000 pphpd. The analysis found that although the BRT option did indeed have higher operating costs than the LRT and HRT options, the much higher capital expenditures (infrastructure and vehicles) associated with these rail-based options meant that the BRT option was cheaper overall. However, the extent to which this analysis applies to the developed world it is not clear. While labor costs are not a significant proportion of overall operating costs in the developing world (typically less than 20 percent (2)), they are a much more significant factor in the developed world (30 to 75 percent of operating costs (2)). LRT, with higher capacity individual vehicles, requires less vehicle operators to carry the same number of passengers, and thus holds an advantage in this case. Wright states that this is a major reason why LRT is often preferred to BRT in the developed world (2). This also possibly relates to the fact that U.S federal transit funding mechanisms tend to favor capital investments over operating cost assistance (13), thus favoring LRT.

ACHIEVING SUSTAINABLE TRANSPORTATION OBJECTIVES THROUGH AN INTEGRATED MOBILITY STRATEGY Mayor Pealosa believes that quality of life is related to a living environment as free of motor vehicles as possible (15) and that transportation investment should be prioritized in inverse proportion to travel speed i.e. pedestrians first, then bicycle facilities, then public transit, and finally private vehicles (15). Thus, the TransMilenio was designed as the centerpiece of a

Cain, Darido, Baltes, Rodriguez, Barrios

city-wide sustainable mobility strategy that provided a network of non-motorized transportation infrastructure such as bicycle lanes, pedestrian walkways, and overpasses. The mobility strategy also actively discourages private vehicles use by banning 40 percent of the citys registered private vehicles from using the roads in the peak periods (7:00 to 9:00 a.m. and 5:00 to 7:00 p.m.) regulated by the last number of the license plate (16). The city also regularly imposes car free days on weekends and public holidays, whereby private vehicle use on major arterials is prohibited, and the roads are opened up to pedestrians and cyclists only. Figure 3 shows the impact that this mobility strategy has had on mode split within the city of Bogot. Figure 3 shows that the mobility strategy has increased transit usage (the sum of BusBuseta-Colectivo and TransMilenio) from 64 percent in 1999 to 70 percent in 2005. However, it should be noted that TransMilenio currently accounts for only around 27 percent of the citys total transit trips. The figure also shows that the proportion of non-motorized trips has increased from around 8 percent in 1999 to around 15 percent in 2005, while during the same period the proportion of vehicle trips has reduced from 18 percent to 11 percent. Overall, this suggests that the mobility strategy has induced some additional corridor transit ridership, and has succeeded in initiating a shift towards more sustainable modes. Mode shift has also been verified by recent survey data. Nine percent of surveyed riders stated that before TransMilenio they made the same trip by private car. However, it has also been reported that overcrowding and service interruptions caused by pavement failure problems has caused many middle-class choice riders to stop using TransMilenio (15). There are several different issues, with applicability to the U.S, that are raised by the success of Bogots mobility strategy. First, it illustrates clearly the benefits obtained from implementing BRT as part of wider package of measures that support sustainable transportation. In modern U.S cities, where the urban environment is often particularly hostile to non-motorized modes, even a high quality BRT system may not be successful in attracting ridership if the link between the system and users ultimate origins and destinations is not considered. Providing high quality access to a BRT system via non-motorized modes takes into account the complete doorto-door trip that transit users make. Experience in Bogot also shows that implementing concurrent restrictions on the private automobile may also enhances the likelihood of mode-shift to transit, though it is unlikely that such restrictions would be politically feasible in the U.S. Secondly, the TransMilenio also illustrates the importance of high quality transit infrastructure to the ability of transit to compete with the private auto. Hidalgo and Hermann argue that, for BRT to accommodate large passenger volumes at high levels of service, buses must be given at least one dedicated lane in each direction (16). Exclusive runningways, with at least one dedicated lane in each direction and passing lanes at stations, are desirable for multiple reasons: Guarantees that service conditions are maintained throughout the day, even during peak travel periods when surrounding general purpose lanes are congested. Guarantees that service conditions are maintained over time, as traffic volumes continue to rise in growing urban areas.

Cain, Darido, Baltes, Rodriguez, Barrios

Allows local, limited stop, and express trunk services to operate efficiently on the same facility (a major advantage over rail-based trunk services), providing high combined corridor service frequencies. As illustrated in Figure 4, such features offer the potential for a competitive advantage over the private auto, in terms of travel time, in congested urban areas. This is important in relation to the U.S, where transit is increasingly viewed as a means of mitigating congestion in urban areas. However, transits usefulness as a congestion reduction tool (and thus its ability to attract federal funding for such initiatives) rests on its ability to attract choice riders. With bus-based transit in the U.S typically viewed as the mode of the low-income, transit dependent traveler, rail based transit is often preferred due to its perceived ability to attract such choice riders. However, the prohibitive cost of rail, as discussed in the previous section, has lead to the current interest in BRT, characterized as a rapid transit mode capable of attracting choice riders at a more affordable cost. The recently published National Strategy to Reduce Congestion on Americas Transportation Network (USDOT, 2006) states that the creation or expansion of express bus services that benefit from free-flow traffic conditions is an important aspect of the goals of reducing urban congestion. Evidence from the recently implemented Orange Line BRT project in Los Angeles suggests that BRT can attract choice riders and have an impact on congestion. A study by UC Berkeley found that 18 percent of riders had previously made the trip by car, and that on parallel highways that average morning travel speeds had increased from 43 to 46 mph, that time spent in congested conditions (below 35mph) had decreased by 14 percent, and that the morning peak period now begins 11 minutes later (17).

BRT AND URBAN RENEWAL

As well as being the centerpiece of a city-wide mobility strategy, the TransMilenio also acts as the anchor for a city-wide urban renewal strategy that aims to stimulate urban renewal through public space improvements that center around the provision of major new station buildings, the re-development of public space around these stations, and corridor enhancements such as new medians and non-motorized mode infrastructure. Such large-scale improvements have succeeded in improving the aethetics of the city, and research suggests that there are initial signs of quantifiable improvements in land value in the vicinity of TransMileno corridors (18), (19). Figure 5 illustrates the changes to Bogots urban form that have been made in recent years. The strategys success demonstrates that: BRT systems are capable of anchoring public sector-led urban renewal programs. This is significant because bus-based transit is typically not viewed as being able to influence land use and stimulate economic development. However, it should be noted that Bogots urban renewal is financed by the public sector, and it is still unclear whether this will in turn lead to significant private sector-based changes in land use. BRT systems are capable of generating civic pride and making a positive impact on the social capital of a city. TransMilenio now is widely recognized by its population as one of Bogots defining cultural icons, similar to the way in which the subway systems in London and New York help to define those places.

Cain, Darido, Baltes, Rodriguez, Barrios

10

BRT systems can be used to achieve social inclusion objectives. TransMilenio has raised the level of access between the citys centrally-located employment centers and the deprived, peripheral areas of Bogot.

THE TRANSMILENIO BUSINESS MODEL

One of the greatest achievements of the TransMilenio system was the successful implementation of a concession contract-based system for regulating service operations. Requiring prospective operators to bid for exclusive rights to specific service routes, and paying them on a per-kilometer basis as opposed to a per-passenger basis, facilitated healthy competition for the market as opposed to the unhealthy competition in the market that characterized the previous traditional system. Allocating system operation to the private sector and system regulation and management to the public sector has allowed the respective strengths of each sector to be utilized. Operating efficiency has undoubtedly been enhanced, while reducing the fiscal risk imposed on Bogots city government. In contrast, the U.S transit industry is typically regulated and operated by the public sector as a Single Public Monopoly, and has been criticized for permitting too little private sector competition. Thus, Bogots traditional system and the U.S approach to transit provision represent two inefficient extremes in the spectrum of private sector involvement, suffering from either too much or two little private sector competition, while the new TransMilenio business model is more representative of the correct optimal balance of public sector involvement. If some form of regulated competition offers the greatest potential for efficient service provision, how can this be introduced in a country where fare box recovery ratios are well below 1.0? Experience in London suggests that the concession contract approach employed in Bogot may also be employed in locations where farebox recovery ratios are less than 1.0. In London, the government defines the bus routes, service characteristics, and fare levels, and then invites private contractors to submit bids that state the fixed annual subsidy that they would require to provide service. The winning bidder makes a profit by keeping costs below the fixed price paid by the government (20). Thus, a Win-Win situation is achieved by removing fare box returns from the profitability equation, offering the potential for private companies to make a profit, while still reducing the subsidy required from the public sector (due to the efficiency gains afforded by competitive private sector involvement). While relatively rare, such arrangements do exist in the United States. Foothill Transit in Los Angeles is a private company providing transit service in east Los Angeles under a concession contract arrangement with Los Angeles County Metropolitan Transportation Authority. Barriers to more widespread use of this approach in the U.S include the problem of low farebox recovery and cross-subsidy, the difficulty in providing socially desirable transit services in a non-self-sustaining economic context, and opposition from transit industry employee unions.

Cain, Darido, Baltes, Rodriguez, Barrios

11

POLITICS
Historically, most Colombian presidents and Bogot mayors, including Mayor Pealosa, were pro-metro. Hidalgo (14) cites a number of reasons for political support of heavy rail transit proposals. Table 4 lists these reasons and, in each case, comments on their applicability in the context of the U.S transit industry. In summary, bus-based rapid transit projects in the U.S do not currently capture popular imagination to the same degree as rail-based projects and are not assumed to be capable of generating or directing economic development. To a large degree, it appears that many people, and their political representatives, are not even aware that BRT alternatives exist while many Americans have experienced extensive metro networks in World Cities like New York, London, and Paris, relatively few by comparison have visited Curitiba or Bogot. While this problem may be beginning to be addressed by the emergence of high-profile BRT projects in U.S cities such as Boston, Los Angeles, and Las Vegas, it is likely that lack of political awareness will continue to be a barrier to the more widespread implementation of BRT systems. Other political issues raised by TransMilenios implementation in Bogot include: Decision-makers need to be encouraged to make public transit planning decisions based first on service requirements, and not by simply by promoting their preferred mode. It is important to identify a high-profile political figurehead to champion BRT projects. Such figures need to have the political power, courage, and commitment displayed by Mayor Pealosa to carry BRT projects through to successful implementation. The first section of Phase I opened in December 2000, less than three years after the system was conceived, and 12 days before the end of Mayor Pealosas term of office. Wright estimates that BRT projects typically take 1 to 2 years in design and 1 to 2 years I construction, while rail based projects typically require 3 to 5 years in planning and 3 to 5 years in construction (2). Thus, while rail-based projects tend to require support over several consecutive administrations, it is possible to implement a BRT project within one term of office, offering a distinct political advantage (assuming the project is regarded as a success!). TransMilenio could not have been implemented with out national government support. Similarly, in the U.S, federal government funding is essential to the high level BRT systems capable of attracting choice riders. The success of the TransMilenio system has resulted in a paradigm shift in governmental attitudes towards BRT, and the government is now backing the implementation of similar BRT systems across Colombia.

REPLICATING THE BOGOT MODEL

TransMilenio provides insight into what BRT systems are capable of achieving in terms of passenger capacity, capital cost effectiveness, and urban renewal. While TransMilenio undoubtedly provides an excellent example of what can be achieved in a developing world urban

Cain, Darido, Baltes, Rodriguez, Barrios

12

area, there is still the issue of whether these achievements can be replicated in developed world countries such as the United States. It is appropriate then that the paper should close with a discussion of the issues associated with replication of the Bogot Model. This is provided in Table 5, based on Hidalgos commentary on the applicability of the Bogot Model to other developing world countries (16). CONCLUSION Bogots TransMilenio is the highest capacity BRT system in the world and is one of the worlds best examples of a high level BRT system. As such, it demonstrates what can be achieved with BRT if certain high capacity design features and operating characteristics are adhered to. This paper has attempted to highlight the different capabilities of BRT that are demonstrated by the TransMilenio, taking into account the variety of differences between the city of Bogot and typical U.S cities. The following list is a summary of the main observations that have been made: The TransMilenio carries as much as 41,000 pphpd on its busiest sections, which equates to 28,000 pphpd under North American passenger loading conditions. This figure is similar to the assumed capacity of LRT systems in the U.S, proving that BRT should not be rejected in favor of LRT on the grounds of insufficient capacity. The TransMilenio demonstrates the wide cost effective capacity range of BRT, estimated at 4,000 to 40,000 pphpd The TransMilenio demonstrates the vast difference in capital cost between Bus Rapid Transit and Heavy Rail Transit. The experience in Bogot suggests that BRT offers a citywide rapid transit solution for a similar capital cost to one heavy rail corridor TransMilenios capital cost is similar to other high level BRT projects in the U.S that feature dedicated runningways, which represent the upper end of the BRT capital cost range TransMilenio demonstrates the importance of using an integrated mobility strategy to achieve modal shift objectives. This is particularly important in the U.S, where BRT is promoted as an affordable rapid transit mode that is capable of attracting choice riders and thus addressing urban congestion problems. BRT systems are capable of anchoring public sector-led urban renewal programs. BRT systems are capable of generating civic pride and making a positive impact on the social capital of a city. The TransMilenio demonstrates the benefits that can be obtained by achieving the correct level of public sector involvement in service operation. Too little as well as too much public sector involvement can be detrimental to operating efficiency. While rail-based projects tend to require support over several consecutive administrations, a BRT project may be initiated and completed within one term of office, offering a distinct political advantage.

Overall, TransMilenios great strength is the fact that it is a true rapid transit network. Constructing such a network, in a relatively short time period, has given the TransMilenio an economy of scale that has maximized operating efficiency and yielded city-wide improvements in mobility and urban renewal. The relatively low capital costs associated with BRT also provides U.S cities with the opportunity to implement rapid transit networks capable of addressing city-wide mobility objectives.

Cain, Darido, Baltes, Rodriguez, Barrios

13

REFERENCES 1. 2. TRANSMILENIO, S.A. http://www.transmilenio.gov.co. Accessed June 15, 2006. Wright, L. Module 3b: Bus Rapid Transit. Sustainable Transport: A Sourcebook for Policymakers in Developing Cities. Institute of Transportation and Development Policy. 2004. Cain, A., G. Darido, M.R. Baltes, P. Rodriguez, and J.C. Barrios. Applicability of Bogots TransMilenio BRT System to the United States. National Bus Rapid Transit Institute (NBRTI), University of South Florida, FL-26-7104-01, Tampa, FL, 2006. Bertaud, A. A Web Page Dedicated to the Study of Urban Spatial Structures. http://alain-bertaud.com. Accessed October 20, 2006. Cain, A., Hamer, P., & Sibley-Perone, J. Teenage Attitudes and Perceptions Regarding Transit Use. Final Report. National Center for Transit Research, University of South Florida, Tampa, Florida. 2005. http://www.nctr.usf.edu/pdf/576-14.pdf. Hidalgo, D. Bus Rapid Transit (BRT): A Technical and Institutional Instrument for Urban Transport Improvement. The Case of Bogots TransMilenio. Infrastructure and Financial Markets Review. Inter-American Development Bank, Sustainable Development Department, Infrastructure and Financial Markets Division. 2004 Steer Davies Gleave (SDG). (2000). Diseo Operacional del Sistema TransMilenio. Proyecto de Transporte Urbano para Santa Fe de Bogot, BIRF 4021-FONDATT-10, Bogot D.C. TCRP Report 100. The Transit Capacity and Quality of Service Manual. Transportation Research Board. Washington, D.C. 2003 Vuchic, V.R. Public Transportation. New Jersey: Prentice Hall. 1992 Samuel, P. Busway vs. Rail Capacity: Separating Myth from Fact. Policy Update 16. Reason Public Policy Institute. 2002 Hidalgo D. TransMilenio: A High Capacity Low Cost Bus Rapid Transit System developed for Bogot, Colombia. In Proceedings of the Tenth International CODATU Conference, Lome Togo, 12-15 November 2002. Hidalgo, D. Personal Communication. November December 2005. Hess, D.B., Taylor, B.D., & Yoh, A.C. Light Rail Lite or Cost-Effective Improvements to Bus Service? Evaluating Costs of Implementing Bus Rapid Transit. Transportation Research Record: Journal of the Transportation Research Board No. 1927. Washington, D.C. pp 22-30. 2005.

3.

4.

5.

6.

7.

8.

9. 10.

11.

12. 13.

Cain, Darido, Baltes, Rodriguez, Barrios 14.

14

Hidalgo, D. Comparing Transit Alternatives After Recent Developments in BRT in Latin America. Presented at the 85th Annual Meeting of the Transportation Research Board, Washington, D.C. January 2006. Cervero, R. Progressive Transport and the Poor: Bogots Bold Steps Forward. Access Journal of Transportation Research, at UC Berkeley, CA. No 27, Fall 2005. Hidalgo, D. & Hermann, G. The Bogot Model for Sustainable Transportation; inspiring developing cities throughout the world in: TRIALOG - Zeitschrift fr das Planen und Bauen in der Dritten Welt / A Journal for Planning and Building in the Third World, 82 / 3-2004. 2004 Liu, C. Orange Line Eases A.M. Rush on 101 Freeway. Los Angeles Times. December 30, 2005. Targa, F. Examining Accessibility and Proximity-Related Effects of Bogots Bus Rapid System Using Spatial Hedonic Price Models, Department of City and Regional Planning, University of North Carolina at Chapel Hill. 2003. Barrios, V.P. Influye La Accesibilidad en la Renta Inmobiliaria? - Estudio Del Caso Del Sistema De Transporte Masivo TransMilenio Bogot. Masters Thesis. Economics Department. Universidad de los Andes. 2002. Savas, E.S. & McMahon, E.J. Competitive Contracting of Bus Service: A Better Deal for Riders and Taxpayers. Civic Report 30. Center for Civic Innovation. Manhattan Institute. 2002.

15.

16.

17.

18.

19.

20.

Cain, Darido, Baltes, Rodriguez, Barrios

15

LIST OF FIGURES

FIGURE 1: FIGURE 2: FIGURE 3: FIGURE 4:

Phases I and II of the TransMilenio System Typical Trunk Corridor Terminal Station Mode Split from 1998 to 2005 in Bogot (Source: TransMilenio S.A) The Importance of Dedicated Runningways to Achieving a Competitive Advantage Over the Private Auto Urban Renewal in Bogot

16 17 18 19

FIGURE 5:

20

LIST OF TABLES TABLE 1: TABLE 2: Comparing Bogot with U.S Cities Comparison of the 1997 Heavy Rail Proposal with TransMilenio Phases I and II Comparison of TransMilenio Capital Costs with Different Rapid Transit Projects in the U.S Reasons for Political Support of Rail Transit Replicating the Bogot Model - Issues to Consider 21 22

TABLE 3:

23

TABLE 4: TABLE 5:

24 25

Cain, Darido, Baltes, Rodriguez, Barrios

16

Phase I Phase II: - Americas - NQS (north) - NQS (south)/Suba

FIGURE 1 Phases I and II of the TransMilenio System

Cain, Darido, Baltes, Rodriguez, Barrios

17

FIGURE 2 Typical Trunk Corridor Terminal Station

Cain, Darido, Baltes, Rodriguez, Barrios

18

70% 60% 50% Bus-Buseta-Colectivo 40% 30% 20% 10% 0% 1998 1999 2000 2001 2002 2003 2004 2005 TransMilenio Non-Motorized Private Vehicle-Motorcycle

FIGURE 3 Mode Split from 1998 to 2005 in Bogot (Source: TransMilenio S.A)

Cain, Darido, Baltes, Rodriguez, Barrios

19

FIGURE 4 The Importance of Dedicated Runningways to Achieving a Competitive Advantage Over the Private Auto

Cain, Darido, Baltes, Rodriguez, Barrios

20

Before TransMilenio

After TransMilenio

FIGURE 5 Urban Renewal in Bogot

Cain, Darido, Baltes, Rodriguez, Barrios TABLE 1 Comparing Bogot with U.S Cities
Bogot, Colombia: Population of about 8 million in greater Bogot: Higher density development CBD-focused urban form Lower income groups live on the city periphery High demand for transit (>80% of mode share): Car travel is relatively expensive Majority do not own cars, many captive riders Capital cost of transit infrastructure requires government assistance Fare box recovery ratio > 1.0 Transit service operated by private sector, regulated by public sector Private sector responsible for vehicle purchase Political structure characterized by a powerful executive body U.S Cities:

21

Only a handful of US cities >5 million population: Lower density development Decentralized activity centers Lower income groups live near the CBD Low demand for transit (<5% of mode share): Car travel is relatively cheap Majority have access to a car Capital cost of transit infrastructure requires government assistance Fare box recovery ratio < 1.0 Transit service regulated and operated by public sector Service operations are subsidized City executives and mayors in the U.S generally wield less power and influence over policy implementation

Cain, Darido, Baltes, Rodriguez, Barrios

22

TABLE 2- Comparison of the 1997 Heavy Rail Proposal with TransMilenio Phases I and II
Heavy Rail (1997 Proposal) Capital Cost (Infrastructure) Vehicles / fare collection Total Capital Cost* Length km (Length miles) Cost per km (Cost per mile) Weekday Ridership Coverage of Citys Total Transit Trips
*

TransMilenio Phase 1 $240M $100M $340M 41km (25.6 miles)

TransMilenio Phase II $545M $80M $625M 41km (25.6 miles)

TransMilenio Master Plan $2,300M $1,020 $3,320M 388km (241 miles)

$2,350M $691M $3,041M 29km (18.0 miles) $105M / km ($169M / mile) 795,000** 16 percent****

$8.6M / km $8.3M / km $15.2M / km***** ($13.8M / mile) ($13.3M / miles) ($24.4M / mile) 792,000 16 percent 468,000*** 10 percent 5,000,000 85 percent

The heavy rail costs are provided in 1997 prices while the TransMilenio costs are in 2003 prices. Accounting for inflation would slightly increase the cost the heavy rail option relative to the TransMilenio options. ** Estimated daily ridership without integration with TransMilenio system and traditional buses. *** In May 2006, shortly after Phase II completion, total system ridership was measured at 1,260,000. This figure is continuing to rise and is expected to stabilize at around 1.4 million by the end of the year. **** As stated previously in this report, Ardila-Gomez (2004) stated that the metro proposal would account for only eight percent of the citys total transit trips. A later analysis by Hidalgo (1999) estimated metro proposal coverage at 26 percent (1,325,000 daily trips) if fully integrated with the new TransMilenio system, and 16 percent (795,000 daily trips) without integration. It is the view of the authors that the 16 percent figure is the most appropriate for comparison purposes. ***** Further analysis of Phase II costs show that only 37 percent of the costs are specifically for TransMilenio infrastructure (exclusive lanes, stations, terminals, pedestrian overpasses, and studies and designs). Of the remaining 63 percent, 40 percent is for general traffic lanes, non-grade intersections, sidewalks and utilities; 11 percent is for right of way acquisition; and 12 percent is for environmental and social mitigation, traffic management, detours, and maintenance. Thus, the TransMilenio portion of Phase II costs US$207M, equating to approximately $5M per km.

Cain, Darido, Baltes, Rodriguez, Barrios Table 3 Comparison of TransMilenio Capital Costs with Different Rapid Transit Projects in the U.S*
System / Location Mode (runningway) Capital Cost Average Average Trunk Capital Cost Weekday per Mile Length (2003 USD Weekday Pass / (2003 USD (Miles) Ridership $M) Mile $M / Mile) 14 7.5 2.3 14.0 25.6 25.6 15.0 11.6 16.5 112.0 $.3.2 $20.3 $27.3 $350 $340.0 $625.0 $397.3 $612.6 $5,557.3 $16,300.0 (Estimated) $0.23 $2.7 $11.9 $25.0 $13.3 $24.8 $26.5 $52.8 $337.6 $145.5 6,100 6,300 14,000 21,000 792,000 468,000 20,000 25,000 96,000 956,000 436 840 6,087 1,500 30,907 18,281 1,333 2,155 5,832 8,536

23

San Pablo Rapid BRT Oakland, CA (mixed traffic) Las Vegas MAX BRT Las Vegas, NV (bus lane) SilverLine Phase I BRT Boston, MA (bus lane) Metro Orange Line BRT Los Angeles, CA (exclusive ROW) TransMilenio (Phase 1) BRT Bogot, Colombia (exclusive ROW) TransMilenio (Phase 2) BRT Bogot, Colombia (exclusive ROW) North South Corridor LRT Salt Lake City, UT (surface) Hiawatha Corridor LRT Minneapolis, MN (surface + tunnel) Red Line HRT (underground) Los Angeles, CA Metrorail HRT Washington, D.C

* Includes estimated cost of vehicles and fare collection costs. Sources: 2003 National Transit Database; APTA 2005 Heavy Rail Transit Ridership Report; FTA Heavy and Light Rail Capital Cost Studies (2003 and 2004 updates) by Booz Allen Hamilton; FTA BRT evaluations in Boston and Las Vegas (2005) by Booz Allen Hamilton

Cain, Darido, Baltes, Rodriguez, Barrios TABLE 4 Reasons for Political Support of Rail Transit
Hidalgo 2006* Rail capital costs are usually funded with federal assistance. Municipalities may prefer metros to maximize disbursements to the local economy. BRT may require reorganization of existing bus routes and operations. This is politically difficult, and unpopular among local administrators. Most metros are implemented without any, or minimum, nuisance to existing private operators. Metros are viewed as more modern than buses. Comments on Applicability to U.S Yes, there is a perception that (dollar for dollar) rail maximizes benefits to the local economy. Reorganization of existing routes is not as significant an issue in U.S because transit tends to be publicly operated and managed.

24

Yes, the stigma associated with buses is perhaps even more significant in the U.S. The rail industry actively promotes the Yes, the BRT industry in the U.S is in a relatively implementation of Metros. nascent state. Yes, but this may change as programs and institutions BRT is not yet understood by decision-makers. continue to promote BRT and other bus solutions. Decision-makers typically do not use transit, but may Yes, local decision-makers of cities without rail transit find a Metro in a developed-world city comfortable, may view metros as a way to make their cities world reliable and quiet. class. *Paraphrased from Hidalgo, D. (2006). Comparing Transit Alternatives After Recent Developments in Latin America. Presented at TRB 2006.

Cain, Darido, Baltes, Rodriguez, Barrios TABLE 5 - Replicating the Bogot Model - Issues to Consider
Issues* Defining a vision of the transport system required for the city: Many local administrators have no clear long-term vision for their cities, and therefore continue existing plans that favor individual mobility or large-scale urban rail projects. These type of projects have long maturation time that extends beyond a given election cycle. If plans are changed in the next election, definitive solutions are often postponed for decades. Getting resources for project preparation: BRT projects are complex and require technical, financial, and legal studies to guarantee successful implementation. There is usually a lack of money set aside for such studies. Some local administrations receive international cooperation or multilateral lending, but only after many months of hard work. Completing technical studies: project design itself is not difficult, since the main engineering concepts for bus priority and operations are well known. Nevertheless, large data collection and model calibration studies are needed for successful demand forecasting, which is then used to define the fleet size, required services, and potential system income. All of this takes time, since data, such as origin-destination matrices and transit networks, are seldom readily available. Getting local transit providers to participate in the process (or at least overcoming their opposition): in developing countries, individual vehicle owners usually provide public transportation services. These bus owners and operatorsoften affiliated with trade unions or cooperativesmake their profits by reducing or eliminating vehicle maintenance and forcing drivers to work long hours without benefits. They tend to instinctively oppose, through political influence and strikes, any effort to rationalize the public transport sector out of fear that efforts to improve service will harm them financially. Winning over the traditional transport sector is perhaps the biggest challenge to the successful implementation of BRT systems in developing cities. Making changes in law, regulations and institutional frameworks: this may include the creation or modification of agencies that are responsible for system oversight, bidding and contracting. This requires political bargaining within the elected bodies (city council, state legislature or national congress). Comments on Applicability to U.S.

25

Yes, the political cycle in the US is typically 4-5 years, which is enough time to implement BRT but perhaps not rail. In many ways, this is less of a problem in the U.S cities which in the past have preferred more expensive rail solutions over bus. As we move into a more financially constrained environment for transit federal funding, there seems to be growing support for lower cost options.

This should be less of a problem in the U.S since many cities are required to regularly update regional travel demand models.

This is less of an issue in US since providers and/or managers tend to be public agencies.

Yes, this would include: creation or modification of agencies that are responsible for the systems oversight, bidding and contracting in an efficient and transparent process appropriate distribution of rights, responsibilities and risks

Creating adequate incentives for a sound economic scheme that remains Yes, this is probably more difficult in the U.S over time: designing appropriate distribution of rights, responsibilities and without profit as a motivator. risks is always difficult as long as there are conflicting interests between Appropriate use of concession contracts may maximizing profits and providing quality service. Neglecting this important be a way to achieve this in the U.S. issue significantly reduces the chances of a successful BRT system. Securing financial resources for infrastructure: despite the relative low cost of BRT as compared with rail-based alternatives, developing cities tend not to have capital readily available. Many resort to new taxes (fuel, value Funding programs such as FTA Small capture), which need approval by elected bodies, and grants from the Starts are becoming more accessible to BRT national government. Some also seek finance from multilateral lending projects. institutions, which takes time. Government-to-government concessionary loans would help increase BRT projects, just as they helped propagate railbased alternatives in the past. Bidding service operations and infrastructure: Developing proposals and The current limited size of the outsourced evaluating and awarding contracts in a transparent way is often difficult as operations market in the U.S. may not yet be long as public agencies are subject of pressures from interest groups. able to support healthy competition. Source: Hidalgo, D., & Hermann, G. (2004). The Bogot Model for Sustainable Transportation; Inspiring Developing Cities Throughout the World.