Bus Rapid Transit: Is Transmilenio a Miracle Cure?

Abstract

Successful mass transit solutions are rare in poor cities. When they appear they are lauded across the globe and too often copied uncritically. The latest exemplar of such best practice is the ‘Transmilenio’ rapid bus system in Bogotá. The article describes its main characteristics and applauds the improvements that it has already brought to urban transport in Bogotá. Naturally, the system is not without its flaws and these need to be drawn to the attention of those who might copy the Bogotá example. This is particularly important at the present time when the jewel of Bogotá has come under surprisingly strong local criticism over its cost, its ownership structure, its decreasing effectiveness and, fundamentally, because it has failed to solve the transport chaos of Bogotá. There is a real danger that ‘Transmilenio’ will stagnate as its popularity declines and as demands for a metro increase. Given the strengths of the system that would be something of a disaster and, most certainly, not in the interests of the poor.

Notes

1. UN‐Habitat (2001) estimates that around 40% of all trips to and from work in low‐ and middle‐income countries are by bus. In Delhi, buses cater for half of all passenger travel (Tiwari, 2002, p. 99) and in Santiago and Bogotá most motorized trips are made by bus (Vasconcellos, 2001).

2. According to Perry (2000, p. 397): “While a pedestrian needs 1.5 square metres to stand and 3 square metres to walk, a car requires on average 91 square metres standing, taking into account all the passageways necessary for access to parking spaces, and 914 square metres while moving at 48 kph”.

3. World Bank (2002, p. 113) suggests BRT systems can carry up to 35 000 passengers an hour although Hidalgo et al. (2007, p. 11) cites figures from Bogotá claiming that the Caracas corridor carries 45 000 passengers per hour. The head of Transmilenio recently claimed that the maximum carried was 42 000 per hour (El Tiempo, 3/2/2007).

4. Debate relates to the routes, particularly those along Carrera Séptima and Calle 26, about whether Mayor Garzón should have postponed the decision on Carrera Séptima and whether the city should really be building a metro.

5. A significant issue insofar as Enrique Peñalosa, the former mayor who introduced the system, was a leading candidate. The election took place on 28 October 2007, and Samuel Moreno beat Peñalosa by a wide margin. He will take office on 1 January 2008.

6. Particular thanks are due to the Leverhulme Trust, which financed the project, to María Teresa Garcés who ran the local team in Bogotá and to the people who were willing to discuss transport matters with me in Bogotá including: the current mayor, Luis Eduardo Garzón, and ex‐Mayors Paul Bromberg, Antanas Mockus and Enrique Peñalosa; planning and transport specialists Arturo Ardila, Mario Noriega, and Ricardo Montezuma; but operators Estéban Gómez, Gustavo Gómez, Milena Martinez, Victor Raúl Martínez and Andrés Ortiz; and Astrid Martínez and Angélica Castro and their staffs at Transmilenio S.A.. The paper also benefited from help and additional comments from Ana María Angel Garcés and three anonymous referees.

7. Several questions were modified in the 2005 survey.

8. Using the local press is justified in the sense that it is frequently reporting information that has been derived from local experts or from Transmilenio itself. And, when it is reporting criticisms of the system, it is a perfectly reasonable measure of the frequency and strength of local opinion. In any case, several different press sources have been used and wherever possible every statement made in the press has been checked against other sources. It should also be noted that while Transmilenio was an important issue in the 2007 campaign for mayor, debates about the system have always figured prominently in newspaper coverage of the city.

9. Presentations by Enrique Peñalosa and Samuel Moreno at Club El Nogal on 27 August and 10 September 2007, respectively, presentations on mobility in the city by five mayoral candidates at the University of Rosario on 4 September 2007, and a television debate on 29 September 2007.

10. It remains an independent agency but since January 2007 reports to the new Secretariat of Mobility. Arguably this gives it less direct access to the Mayor.

11. During strikes the system has carried up to 1.6 million passengers in a day.

12. McCormick (2005) claimed that Lima, Mexico City, Quito, Panama City had projects underway and Beijing, Dakar, Delhi, Djakarta, Johannesburg and Cape Town were showing strong signs of interest.

13. Bogotá’s long history of running its own bus or tram company did not provide much in the way of improvement. The last of a series of public companies, EDTU, was established in 1959, partly as a means of avoiding total paralysis in the event of a strike organized by 36 private operators (Rodriguez and Nuñez, 2003, p. 258). At its peak, in the 1960s, it carried around one‐tenth of the city’s bus passengers. Gradually, the private sector began to operate more routes and EDTU passengers fell dramatically. In 1989, the company was carrying only 4.4 million passengers a year, ten times that number when compared with a decade earlier. In 1990, it was operating only 5 of the city’s 430 routes and its fares covered only three and a half months of the annual salary bill (Caicedo, 1992, p. 223). The company was described by Mayor Prieto Ocampo, in 1996, as “showing all the ills bought about by mistaken and improvised management” (Prieto Ocampo, 1976, p. 36).

14. For examples of the difficulties faced by the drivers, see Perea (2002).

15. The average passenger in the middle 1990s was spending 123 minutes each day on the bus (JICA et al., 1996).

16. The transport secretariat has been reorganized many times over the years, most recently in January 2007.

17. Mentioned by Paul Bromberg during interview.

18. Like others I was given a copy of the article during my first interview with him.

19. Seemingly he was still talking about the contract 100 days into his period of office (La Rebeca, 1998).

20. In recent debates with the pro‐metro candidate, Samuel Moreno, Enrique Peñalosa has repeated that he had little choice but to build Transmilenio, given the new president’s reluctance to approve spending on a metro. It is important to note that both Peñalosa and Andrés Pastrana took office at a time of increasing economic recession.

21. Cárdenas (2005, p. 191) offers a different explanation: “The economic technocracy of the national government ruled out the metro and opted for Transmilenio because it was the only fiscally feasible alternative. In fact, the national government actually chose the TM solution while the city administration was still considering the metro as an option”.

22. Even in Caracas, only 15% of passengers are carried on the metro and in Medellín a BRT system is being constructed to complement the metro.

23. Even if some rat runs have developed along nearby streets and some of the bridges, that were built to provide pedestrians with safe routes to and across Transmilenio, are extremely ugly.

24. And, in the light of the chaos associated with the opening of TransSantiago, it operates spectacularly well.

25. They are actually more like a tile because they are made of cement but to English speakers the ‘tile’ tends to be used for roofs and walls rather than for roads and paving stones for pavements or ‘sidewalks’.

26. Some claim that these protests were organized by the bus companies, i.e. the drivers being displaced by the system or even by the FARC guerrillas.

27. The articulated buses are 17.5 metres long and 2.6 metres wide. Reducing that length by 3 metres to allow for the driver and entrances, this gives a theoretical average amount of space of 0.24 square metres per passenger if the bus is carrying a full complement of 160 passengers. In fact, the space available to the 112 standing passengers will be less than that because the seats take up proportionately more room. Similar length ‘bendy buses’ in London have a maximum capacity of 140 passengers and one consulting company suggests that it is undesirable for standing passengers to have less than 0.55 square metres of space (http://www.crowddynamics.com/Egress/Overcrowding%20on%20public%20transport.htm). According to Martinez and Jimeno (2007, p. 189) an ideal measure is more than 0.4 square metres per person, a normal measure is between 0.2 and 0.4, and below 0.2 is bad. But it is not just overcrowding on the Transmilenio buses that is a problem. During rush hours, queues are sometimes formed to enter the stations and there is frequent crowding on the stations themselves.

28. The element of danger was emphasized after two bombs were placed on the feeder system in 2004.

29. No doubt much of the criticism was associated with the 2007 mayoral campaign; Juan Carlos Flores was one of the candidates. As noted above a new Secretariat of Mobility began operating in 2007.

30. Every director of a government agency that I have consulted, including the current mayor himself, attests to taking these surveys very seriously indeed. The heads celebrate news of good ratings and worry about bad results.

31. At the top came the Botanic Gardens with an approval rating of 97%; at the bottom, the Defender of Public Space with 50%.

32. Martínez and Jimeno (2007: 158) note that: “During Phase II, bidding requirements demanded that at least 10% of the concessionary company’s property be in the hands of transporters owning at least two buses.”

33. A recent interviewee told me that several big investors bought share in the Phase I operators but this ceased in Phase II because of the lower profits.

34. Garzón was accused of receiving around US$100,000 million of campaign finance from a well‐known transport operator, Carlos Delgado, who is based in the neighbouring municipality of Soacha. The mayor said that the money was a loan to finance the campaign, not a gift, and returned the money but the news did take a little of the gloss off his image of political independence. Most importantly, it exemplified how the bus companies attempt to influence the authorities.

35. Ciudad Móvil SA has very few shareholders but its feeder company Conexión Móvil has 780 small owners who own 15% of the company. The latter helped them obtain the credit to buy their shares.

36. Although one of the referees believes that this is an irrelevant issue and that the only important issue is whether the system is well run. While the question of efficiency is critical, some concession to the former drivers and owners ought to be made in Bogotá for both social and political reasons. It is difficult to ignore the poverty of many of the former drivers, and Transmilenio will have difficulty in operating effectively if it is being opposed by such a powerful lobby as the transporters. Wider participation in the ownership of the Transmilenio companies represents one way, but not the only way, of addressing these problems.

37. Incidentally, some drivers of the old buses are recruited by the operating companies either as drivers or as cleaners or mechanics. However, one operator is extremely reluctant to use them as drivers.

38. Although it was in surplus in 2006 and should break even now that its share of the receipts has risen from 3% to its current level of 6.95% (data from Transmilenio SA).

39. All bus companies were buying diesel from Ecopetrol in August 2005 at 4000 pesos (cUS$1.72) per gallon compared with the international price of US$2.3. Decree 2988 of 2003 ordered the phasing out of that subsidy from 1 January 2005. It is important to note however that all consumers of diesel benefit from this subsidy not just the bus companies.

40. The original estimate for building 25 corridors was U$2.94 billion, of which US$1.97 billion would be for infrastructure (CONPES, 2000, p. 4); a total cost of US$7.57 million per kilometre at 2000 prices for the whole 387.9 kilometre system or US$5.08 million per kilometre for the infrastructure alone (Table 3). Hidalgo (2007, p. 15) estimates the combined infrastructure and equipment costs at US$8.2 million per kilometre. Martínez and Jimeno (2007, p. 156) cite an estimate made by Jorge Acevedo that infrastructure costs during Phase I were US$7 million per km and reached US$25 million during Phase II. The internal rate of return calculated for the metro and for Transmilenio was extremely favourable to Transmilenio: 61% calculated by UNECLAC (Gómez, 2004, p. 101) compared with 15.8% for the metro (calculated by CONPES and cited in Ardila‐Gómez, 2004, p. 316). Even the World Bank does not seem to have calculated the full subsidy, although it did calculate the internal return on the investment, which again proved much higher than that on a metro. Nevertheless, the construction subsidy is certainly not insignificant. An early estimate of the cost of building 24 bus lanes across the whole city was US$2.387 billion, about the same as building a single metro line. This does not cover the cost of the road space dedicated to the bus lanes. Later estimates of the infrastructure cost of the first three bus routes lanes, 41 kilometres in length, was US$213 million. This included about US$80 million for the rehabilitation of the mixed traffic lanes adjacent to the busways but not US$38.8 million for property takings. Including them, the total cost of the civil works is approximately US$254 million (Ardila‐Gómez, 2004, pp. 368–369).

41. In September 2007, Enrique Peñalosa was claiming that Transmilenio cost US$5 million a kilometre while Samuel Moreno was claiming US$20 million. According to the former, a metro would cost US$100 million a kilometre, and the latter, a maximum of US$50 million (El Tiempo, 5/9/2007). In January, 2007, the Minister for Transport claimed that the national government would not pay for a metro at a cost of US$200 million a kilometre (El Tiempo, 26/1/2007).

42. Bids are invited for particular routes at set fare levels and the lowest tenders win. One referee suggests that whether or not ex‐post level of profits have proved to be high is irrelevant; the only significant issue is how the concessions were tendered. Most local observers would disagree and UN‐CID (2005, p. 12) notes that the Phase I contracts were negotiated during an economic recession which naturally affected the operators’ perceptions of risk. Because the economic situation improved markedly after the formula had been agreed, this reduced the actual level of risk and allowed the companies to make higher than anticipated profits. In any case, tendering processes are not foolproof because they can be manipulated by bidders to avoid real competition (Ardila‐Gómez, 2004, p. 360; Echeverry et al., 2005, p. 182). As such, Transmilenio should be checking the profitability of the operators before tendering begins for each phase of the programme. If profits are too low, as with the Phase II feeder contracts, then the formula needs to be made more generous. If these are too high, as with the Phase I corridor contracts, then the formula should be made less generous. This is precisely what happened with the contracts offered for the Phase II corridor routes.

43. Ardila‐Gómez (2005) notes that it is difficult to find figures on the level of profits.

44. Although CID (2005, p. 38) provides a series of lower figures, none exceed 17%.

45. Transmilenio SA claim that four companies providing feeder services have been making losses and may give up their franchises unless they are helped. Indeed, I have since been informed that negotiations took place during October 2007 which increased the sum paid per passenger by one‐third. However, another interviewee could not understand how feeder companies could make a loss given what he considered to be a reasonable payment per bus. One answer is that the companies carry more passengers than they are paid for because people get on the feeder buses and get off before the bus arrives at the interchange where the passenger numbers are calculated (Martínez and Jimeno, 2007: 163). However, one interviewee could not understand how they could make a loss given what he considered to be a reasonable payment per bus.

46. Of course, Transmilenio is much equitable than the traditional system insofar as it helps women and the disabled. A Gallup poll in 1998 found that 83% of women used public transport compared with only 54% of men, so any improvements in speed and convenience through the new system will benefit women more than men (El Espectador, 23/8/1998). The term ‘middle‐income’ fails to indicate just how poor many of those people are.

47. A possible alternative is to replicate the MetroCable system used in Medellín—a system of cable cars that links the slums on the high slopes to the metro.

48. Construction of the route along Séptima was put on hold by Mayor Garzón during 2007 and incoming Mayor Moreno does not seem inclined to reactivate it. There is also opposition to building Calle 26 towards the airport, mainly on grounds that it will generate too few passengers.

49. EIDHB (2006, p. 60) estimate that people living in the poorest settlements in Bogotá have the longest average journey times (Ciudad Bolívar: 11.3 km, Usme: 15.0 km and Bosa: 11.9 km).

50. Interviews with two officials of Transmilenio.

51. Bogotá’s bus fares are determined according to the nature and age of the bus. In July 2007 Transmilenio fares rose to 1400 pesos, ordinary colectivo fares to 1200 pesos and buses and busetas older than six years remained at 1000 pesos. Subsequently, the cheapest fares were raised to 1100 pesos.

52. According to El Tiempo (26/2/2007), the fare would have risen to 1700 pesos but for a subsidy on the price of diesel that the national government conceded to the bus companies. EIDHB (2006, p. 55) estimates that, in 2004, the poorest stratum spent 19.1% of their household budget on transport and communications, stratum 2 16.9% and stratum 3, 15.5%. They also claim that the average Bogotá home spent 14.2% on transport and communications in 2004 compared with 7.7% in 1980s.

53. I calculated the figure for this 19‐week period as it omitted major holidays. The comparative figure for the same 19‐week period of 2006 was 1.1 million passengers a day.

54. Local transport expert, Darío Hidalgo, is reported to have argued that the difference in fares between Transmilenio and the ordinary buses encourages people to use the latter (El Tiempo, 19/7/2007). Usage is likely to increase further as many of the old buses are carrying very few passengers and because so many routes are still being operated, the traditional buses often eliminate the need to change buses as is necessary with some Transmilenio routes. BCV (2007) surprisingly shows that the old buses are now more popular than Transmilenio because they are more direct, the service friendlier and most passengers can find a seat.

55. Figures kindly provided by Angélica Castro.

56. The average journey distance during morning rush hour rose from 11.7 to 12.9 kilometres.

57. Since the number of passengers using the buses determines how many buses actually leave the garages, companies may not operate all of their buses.

58. One interviewee claimed that the cost of an old bus had doubled since the District had begun to purchase some buses directly, another that the TM companies now had to pay $80 millions. In Phase II, the companies had to buy 7.7 buses for every articulated bus, which increased their investment costs considerably. An articulated bus costs approximately US$200,000. To put a feeder bus into operation requires the scrapping of two old buses.

59. A temporary solution was to borrow buses from Pereira whose own system had not begun operations.

60. Decree 115 of 2003.

61. The Administrative Tribunal of Cundinamarca annulled the decree on 28 April 2005.

62. I understand that in November 2007 the fund contained 100 million pesos but only 27 buses had been scrapped.

63. Although they did finally approve it towards the end of 2006.

64. Construction of the troncal along Carrera Séptima has been temporarily postponed, Mayor Garzón having decided to leave the decision to his successor.

65. Perhaps this is why Mayor Garzón and his Secretary of Mobility have recently been discussing the desirability of an operating subsidy (El Tiempo, 20/8/2007).

66. Gómez (2004, pp. 18–19) estimates that the number of cars using Bogotá’s roads is increasing annually at 10% while Montezuma (2005) thinks that the growth is nearer to 7% per annum. Either way, with the number of buses and taxis also growing rapidly, it is no surprise that the roads cannot cope.

67. The poor quality of the diesel coming from the national petrol company is a major contributory factor but the contamination from the old buses is an important source of pollution and the long traffic jams are another. The fact that some bus companies do not pay the fines that are imposed on them does not help control pollution.