SPECIAL ARTICLEEconomic & Political Weekly january 26, 200841Mythologies, Metro Rail Systems and Future Urban TransportDinesh MohanThere is still no clear vision among planners, policymakers and transport experts about what will make Indian cities better places to live in as far as mobility and access are concerned. The prevailing mythology is that construction of metro rail systems will somehow solve the problems of the future. A review of urban mass transport systems over the past century shows that metro systems were the obvious choice when relatively inexpensive cars and two-wheelers were not available. With the introduction of efficient buses, computer and information technologies to manage large fleets, and the need to have flexible, medium capacity systems that go close to homes and destinations, bus rapid transit systems with dedicated lanes seem to be the only choice for providing affordable mass transport in our cities.Consider the following statements on urban transport: A comprehensive bus system – which would help remove thou-sands of cars from the streets – can be set up for the same cost as constructing a flyover, which often only serves to shift a traffic jam from one point to another… It is crucial to give due consideration to the magnitude of a project in order to avoid the risk of presenting “show-case” solutions which are conceived for the media and only benefit a minority of the inhabitants [Lerner 1996].1Unfortunately sometimes rail systems are also chosen for the worst possible reasons… Rail system salesmen are legendary for the proce-dures they utilise for selling their expensive wares…bus systems are more flexible, an important asset in developing countries’ dynamic cities. As a city attraction centre shifts, it is easier to adjust a bus system than a rail one [Penalosa 2004].2A sustainable city is one that wastes the least and conserves the maxi-mum. Most importantly, it means making the existing system of people and resources work better – rather than throwing it away and trying to replace it with a single, capital-intensive project such as a subway or a rail-based system [Rabinovitch 1995].The demand for rail has continued to shrink because transit networks are unable to keep up with changing land use and travel patterns that have decentralised residences and employment...Unfortunately, transit systems have been able to evolve because their supporters have sold them as an antidote to the social costs associated with automobile travel, in spite of strong evidence to the contrary. As long as rail transit continues to be erroneously viewed in this way by the public, it will continue to be an increasing drain on social welfare [Winston and Maheshri 2007].The above comments come from a wide spectrum of interna-tional experts, planners and politicians covering a period of more than a decade. Now examine the recent statements on need for metro rail projects for Indian cities from policymakers in India:A rail based metro system is inescapable…World-over the practice is that when the population of a city reaches 1 million mark, the studies and investigations needed for a metro system are taken up [Sreedharan 2004].Punjab took a vital step towards the Mass Rapid Transit System (MRTS) era on Tuesday with the state government inking a Memorandum of Understanding (MoU)… Similar rail systems would be set up at Amritsar and Mohali, steps for which are being taken on a priority basis… Speaking on the occasion, chief minister Parkash Singh Badal described it as a historic event which would completely revolutionise the development scenario in the state.3The Bangalore metro comes with a package deal. Comfortable, quick, safer and economical, its energy requirement per passenger km is only one-fifth of that of road-based systems. Commuters can breathe easy, there will be no air pollution as the system runs on electric power. The economic rate of return is 22.3 per cent.4The city is all set to have a metro rail of international standards to meet the increasing traffic congestion. The Andhra Pradesh cabinet today approved the Rs 8,760 crore Hyderabad metro rail project covering 66 km and involving the construction of 63 stations.5Dinesh Mohan (dmohan@cbme.iitd.ac.in) is at the Transport Research andInjuryPreventionProgramme and the WHO Collaborating Centreat the Indian Institute of Technology, Delhi.

SPECIAL ARTICLEjanuary 26, 2008 Economic & Political Weekly42Why are Indian policymakers out of sync with international expert opinion on the choice of urban transport technologies and policies? Are Indian cities completely different from the rest of the world? Or, are the Indian experts oblivious of international developments in urban transport theory and practice? Is it possi-ble that metro rail projects are in favour because they are extremely capital intensive and suit most policymakers and contractors? Finally, even if all these explanations are partially valid, will urban metro rail projects help provide public transport at affordable prices, reduce congestion on roads, and reduce pollution and road traffic injuries and deaths?In this article we trace the history of urban transport systems, successes and failures around the world, and the lessons thereof. We also describe the attributes of surface public transport systems (bus, light rail, tram) and grade separated systems (metro rail, skybus, monorail, light rail) and their suitability for 21st century cities. Finally, we comment on urban forms and other issues that affect accessibility and mobility in modern urban areas.1 History, Technology and Urban TransportationUrban transport can be roughly divided into the following periods:Pre-1850: Pre-mechanisation era. Travel speeds (walking)tabout five km per hour. This limited city diameter to less than five km and influenced city form as the rich and the poor travelled at similar speeds.1850-1920: Steam engine/rail/tram/bicycle era. Rail-basedttransport is the first mechanised form to appear in the mid-19th century, the bicycle in its present form at the end of the 19th century and the commercial motor car soon after. Before the perfection of the pneumatic tyre and improve-ments in road building technology, large vehicles had to run on rails and so public transport was completely rail based. In the last two decades of this period the bicycle and the car appear as personal modes of transport for mainly the upper middle and rich classes. Average speeds go up to 10-15 km/h, city form decided by existence of rail tracks, and city size increases to about 10 km in diameter.1920-50: Metro-bus-car era. The diesel engine, goodtpneumatic tyres (making possible large buses) and mass produced cars become available in the first two decades of this period and change transport modes especially in the United States (US). Average urban speeds increase to about 30 km/h, buses start replacing rail-based trams and metro goes mainly underground. Bicycles as personal mode starts reducing.Post-1950: Car era. In this period cars become the dominanttmode of travel in all rich country cities and for the upper-mid-dle class everywhere. Underground metro systems also expand, but mainly in rich cities. Buses and para-transit remain the main modes for pubic transport in low income cities. Average car speeds on arterial roads go up to about 50 km/h, but door-to-door speed rarely exceeds 25 km/h. Cities can now be 20-30 km in diameter and form dictated by car travel.Most large cities in high income countries (HIC) grew to their present size between 1850 and 1950. Technological developments were critical in changing the shape and form of the city. Before the invention of the steam engine everyone had to walk, ride horses or horse carriages to get around. This meant speeds around five km/h for the rich and poor alike, and consequently city size was limited to about five km in diameter. Further, the city centre was very important as all important buildings and facilities had to be in the centre and that is where the rich preferred to live. City centres in Europe could be very grandiose as they represented the success of empires and the capital availa-ble from the colonies. These city centres developed as central business districts and remain so as the influential sections of society take pride in maintaining and improving these locations. Cities that have grown after 1950 do not have such characteristics and the elite of these cities do not have any particular emotional attachment to the old city centre.Rail-based transport starts in the second-half of the 19th century. The first rail-based metro line was completed in London in 1853, the first elevated railway in New York city in 1868, the first line of the Paris network in 1900, and in Mumbai the first suburban line, between Virar and a station in Bombay Backbay, in the year 1867. At the beginning of the 20th century road surfaces were not very smooth and roads were paved with the same materials and technology used by the Romans: bricks laid on a well-prepared foundation of sand, gravel, and stone [Whitten 1994]. The use of asphalt and bitumen only gets perfected between 1910 and 1930. The pneumatic tyre for large vehicles takes shape after 1930 and so does the heavy duty diesel engine. Therefore, mechanised transport could be comfortable only if vehicles moved on steel rails up to 1920 or so. This is why street trams became very popular as they were more flexible in opera-tion and cheaper to build than underground rail systems. The golden age of the tramway system was the first quarter of the 20th century but starts declining in the 1920s. This was because systems resulting from the initial capital investment were reaching the end of their useful lives and the income from the system was not adequate enough for large investments necessary for renewal [Chant 2002]. This decline was aggravated by the appearance of the mass produced motor car and more efficient buses. The onset of the financial depression in 1929 in the US and Europe followed by the war in 1939 put the trams at great risk and they got replaced by buses (cheaper to introduce and run) in many cities helped along by the lobbying power of vehicle manufacturers.But, underground or elevated rail and surface trains/trams remained the only mode of mechanised transport well into the 1920s for those who could afford it in all large cities of the world. This is when the largest cities expanded to around 10 km in diameter with populations around one to two million. If one lived away from the centre it was essential to live along the rail lines, and all factories and employment centres were built along these tracks. Rail technology decided the shape and form of these 19th century large cities in Europe and the US, also giving a great deal of importance to the central business district (CBD) as it could be fed by these trains.Car ownership started increasing in the 1920s, but most families did not own a car until the middle of the 20th century. By then the essential land use and transportation patterns of large cities in HICs except in the US were well set with large CBDs. This encour-aged building of high capacity grade separated metro systems, and in turn, the transport system encouraged densification of CBDs

SPECIAL ARTICLEEconomic Political Weekly january 26, 200843as large numbers of people could be transported to the centre of the city. The non-availability of the car to the middle class decided the widespread use of public transport and city form. However, this period when the bus and roads became comfortable (1920-50), was also the period of great economic and social upheaval and little thought could be given to improving urban transport with innovations in management systems and technology. Car owner-ship increased much faster in the US and so many cities did not have the political pressure to provide for public transport.Most Indian cities have expanded after 1960 and all have planned for multiple business districts. In addition, in the second half of the 20th century most families did not own a personal vehicle and so all leisure activity revolved within short distances around the home. In the past two decades motorcycle ownership has increased substantially in Indian cities, as a result about 50 per cent of Delhi’s families own a car or a motorcycle at a very low per capita income level of about $ 1,400 per year. Such high levels of private vehicle ownership did not happen until incomes were much higher inHIC cities. Therefore, the high ownership of motorcycles, non-availability of funds to build expensive grade separated metro systems and official plans encouraging multi-nodal business activity in a city have resulted in the absence of dense high populationCBDs and city forms which encourage “sprawl” in the form of relatively dense cities within cities.1.1 Declining Demand for Public Transportation Most middle class families did not own air-conditioned cars with stereo systems inHICs before 1970. The cars were noisy and occupants were exposed to traffic fumes as windows had to be kept open. Under such conditions, the train was much more comfortable. This created the condition in which there would be a political demand for metro systems that came from the middle class and could not be ignored. On the other hand, brand new, quiet, stereo equipped, air-conditioned cars are being sold in India now at prices as low as $ 5,000 to 6,000, and used ones for quarter the price. This has made it possible for the middle class first time car owner to travel in cars with comfort levels Europeans had not experienced till the late 20th century. Air-conditioned, comfortable, safe and quiet travel in cars with music in hot and tropical climates cannot be matched by public transport. Owners of such vehicles would brave congestion rather than brave the climate on access trips and the jostling in public transport. If public transport has to be made more appeal-ing, it has to come closer to home, reduce walking distances and be very predictable. These conditions would favour high density networks, lower capacity, surface transport systems (to reduce walking distances) with predictable arrival and departure times aided byITS information systems.Wide ownership of motorised two-wheelers (MTWs) has never been experienced byHIC cities. This is a new phenomenon, especially in Asia. The efficiency of MTWs – ease of parking, high manoeuvrability, ease of overtaking in congested traffic, same speeds as cars and low operating costs make them very popular in spite ofMTW travel being very hazardous. Avaialability of MTWs has further reduced the middle class demand for public transport. In addition it has pegged the fare levels that can be charged by public transport operators. It appears that public transport cannot attract these road users who can afford anMTW unless the fare is less than the marginal cost of using aMTW. At current prices this amounts to less than Rs 1 per km. The only option available is to design very cost efficient public transport systems that come close to matching this price.The above discussion shows that Indian cities in the 21st century are growing under very different conditions from those in HICs in the first half of the 20th century. The political and ideological forces combined with changes in technologies will make it difficult to provide efficient transport systems in the old manner. It will also be very difficult to move away from multi-nodal city structures with future job opportunities developing on the periphery.Higher education and trade obviously have a reasonable amount to do with size of cities and form of urbanisation. The more “educated” we are, the larger the pool of resources we need both for work and human contact. Therefore, a large city becomes essential for a reasonable section of the population for finding “optimal” employment and friends. Inverse of the same issue is that trade and industry needs a large pool to select employees. This forces Indian cities to become larger thanHIC cities. This is because for each rich person there is a larger number of poor people to serve her as compared to that inHICs. So, the same number of professionals in an Indian city will coexist with a much larger number of poorer residents than that in a HIC. For the foreseeable future, this will make Indian cities much larger than the “mature” cities of Europe. The existence of a large number of low income people pursuing informal trade and income generat-ing activities places different political pressures on the rulers, increases demand for low cost mobility and short distance access to jobs and trade.This is offset by the middle and upper classes wanting to live away from the poor and form gated communities at the periphery of the city. These developments set up a powerful political demand, aided and abetted by contractors and consultants to provide infrastructure. The upper middle class of the post-colonial nations mainly have theUS as a model for the good life. All Asian, African, South American cities are more influenced by the US than any other society. For example, American town planners were sitting in Delhi helping us plan our cities in the 1950s [Breese 1963]. So all these cities have tried hard zoning, broad avenues with highways running through them. If it has not happened it is due to inefficiency and shortage of finances! In the face of all these changes and constraints, the Indian upper class and policymakers still seem to think that just provision of rail-based metro systems in our cities will solve all our problems.2 The ProblemPressures of global warming, rising pollution and road traffic injury rates, and difficulties in moving around in the city are putting pressures on the government to find solutions for “sustainable urban mobility”. In response to such pressures the government of India announced a National Urban Transport Policy [Ministry of Urban Development 2005] and launched the Jawaharlal Nehru Urban Renewal Mission in 2006 for central

Education Education Education Education Shopping Shopping Shopping Social Social Personal Business Work Work Work Work (Trips/person/day)

1720412529

Central Paris C 75% PT 25% C 35% PT 65% C 42% PT 58% C 90% PT 10% C 78% PT 22% C 69% PT 31% Second Ring First Ring

SPECIAL ARTICLEEconomic Political Weekly january 26, 20084710 km if walk trips are taken into account. An analysis of trip times for Delhi shows that 26 per cent of the trips by bus take less than 30 minutes and another 23 per cent take 30 to 45 minutes [RITES 2005b]. The former trips are likely to be less than three km and the latter three to six km in length. The same study shows that 60 per cent of the car trips took less than 30 minutes.9 These data clearly show that even in very large cities in India a vast majority of trips remain less than 10 km in spite of large city size. This is likely to be true for many of the other cities mentioned in Table 3 also as all of them have significant proportion of non-motorised and bus trips. These data point to the fact that large cities in LMICs may have different travel patterns in terms of trip lengths than those in HICs as a majority of the people do not own cars. This is supported by studies form Europe which show that trip length increases significantly only when people are able to own cars and travel at high speeds or use long distance commuter trains as time budgets remain relatively constant [Knoflacher 2007a]. It seems that a majority of people in LMIC cities chose to live closer to work as they do not have the option to travel at high speeds. This would mean that these large cities function as cities within cities and the effect of sprawl is different in character from the sprawl of HIC cities and we should be careful when evaluating transportation policy literature form HIC cities. If large Indian cities have high density living in spite of the sprawl compared to western cities [Urban Age 2007] and trip lengths are relatively short, then it becomes an ideal case to plan for walking and bicycling facilities for the healthy and efficient bus transport systems.10The data presented above shows us that contrary to the widespread perception of most urban policymakers and planners, rail based metro transportation systems play a limited role in most large cities of the world except Tokyo. The cities where metro systems have a wide network and are successful in attract-ing a significant proportion of commuter trips are those where the system was initiated in the first half of the 20th century and have very dense and large central business districts [Cox 2004; TCRP 1996]. This is probably because the only form of mechanised transport the middle class commuters could afford had to be public transport. Technological choices were limited before the 1940s and the only system available had to be rail based. Therefore, workplaces and homes had to be along these lines which fed city centres and the large cities of the early 20th century had little choice but to take these forms.Another characteristic common to most large and mid-sized cities is that use of personal modes is low only in those cities where the combined share of walking, bicycling and public transport trips is high. Provision of public transport facilities alone is not sufficient to ensure low use of personal modes. Provision of efficient and widespread bus transport systems seems to be essential for significant transport use, even for the use of metro systems. For example, in Singapore 25 per cent of the commuters use the metro systems but 60 per cent of them combine it with a bus trip [Statistics Singa-pore 2006].Cities inLMIC that have grown after the 1950s seem to be different in character with multiple business districts, mixed land use (largely by default, illegally), relatively short trip distances and large share of walking and public transport, even if the latteris not provided by the city authorities.11 Car share remains below 20 per cent even at per capita incomes of $6,000-8,000 (Bogota, Mexico City). This is an important point to note as Indian incomes are not likely to reach these levels in the next 20 years. It is also clear that no city inLMIC has been able build a metro system that attracts a majority of public transport passengers. This is partly because no city that has grown after 1950 has large and dense central business districts. All large Indian cities are growing around the periphery and will not have dense centres in the future.In the next section we review the performance of rail based grade separated (elevated or under-ground) metro systems to under-stand why their presence in growing low and middle income cities has not been as successful as in the large cities of somelmIC.4 MetroRailPerformanceMetro systems in India have been justified on the basis of the following criteria [Sreedharan 2004]:Need to be in place when city size exceeds threemilliontpopulation.Very high carrying capacity of 40,000-60,000 persons perthour per direction (pphpd).Reduction in time of travel, per-tsonal transport use, traffic congestion, pollution, road traffic crashes, and buses on the road.Increase in social welfare.t4.1 City Size, Income and Metro PerformanceOne of the main justifications for intro-duction of metro systems is that all large cities need a grade separated metro system as there is enough demand for transporting more than 40,000 passen-gers in one hour per direction during peak time. The website of the Delhi Metro Rail Corporation states: “It has Table 1: Modal Share of Trips by Different Modes of Transport in Medium Sized Cities in Europe Modal Share (in %)City Car + MTW PT WC Bristol, UK 65 12 23 Leeds, UK 61 36 3 Nantes, France 58 14 28 Helsinki, Finland 54 20 26 Marseille, France 53 12 35 Edinburgh, UK 52 29 19 Newcastle, UK 48 19 33 Brussels, Belgium 44 18 38 Frankfurt, Germany 42 21 37 Stuttgart, Germany 36 25 39 Amsterdam, Neth's 32 16 52MTW – motorised two-wheeler, PT – Public transport.WC – Walking and cycling.Source: Adapted from Commission for Integrated Transport UK 2001.Table 2: Modal Share of Trips by Different Modes of Transport in Cities in Low and Middle Income Countries City Modal Share (in %) Walk + Car + Bus Paratransit Metro/ BicycleMTW + TaxiRailBangalore, India 18 35* 41 6 -Bogota, Colombia 13 11 73 3 -Delhi, India 37 18* 40 3 2Hanoi, Vietnam 32 60* 8 -Johannesburg, S Africa 33 33 5 24 5Lahore, Pakistan 37 35* 14 14 0Manila, Philippines@ - 21 24 53 2Mexico City, Mexico@ -20 61 5 14Mumbai, India 47 11 16 3 23Shanghai, China 54 14 28 -4MTW – motorised two-wheeler, * Majority MTW, @ Walk and bicycle not included.Source: Baker, Basu, Cropper, Lall and Takeuchi 2005; de la Torre 2003; de Villa and Westfall 2001; Hidalgo 2004; Roth 2000; Tiwari 2002; Urban Age 2007; World Business Council for Sustainable Development 2001.Table 3: Trip Length Percentage Distribution in Mumbai and Delhi Trip Length Mumbai Delhi (km)0-3 57 } 403-5 12 5-10 123510-15 8 } 25 15 11 Source: Adapted from Baker, Basu, Cropper, Lall and Takeuchi 2005; RITES 1998.

SPECIAL ARTICLEjanuary 26, 2008 Economic Political Weekly48been observed that in developed countries, planning for mass transit system starts when city population size exceeds one million; the system is in position by the time the city popula-tion is two to three million and once the population exceeds four million or so, planned extensions to the Mass Rapid Transit Systems is vigorously taken up… The city of Delhi with a popula-tion of round 16.2 million should have had anMRTS network of at least 300 km by this time”.12 The original feasibility study for developing a metro system for Delhi justified the economic feasibil-ity of the system projecting a daily ridership of 3.1 million passengers by 2005 [RITES 1995]. This was later reduced to a projected demand of 2.18 million passengers to be transported per day on the first three corridors (65.8 km) when completed in Decem-ber 2005 [Sreedharan 2003], and then in 2005 further reduced to 1.5 million a day.13 The system is actually operating at around 0.6 million passengers per day at the end of 2007, less than 20 per cent of projected capacity. Similarly, the Kolkata metro is operating at about 10 per cent projected capacity [Singh 2002].The first few metro corridors are always selected to be on the most heavily travelled stretches in a city, and so it should be surprising that the Indian metros are operating at such low levels. There are very few cities in LMICs that have metro systems. Table 4 summarises the experience of metro (and light rail) and bus rapid transit use inLMIC cities [Cox 2001; Bus Rapid Transit Policy Centre 2007; World Bank 2006; UrbanRail.Net 2007]. MostLMIC cities have small systems and the largest systems are much smaller than those in the cities included in Figure 4. Metro systems inLMIC cities are operating between 10,000 to 20,000 passengers per day per route km (tppdpk) with Mexico City and Beijing having the highest rates (around 20,000 tppdpk). Mexico City has the oldest (30 years), the most extensive (202 km) and the cheapest system. Table 4 also shows the experience of cities with light rail systems (monorail included) and all of them have very low rates.In Figure 6 (a-f, p 49) we examine the influence of city popula-tion, per capita income and total transit length on the systems’ productivity in terms of tppdpk. Figure 6 a shows that even in large cities ( 8 million) there is large scatter and no strong relationship with size of the population. The rail systems are operating with an average productivity of 11.5 tppdpk. This should be examined in light of the fact that the projection for phase I of the Delhi metro for 2005 was 33.5 tppdpk and the projection for 2011 at the end of phase II is 16.1 tppdpk. Large city size does not ensure high produc-tivity as Shanghai has half the productivity level as Mexico City at similar large populations (18m). Similarly, per capita income does not have any strong relationship with metro productivity levels (Figure 6c) and some low income cities have higher rates than high income cities and vice versa. It is often argued that as the system length increases ridership levels will improve. Figure 6e gives no evidence for this. Shanghai (228 km) has less than half the tppdpk than both Mexico City with a large system (202 km) and Beijing with a small system (96 km). It is also likely that tppdpk producti-vity decreases as the system expands because it starts covering less dense corridors. In Figure 6 (b,d,f) we also examine BRT tppdpk productivity levels. For the cities examined BRT systems seem to compete well with metro systems with levels of productivity similar to the best rail systems. Population size, per capita income, and length of the BRT system seem to have little effect on productivity. This means that BRT systems are more likely to have success than rail systems on any heavily travelled route in any city.Figure 7 (p 50) shows all the productivity (tppdpk) of metro systems in all cities with MRT in Table 4 and also New York and London. This again shows a wide scatter and just high density is not enough to project metro use with certainty. What is of note is that even in the very large HIC cities the metro share of commuters is less than 30per cent except in Tokyo. In any case Tokyo and Hong Kong are exceptions, where urban rail produc-tivity (passengers carried per km of rail) is around 27,000 passengers per day per km [Cox 2001]. In almost all other cities in LMIC this number is less than 40 per cent of that in Tokyo. For example, London, Paris and New York metros operate at less than half the productivity of Tokyo, and all other cities at even lower factors. Light rail productivity factors are gener-ally less than one-tenth that of Tokyo in almost all cities. Table 1 also shows that medium sized HIC cities have very low public transport use (which includes bus use). In the largest metro networks of around 200 km in Mexico City and Shanghai, the metro system accounts for less than 15 per cent of motorised trips, and if walk and bicycle trips are included it could be less than 10 per cent. It appears that metro systems in LMIC mega cities are not likely to be very successful in attracting even a significant proportion of the total number of commuter trips and the possibility of the same is even lower.Therefore, it is not surprising that the projected ridership of 34,000 passengers per day per km of the metro in Delhi did not come true as there is no empirical evidence for such high ridership values anywhere in the world. The Delhi metro system is expected to complete 186 km of rail lines in 2011 Table 4: Performance of Metro, Light Rail and Bus Rapid Transit Systems in Low and Middle Income Cities City PopulationCountryTypePassen-LengthPassen- (million) Per Capita gers/Day (km) gers Per Income (million) km Per Day ($ Per Year) ThousandMexico City, Mexico 18 6,760 MRT 3.90 202 19.31Shanghai, China 18 1,500 MRT 1.80 228 7.89Kolkata 16620MRT0.16179.41Delhi 15620 MRT 0.6065 9.23Istanbul, Turkey 12 3,750 MRT 0.42 28 15.00Beijing, China 12 1,500 MRT 2.10 96 21.88Rio de Janeiro, Brazil 11 3,000 MRT 0.50 45 11.11Manila, Philippines 10 1,170 MRT 0.40 29 13.79Bangkok 9 2,490MRT 0.584413.18Santiago, Chile 6 5,220 MRT 0.57 60 9.50Kuala Lumpur, Malaysia 5 4,520 LRT 0.17 55 3.09Medellin, Colombia 4 2,020 LRT 0.29 29 10.00Guadalajara 46,760LRT0.13245.42Tunis, Tunisia 2 2,650 LRT 0.40 32 12.50Mexico City, Mexico 18 6,760 BRT 0.28 20 14.21Sao Paulo, Brazil 16 3,000 BRT 0.80 78 10.26Bogota, Colombia 8 2,020 BRT 0.79 42 18.81Curitiba, Brazil 3 3,000 BRT 2.43 122 19.92Quito, Equador 2 2,210 BRT 0.38 36 10.50MRT: Metro, LMRT: Light Rail, BRT: Bus Rapid Transit.Source: Data from Cox 2001; Bus Rapid Transit Policy Centre 2007; World Bank 2006; UrbanRail.Net 2007; official web sites of metro systems; official web sites of BRT systems.Table 5: CO2 Emissions by Mode of Travel(in gram/passenger km)Mode EmissionsCar Petrol 186 Diesel 141 Hybrid 125Bus 56Metro 107Source: Bannister 2005.

5 10 15 20

2 000 4 000 6 000 8 000

Elevated Expressway Metro BRT

SPECIAL ARTICLEEconomic Political Weekly january 26, 200851The Delhi MetroThe experience of the Delhi metro substantiates all of the above observations in terms of low ridership rates, and no positive welfare in terms of actual reductions in pollution, road traffic crashes or congestion. However, studies claiming welfare benefits are based on notional time and energy savings [Murty, Dhavala, Ghosh, and Singh 2006; RITES 2005a]. These studies calculate environmental benefits based on the theoretical reduc-tion of motor vehicles and not the actual experience. They also do not take into account any increase in motor vehicles or trips due to induced and latent demand.14 Time savings are calculated on the basis of only the train trip and not door-to-door time. There are generally no door-to-door time savings by metro unless the trips are greater than 12 km as shown in Figure 4. This is mainly because access and egress times for underground or elevated systems can average more than 20 minutes [Krygsman, Dijst, and Arentze 2004]. Actually, pollution in Delhi has increased over the past three years [CSE 2007] especially NOx and respiratory particle matter. The number of vehicles and road traffic crashes also increased by about 20 per cent [Delhi Traffic Police 2007]. Traffic monitoring on the metro corridors shows that the number of motor vehicles increased and the number of buses remained constant over the past three years [Tiwari 2007]. Metro authorities projected a decrease in buses but there has been none, and now extra feeder buses are being added to the fleet by them to bring passengers to the metro stations.Exaggerating benefits and underestimating costs are not confined to the Indian experience. A study of more than 210 transportation infrastructure projects worldwide demonstrates that cost underestimation and exaggeration of benefits (both by an average factor of two) are common, especially for rail projects [Flyvbjerg, Holm and Buhl 2002; Flyvbjerg, Holm, and Buhl 2005]. The authors conclude that “Underestimation cannot be explained by error and is best explained by strategic misrepre-sentation, that is, lying”. They also show that forecasts have not become more accurate over the 30-year period studied, despite claims to the contrary by forecasters. Decision-makers should not trust cost estimates and cost-benefit analyses produced by metro project promoters and their analysts. In the case of the Delhi metro the cost of capital alone accounts for subsidy of Rs 35,000 per passenger per year.15 This is more than the per capita income (Rs 28,000 per year) of India and more than 60 per cent of the estimated per capita income of Delhi (Rs 56,000 per year). This is obviously not sustainable. At this cost we could send over a million children to a good private school every year. The policy implications are clear, if the risks generated from misleading forecasts are ignored or downplayed it will be to the detriment of social and economic welfare.5 ConclusionsInternational empirical evidence and the Delhi experience indicate that metro rail systems (elevated or underground) have not delivered the goods in terms of passengers carried or social welfare and are unlikely to in cities that do not have a very dense and large central business district.All Indian cities are developing on the periphery, have multi-ple business districts, and are not suited to fixed line very high capacity rail systems. The demand will never come up to the theoretical capacity of these systems partly because metro rail is not time saving as a vast majority of trips in Indian cities are lessthan 10 km in large cities. The presence of motorised two-wheelers makes it even more difficult as the marginal cost of travel amounts to less than Rs 1 per km. Public transport cannot charge more than this amount without losing ridership. Therefore, we have to promote an efficient and economical public transport system that has a dense network, is flexible, on the surface and of medium capacity (15,000-30,000 passengers per hour per direction). TheBRT with dedicated bus lanes seems to be the only option left as it can be built at 5 per cent of the cost of metro systems [Allport and Thomson 1990; Ben-Akiva and Morikawa 2002; Fouracre, Dunkerley and Gardner 2003; Fulton, Hardy, Schipper, and Golub 2007; GAO 2003; Halcrow Fox 2000; Levinson, Zimmerman, Clinger, Rutherford, Smith, Cracknell and Soberman 2003; Penalosa 2004; Ridley 1995; The Bus Rapid Transit Policy Centre 2007; Tiwari 2002a, 2002c; Winston and Maheshri 2007; World Business Council for Sustainable Development 2001].16The cost of building elevated rail systems (including monorailand light rail) is around Rs 1,500 million per km and for underground systems Rs 2,000-2,500 million per km. On the other hand, BRT systems cost about Rs 50-100 million per km. About 20-30 km of BRT can be built for each km of the metro. This cost of theBRT system includes shifting of water and electric services, providing better footpaths and bicycle lanes and install-ing modern lighting and other road furniture. This happens because when the road layout is altered you get an opportunity to redevelop the corridor. So a BRT project ends upbeing a urban rejuvenation project. Since the metros are underground or elevated they do not have this effect on the ground.17There are strong reasons why surface BRT is preferable in modernLMIC large cities [Knoflacher 2007b]:Economic: Public transport on the ground in form of buses and street cars is cheaper to build, maintain and to operate.Efficiency: Public transport is one of the most efficient modes with respect to energy consumption, use of space and safety. Therefore, there is no reason to remove it from the road surface.Accessibility: Elevated or underground public transport loses half or even two-thirds of potential customers compared to street level public transport modes. Further, if public transport is separated from the street level, it becomes necessary to build and operate escalators, lifts, etc. This enhances the costs for construc-tion, maintenance and operation.Security: The entire transport system on the street level is under public social control and is, therefore, much safer. Urban Economy: Street level public transport is good for the urban economy. The experience of European cities shows that

SPECIAL ARTICLEjanuary 26, 2008 Economic Political Weekly52replacing street level public transport by underground systems has a negative effect on local shops. Underground or grade separated public transport systems increase both disparities and the need for longer travel.Structural: Public transport on street levels keep people moving without fundamental changes of urban structures and the system provides flexibility as land use changes. Urban Vision: It is crucial to integrate public transport also in the mental map of people and visitors. Public transport on the streets tells the people that it is a socially balanced city.Environmental: Public transport on the street level serves as an indicator for an environment-friendly transport policy of the city. To integrate public transport in the human society it is necessary to keep it on the road surface instead of the sky or underground. It is clear that urban metro systems appeared at the end of the 19th and early 20th century when there were no choices except rail available for the middle class commuters for mechanised travel. They got a further boost during the cold war when deep tunnels were justified as air raid and nuclear shelters and cost was no criterion. The evidence seems to be clear that elevated and underground rails systems are far too expensive and not very successful in 21st centuryLMIC large cities. Surface light rail systems may have a role in connecting megacities with surround-ing towns on existing rail alignments. With the challenges of global warming staring us in the face, the only choice we have is to plan our cities for safe walking and bicycling facilities on all roads, bus rapid transit systems on all major roads, supplemented by low energy consuming economical taxi systems.Notes 1 Jamie Lerner was the mayor and chief architect of the Curitiba (Brazil) master plan. During his 12 years in office, Lerner devised many of Curitiba’s innovative and inexpensive solutions to city prob-lems and is credited for introducing the first ver-sion of the bus rapid transit system in his city. 2 Enrique Penalosa was mayor of Bogota, Colom-bia, and chose to implement a bus rapid transit system instead of a metro system. He is in demand worldwide to share his views on city planning and development.3 ‘Ludhiana to Have Its Metro Rail Service’, The Hindu, Wednesday, October 10, 2007, online edition, http://www.thehindu.com/2007/10/10/stories/2007101053990500.htm. 4 Bangalore Mass Rapid Transit (BMRTL), Benefits of Namma Metro, http://www.bmrc.co.in/benefits.html, accessed October 15, 2007.5 ‘Hyderabad to Have Metro Rail’,The Tribune, online edition, Sunday, January 21, 2007, Chandigarh, http://www.tribuneindia.com/ 2007/20070121/nation.htm#4, accessed October 16, 2007. 6 See the web site of the ministry of urban develop-ment to get an insight into this confusion (http://urbanindia.nic.in/moud/programme/ud/main.htm). The ministry has to repeatedly write to the states and cities to clarify matters, eg, “…This ministry is receiving proposals for flyovers road widening, metros etc, under JNNURM Viability Gap Funding budgetary support from government of India… while proposing any option, all other options should also be evaluated specially when high cost options are being proposed” (D O No K-14011/07/2007-UT, May 22, 2007, from secre-tary urban development to chief secretaries of all states). 7 The managing director of DMRC, E Sreedharan, is today convinced that all cities above three million population should have a metro system and is in the process of preparing detailed project reports for implanting such systems (seeProject Monitor, October 20, 2007: http://www.projects-monitor.com/detailnews.asp?newsid=7734).8 The ministry of urban development web site http://jnnurm.nic.in/missioncities.htm includes all the city development plans submitted under the JNNURM. Almost all of them use this argument of increase in demand for widening roads. 9 At average urban speeds of 20 km/h in Delhi, this would mean that even 60 per cent of the car trips are less than 10 km.10 The average density in persons per square km for metropolitan regions of low and middle income cities are: Mumbai – 4,080, Delhi – 1,227, Kolkata – 7,978, Bangalore – 1,050, Shanghai – 2,619, Mexico City – 3,796. Similar statistics for some HIC cities: London – 679, Berlin – 818, New York – 783.11 When public transport is not provided officially, informal systems using mini-buses, three-wheel-ers and vans operate semi-legally or illegally and provide a majority of the motorised trips. No low or middle income city is without such systems. 12 Delhi Metro Rail Corporation web site http://www.delhimetrorail.com/corporates/needfor_mrts.html accessed on 2007-11-03.13 Delhi Metro Rail Corporation web site http://www.delhimetrorail.com/corporates/projectupdate/ expected_ridership.html, accessed on 2005-12-02.14 Newspapers reported the results of study done by CRRI (‘Metro Brings Down Pollution Levels’, Times of India, June 5, 2007) claiming a benefit of Rs 20,725 million by the end of 2007 based on savings in petrol, diesel and CNG, passenger time due to faster mode, reduction in environmental pollutants and the like.15 Based on a capital expenditure of Rs 105,710 million for phase I, 600,000 passengers day (300,000 roundtrips), depreciation and interest rate at 5 per cent each.16 Details of and studies regarding bus rapid transit system are available at www.go.brt.org. In India, the cost of a BRT system is Rs 50-100 million and metro system 1,500-2,500 crore per km.17 For BRT the costs are based on the project being implemented in Delhi and others being sanctioned by the ministry of urban development. For metro systems, the costs are based on those quoted by the Delhi Metro Rail Corporation from time to time. The average cost of Metro Phase I in Delhi is estimated at Rs 1,626 million per km (http//www.delhimetrorail.com/corporates/projectupdate/project_cost.html). Only 20 per cent of this is underground.ReferencesAllport, R J and J M Thomson (1990): Study of Mass Rapid Transit in Developing Countries, Report 188, Transport Research Laboratory, Crowthorne, UK.Baker, J, R Basu, M Cropper, S Lall and A Takeuchi (2005): ‘Urban Poverty and Transport: The Case of Mumbai’, World Bank Policy Research Working Paper 3693, World Bank, Washington DC, pp 1-81.Bannister, D (2005): Unsustainable Transport: City Transport in the New Century, Routledge, New York.Ben-Akiva, M and T Morikawa (2002): ‘Comparing Ridership Attraction of Rail and Bus’, Transport Policy, 9(2), pp 107-16.Breese, G (1963): ‘Urban Development Problems in India’, Annals of the Association of American Geographers, 53(3), pp 253-65.Brons, M, P Nijkamp, E Pels and P Rietveld (2003): ‘Railroad Noise: Economic Valuation and Policy’, Transportation Research Part D: Transport and Environment, 8(3), pp 169-84.Bus Rapid Transit Policy Centre (2007): ‘BRT Database’, Breakthrough Technologies Institute, Washington DC, http://www.gobrt.org. Cerin, E, D J Macfarlane, H H Ko and K C Chan (2007): ‘Measuring Perceived Neighbourhood Walkability in Hong Kong’,Cities, 24(3), pp 209-17.Cervero, R (2002): ‘Induced Travel Demand: Research Design, Empirical Evidence, and Normative Policies’, Journal of Planning Literature, 17(1), pp 3-20.Chang, J S K (2007): ‘Integration of Ring Road and BRT System Urban Transport’, International Workshop Urban Transport: Today and Tomorrow, TRIPP, Indian Institute of Technology Delhi, New Delhi.Chant, C (2002): Rapid Transit Systems and the Decline of Steam, Grange Books, Rochester, Kent.Christopher Zegras, P (2007): ‘As If Kyoto Mattered: The Clean Development Mechanism and Trans-portation’,Energy Policy, 35(10), pp 5136-50.Commission for Integrated Transport UK (2001): European Best Practice in the Delivery of Integrated Transport, WS Atkins Transport Planning, Epsom, Surrey.Cox, W (2001): ‘Urban Transport Fact Book – Urban Rail Productivity’, http://www.publicpurpose.com/ut-intlrrprodtok.htm, Wendell Cox Consul-tancy, accessed 2007-07-11. – (2004): ‘The Public Purpose, a National Journal’, http://www.publicpurpose.com/, Wendell Cox Consultancy, accessed 2007-07-11. CSE (2007): Is Delhi Losing the Pollution Battle?, Centre for Science and Environment, New Delhi.Daniels, P W (1972): ‘Transport Changes Generated by Decentralised Offices’,Regional Studies: The Journal of the Regional Studies Association, 6(3), pp 273-89. – (1981): ‘Transport Changes Generated by Decen-tralised Offices: A Second Survey’,Regional Studies: The Journal of the Regional Studies Associ-ation, 15(6), pp 507-20.de la Torre, G Q (2003): Mexico City: Energy, Mobility and Sustainable Urban Development, Business Council for Sustainable Development – Mexico, Mexico City.de Villa, V and M S Westfall (2001): Urban Indicators for Managing Cities: Cities Data Book, Asian Development Bank, Manilla.Debrezion, G, E Pels and P Rietveld (2004): ‘The Impact of Railway Stations on Residential and

SPECIAL ARTICLEEconomic Political Weekly january 26, 200853Commercial Property Value: A Meta Analysis’, Tinbergen Institute Discussion Paper, TI 2004-023/3, Tinbergen Institute, Amsterdam.Delhi Traffic Police (2007): ‘Road Accidents in Delhi – 2006’, Accident Research Cell, Delhi Police, New Delhi.Dora, C (1999): ‘A Different Route to Health: Implica-tions of Transport Policies’,British Medical Journal, 318(7199), 1686-89. – (2007): ‘Health Effects’, Seminar, 579, pp 26-30.Edwards, M and R l Mackett (1996): ‘Developing New Urban Public Transport Systems: An Irrational Decision-making Process’,Transport Policy, 3(4), pp 225-39.Flyvbjerg, B, M K S Holm and S L Buhl (2002): ‘Under-estimating Costs in Public Works Projects – Error or Lie?’,Journal of the American Planning Associa-tion, 68(3), pp 279-95. – (2005): ‘How (In) Accurate Are Demand Forecasts in Public Works Projects? The Case of Transporta-tion’, Journal of the American Planning Associa-tion, 71(2), pp 131-46.Fouracre, P, C Dunkerley and G Gardner (2003): ‘Mass Rapid Transit Systems for Cities in the Developing World’,Transport Reviews, 23(3), pp 299-310.Fulton, L, et al (2007): Bus Systems for the Future: Attaining Sustainable Transport Worldwide, Inter-national Energy Agency, Paris.GAO (2003): Bus Rapid Transit Offers Communities a Flexible Mass Transit Option, GAO-03-729T, United States General Accounting Office, Washington DC.Halcrow, Fox (2000):World Bank Urban Transport Strategy Review – Mass Rapid Transit in Developing Countries, Final Report, Department for Inter-national Development, World Bank, Washington DC.Hidalgo, D (2004): ‘Bogota Experience in Sustainable Transportation’, Proceedings of International Conference on Reducing the Impact of Vehicles on Air and Environment Quality in Cities, http://iuappa.com/conference/seminar-mexico. htm#Summary, IUAPPA, Brighton, UK, accessed 2007-10-11.IPCC (2007):Climate Change 2007: Synthesis Report, Intergovernmental Panel on Climate Change, Geneva.JBIC (1999):Urban Public Transportation in Vietnam: Improving Regulatory Framework, JBIC Research Paper No 4, Research Institute for Development and Finance, Japan Bank for International Cooperation, Tokyo.Kågeson, P (2001): The Impact of CO2 Emissions Trading on the European Transport Sector, VINNO-VA Report VR 2001:17, Nature Associates, Stock-holm.Kawabata, M and Q Shen (2006): ‘Job Accessibility as an Indicator of Auto-Oriented Urban Structure: A Comparison of Boston and Los Angeles With Tokyo’, Environment and Planning B: Planning and Design, 33(1), pp 115-30.Knoflacher, H (2007a): ‘Success and Failures in Urban Transport Planning in Europe – Understanding the Transport System’,Sadhna – Academy Proceed-ings in Engineering Sciences, 32(4), pp 293-308. – (2007b): ‘From Myth to Science’,Seminar, 579, pp 40-44.Krygsman, S, M Dijst and T Arentze (2004): ‘Multi-modal Public Transport: An Analysis of Travel Time Elements and the Interconnectivity Ratio’, Transport Policy, 11(3), pp 265-75.Lerner, J (1996): ‘Change Comes from the Cities’,Our Planet: Human Settlements, 8(1).Levinson, H, S Zimmerman, J Clinger, S Rutherford, RL Smith, J Cracknell and R Soberman (2003): Bus Rapid Transit – Volume 1: Case Studies In Bus Rapid, TCRP Report, 90, Transportation Research Board, Washington DC, pp 1-62.Mackett, R and E Babalik Sutcliffe (2003): ‘New Urban Rail Systems: A Policy-Based Technique to Make Them More Successful’,Journal of Transport Geography, 11(2), pp 151-64.Mackett, R I and M Edwards (1998): ‘The Impact of New Urban Public Transport Systems: Will the Expectations Be Met?’,Transportation Research - A, 32(4), pp 231-45.McGuckin, N and N Srinivasan (2003): Journey to Work Trends in the United States and Its Major Metropolitan Areas 1960 – 2000, FHWA – EP-03-058, Office of Planning, Federal Highway Admin-istration, US Department of Transportation, Washington DC.Ministry of Urban Development (2005): National Urban Transport Policy, Ministry of Urban Devel-opment, Government India, New Delhi, pp 1-22.– (2006): ‘Jawaharlal Nehru Urban Renewal Mission’http://jnnurm.nic.in/defaultud.aspx#, Ministry of Urban Development, Government of India, New Delhi, accessed 2007-11-11.Mogridge, M J H (1997): ‘The Self-Defeating Nature of Urban Road Capacity Policy: A Review of Theories, Disputes and Available Evidence’,Trans-port Policy, 4(1), pp 5-23.Morichi, S (2005): ‘Enhancement of Urban Transport Sustainability in Japan’, Ministry of Land, Infra-structure and Transport, Japanese government, Tokyo, Murty, M N, K K Dhavala, M Ghosh and R Singh (2006):Social Cost-Benefit Analysis of Delhi Metro, Report for Planning Commission, Government of India, Institute of Economic Growth, Delhi, pp 1-30.Nelson, A C (1992): ‘Effects of Elevated Heavy-Rail Transit Stations on House Prices with Respect to Neighborhood Income’,Transportation Research Record, (1359), pp 127-32.Newman, P and J Kenworthy (1999):Sustainability and Cities: Overcoming Automobile Dependence, Island Press, Washington DC.Noland, R B (2001): ‘Relationships between Highway Capacity and Induced Vehicle Travel’,Transporta-tion Research Part A: Policy and Practice, 35(1), pp 47-72.Penalosa, E (2004): ‘Social and Environmental Sustainability in Cities’, Proceedings of Interna-tional Mayors Forum on Sustainable Urban Energy Development, Kunming, P R China.Rabinovitch, J (1995): ‘A Sustainable Urban Transpor-tation System: The “Surface Metro” in Curitiba, Brazil’, Working Paper No 19, United Nations Development Programme, New York.Ridley, M A (1995): World Bank Experiences with Mass Rapid Transit Projects, World Bank, Washington DC.RITES (1995)Integrated Multi-Modal Mass Rapid Transport System for Delhi, Economic Analysis for Modified First Phase, RITES, Gurgaon, Haryana. – (1998):Traffic and Transportation Policies and Strategies in Urban Areas in India, Final Report, Ministry of Urban Affairs and Employment, Government of India, New Delhi. – (2005a):Environmental Impact Assessment for Phase II Corridors of Delhi Metro, Delhi Metro Rail Corporation, New Delhi. – (2005b):Integrated Multi-Modal Public Transport Network for the National Capital Territory of Delhi, Final Report, Transport Department, Government of NCT of Delhi, New Delhi.Rodrigue, J P (2005): ‘Urban Mobility’ in J P Rodrigue, C Comtois and I Slack (eds),The Geography of Transport Systems. Ross, H, A Poungsomlee, S Punpuing and K Archavan-itkul (2000): ‘Integrative Analysis of City Systems: Bangkok “Man and the Biosphere” Programme Study’, Environment and Urbanisation, 12(2), pp 151-61.Roth, L C (2000): ‘An Economic Approach to Urban Development and Transportation in Metro Manila’, Master Thesis dissertation, Programme for Business Administration and Economics, Linkoping University.Santos, G and B Shaffer (2004): ‘Preliminary Results of the London Congestion Charging Scheme’, Public Works Management and Policy, 9(2), pp 164-81.Schrank, D and T Lomax (2007): The 2007 Urban Mobility Report, Texas Transportation Institute, The Texas AM University System, College Station, Texas, pp 1-138.Singh, Y P (2002): ‘Performance of the Kolkata (Calcutta) Metro Railway’ in X Godard and Fanto-zoun I (eds),Proceedings Conference CODATU X: Urban Mobility for All, A A Balkema Publishers, Rotterdam.Sreedharan, E (2003): ‘Delhi Metro Rail...2,500 Less Buses on Roads’, http://www.indiainfoline.com/infr/spfe/delh.html, India Infoline, accessed 2005, pp 12-02. – (2004): ‘Need for a National Policy for Metro Rail Projects’, mimeo, Delhi Metro Rail Corporation, New Delhi.Statistics Singapore (2006): ‘General Household Survey 2005, Statistical Release 2: Transport, Overseas Travel, Households and Housing Characteristics’, Department of Statistics, Minis-try of Trade and Industry, Singapore.Stopher, P R (2004): ‘Reducing Road Congestion: A Reality Check’, Transport Policy, 11(2), pp 117-31.Taylor, B D (2004): ‘The Politics of Congestion Mitiga-tion’,Transport Policy, 11(3), pp 299-302.TCRP (1996):Transit, Urban Form, and the Built Environment: A Summary of Knowledge (Volume 1, Part I), Washington DC, TCRP Report 16, Trans-portation Research Board, National Research Council, National Academy Press.Tiwari, G (2002a): ‘Bus Priority Lanes for Delhi’ in G Tiwari(ed), Urban Transport for Growing Cities: High Capacity Bus Systems, pp 200-22, Macmillan India, New Delhi.– (2002b): ‘Urban Transport Priorities – Meeting the Challenge of Socio-Economic Diversity in Cities, A Case Study of Delhi, India’, Cities, 19(2), pp 95-103. – (2002c):Urban Transport for Growing Cities: High Capacity Bus Systems (ed), Macmillan Indian, New Delhi. – (2007): Traffic Monitoring at Selected Locations in Delhi, Transportation Research and Injury Prevention Programme, Indian Institute of Technology Delhi, New Delhi.Transport Department (2007): ‘Transport in Hong Kong’, The Government of the Hong Kong Special Administrative Region of the People’s Republic of China, http://www.td.gov.hk/transport_in_hong_kong/transport_figures/monthly_traffic_and_transport_digest/index_t.htm, accessed 2007-11-19.Transport for London (2006): London Travel Report 2006, Mayor of London, London, pp 1-74.Urban Age (2007): ‘Urban India: Understanding the Maximum City’, The London School of Economics and Political Science, London. UrbanRail.Net (2007): ‘A Complete Guide to Rail Metro, Subway, or Underground Networks’, Robert Schwandl, http://www.urbanrail.net, accessed2007, pp 11-19.Vincent, W and L C Jerram (2006): ‘The Potential for BRT to Reduce Transportation-Related CO2 Emissions’,Journal of Public Transportation, 9(3), pp 219-37.WBCSD (2001): Mobility 2030: Meeting the Challenges to Sustainability, World Business Council for Sustainable Development, Geneva.Whitten, D O (1994): ‘Rollin on Rubber: 120 Years of Rubber Road Surfacing’ in E J Perkins,Essays in Economic and Business History(12): pp 414-27. http: //www.business. auburn.edu/~whittdo/Rollin%2 0on% 20Rubber.pdf, accessed 2007, pp 11-19.Winston, C and V Maheshri (2007): ‘On the Social Desirability of Urban Rail Transit Systems’, Journal of Urban Economics, 62(2), pp 362-82.World Bank (2006): World Development Indicators 06, World Bank, Washington DC.World Business Council for Sustainable Development (2001):Mobility 2001: World Mobility at the End of the Twentieth Century and Its Sustainability, WBCSD, EY Direct, Geneva.