1.1. The cities of Chennai (Tamil Nadu) and Bangalore (Karnataka) are mounting major efforts to deal with urban transport problems generated by exceptional rates of demographic, spatial and economic growth experienced therein over the last decade. The World Bank has a long history and a current presence as a partner in development endeavors of these cities and states. The areas of the current Bank activity include both urban and transport projects, but not specifically focused on urban transport.7 Given the perception of a growing importance of urban transport activities in both growth and poverty dimensions, an expansion of the Bank’s assistance into this field, in the form of advisory and lending activities, is now being considered by all concerned parties. The report in hand is intended to facilitate the discussions in this context, by providing an external angle of the urban transport problems, prospects and possible ways forward.
1.2. The main body of this report: (i) provides a brief diagnostic of the urban transport infrastructure and services in Chennai and Bangalore; (ii) identifies the underlying strategic issues; (iii) proposes an amended strategy, and (iv) outlines an agenda for the involvement of the World Bank in the short-to-medium term. The case studies of urban transport in Chennai and Bangalore, on which the main report is based, as well as a bibliography, are provided as attachments.
1.3. The report is a first-cut attempt to understand and address a complex subject. It is based on a brief field visit and desk research, both of which have disclosed serious lacunae in data. Other limitations have to do with a narrow focus on urban transport, adopted to make this initial attempt doable. For example, the report does not touch on the environmental aspects of urban transport, even though vehicle-produced air pollution is a major and increasing problem in both Chennai and Bangalore, indeed in all urban India. This omission is not likely to invalidate the proposals made herein, since they focus on potential increases in public transport patronage and on traffic restraint, both of which are unequivocally beneficial with regard to emissions. Conversely, the most important decision variables from environmental point of view (re vehicle emissions and fuel prices) apply at any level of modal split.8 A more serious limitation is that the report stays away from urban planning, land markets and municipal funding issues. Major analytical work is being done by the Bank in these areas, and its results are being incorporated into the design of lending operations. In the next stage of the work on urban transport, stronger links will need to be established between this subject and that of local government organization, funding and planning processes.
2. The Background
A. Transport Demand Characteristics
2.1. The main features of Bangalore and Chennai are shown in the following boxes. The two cities have similar population “masses,” just above 4 million within city boundaries and about 7.-7.5 million in the urbanized area. Chennai is much more dense, but Bangalore is growing at a much greater rate (4.9% per annum in the 1990s). Chennai is a long-established port city, with two adjacent centers also of older vintage - the traditional commercial hub next to a pre-independence administrative and military complex. Development spread from these centers and the port along a few major road and rail radials. Its industries include petrochemicals, machine manufacture, and automotive equipment (both cars and rail rolling stock). Bangalore is land-locked, but at an important cross-roads of state/national roads and rail lines. Better known in the past as a city of gardens and lakes, whose moderate climate attracted pensioners and vacationers in large numbers, it has become a world-known center of information (software) technology, a synonym for outsourcing services for the U.S. and Western European countries. Bangalore’s economy is much broader than its international image: most employment is in fact provided by trade and commerce (60% in 1995), and manufacturing (37%). Traditional activities like silk weaving and garments are also vibrant. Though Bangalore also has inherited two strong centers, it is much more poly-nuclear than Chennai and its road system is more diffuse and complicated. This is in tandem with the fact that rail lines entering Bangalore were neither designed nor operated to cater for urban and regional traffic, so the city’s growth and mobility patterns have been very much road-dependent. Chennai’s transport system, though greatly road dependent, also leans heavily on its commuter rail services and (soon) on its first urban rail line, now only open on a short link.
Chennai at a glance
2003 population 4.2 million (city), 7.5 million (metropolitan area)
port city, major industrial and commercial center
population growth in the 1990s: 0.9% per annum;
density in Chennai City: 250 people/hectare, double in sub-areas
urban pattern: higher-density historical center with developments along major radials
economic growth (Tamil Nadu state) 6.1% per annum (1997-01);
60% households have incomes under Rs.5,200/month, 37% under Rs.3,100/month (1998);
one million people live in slums (city only)
informal employment dominant;
transport system: road-based but with strong commuter rail network
travel by mode (adjusted data from early 1990s): walking (30%), bikes (14%), MTC buses (38%), urban & suburban rail (4%), motorized 2-wheelers (7%), cars (2.5%);
motorization: 1.5 million vehicles of which 1.1 million 2-wheelers, 250,000 cars;
main public transport providers: CMTC (2,400 buses in peak service at 16 km/h, 3.5 million daily passengers); Southern Railway (3 commuter rail lines carrying 643,000 psgrs/day and 1 short urban metro rail line, 9,000 psgrs/day ).
2.2. Both cities have in the last 20 years experienced a combination of demographic, spatial and economic growth that has catapulted them into the forefront of India’s great jump forward. These same processes have placed a tremendous strain on their public infrastructure and services. For transport management and planning purposes, it is of essence to understand the divergent patterns in population, location and income changes.
Bangalore at a glance
2001 population: 4.1 million (city), 5.7 million (metropolitan area) + floating population of about 1 million;
population growth in 1990s: 4.9% per annum, expected to reach 10 million by 2011;
poly-centric, land-locked city, major cross-roads in Southern India,
gross residential density in the city of Bangalore: 113 people/hectare;
economic growth (Karnataka state): 7.5% per annum (1990s);
leader in India’s information technology, electronics, consumer goods;
multi-ethnic, multi-layered urban society
median monthly income (1998): Rs 5,200 per household; 28% have income less than Rs 3,100/month;
2.2 million people live in about 750 slums (1998-99 data), sharply up from 1991 (estimates vary in scale);
motorization: city 1.6 million of which 1.2 million 2-wheelers and 279,000 cars; agglomeration 2 million vehicles, of which 1.6 million 2-wheelers;
transport system: road based; major railway network is in place but not significant for urban/regional travel;
main public transport providers: BMTC (2,200 buses in peak service, 675 buses sub-contracted to BMTC), carry 2.6 million trips per day; plus company buses;
modal split: walk and bike 17%; BMTC buses 41%; other buses 3%; auto-rickshaws 4%; cars and 2-wheelers 38%.
2.3. Economic growth has raised incomes of a large number of people and their expectations as to the services they deem essential. In the transport dimension, the most visible impact of rising incomes is accelerated motorization (vehicle ownership and use), accompanied by a shift from public transport services to individually or company owned vehicles. In the spatial dimension, this means an increase in the degrees of freedom to locate residences. At higher income brackets, this typically means a choice of more distant spots of greater environmental and other types of amenity.
2.4. Motor vehicle ownership in Bangalore and Chennai has been increasing at unprecedented rates, between 10 and 20% per annum. The current ownership level is about 324 individual passenger vehicles per 1,000 population in Chennai, and 298 in Bangalore. These are high rates, similar to those in the wealthiest cities of Eastern Europe and common in Western Europe, but at vastly lower level of incomes than in Europe. The explanation for this seeming anomaly lies in the structure of the passenger vehicle fleet. Motorized 2-wheelers are the main growth category, with about 1.1 million registered in Chennai and 1.2 million in Bangalore.9 Cars are a distant second: about 250,000 are registered in Chennai and 267,000 in Bangalore. This motorization pattern is similar to that experienced elsewhere in South and East Asia, e.g. Hanoi and Ho Chi Minh City in Vietnam; Kuala Lumpur in Malaysia. The consequences of 2-wheeler primacy, while a boon for the mobility of many people, are unfortunately quite negative for traffic flow, safety and air pollution. In terms of relations between motorization and incomes, car-based motorization is linked to higher and high-middle income households (in addition to business owners). Motorcycles, on the other hand, are bought by low-middle and low-income households. From transport planning point of view, they are bought by households who are “normally” major users of public transport services. Just how deep down the income ladder is motorcycle ownership was illustrated 10 years ago in a survey of bus passengers in Bangalore: 27% of households with monthly income of Rs.500 or less owned a motorcycle (71% owned a bike).10 For monthly incomes in the range Rs.500-1,500, 47% owned a motorcycle. These numbers must have changed significantly since 1993, but the main point has not: many bus users are not captive and make their modal choice on the basis of some calculus of price, travel time, comfort, convenience, etc.
2.5. This said, the split of daily travel by mode is still not dominated by motorcycles and cars, but by public transport services, walking and biking. According to admittedly aged surveys in Chennai (probably 1992, with modifications based on more recent small-scale surveys), walking and biking accounted for 44% of all trips, and public transport modes carried 42%. The share of cars (2.5%) is downright minor in comparison. In Bangalore, where data are even weaker but of more recent vintage, walking and bikes carry about 17% of all trips, and public transport carried about 41% (up to 60% of all trips longer than 1 km), and individual motor vehicles carry 38%.11 Even after newer and better data adjust these numbers downward, the visual evidence of unrestrained dominance of 2-wheelers, 3-wheelers and cars on the traffic scene in these two cities is misleading. The bias comes from focusing the visits and surveys on major street traffic. Urban transport also takes place elsewhere.
2.6. One of the reasons for the importance of non-motorized and public transport modes in Chennai, and somewhat less in Bangalore, is that economic growth has left many people behind. The new wealth is in sharp contrast to concurrent poverty, with inherited inequalities deepened by the growth processes, or new ones generated by them as the migrants from the countryside pour into cities. The population growth has taken place largely at the low-income end of the economic spectrum. In spatial terms, many of the lowest income people live in informal settlements in peri-urban areas, in older city slums, or encroach any place where development by leapfrogging has left some land unused. It is not that lower-income groups have not benefited from economic growth. Many did, but growth for this stratum of urban residents is in the informal sector, low-paid and unstable jobs held by unskilled workers in construction, diverse services, and informal manufacture.
2.7. Different income strata have different expectations of the urban transport system. Those owning individual motor vehicles, be they households or businesses (the latter including freight vehicles) expect a good road system: well-maintained pavements, efficient traffic control, high travel speeds, easily available parking. Rising incomes have also increased service expectations of some public transport passengers, especially if they own or aspire to own a motor vehicle. They expect higher-quality services: easy access, a seat, high travel speed, air conditioning (especially in Chennai with its humid and hot climate). Since the majority of public transport services in both cities operate on city streets, public transport passengers are also interested in the performance of the road system, as are public transport operators. Finally, and certainly not the least important aspect, a good-quality road system and good-quality public transport services are essential parts of a “package” that Chennai and Bangalore offer to potential investors from outside, in competition with other cities in India and elsewhere.
2.8. Transport expectations of people at the low end of the income distribution are very different from those holding formal and/or better paid jobs: they rely on walking, some in addition have bicycles, and those holding or seeking distant jobs rely also on public transport services. This implies, first, the demand for a basic network of all-weather roads in the secondary and tertiary category, linked to the arterial road system. Second, it implies minimally-priced and easily accessible public transport services.
2.9. This simple 3-way segmentation of the travel market in Bangalore and Chennai does not capture the richness of what takes place on the ground. For example, the high-tech and engineering businesses of Bangalore have quite different transport habits and requirements than those than the traditional businesses, e.g. small-scale manufacture, silk weaving, commerce and services. The former are highly motorized, their job and familial networks are spread widely (well beyond Bangalore, in fact). As a caricature, it is this group that is conscious of traffic speeds and delays, and seeks flyovers, urban expressways and multi-level garages. The traditional businesses are more location-bound, with kin businesses locating in close proximity, and walking retaining importance for interaction between partners and with clients. These businesses may also be concerned for the ease and cost of longer-distance urban transport by motor vehicles, but within their large activity areas they do not seek to “reduce congestion” but thrive on it.12
The travel markets in Bangalore and Chennai are heterogeneous: car owners are at one end of the spectrum, and slum dwellers are at the other. Between these extremes are two partially overlapping groups which use public transport services and/or own motorized 2-wheelers. This is where the battle for modal dominance is being fought and where a strategic approach is called for.
B. The Performance of Urban Transport Systems
2.10. How well are the transport systems of Chennai and Bangalore serving their diverse client populations? Answers should be sought both from the service providers (the supply side) as well as those for whom the services are provided (the demand side).
2.11. A comprehensive and rigorous evaluation from the supply side is not available. The urban transport institutions in Chennai and Bangalore have not yet focused on the question of service to citizens in a systematic manner.13 The following evaluation is culled from various technical studies consulted for this report, complemented by visual evidence from a recent, but all-too-brief exposure to on-street conditions in the two cities. The overall conclusion is that the performance of urban transport systems in Bangalore and Chennai leaves much to be desired across all economic and spatial strata.
2.12. The worst off are the pedestrians in all parts of the urban areas, due to non-existent, broken-down, and/or obstructed sidewalks; large height differences between sidewalks and frequent driveways/alleyways; danger at street crossings and distance between crosswalk locations; and flooding in monsoon seasons. The next on the list of poorly-served travelers are by bicycle riders, who have few exclusive-use lanes while gradually being pushed out of busy roads by motor vehicles, be these 2- or 3-wheelers, buses or cars. Traffic accident data from Chennai show that pedestrians and bike riders are second- and third-highest group among those killed in traffic accidents, with 190 and 126 killed in 2001, respectively (topped only by 208 dead riders/passengers of 2-wheelers).
2.13. Traffic studies cite poor condition of pavements (30% of Bangalore’s road network is in that shape), low travel speeds (down to 10-12 km/h), high intersection delays, and poor or non-existent parking facilities. Traffic accidents are high at about 50 and 40 per 10,000 registered vehicles in Bangalore and Chennai, respectively, with about 700-800 fatalities (Bangalore is responsible for the upper range).
2.14. Bus services are infrequent and slow moving; buses are hard to get on/off, overcrowded (up to 150% of the nominal capacity), with uncomfortable ride, and polluting. Suburban rail services have low frequencies, and difficult access to/from stations. These generalizations apart, a 1997 survey of MTC passengers in Chennai found 75% satisfied with service frequency, 80% satisfied with punctuality, 89% satisfied with reliability, 93% satisfied with safety, and 89% satisfied with vehicle condition. The lowest score (48%) was on “route condition” which probably refers to the road condition and possibly traffic delays. The same survey covered some potential and/or ex-passengers. The ranking of “push-away” factors was as follows: low travel speed, lack of punctuality, poor connectivity and low frequency.
2.15. Are public transport services affordable? A simple analysis of travel fares and passenger incomes for CMTC (Attachment I), based on the price of the monthly fare, concluded that bus fares were onerous at monthly household incomes of less than Rs.1,000 (roughly 10-13% of passengers). At an income of exactly Rs.1,000, a monthly bus pass accounts for 14% of the household income for a 10-km trip by one person, and 26% for a 30-km commute. Commuter rail monthly passes were significantly more affordable. At Rs.2,500 a month per household, a monthly bus pass for one person would be under 10% for most distances, and rail passes were half that.14 The conclusion is that fares are set at levels acceptable for a majority of passengers.
The worst-served by the transport systems in Chennai and Bangalore are people who walk or ride bikes, who account for more than 40% of all trips, and come mainly but not entirely from lower-income and poor strata. The best served are bus system captives, since the level of service is reasonable and fares are low.
2.16. Answers from the demand side come from two recent surveys. These covered several cities, including both Bangalore and Chennai. A 2003 Confederation of Indian Industry survey of urban populations in Southern India showed 90% dissatisfied with roads, and 58% dissatisfied with public transport services. It is noteworthy that 65% of the respondents were willing to pay higher public transport fares to get more comfort and frequency, and 89% of the respondents were willing to pay for good-quality toll roads.15 A 2003 study by the National Association of Software and Service Companies, done to evaluate the relative attractiveness of major Indian cities from IT business point of view, cited Bangalore’s ”weak public transport infrastructure (that) resulted in many people buying their own vehicle” and generally low infrastructure availability.16 The same study also cited Chennai as lacking in infrastructure. In other words, the dissatisfaction with infrastructure in Bangalore and Chennai is shared between the population and the business community. The evidence of merely two surveys cannot be taken as conclusive. Still, this is a serious situation since both cities perceive their chances of continued economic growth hinges on having much better infrastructure and services then at present, not to mention the satisfaction of their own citizens.