PATANJKAR, P G and P R FOURACRE, 1986. Public transport in second order cities in India. Journal of Transport Management, 10 (7), 5-10. Public tan.,sport ~~in ,second ordeor~ citis in India Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom PUBLIC TRANSPORT IN SECOND ORDER CITIES IN INDIA P. G. PATANKAR, Director, CIRT, Pune P. R. FOURACRE, TRRL, U.K. SUMMARY The development of public transport in' second order Indian cities (with a population of 0.5 to 1.0 million) thas followed different paths. Some cities hav e adopted a more conventional approach in which State operated stage-carriage buses are the main carrier. Others have left the provision of public transport services to the activities of private enterprise, which has led to domination of the sector by intermediate public transport (minibuses and rickshaws). - This paper compares and contrasts the provision of public transport in three second order cities in India which reflect these wide differences. Performance and costs are compared in order to assess some of the relative merits of these different applroaches. Introduction The total population of India is 685 million. almost a quarter of which live in urban settlements. While there are over 200 cities having populatiors in excess of 0. I million, over one third of the urban population is concentrated in the 36 largest cities. Twelve of these cities have populations in excess of one million, the four largest (the so-called jumbo cities) having populations of more than four million, viz: Bombay, Calcutta, Delhi and Madras. As in most Third World cities, road-based, public transport plays a key Tole in the movement of people in Indian cities. The development of public transport facilities has not kept pace with demand, and further has followed a varied pattern. While the four jumbo cities have comparatively well organised mass-transit systems (road and rail based), the other metropolitan centres appear to have no planned and rigid approach to the development of public transport. Some, like Bangalore, Ahmedabad and Pune, have opted for what might be termed the more conventional approach to public transport provision: standard buses operated by public undertakings are used to provide the principal road-based, mass-transit This paper was presented in the National Seminar on Mass Trans it in Second Order Cities in India, held (it GIRT fromt 28-29, June 1986. network, without any supporting rail system. These' operators have a near-monopoly in the provision of stage-bus services, and are invariably dependent on some form of subsidy to maintain the level of service that they provide, fares being divorced from operational costs. Other cities, like Lucknow and Kanpur, have adopted a less rigorous approach in which the provision of public trans- port is left largely to the activities of private enterprise, with the minimum of regulation and supervision. Probably as a result, intermediate public transport (IPT - rickshaws and various forms of minibus) has become the dominant mode of public transport in these cities. The same variety of development is also seen in the smaller cities of India. It is a matter of some importance for future development of both the city and its public transport system to know. whether such differences confer any special benefits or costs. In a sutuation where the majority of operators of conventional public transport lose money (and hence have to be subsidised by some means) it is important to know whether there is any advantage to be gained from encouraging further development of such services in those towns. and cities which currently rely on the private sector and IPT. This paper addresses this significant issue using survey material gathered from three second-order cities (having populations in the range 0.5 to 1.0 Journal of Transport Management, July 1986 55. million) which have public transport systems representing the broad range of options which 'have come to be developed in India. Vadodara presents one end of the spectrum of options, having a conventional public transport system 'based on standard single-deck and articulated double-deck vehicles. Autorickshaws (the poor man's taxi) are used to provide supporting taxi s.ervices. At the other extreme is Patna, which has only a skeleton conventional bus service but asignificant number of minibuses and auto- rickshaws providing bus-like services, and a large number of cycle-rickshaws. In between these extremes is Jaipur, which at the time of the survey ( 1 982) had a conventional bus system, a minibus network and a taxi-like service provided by both Cycle and auto-rickshaws. A cross-sectional comparison of these pubflic transport systems is presented in order to establish some of the merits of the alternative approaches. (Note: at the time of compiling th is paper the analysis of the survey material from Patna was still in hand. The data quoted is provisional and subject to amendment). Criteria for Comparison A considerable debate has been generated from the basic question of what type of public. transport s;ystem should be encouraged in a city. The debate is often confused because it involves two quite distinct (though dependent) issues:- -the extent to which the public sector should be involved in the provision of a public trans- port system (by ownership, regulation etc). -the technical and economic benefits of different types of vehicle. Regulatory policy influences the type of vehicle employed in public transport provision, as well as operating costs in general. Under a deregulatted system private operators flourish; they tend to invest in low-cost capital equipment (minibuses and rickshaws). They are also able to operate a given conventional bus service more cheaply than nationalised undertakings, because they are less constrained by labour laws, union pressures on wage rates, tax demnands and safety standards. Clearly in a comparison of public transport systems in different cities it is necessary to try to isolate the effect of regulatory policy on unit costs. It would be wrong to assume that because a na- tionalised undertaking makes large losses on a city .6 service, that the use of conventional buses is thus automatically invalid. The economic benefits from operating large buses may outweigh the base costs (those that a private operator might incur for the same level of service) of operating the service. (WVhether the additional costs incurred by State operators are worthwhile will not be considered in this paper. The argument would be that such costs may be off-set by better safety, reliability and co-ordination of services and more fulfilment of social obligations like concessions to the poor and handicapped). The comparison of different systems is also com- plicated by the fact that public transport is not used in a homogeneous role. Cycle-rickshaws and the small 2-seat auto-rickshaws are used to provide a taxi-like service, giving a high degree of acces- sibility at relatively high cost. Minibuses (includ- ing the larger 6-seat auto-.rickshaws) and conven- tional buses are essentially used to provide stage- carriage services, where costs per unit of output are lower, th'ough journey lengths are likely to be longer, and accessibility poorer. The trade-off be- tween time (of travelling) and cost (of providing alternative vehicle types and services, together with any identifiable external costs like congestion and safety) provides the main criteria for judging al- ternative options. There are, however, other criteria which go beyond the purely transport factors, and which mnight be considered important, particularly in the Third World context. Energy and employment are two obvious candidates for consideration., Another factor which should be given due attention is the development of the city itself and the extent to which a particular public transport policy will in- fluence that development. In this paper the data currently available limits the extent of comparison to the following factors:- - the operational costs of providing public transport and related issues of relative pro- ductivity. and subsidy. - the energy and employment implications of each policy. Thus the comparison is not yet comprehensive in the sense that it is possible to say that one system is more cost-effective than another. It is possible, however, to draw some conclusions on relative advantage of different systems and to make informed speculation about the consequences of Journal of Transport Management, July 1986 introducing conventional systemns into a city like P atna. Cit y Characteristics Jaipur and Patn a are the State capitals of Rn- jasthan and Bihar respectively. Vadodara is the third city of Gujerat State. All three have expe- rienced rapid growth within the last decade and Jaipur is now a metropolitan city with a population just in excess of one million. However, for the. purpose of this study it is deemed a second order city. Table I shows basic census data for the* three cities. Table 1 - Census daka for three cities 'adodara jaipm. Paina- population (Millions) 0.74 1 1.004 0.91 6 1981 Density (persons% per. (68,4( i90(0 1 0100 s~q. kin) 1981 Populaition grovthi ( %/ ) 59, 5 A 87 19? 1-1981 .Source: Ccnsus of I ndia Public Transport Characteristics Supply Table 2 shows the inventory of the main pub- lic transport modes in the three cities. They. are sub-divided into two main groups: those vehicles pro viding bus-like services and those providing taxi-like services. In each .of the three cities conventional, pub- lic transport, which is organised by the State, loses money. All other modes are operated by the pri- vate sector and are profitable. The estimated daily cost of providing public transport in each city is shown in Table 3. (These costs are the cumulative total of the estimated daily operating costs of each vehicle being used to provide a public trans- port service). Use made of Public Transport The importance of public transport in the three cities may be gauged from the modal split figures given in Table 4. It is evident that public transport is much more important in Vadodara and Patna than in Jaipur, a fact that is reflected in the cost of providing public transport in the three cities (Table 3). Esti- mates of the use of public transport in each city are shown in Table 5. Although the number of trips per head of population is about the same for Patna and Vadodara, trip lengthss are longer in the latter and hence the demand for public transport (measured in terms of passenger kin) Is correspondingly greater. Table 2 - Public Transport Inventory in three cities Nos. of vehicles Vadodara Jaip,:r Patna per 1000 population (1983) (19112,) (1983) Bus Services: - cons entional buses 0.25 0.09 0.08 - Ininibuses - 0.18 0.09 - auto-rickslhaivs -n eg. 0.117 Taxvi .Services: -auto-rickslhaivs 7.9 2.4 - -cycle-ricksliaxi s - 7.1 39 Table 3 - Cost of providing Public Transport three cities Rs.'000 per day Vadodara Jaipur Patria Bus services 1 12) 113 165 Taxi services 22 1 96 :406 All public transport 13'J) 209 471 Subsidy paid to State 30 20 21 operators Source: Study estimates Journal of Transport Mangemnent, July 19867 7. Table 4 - Modal split for main Transport Options in three cities % of trips by Vadodara Jaipur Patna Walk 40.1 39.5 35.8 Cycle 15.1 26.5 12.5 Public Transport 32.1 21.0 33.2 Personal Motor 11.6 12.0 17.2 Vehicle Source: household surveys Table S - Use.,of Public Transport, Vadodara Jaipur Patnia Public transport trips/head of pop/yr 213 94 222 Per cent by bus 70 69 39 Per cent by con- 70 32 8 ventional bus Public transport pass. km/head of pop/yr 760 366 618 Per, cent by bus 74 82 56 service Per cent by conven- 74. 42 10 tional bus Sourice: Study estimates Quality of service There is a, high degree of variability in the quality of service offered by different modes, both within and between cities. Table 6 lists some of the attributes which can be used to describe qua- lity, together with some qualitative remarks about the main modes in use in the three cities. (Quan- titative analysis of these indicators is yet to be completed). Table 6 - Quality assessment of main modes in use in the three cities Minibus Convcn- Auto Cycle tional bus rickshaw. rickshaw Speed usually good usually good average poor Regularity variable 11n/a n/a Wait times usually good average good good Load factor high high n/a n/a Accessibility average average good good Energy consumption Estimates have been made of the expenditure on petrol, diesel and lubricating oil by the main mechanised modes of public transport. Table 7 records these costs. Table 7 - Estirriated energy costs of Public Transport Vadodara Jaipuir pa tna Energy costs Rs'ooo per day 117 36 48 as a % of total public transport costs 35 27 10 Source: Study estimates The significantly higher energy costs in Vado- dara and Jaipur (as a proportion of total costs) reflect the high content of auto-rickshaws in the total public transport fleet. The incidence of tax on petrol (used in auto-rickshaws) is much higher than on diesel. A more detailed analysis of energy costs, net of tax (i.e. resource costs, which reflect the real cost of consumption) would undoubtedly show Jaipur and Vadodara in a somewhat better light. 8 ~~~~~~~~~~~~~Journal of Transport Management, July 1986 8 'Employment Generation Table 8 contains estinmates of the number of employees directly involved in the public trans- port sectors of the three cities. Those involved in ancilliary and support services have not been in- cluded. 'Table 8 - Estimated number of Puiblic Transport Workers in three cities sidered that the minibus is likely to be more in- efficient in its use of road space than larger buses: to achieve the same level of capacity as a small number of large buses would require a larger number of minibuses. However, in the centre of Patna, where traffic congestion is conditioned by the disordered activities of numerous cycle-rick- shaws, the small auto-rickshaw may well have the advantage over larger vehicles of greater case of penetration. Vadodara Jaipur Pat na Number 6,400 7,700 28,000 Source: Study estimates The domination of the public transport sector in Patna by cycle-rickshaws is reflected in its very high employment figure. Discussion The cost of public transport per passenger car- ried varies little between the three cities despite the different inputs to provide public transport services. Table 9 shows unit costs for the main mrodes in use, as well as for the total system. (These costs include all direct operating costs plus fixed overheads, where appropriate). The better cost performance of conventional buses in Vadodara reflects the high utilisation of vehicles and high number of passengers using the system, as compared to Jaipur and Patna. The hig-her utilisation of vehicles is partly a result of better operational efficiency and also possibly the less congested nature of the street network in Vadodara. The higher throughput of passengers will itself be dependent on the better vehicle utili- sation, as well as the absence of any competing stage-bus service in Vadodara. Furthermore, pas- sengers in Jaipur, and to a lesser extent Patna, mnake longer journeys on conventional buses than they do in Vadodara; as a result the effective capacity of the vehicle is less. Minibuses in Patna are a relatively high cost mnode. This is accounted for solely by the exten- sive use of the 6-seat auto-rickshaw, a high cost vehicle when used in the bus-like role. (The lar- ger 25 seat minibuses in Patna are equally as cost ,efficient as those in Jaipur). It is generally con- In terms of costs per passenger km. public transport in Patnia is significantly more expensive than in the other two cities, due to the high use of cycle-rickshaws for short journey lengths. Cycle- rickshaws are a costly form of public transport; although their capital investment and operating costs are low, their output is correspondingly small. The similar cost of public transport in Vado- dara and Jaipur is not reflected in similar use, which is much lower in the latter. The reasons for this are not immediately clear and further analysis is required to identify what other modes are being used in preference to public transport and the reasons for such use. It might be expected that the high cost of public transport in Patna. compared to Vadodara, would be reflected in lower use; this is the case, with demand (measured in terms of passenger km per head of population) being some 20 per cent greater in Vadodara. Subsidy given to State bus operators is higher in Vadodara than the other two cities, both in absolute terms and per head of population. (The subsidy ranges from Rs. 15 per person per year in Vadodara to about Rs. 8 in both Jaipur and Patna). It seems probable, however, that more of the Vadodara subsidy goes towards maintaining cheap fares and/or better services than is the case in the other two cities. (If it is assumed, conservatively, that private operators could pro- vide the same services, using the same vehicles, for 1 0 per cent less cost than the State operators- for the reasons already discussed in Section 2 -then the total cost savings would range from Rs. 1 1,000 per day in both Vadodara and Jaipur to Rs. 1 6, 000 per day in Patna. The difference between this and the actual subsidy paid represents the cost of main- taining low fares and/or socially desirable services. It represents about two thirds of the 'subsidy in Vadodara, one half in Jaipur, and less than one quarter in Patna). Journal of Transport M~anagement, July 19869 9 Table 9 - Operating costs per Passenger and per Passenger Km. per- passkm per pass pcr passkm Couventiorial bus 3 7 'U0 83 1 7 11 iiibus 1 0 Auto-rickshaw(2-se-at) 167 5 6 2126 58) Cycle-rick~shaw 76 5 1 4.5 90 All public transport 76 2 1 - 80 2 1 Sour-ce: Study estimates Th~e Vadodara public transport system undoub- tedly has high energy costs. It is an all-mechanised system, and a high proportion of its content is auto-rickshaws. The Jaipur system is little better in energy terms, having energy costs per passen- ger km. double that of the Patna system. There is a clear trade-off between the high energy costs of Vadodara and Jaipur with the high labour employment (and hence costs) of Patna. Labour costs represent something like 60 percent of the total operating costs of public transport in Patna. A direct comparison of the two extreme options for public transport provision does suggest that operating expenditure could be reduced in Patna by the introduction of more conventional services and/or larger minibuses, with a corresponding re- duction in cycle-rickshaws and small minibuses. Whether users of public transport in Patnia cur- rently have a better quality of service than those in Vadodara is yet to be established. (This task is part of the current study). journey times by cycle- Tickshaw may be faster overall (because cycle- -rickshaws eliminate the long wait times and walk ,times associated with conventional bus activities). It seems unlikely, however, that any such advan- tage would outweigh the high cost penalty of providing this form of transport. The indications are that the conventional bus system should be encouraged to provide the back- bone of the overall public transport system in se- cond order Indian cities. At the same time, there is a role for IPT: rickshaws providing a relatively cheap taxi-service and minibuses providing for des- tination oriented. route patterns on narrow roads and for low demand routes. This is vividly brought out in the case of Jaipur where all forms of pub- lic transport combine to produce a low cost system catering for a wide cross-section of trips. Vadodara indicates how a combination of conventional buses and auto-rickshaws can still be effective. Patnia is an example where too much stress on the auto- rickshaw has increased the cost per passenger km., though has contributed to the much needed em- ployment potential and energy saving. Acknowledgements The observations recorded in this paper are the outcome of a co-operative research programme undertaken by the Central Institute of Road Transport (GIRT -of which Dr. Patankar is the Director) in Pune, the Town and Country Plan- nin~g Organisation (TCPO) in N. Delhi, and the Overseas Unit (Unit Head : Mr. J. S. Yerrell) of the Transport and Road Research Laboratory (TRRL) in Crowthorne. The authors wish to ac- knowledge the contributions of Dr. D. A. C. Mauni- der (TRRL), Dr. M. V. Bagade (CIRT) and Mr. P. G. Valsangkar (TCPO), in the survey work and subsequent analysis. The paper is pub- lished by permission of the Director. 0@0 10 ~~~~~~~~~~~~Journal of Transport Management, July 1986 Custs-paise Vadodara per I)ass jaipuir Patrnia per pass per passkm 7 1 18a 7 7 18f 84e 30 1 0