ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Unravelling Myths about Subsidies in Urban Transport

Is urban public transport subsidised more than its private counterpart? Through a case study of urban transport in Pune, this article demonstrates that car and two-wheeler users receive larger subsidies than bus users when all costs imposed by transport modes are considered.

INSIGHTEconomic & Political Weekly EPW august 2, 200837This article has been greatly improved by inputs and comments from Ashok Datar, Shantanu Dixit, Sharadini Rath, Sriraman Siva and Sudhir Chella Rajan.Ashok Sreenivas ( and Girish Sant ( are with the Prayas Energy Group, Pune.Unravelling Myths about Subsidies in Urban TransportAshok Sreenivas, Girish SantIs urban public transport subsidised more than its private counterpart? Through a case study of urban transport in Pune, this article demonstrates that car and two-wheeler users receive larger subsidies than bus users when all costs imposed by transport modes are considered.This article challenges the popular perception that urban public trans-port, especially bus services in In-dia are subsidised while cars and two- wheelers are not.Itdemonstrates through a case study of urban transport in Pune that a bus user is much less subsidised than a car or two-wheeler user when all costs imposed by transport modes are con-sidered. In the process, it also examines some issues regarding municipal budge-ting in Indian cities. Costs Imposed by Urban Transport Costs imposed by transport have been well studied in research [Lindberg 2002; Sen et al 2005; Shoup 2005; VTPI-TCA 2007]. For example, Litman lists 20 dif-ferent costs inVTPI-TCA (2007). These include vehicle ownership, operation, road land value, road construction and maintenance costs, congestion caused, etc. Some of these costs are internal in that they affect only the user of the service without any impact on the rest of society, while many are external, i e, imposed by the user on the rest of society[Lindberg 2002; Sen et al 2005]. Thus, while the cost of owning a vehicle is internal, costs such as air pollution, congestion and safety risks are external as the rest of society is also affected by them, often more than the owner of the vehicle. Moreover, external costs differ across different modes of transport. For example, air pollution costs imposed by a car differ from those imposed by a bus. For this article, we define “subsidy” given to a mode as the sum of the external costs imposed (and unpaid) by the mode plus any explicit subsidy to the mode. Considering such externalities, which are typically ignored, drastically changes our understanding of subsidies. Our primary interest is in exposing the relative subsidies enjoyed by the three modes rather than the actual values. Therefore, we perform a simple and approximate analysis to estimate some of the external costs that are usually not considered and arrive at roughly indica-tive numbers. Computing precise numbers requires more complex modelling [Sen etal2005] that is beyond the scope of thisarticle. We focus only on three modes of passenger transport, namely public transport buses, two-wheelers and cars. Moreover, we consider just the land and road construction costs. Land Cost: Since land is a very precious urban resource, its opportunity cost is one of the costs of providing transport servi-ces.1 For our purposes, we assume that road space is leased for transport services at a nominal rate of 3 per cent per annum.2 We consider land used in two ways. The first is road land cost. This estimates the opportunity cost of land used to build roads. This cost is divided among different modes according to the amount of space they occupy at peak hours, since the amount of road space required is dictated by peak hour traffic. The relative road space occupied by each mode is obtained by multiplying its passenger car unit (PCU) number3 with the share of vehicles of that mode in the total number of vehicles on road at peak time. The second is depot land cost. This is the cost of land given by the city to be used as bus depots. This cost is allocated only to the bus service.Road Construction Cost: This covers the cost of building and maintaining roads. For simplicity, this cost is also divided among modes according to the space they occupy at peak hours, and ignores other factors such as the weight of vehicles, their speed, distance travelled, etc. In addition to the two implicit subsidies considered above, we also consider the ex-plicit subsidy given to the bus service. Therefore, the total subsidy (Rs lakh/day) availed by buses is the sum of the explicit subsidy, depot land cost and their share of road land and road construction costs. The total subsidy availed by cars and two-wheelers is the sum of their share of road land and road construction costs.
INSIGHTAugust 2, 2008 EPW Economic & Political Weekly38The percentage of passenger-km travel-led by a mode is calculated from the avail-able data for the percentage of vehicle-km travelled by the mode and its average occupancy. The subsidy per passenger-km for a mode relative to buses is obtained by dividing the total subsidy availed by a mode by its percentage of passenger-km and scaling it relative to the subsidy re-ceived by buses. Users (of both private and public trans-port) pay only a small fraction of these costs as price through a one-time vehicle registration tax, octroi on fuel, parking fees where applicable and (part of the) bus ticket fare.4 The price paid being a small percentage of the total cost, it is appropri-ate to treat the entire cost considered here as a subsidy. Subsidy CalculationsCosts: Pune has 10.4 sq km of land under roads [Environmental Status Report (ESR),Pune 2007]. Valuing commercial land in Pune at a conservative Rs 20,000 per sq m, the total road land cost for Pune is Rs 171 lakh/day. We consider the aver-age of the amount explicitly spent on/al-lotted to road works in the last three years by thePune Municipal Corporation [PMC Budget 2008-09].5 This comes to an average of Rs 566 crore per year or Rs 155 lakh/day.About seven acres of land are being used by Pune’s bus service Pune Maha-nagar Parivahan Mahamandal Ltd (PMPML) for their depots. So, the depot land cost is Rs 0.52 lakh per day. The ex-plicit amount allocated for PMPML is Rs 25 crore or Rs 6.85 lakh/day (ibid).Vehicle Data: Table 1 calculates the rela-tive road space occupied by different modes using data from Pune City Devel-opment Plan [Pune CDP 2006]. This data is based on a detailed survey carried out in Pune in 2005. ThePCU numbers are standard in transportation literature.6Table 2 presents each mode’s share of passenger-km, calculated from its share of total vehicle km (ibid). For simplicity, we assume that “other vehicles” do not carry any passengers, since our interest is in the other three modes.Analysis Findings: Using the data for Pune, the total subsidies per day and rela-tive subsidy per passenger-km can be cal-culated for each mode. Figure 1 (p 39) presents the overall subsidy enjoyed for each mode and Figure 2 (p 39) presents the relative subsidy per passenger-km for them. The following observations can be made from this. First, using just the exter-nal costs considered here, the total subsidy per day amounts to Rs 333 lakh, out of which explicit subsidies account for less than Rs 7 lakh a day – only 2 per cent, while the remaining 98 per cent is an implicit subsidy.Second, comparing the total subsidy across modes, buses are subsidised about Rs 33 lakh per day, while cars are subsi-disedRs 72 lakh and two-wheelers Rs 151 lakh. That is, two-wheelers as a mode re-ceive 4.6 times the subsidy buses get, while cars get 2.2 times that. However, note that two-wheelers also carry about twice as many passenger-kmascars,as reflected in the per passenger-kmsubsidy analysis which follows. Third, making a comparison on a per passenger-km basis, two-wheelers are subsidised 10.4 times as buses and cars are subsidised 9.2 times. It is slightly counter-intuitive that two-wheelers are subsi-dised more per passenger-km than cars. We believe this is because both the costs we consider are determined by the share of the mode in peak hour traffic and it is likely that cars use roads more inoff-peak hours (such as the cars ferrying business process outsourcing (BPO) employees). The data we use is from 2005 and the number of cars on road (at peak and non-peak hours) has grown faster than the number of two-wheelers since then. Therefore, it is likely that even for these two costs, cars are more subsidised if one could use current data. Moreover, if one considered only the peak hour passenger-km by different modes, cars would be more subsidised than two-wheelers and buses would be far less subsidised than seen here. This is because bus occupancy at peak hours is more than 100 per cent rather than the 50 per cent considered here while car and two-wheeler occupan-cy would be pretty similar and peak hour passenger-km by two-wheelers would be considerably more than cars. DiscussionThe most obvious conclusion is that public transport buses are far less subsidised than private motorised transport, both in abso-lute terms and terms of per passenger-km. This is highly regressive considering that cars and two-wheeler users are generally more affluent than bus users [Badami et al 2004; Baker J R et al 2005].Moreover, 98 per cent of the subsidies considered are implicit. While the entire subsidy to cars and two-wheelers is im-plicit, only 80 per cent of the subsidy to buses is implicit. This explains the popu-lar perception that public transport is sub-sidised while private transport is not be-cause the only small visible element of the subsidy applies to buses while the much larger hidden subsidy is consumed mainly by cars and two-wheelers. Other Costs: We briefly examine whether considering other costs paid by users of cars, two-wheelers and buses change the analysis conclusions significantly. A more detailed analysis is beyond the scope of this article. Some external costs not con-sidered by us are now discussed. First, cars and two-wheelers impose a greater air pollution cost than buses, both overall and per passenger-km. This cost includes the cost of healthcare due torespiratory illnesses, cost of lost productivity due to illnesses, etc. Second, cars and two-wheelers impose a far Table 1: Relative Road Space Occupied % of Total Vehicles PCU % Space Occupied on Road at Peak Hours at Peak HourCars 171.0022Two-wheelers 71 0.5046Bus 2 3.008Autorickshaws 6 1.008Others 43.0016Table 2: Relative Passenger-km Travelled by Different Modes Average % of Total Passenger- Occupancy Vehicle Km Km (%)Cars 1.2a3013Two-wheelers 1.2b 55 24Bus 25c 6 56Autorickshaws 2 9 7Others 0 0 0(a) This figure is taken from Badami et al (2004). (b) This figure is considered equal to that of a car, thought it would be lower. (c) Pune buses have a load factor of 51%. Thus 25 is a conservative estimate of occupancy per bus.
Explicit Road con-struction Land cost
Explicit Road con-struction Land cost
INSIGHTAugust 2, 2008 EPW Economic & Political Weekly40accessible to the poorer sections of society, there is also a socio-economic justification for subsidising it. However, our study shows that the av-erage car and two-wheeler user is subsi-dised more than the average bus user through underpriced or free infrastruc-ture. Such underpricing only increases the demand for cars and two-wheelers as they do not bear the full cost of the service [GOI 2006;MRTS2006]. It is interesting to contrast the existing urban transport pricing situation with the electricity sector. The National Electricity Act and electricity reforms insisted on cost-based pricing in order to expose all the costs involved in electricity supply [E-Act 2004] in spite of the fact that this would affect the poorer sections of society the most. However, the urban transport sector seems to ignore cost-based pricing though the beneficiaries are the better-off sections of society. If urban transport planning were to minimise the total (external and internal) societal cost, it would have the following advantages: (i) correct the regressive subsidies that exist today and make it fairer for the poorer sections of society; (ii) make more funds available for public transport; (iii) encourage greater use of public transport; (iv) provide the commut-er a greater choice based on the true cost of each mode; (v) decrease the internal cost borne by each commuter, since there would be lesser need to invest in buying and running a vehicle; and (vi) greatly re-duce the need for infrastructure expenses by the city. This is not to say that public transport systems in India do not need improve-ment. In fact, they need drastic and urgent improvement. However, the ills of our public transport systems and their govern-ance are separate topics, beyond the scope of this article. But what is clear is that cars and two-wheelers getting greater subsidies than bus users cannot be justified on any eco-nomic, social, environmental or policy grounds. Therefore, there is an urgent need for correction in the structure of urban transport pricing in Indian cities. Different cities around the world have adopted different ways of addressing the hidden subsidies to private motorised transport. For example, London and Stockholm have introduced congestion charging. Singapore has severe restric-tions on ownership and usage of cars, while Bogotá restricts their use on differ-ent days. Other pricing options available include road usage charges, fuel cess, greater vehicle taxes, unsubsidised park-ing etc. Each Indian city can and must choose the solution best suited to its local context from the available bouquet of options. As a first step, the city adminis-tration must explicate all such hidden subsidies and place them in the public do-main. Thereafter, a transparent and parti-cipative process can help identify the best solution. This will not only send the right signals to transport users but also improve the city’s economic, social and environ-mental health. In other words, ration-alising the pricing can start off a virtuous cycle, which would be in the larger public interest. ConclusionsThis article exposes the hidden subsidies enjoyed by users of cars and two-wheelers, thus marginalising the needier sections. Since such a subsidy structure defies all rationale, there is an urgent need for reform in urban transport pricing in our cities. Rational pricing of urban transport can not only move people away from private modes to more desirable public modes but also help improve access and mobility of the poorer sections of society and provide more funds for other social expenses. The analysis in this paper also high-lights some misplaced priorities and a cer-tain lack of transparencyinmunicipal budgets. For example, Pune’s budget allo-cates less money to public health than to providing signals at junctions and divid-ers on roads, and proposes anSPV ostensi-bly to improve public transport but actu-ally containing a large number of road improvement projects that will primarily benefit private motorised vehicles. This highlights the need for greater trans-parency and public participation in mu-nicipal budget preparation, as this will help in allocating funds according to pub-lic needs and priorities and help citizens better understand how their money is being spent. Notes 1 For example, could the space have been better used for, say, a primary school or a public hospital? 2 The figure of 3 per cent is chosen to reflect a nomi-nal lease rate of a non-depreciating asset. 3 The PCU number of a mode reflects the road space required for a vehicle of that mode, considering the road space requirement of a car to be one. 4 Other prices paid by users such as fuel, insurance, etc, are internal, i e, they do not compensate for the external costs imposed. Hence, they are not considered. 5 Note that this does not include many other road related expenses such as road lighting, signals, speed breakers, etc, all of which may contribute another Rs 150 crore. 6 We assume that all the “other” vehicles have a PCU of 3, to minimise the road space share occu-pied (and therefore subsidy enjoyed) by cars, two-wheelers and buses. 7 This is not the amount budgeted to be spent in the current year, which is only Rs 300 crore. 8 In comparison, buses get just Rs 44 crore a year.ReferencesBadami, Madhav, Geetam Tiwari and Dinesh Mohan (2004):Access and Mobility for the Urban Poor in India: Bridging the Gap between Policy and Needs, forum on Urban Infrastructure and Public Service Delivery for the Urban Poor.Baker, J R, M Basu, Cropper S Lall and A Takeuchi (2005): Urban Poverty and Transport: A Case Study of Mumbai, World Bank Policy Research Working Paper, September.Eleventh Plan Working Group (2006): Report of the Working Group for the 11th Five Year Plan on Urban Transport Including MRTS.Government of India (2004):National Electricity Act, Government of India. – (2006): National Urban Transport Policy, Ministry of Urban Development, Government of India. – (2008): National Action Plan on Climate Change, Prime Minister’s Council on Climate Change, Government of India.Lindberg, Gunnar (2002): Recent Progress in the Measurement of Eexternal Costs and Implications for Transport Pricing Reforms, IMPRINT Europe Report.Litman, Todd (2007): Transportation Cost and Benefit Analysis: Techniques, Estimates and Implications, Victoria Transport Policy Institute, May. Mohan, Dinesh and Geetam Tiwari (1999): ‘Sustain-able Transport Systems, Linkages between Envi-ronmental Issues, Public Transport, Non-motor-ised Transport and Safety’,Economic & Political Weekly, Vol 34, No 25.Ministry of Urban Development (2006): National Urban Transport Policy, Ministry of Urban Deve-lopment, Government of India.Pune (2006):Pune City Development Plan. – (2007): Environmental Status Report 2006-07, Pune Municipal Corporation. – (2008): ‘Pune Municipal Corporation Budget Pro-posal for 2008-09’, accessed July.Sen, Akshaya, Geetam Tiwari and V Upadhyay (2005): Optimal Pricing of Urban Road Transport – An Eco-nomic Approach, National Conference on Urban Transport Planning and Management, CIRT, Pune.Shoup, Donald (2005):The High Cost of Free Parking, American Planning AssociationSingh, S K (2005): ‘Review of Urban Transportation in India’, Journal of Public Transportation, Vol 8, No 1.VTPI-TCA (2007):Transportation Cost and Ben-efit Analysis: Techniques, Estimates and Implica-tions, Todd Litman, Victoria Transport Policy Institute, May.

To read the full text Login

Get instant access

New 3 Month Subscription
to Digital Archives at

₹826for India

$50for overseas users


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top