View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Loughborough University Institutional Repository

This item was submitted to Loughborough’s Institutional Repository by the author and is made available under the following Creative Commons Licence conditions.

For the full text of this licence, please go to: http://creativecommons.org/licenses/by-nc-nd/2.5/

68 DRT – lessons from LDCs Demand Responsive : lessons to be learnt from Less Developed Countries Marcus Enoch examines three Demand Responsive Transport (DRT) schemes which are being successfully operated in Turkey, Kong and Mauritius and asks what the UK can learn from their experiences.

ith the rapid increase in car transport agencies and operators to de- to be kept as low as possible so as to en- ownership, the market share velop systems able to operate effec- able the socially excluded to afford to W of systems tively with lower and more dispersed use them. Unfortunately, this has had has been gradually declining through- patterns of demand than the , ie the unintended outcome that many out much of the industrialised world Demand Responsive Transport (DRT1). such schemes may be alarmingly vul- since the 1950s. This has a conse- In Britain, such DRT schemes have nerable should this financial crutch be quence not only of the car directly re- tended to have been driven by public removed. Furthermore, given the high placing specific trips, but also of indi- policy goals (particularly improving so- levels of subsidy required, it is unlikely rect mechanisms that affect the de- cial inclusion) backed by the ready that such DRT schemes will ever be ex- mand for transport through dispersing availability of Government cash in the tensive enough to offer more than a activity and land use patterns making form of grants for which local authori- marginal contribution to reducing car them less suitable for and rail- ties had to compete. This has had the use and hence to mitigating congestion The dolmus based modes to serve. effect of setting up ‘innovative’ and environmental problems in the service operates One response enthusiastically schemes that are relatively high tech wider transport context. in urban areas adopted by a number of Governments and high quality (such technological It is true that there are exceptions to across Turkey. such as in the Netherlands, United innovation being necessary to secure this high-tech, high cost approach to States and the UK, has been for public the funding), in areas where need DRT, even in the UK. For example, the Black Taxibus has been operating in areas of Belfast since 1969, while in summer 2003 public transport operator Stagecoach introduced the Yellow Tax- ibus service linking Dunfermline with central Edinburgh. And, as of August 2004, the entrepreneur behind Easyjet, Stelios Haji-Ioannou, launched a DRT service connecting Milton Keynes, Bed- ford and Luton with Brent Cross Shop- ping Centre and transport interchange in north London. So far though, only the first of these has proved itself over a period of time, and even then only op- erates in an environment that could never really be described as being ‘typical’. However, elsewhere in the world there are many examples of DRT opera- tions that have been established for many years that make a significant contribution to the transport systems tec FEBRUARY 2005 69 DRT – lessons from LDCs Three columns features 69

In Istanbul more they form a part of, without any need than half of for public subsidy. These include the public transport Jitneys of Atlantic City, USA; the Jeep- trips (400m neys of Manila, ; the Taxi Collectifs of Havana, Cuba; the Sherut passengers a of Tel Aviv, ; the Kombi shared year) are carried taxis of Cape Town, South Africa; the by dolumus. Matatu shared of Nairobi, ; and the shared minibus of , ; to name but seven of a huge number of subtly different types of DRT operation across the world. This article therefore aims to draw on the experiences of three slightly dif- ferent schemes, to see if any lessons can be learnt that may be applicable in the UK context. These are, the Dolmus of Istanbul, Turkey; the Public Light over fares. A second problem is that be- Kong Transport Department. Bus of ; and the Taxi Train cause the Dolmus picks up and drops PLBs are composed of two types, of Mauritius. off passengers almost anywhere there is those with red roofs (RMBs) and those a heightened risk of collisions and con- with green roofs (GMBs). Of these, the Dolmus, Istanbul, Turkey gestion caused by drivers veering RMBs provide non-scheduled services The dolmus (pronounced ‘dolmush’) is sharply or stopping suddenly to max- that are purely ‘demand responsive’ ie a form of hail-and-ride shared taxi or imise their revenue. they operate flexibly subject to the minibus service that operates in urban Overall, more than half of public market demand, and the Transport De- areas across Turkey. Vehicles typically transport trips (400 million passengers partment has no control over routes or operate on an owner-driver model and a year) in the city are carried by around fares. Thus, when demand is high in are licensed to run on specific routes 4,000 14-seat and 16,000-or- peak periods, or when rail or other bus (widely defined) by the local authority so Dolmus licensed shared taxis nor- services are not working properly, or – in Istanbul this is the Department of mally carry eight or nine passengers even when there is bad weather, fares Transportation Coordination of the (Janes Information Group, 2004). charged will be higher than at other Metropolitan Municipality (IMM) – for times. RMBs are also allowed to operate a profit. Despite this though, as with , Hong Kong, China anywhere within their existing service taxi drivers there appears to be some Public Light Buses (PLBs) are minibuses areas but not in new towns or new kind of cooperative relationship occur- with not more than 16 seats, and were housing developments in Hong Kong. ring on a route by route basis. It also introduced in 1969 to regulate the ille- There are also local stopping restric- In Hong Kong the tends to be the case that similar gal minibus trade that existed at that tions on RMBs to relieve traffic prob- Public Light Bus makes/colours are employed on partic- time. Altogether PLBs account for 15% lems caused by their aggressive manner fleet is largely ular routes (eg Tacsim-Besiktas are all of all public transport journeys in of stopping and waiting for passengers unregulated, with yellow , while other Hong Kong, a figure which compares at kerbside and at road junctions. fares set only by routes may use light blue or beige favourably with that of taxis which op- Overall, the size of the PLB fleet has what the market minibuses). Strictly speaking, the num- erate 12% of all journeys. PLBs are op- been frozen at 4,350 since 1976 by an will bear and ber of licences is rationed for each cor- erated either by individuals or by order of the Hong Kong Executive routes relatively ridor. minibus companies under passenger Council. The limitation order has been unrestricted. There are formal ‘terminus stops’ service licences issued by the Hong extended from time to time since then (where drivers generally wait until they are almost full) but elsewhere along a corridor Dolmus’s stop where they are requested to do so. Fares are set by the IMM and are comparable to bus. The advantages that Dolmus’s offer over the bus are that frequencies can be far higher in the peak, journeys tend to be quicker, there are usually seats (defi- nitely seats on some vehicles but not always on others), and drivers will al- ways accept cash whereas on the buses either a fixed or pre-bought tickets are often required. However there are also problems. One is that while the numbers of Dolmus and minibuses have been frozen for forty years or so, the number of trips has risen dramati- cally as Istanbul has continued to grow. As a result there is a significant level of illegal operation which has led to ser- vices being overcrowded and disputes

tec FEBRUARY 2005 70 DRT – lessons from LDCs

any subsidies to operate as a taxi-train, although they are offered an 80% re- bate by Customs and Excise on the pur- chase taxes of their vehicles (a substan- tial incentive given that this can be set as high as 200% of the value of the ve- hicle), and they pay only half the an- nual road tax. In brief, Section 103 of the Road Traf- fic Act 1962 first permitted licensed taxis to operate as so-called ‘taxi-trains’, which are effectively supplementary buses that are able to stop and pick up passengers along a particular route and charge separate fares of each passenger. While initially these taxi-trains were unable to collect passengers within 60 metres of a , the public trans- situation became so chaotic during the mid-1970s, with many passengers waiting for inordinately long periods for a bus, that the 60-metre rule was re- Above and below through resolutions passed by the Leg- actly the right number of seats (as the scinded. Interestingly, the separate fares right: islative Council, and was last extended more accurately they do this the higher charged are set at the same level as for a The taxi-trains of in June 2001 for five years up to June the fare they can charge and the more bus operating the same route, despite Mauritius operate 2006. money they will make). the rather quicker and more comfort- on a largely Interestingly the Government’s es- able ride offered by the taxi-train. Un- tablished policy is now to encourage ‘Taxi-train’ shared taxis, Mauritius, derstandably, there are many people network rather the provision of scheduled PLB service Indian Ocean who prefer to wait for a taxi-train, even than on corridors in the form of GMBs to supplement the The final example, the taxi-trains of if a bus to the same destination arrives in very large higher capacity rail and bus-based pub- the Indian Ocean island of Mauritius, is in the meantime. cities. lic transport modes. GMBs were first in- slightly different from the previous two With the taxi train the crucial point troduced in 1972 through the conver- in that the taxi-trains operate on a seems to be that the service is an addi- sion of RMBs. They operate scheduled largely interurban network instead of tional option for the taxi driver – almost services on fixed routes at fares ap- on corridors in very large cities. The a job of last resort that can be employed proved and regulated by the Transport service therefore tends to operate with when there are no ‘conventional’ (and Department. taxi vehicles carrying up to four pas- lucrative) tourist fares in the offing. Thus, as of October 2003, there were sengers rather than the larger capacity Also, as with the Dolmus, the route fran- 1732 RMBs (down from 1835 a year minibuses of Istanbul and Hong Kong. chises are fairly widely defined to allow earlier) and 2618 GMBs (up from Once again the mode was intro- flexibility, but tight enough to maintain 2515). Meanwhile the daily patronage duced as a result of a public policy deci- the semblance that the service is ‘local’. of RMBs was around 500,000 in 2003 sion rather than as a market response This means the likely demand at various and for GMBs about 1.2m. In October to a perceived gap in the transport mar- times of the day/week is likely to be 2002 there were 71 GMB routes on ket by commercial operators, this time served with roughly the right number of Hong Kong Island, 76 in Kowloon and to try and help address the problem of vehicles and that the driver is often 193 in the . Unfortu- a severe shortage in the supply of pub- known at least to the more regular users nately there are no data relating to the lic transport. And, once again the which engenders a feeling that the ser- operating costs of PLBs owner-drivers are not provided with vice is ‘safe’. Augmenting franchised buses and PLB fleets are about 9800 buses, minibuses and coaches used mainly on contract for carrying tourists, factory workers, school children and com- muters. In summary, the Hong Kong RMB case is fascinating because it illustrates that even in perhaps the most attrac- tive city in the world in which to oper- ate high capacity rail and bus based public transport systems (very many well off people owning very few cars, living and working in very high densi- ties along radial corridors etc.) there are still significant markets to be served (profitably) by DRT. Here, the key ele- ment seems to be that PLB drivers use their (combined) local knowledge to match the predicted demand with ex- tec FEBRUARY 2005 71

Overall, it is not known how many or what type of trips are made using taxi-trains in Mauritius, or what type of people are using the mode. Around 5300 taxis are registered nationwide.

LESSONS

In conclusion there a number of key themes that emerge from the cases ex- amined. Firstly, each of the systems is operated and used in a consistent man- ner across the region it serves. Thus, only one ‘system’ needs to be learnt to use any RMB in Hong Kong (unlike with DRT in the UK where every sys- tem seems to be subtly different from its neighbour). Second, service quality is higher with the DRT vehicles being more comfort- with the exception of the Yellow Tax- Department and the Istanbul Metropolitan The launch of the able than the bus, and journeys often ibus and Easybus schemes referred to Municipality for their help. Yellow Taxibus being individually tailored to meet the earlier, the UK DRT market is still rela- It is also necessary to offer my gratitude to service linking needs of the customers on board. tively immature. the Department for Transport and Greater Dumferline with Moreover, journey times are often less This is the case largely because of the Manchester Passenger Transport central Edinburgh than a bus due to a reduced waiting high degree of technical and opera- Executive who commissioned the is a recent time (RMBs, Dolmus and taxi-trains are tional experimentation and innovation Intermode Study on which much of this addition to the generally more frequent than the bus) stimulated by significant levels of Gov- work is based, as well as my colleagues on UK’s innovative and reduced in vehicle time (smaller ernment funding through the Rural the project, Stephen Potter of the Open DRT solutions. vehicles are more manoeuvrable on and Urban Bus Challenge grants. Un- University, Graham Parkhurst of the congested streets and stop less often, fortunately, while this investment has University of the West of England and for a shorter period). demonstrated that DRT is a viable Mark Smith of Sustainable Business Finally, each of the services operates mode in a technical sense, on an opera- Solutions. commercially at a profit in high vol- tional level the resultant focus on the ume markets where they provide addi- marginal, small scale, high tech, high REFERENCES tional capacity to the existing public cost, low fare end of the market means transport network in places and/or at that many current UK schemes are fi- times when bus or rail based systems nancially vulnerable. In short, these Brake J, Nelson J D, Wright S (2004) would not be viable. This is despite schemes reside beyond (sometimes way Demand Responsive Transport: Towards both the RMB and the taxi-train being beyond) what would be considered an the emergence of a new market segment, introduced for public policy reasons. acceptable level of subsidy measured Journal of Transport Geography, 12, 323- Of course there are also some differ- on a ‘per passenger trip’ or ‘per passen- 337. ences. Relating to fares, while the Mau- ger trip kilometre’ basis for a conven- Enoch M P, Potter S, Parkhurst G and ritius and Istanbul DRT systems charge tional bus service. Smith M T (2004) Exploratory regulated fares only very slightly Of course, some of the current style Assessment of and Innovations in Demand higher than the buses they compete of UK DRT systems will become in- Responsive Transport Services - with, in Hong Kong the fares are set creasingly financially viable as they Intermode, Final Report, Department for only by what the market will bear. Sim- begin to grow, and/or as DRT technol- Transport and Greater Manchester ilarly with the routes – the Mauritius ogy begins to offer a value-for-money Passenger Transport Executive, London, and Istanbul cases are effectively lim- replacement for some über-expensive August. Visit http://www.dft.gov.uk. ited (though fairly loosely set), whereas social-service, dial-a-ride or patient Janes Information Group (2004) Janes the RMBs of Hong Kong are rather less transport services. But really perhaps Urban Transport Systems 2004-2005, Author: restricted and can almost go where the the greatest scope is at the more com- Transport Library CD ROM, Janes demand is. mercial end of the spectrum, where ei- Information Group, Coulsdon, Surrey. Dr Marcus Enoch And there are a few ways that each of ther a more mainstream low tech, low is a lecturer in these services could be improved. In cost, low fare approach, or the niche 1. Brake et al (2004) defines DRT as being transport studies particular, none of the three DRT high quality, premium fare model (eg ‘an intermediate form of public transport, at Loughborough modes is geared towards attracting oc- like shuttles found across the somewhere between a regular service route University, a casional users2 (a situation that is also USA, or employer shuttles eg at Voda- that uses small low floor buses and variably visiting research sadly true in UK DRT systems). There is fone in Newbury, Berkshire) may be the routed, highly personalised transport fellow at The also clearly a need for dangerous dri- real way to mainstream DRT systems. services offered by taxis’. Open University ving habits to be discouraged. 2 Indeed, even finding the most basic and a freelance Overall though, the three systems ACKNOWLEDGEMENTS information about how to use the Dolmus, journalist. He can described almost provide a taxi-level of RMB or taxi train is almost impossible for be contacted by service but for almost a bus level of fare the casual user and not especially easy for email at – a niche that in Britain remains largely Thanks are due to Dhiraj Romooah of the a seasoned public transport researcher m.p.enoch unexploited. Mauritian National Transport Authority, despite some prior knowledge of what was @lboro.ac.uk. As for transferring such experiences, Ming Pang of the Hong Kong Transport to be expected.

tec FEBRUARY 2005