World transport policy


Published on

Published in: Travel, Technology
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

World transport policy

  1. 1. World Transport Policy & Practice Picture courtesy of: Millie Rooney Volume 14, Number 2 Sustainable intermediate transport in West Africa Bicycle Ambulances in rural Uganda International review of the significance of rail in developing more sustainable urban transport systems in higher income cities Simplified travel demand modelling for developing cities: the case of Addis AbabaEco-Logica Ltd. ISSN 1352-7614 World Transport Policy & Practice___________________________________________________ 1 Volume 14. Number 2. July 2008
  2. 2. World Transport Policy & Practice Volume 14, Number 2© 2008 Eco-Logica Ltd. Mikel MurgaEditor Leber Planificacion e Ingenieria, S.A., Apartado 79,Professor John Whitelegg 48930- Las Arenas, Bizkaia, SPAINStockholm Environment Institute at York, Departmentof Biology, University of York, P.O. Box 373, York, Paul TranterYO10 5YW, U.K School of Physical Environmental & Mathematical Sciences, University of New South Wales, AustralianEditorial Board Defence Force Academy, Canberra ACT 2600,Eric Britton AUSTRALIAManaging Director, EcoPlan International, TheCentre for Technology & Systems Studies, 8/10 rue PublisherJoseph Bara, F-75006 Paris, FRANCE Eco-Logica Ltd., 53 Derwent Road, Lancaster, LA1 3ES, U.K Telephone: +44 (0)1524 63175Professor John Howe E-mail: j.whitelegg@btinternet.comIndependent Transport Consultant, Oxford, U.K 3John WhiteleggAbstracts & Keywords 6Sustainable intermediate transport in West Africa: 8Quality before quantityBryan DorseyAn International Review of The Significance of Rail in Developing 21More Sustainable Urban Transport Systems in Higher Income CitiesJeffrey KenworthyBicycle Ambulances in rural Uganda: 38Analysis of factors influencing its usageCorinna Wallrapp and Heiko FaustSimplified travel demand modelling for developing cities: 47The case of Addis AbabaBinyam Bedelu & Marius de LangenWorld Transport Policy & Practice___________________________________________________ 2Volume 14. Number 2. July 2008
  3. 3. Editorial Introduction pedestrian fatality risk and the impact speed of a car.If a residential area in the full sense ofthe word (a collection of men, women,children, teenagers, parents withbabies, those over the age of 70,healthy, not healthy, worried,over-confident) could select thelevel of probability of death andserious injury would they go fora high risk or a low risk?On June 17th 2008 a group of 20elected representatives inLancashire (UK) encouraged by aself-selected group of localresidents chose the high riskoption. The committee was asked to Figure 1: Pedestrian fatality risk as adetermine a proposal to introduce a function of the impact speed of a car20mph zone in Silverdale in North Source: World report on traffic injuryLancashire. Those opposing the idea preventionexpressed a strong dislike of humps and that would be introduced to slow ention/publications/road_traffic/world_retraffic. They also expressed concerns port/chapter3.pdfabout losing car parking places. Theopposite argument was put about the The decision of elected representatives tobenefits of a 20mph speed limit (even increase the chances of death and injuryone without humps and bumps) but the in this community tells us a great dealcommittee dismissed the whole idea of about transport policy and the enormous20mph in Silverdale and the scheme was barriers we have to overcome to achievescrapped. progress. This has been a recurrent theme of the material in this journal forThe committee of elected representatives 14 years and if anything the last 14chose to impose a higher degree of risk years have seen a decline in intelligenceof death and injury on the streets of and ethics in the road traffic environmentSilverdale. alongside an increase in fine policies and fine language that is never translatedThe evidence on risks and probability into improvements on the ground. Thecould not be clearer and the World Health decline in intelligence is patchy. It isOrganisation has reinforced the particularly severe in the UK where trafficimportance of speed limitation to growth and economic development have20mph/30kph, as a major part of the been elevated into high level theologicalglobal effort to reduce the 3000 deaths objectives with no grasp of alternativeeach day in road crashes. Figure 1 shows scenarios and alternative ways ofa very clear relationship between creating sustainable economies at lower levels of carbon output and lower levelsWorld Transport Policy & Practice___________________________________________________ 3Volume 14. Number 2. July 2008
  4. 4. of demand for transport and fewerdeaths and injuries in road crashes. Don’t change your light bulbs, changeSweden is at the opposite end of the your politicians…spectrum with high level policy objectivesthat have been captured in the road History is full of major shifts in mindsetssafety policy known as “Vision Zero” and policy including:(there will be zero deaths and seriousinjuries in the road traffic environment) The abolition of slaveryand it’s oil-free by 2020 policy. Stopping children working down coal mines and in factoriesSo what do we do next? Providing clean drinking water to working class homes in cities inFirst of all we have to acknowledge that the 1850s and 1860s in the UKthe problem is a mindset problem. Large Introducing the National Healthamounts of science and research and Service (UK)even best practice will not persuade Clean air legislation in the 1950spolicy makers who live in their cars and that got rid of yellow smogenjoy a high carbon/high mobility lifestyle to visualise a different world. The task for all of us in transport is toScience and rationality does not work. identify the virtuous DNA or virus thatSecondly we have to find ways to work brought about these enormous changesfrom the bottom up. Every street and and infect the body politic with the sameevery community should be empowered and do it articulate its own view of a desirablefuture and then have the capacity to Are we up to the task?deliver it. If this means inventing Utopiathen so be it. Thirdly we could all becomepoliticians. I don’t recommend this forhealth reasons but someone once said ina UK political broadcast when speakingabout how to sort out “the environment”:Note:The decision of elected representatives to abandon a 20mph scheme in Silverdale in North Lancashire(UK) can be followed through the minutes of the relevant committee. The committee is “LancashireLocals, Lancaster” and it met at the Midland Hotel in Morecambe on 17th June 2008. The 20mph itemis Agenda item 6 “Emesgate Lane Area, Silverdale Proposed 20mph Zone”. Minutes and agendas forthis committee can be found on: Transport Policy & Practice___________________________________________________ 4Volume 14. Number 2. July 2008
  5. 5. Abstracts & KeywordsSustainable intermediate transport in West Africa: Quality before quantityBryan DorseyThis research provides a brief review of countries for development of the marketthe scant previous studies of non- for quality bicycles, yet some importantmotorised modes of sustainable secondary cities have been thus far,transportation in Sub-Saharan Africa overlooked. This study identifies theurban and peri-urban areas. Particular potential market for ITDP’s quality, yetattention is given to the need to affordable, “California Bike” in Togo’sencourage bicycle transit in West African second largest city, Sokodé. It is arguedsecondary cities. Although the Institute that ITDP’s strategy to focus on qualityfor Transportation and Development rather quantity of bicycles is wellPolicy (ITDP) has made considerable founded, but significant marketingprogress in developing improved bicycle obstacles have yet to be overcome.transportation, much work remains. Intheir outstanding effort at decentralised Keywords: Sustainable transportation incooperation, ITDP has identified key Sub-Saharan Africa, ITDP, bicycle transit.An International Review of The Significance of Rail in Developing MoreSustainable Urban Transport Systems in Higher Income CitiesJeffrey KenworthyThe significance of urban rail systems in of cities and discusses the findings. Thecities, especially in comparison to buses, study finds generally that cities withis a widely discussed topic. This study more strongly rail-oriented publicexamines 60 high-income cities in North transport systems experience a wideAmerica, Australia, Europe and Asia, range of positive impacts at an urbandividing the sample into strong rail, weak system level in all of the above importantrail and no rail cities using three criteria areas. Explanations are offered for theserelated to the significance of rail within observations and the overall results arethe public transport system of each city found to be in line with otherand the rail system’s speed comparative research on this topic thatcompetitiveness with cars. Trams, LRT, has focussed on European and US cities.metro and suburban rail are all The paper suggests that urban railconsidered. It then looks systematically systems are a critical element in buildingat key comparative urban form, public effective multi-modal public transporttransport operational features, transport systems that create a ‘virtuous circle’ ininfrastructure and performance, private public transport and compete moretransport patterns, economic features of successfully with the car.the transport systems and environmentalfactors in each of the three groups of Key words: Urban rail systems, urbancities. It examines whether there are any form, public transport operation,statistically significant differences in infrastructure and performance,these factors between the three groups economic and environmental factors.World Transport Policy & Practice___________________________________________________ 5Volume 14. Number 2. July 2008
  6. 6. Bicycle Ambulances in rural Uganda: Analysis of factors influencing its usageCorinna Wallrapp and Heiko FaustSince 1997 about 400 bicycle circumstances. However, factors wereambulances were distributed to villages defined influencing its usage, such as thein rural Uganda to improve their situation in the villages, the features ofaccessibility of health centres. For the bicycle ambulance, the system ofgeneral understanding, a bicycle distribution and the organisation of theambulance is a bicycle with an attached groups. The only significant influence ontrailer especially to transport seriously the frequency of usage could besick patients to the nearest health observed between the differentcentre. This paper presents the main organisational structures of the bicycleresults of a study carried out to analyse ambulance groups. Furthermore, highthe usage of the bicycle ambulances in distances, costs of repairs and otherUganda. The research was undertaken barriers could be overcome through themainly through guided interviews with positive attitude of group members.receivers and the distributors of bicycleambulances in selected villages. In Keywords: Bicycle ambulance, Uganda,referring to the technology, the bicycle access to health centres, intermediateambulance can be described as an means of transport, rural development,appropriate technology within the Sub-Saharan Africa.Simplified travel demand modelling for developing cities: the case of Addis AbabaBinyam Bedelu & Marius de LangenThis paper presents a simplified travel The key difference between thisdemand model. The model was simplified model and the standarddeveloped as a tool to support long-term current travel demand model is that thestrategic transport system planning, modal-split is dealt with by means of so-specifically for low-income cities with called mobility matrices. These mobilitylimited data availability and the need for matrices show the shares of each modea transparent planning tool that can be of travel per distinct trip-distanceused easily and at a low cost. A study category, with a further segmentation bywas carried out to test the applicability of trip purpose and income of the tripthe model for strategic and sustainable maker. The mobility matrices can betransport planning in Addis Ababa estimated from a household travel survey(Ethiopia). The model consists of of a limited size (1400-2000interlinked spreadsheets with open- respondents).source codes. It requires no specializedlicensed software, and is available free of The test-study divides the Addis Ababacosts, upon request. In case GIS urban area into 35 traffic zones andsoftware is available this can be used to defines an arterial road network of 137facilitate working with maps and for km. Data required for the model wereshowing traffic flows on road network obtained from the municipality, largelymaps, but the model can equally be used from an earlier urban transport study,without. which included a household travelWorld Transport Policy & Practice___________________________________________________ 6Volume 14. Number 2. July 2008
  7. 7. survey. Mobility matrices were estimated manner (correlation coefficient 0.95,from this household travel survey. The RMSE 14%). The test shows that thismodel applies the traditional four travel- simplified model is likely to be ademand forecasting steps: trip valuable, manageable and low-cost toolgeneration, trip distribution, modal split, in support of strategic and sustainableand traffic assignment. transport policy and network planning for low-income cities.In the test, the simplified modelestimates the observed average daily Keywords: simplified travel demandtraffic flows on the main arterial road model, mobility matrix, urban transportnetwork of the city in a very satisfactory planning.World Transport Policy & Practice___________________________________________________ 7Volume 14. Number 2. July 2008
  8. 8. Sustainable intermediate transport in West Africa:Quality before quantityBryan DorseyDepartment of Geography, Weber State University; Ogden, Utah Introduction 22). Some of these same factors,Non-motorised transport (NMT) is central particularly demographics and income,to the issue of sustainable transportation. influence choices in urban transport. JustAmong the more arguably important as secondary African cities are oftenaspects of NMT that are sometimes economically linked to primary cities,overlooked are bicycle transportation there is close interdependence withdevelopment and accompanying policy peripheral rural areas that supplyreform. Given the fact that the majority of agricultural goods, thereby ensuringthe world’s poor do not have access to regional food security. Combined with themotorised transport, it has been well general weakness of rural transportationnoted that this should not be the only systems in Sub-Saharan Africa, it wouldmode considered for development in therefore seem essential to consider peri-Africa, the world’s poorest region urban NMT when discussing urban transit.(Leinbach, 2000; Mozer, 2000; World Indeed, World Bank reports show a clearBank, 2002). Indeed, the last of ten link between NMT and the reduction ofmajor urban NMT strategy elements that poverty in both rural and urban settingsthe World Bank reviews in Cities on the (Starkey et al, 2002; World Bank 2002).Move provides some impetus for thisstudy: “development of small-scale credit Among the more successful decentralised,mechanisms for finance of bicycles in poor cooperative projects focused on bicyclecountries” (World Bank, 2002, 134). transport is the recent work conducted byAlthough the emphasis herein lies not on the Institute for Transportation andinternal credit mechanisms in particular, Development Policy (ITDP). ITDP hasthe market conditions for bicycle sales been active in Africa for over twentyand promotion in two very different, yet years, with particular success in Ghananeighboring countries, Ghana and Togo, (Gauthier, 2005; Gauthier and Hook,are analysed and compared. 2005). Gauthier and Hook (2005) present compelling evidence that the market forThe World Bank has made some studies durable, yet affordable, quality bikes inregarding non-motorised transport in the Africa is primed for development. Basedurban periphery in Sub-Saharan Africa on bicycle marketing studies by ITDP,(Starkey et al, 2002), yet scant academic there may be similar success inresearch has focused on the intermediate developing and marketing bicycle trailerstechnology of bicycles and bike trailers. in selected countries, thereby expandingWorld Bank researchers note that a wide transit mode shares in secondary citiesvariety of factors influence differences in where bicycle use for the movement ofrural transportation: “population density, agricultural goods is significant. Inculture, income, topography, climate, or countries where the transport ofcrops and animals” (Starkey et al, 2002, agricultural goods is critical to nationalWorld Transport Policy & Practice___________________________________________________ 8Volume 14. Number 2. July 2008
  9. 9. economic prosperity and food security, allowed on major roadways that arethe need for alternative mode increasingly dominated by vehicles. Thedevelopment becomes even more acute. cycling mode share in Shanghai forGiven the fact that approximately 42 example, has declined from an estimatedpercent of Togo’s gross domestic product 70 percent in 1990 to only 17 percent incomes from agriculture (Bureau of African 2003 (Summerville, 2005).Affairs, 2005), the country provides anappropriate case study for improving The decrease in ridership within China hasintermediate transport. signalled a shift in production trends. As the Chinese buy fewer bikes, bicycle Bicycle mode shares and transit oriented exports from China are expected to development continue to increase. In fact, theBefore discussing the details of bicycle European Union recently endorsed antransport development in Africa, it may be anti-dumping duty of 48.5 percent on allof assistance to note changes occurring in Chinese built bicycles (begun in July,other world regions. Data analysed by the 2005) (European Union Council, 2005).World Watch Institute (2001) show that The rippling effect of depressed demandbicycle production increased from about in China also led the Economist magazine10 million units in 1950 to 100 million to report that streamlined production inunits in 2000, whereas automobile China has led to a significant decrease inproduction increased from about 8 million the cost of low-end mountain bikes inunits to only 40 million units during the Ghana where the price dropped from $67same time period. Although Holland has in 2001 to $25 in 2003 (author unknown,lead the world in per capita bike sales 2003).with almost 90 bicycles sold per 1000people, China has the greatest total Despite some decline in ridership inbicycle ridership and production in the primary cities such as Shanghai, growingworld (Gauthier and Hook, 2005). While world wide bicycle production and salesonly 20 bikes per 1000 people are sold in noted above give proof to the viability ofChina (op. cit.), approximately 52 percent bike transport. In cities where safe bicycleof the estimated 100 million bikes transportation is encouraged, and whereproduced worldwide in 2000 were bicycle sales are high, it comes as nomanufactured in China (World Watch surprise that a significant percentage ofInstitute, 2001). In terms of urban NMT, the mode share is in bicycles. The city ofChinese cities have the greatest Groningen in the Netherlands, forpercentage of cyclists per total example, has a bicycle mode share oftransportation mode shares. It has been about 50 percent, which is just ahead ofestimated that Tianjin has among the Beijing at 48 percent (Internationalhighest bicycle mode share of any city in Bicycle Fund, 2005). In addition to strongthe world, with an estimated 77 percent. markets for bikes, Holland has some ofClose behind Tianjin is the city of the most well developed bicycleShenyang with a 65 percent bicycle mode transportation master planning andshare (International Bicycle Fund, 2005). supporting transit oriented developmentHowever, bicycle ridership has begun to (TOD) in the world.rapidly decline in some Chinese cities asChina moves toward free enterprise. In Certainly the prevailing transportationsome cities, bicycles are no longer ethics of a community, its level ofWorld Transport Policy & Practice___________________________________________________ 9Volume 14. Number 2. July 2008
  10. 10. economic development, and local cultural community planning involving “walkable”characteristics will influence the and “cycleable,” or “new urbanist” designpopularity of cycling and trailer use. Even elements must be integral to TOD. Wethough these cultural characteristics, as cannot assume that where motorisedwell as socioeconomic conditions vary transportation infrastructure has beendramatically from Europe, to North developed, non-motorised transit willAmerica, to Africa, there is one follow. In fact, some “autocentric”commonality: regardless of dependency planning undermines the safety andon the singly occupied vehicle (SOV), encouragement of non-motorisedmany people are re-evaluating their transportation. The lack of cyclist andtransportation choices, seeking pedestrian safety is endemic not only toalternatives that may be counter to the more economically developed countries,expanding status quo. The European but developing countries in Africa faceCommission (2003) estimates that the similar challenges of modernisation. Fortotal kilometres travelled in the European further reading on best practices forUnion by SOV will increase by 40 percent pedestrian and bicycle planning, seebetween 1995 and 2030. American SOV Litman, Demopolous, Eddy, Fritzel,use has far outpaced bicycle Laidlaw and Maddox (2000) with thetransportation with less than one percent Victoria Transportation Policy Institute inof the mode share occupied by cyclists British Columbia ( Velo(American Public Transportation Mondiale is another key decentralisedAdministration, 2002; US Census Bureau, cooperative organization that provides2000). Nevertheless, in some American valuable information on bicycle mastersecondary cities where traffic conditions planning ( While countriesallow, bicycles are being used as a viable like China and the Netherlands havealternative mode of transport. significant percentages of their population moving by bicycle, how do Sub-SaharanIn American primary and secondary cities African countries compare?where TOD has integrated infrastructuraldevelopment to support bicyclists and Bicycle promotion in Africapedestrians (e.g., Portland, Oregon and While attitudes toward bicycling in AfricanBoulder, Colorado), some of the better cities vary from country to country, manydeveloped bicycle pathway systems in the view bikes as an antiquated form ofworld are emerging. Frustrated by traffic transport, and far prefer motorisedcongestion, degrading air quality, and the vehicles; yet the reality is that bicycleslack of government initiatives to reduce are a sustainable form of transport,carbon dioxide emissions, local particularly for those of low incomecommunity planners and commuters are (Howe, 1997; Mozer, 2000; ITDP, 2005).turning to the bicycle and, or mass Researchers have found that bicycle userstransit. Cyclists are increasingly prevalent in Accra and Ouagadougou preferredin the Front Range communities of motorised two-wheeled vehicles (mopeds,Colorado where TOD includes cycling, scooters, or motorcycles) to bikes.whereas Wasatch Front communities in Nevertheless, some found bicycles moreUtah have many recreational cyclists, but appealing than buses due to low cost,far less development to support bicycle speed, reliability, and more flexiblecommuting. If a revival of bike use is to routing (World Bank, 2002). In an effortoccur that ensures safe movement, to improve non-motorised transportation,World Transport Policy & Practice___________________________________________________ 10Volume 14. Number 2. July 2008
  11. 11. non-government donor agencies and 9.5 bikes/1000 people in 2001 following aindividuals have generously donated used reduction in bicycle import tariffs. As abicycles from the U.S. and Europe to result, the CBC has identified a bicycleAfrica. Although well meaning, donations marketing strategy targeting thoseare problematic in that used bikes are countries with low import duties such asoften of low quality, and are therefore Ghana and Kenya (op. cit.).prone to mechanical failure. When theseforeign bicycles are introduced without While devising a strategy for theproper marketing, there are few services Coalition, ITDP also considered annualdeveloped for repairs and parts made growth rates in gross domestic productavailable to African cyclists. Gauthier and (GDP), as well as the level of existing bikeHook bluntly state the problem: “it was ‘a use and sales in selected Africanrace to the bottom’ in terms of the quality countries. The CBC strategy to target[of bicycles], and this undermined cycling countries with low tariffs seems logical;as a legitimate form of transport” (2005, however, the reliance on GDP growth9). rates as an indicator of success for the development of the California Bike marketHaving experienced the drawbacks to may be constraining. Kenya provides arelying on used bikes, ITDP staff then strong example with an annual growth indecided to take a new approach to the GDP from 2000-2004 of less than 1bicycle commuting problem. ITDP percent, yet as previously noted, bicycletechnical team members worked with sales almost doubled from 2001 to 2002.bicycle manufacturers, Trek and Sram to Indeed, the use of per capita GDPdevelop the California Bike. The California estimates and fluctuating growth rates asBike, a six-speed, low end mountain bike indicators of actual prosperity is being produced by Giant Inc. (ironically, GDP figures mask differences in wealthin Shanghai) and made available to within countries, and do not account forconsumers in Senegal, Ghana, Tanzania the informal sector. Even the casualand South Africa at an average price of observer in Sub-Saharan Africa can attest$100 (Gauthier and Hook, 2005). Once to the significance of the informal sectorITDP identified bike dealers and formed of these economies where muchthe California Bike Coalition (CBC), the commerce goes undocumented andbikes were sold at a rate many would not untaxed.have anticipated. Though the correlation between GDPIn spite of low average per capita incomes growth rate and per capita GDP is weak,in most African countries, consumers in bicycle ownership appears to be closelyGhana, Kenya and Tanzania are buying correlated with per capita income (Hook,bikes at unprecedented rates. Indeed, 1995; Gauthier and Hook, 2005).bicycle sales in Ghana in 2002 were However, bicycle sales and ownership doestimated at almost 29 per thousand not necessarily equate with bicycle use.people, higher than bicycle consumption Hook (1995) showed that the relationshipin China (Gauthier and Hook, 2005). between per capita income and bicycleKenya had the second highest bike sales commuting, or the total number of workper 1000 people in Africa in 2002 (16.4 trips by bike, is not statisticallybikes/1000 population). It is important to significant. Table 1 shows relationshipsnote that sales in Kenya increased from between GDP and the prevalence ofWorld Transport Policy & Practice___________________________________________________ 11Volume 14. Number 2. July 2008
  12. 12. bicycle ownership and sales for selected GDP/capita. Unfortunately no data on countries, yet no definite conclusions can bicycle ownership or sales are currently be drawn for actual bicycle use. Clearly, available for Togo. Regarding Togo’s low there is positive correlation between GDP growth rate, it should be noted that Ghana’s GDP and the number of bikes per Togolese face considerable constraints thousand people (Table 1). One would given political unrest and economic expect to see low bicycle ownership and decline since the early 1990s. sales in Togo given the country’s low Table 1: Economic Indicators and Bicycle Prevalence, 2005 Estimates Population GDP/capita GDP growth # of bikes/ # bikes sold* (million) (PPP$) (% increase) 1000 people* Ghana 20.9 2500 4.3 28.7 588,048 Kenya 33.8 1200 5.0 16.4 517,302 Senegal 11.1 1800 6.1 2.0 20,253 Togo 5.7 1600 2.8 N.A. N.A. Uganda 26.9 1700 9.0 2.4 60,675 GDP = Gross Domestic Product N.A. = Not Available PPP = Purchasing Power Parity Sources: Bureau of African Affairs, US Dept. of State, 2005; *Gauthier and Hook, 2005.Economic factors alone do not explain the Road safety and bicyclists in Sub-Saharanlack of cycling in Africa, where most trips Africa: reality and policyare made on foot, yet the bicycle is The danger of promoting greater relianceunderutilised (Howe, 1997; Mozer, 2000; on bicycles in West Africa, and Sub-World Bank, 2002). Although extensive Saharan Africa in general is inherent in thepedestrian movement and some bicycle risks of death and injury of cyclists in theuse are widespread in most West African region. Although statistics can becities, little attention is given to safe misleading, and in some casesmovement. Indeed some villages and contradictory from one source to another,cores of secondary cities are divided by it is important to consider safety issuesmajor roadway corridors, e.g., Togo’s and related policies that could assist toRoute Nationale runs through the centre of reduce death and injury amongst thistowns where pedestrians and cyclists are vulnerable group. To gain someat risk. Drawing again from World Bank perspective on the safety issues thatpolicy outlined in Cities on the Move, cyclists face in Africa, it may be helpful toemphasis must be placed on consider overall road-related trafficinfrastructural development for “safe dangers, both world-wide and in particularmovement” of non-motorised transport African countries.(World Bank, 2002, 134). Although someAfrican city governments are embracing According to the World HealthTOD with safe bike route development, the Organisation (WHO), 23 percent of allissue is generally neglected. injury related deaths world wide can be attributed to road traffic accidents. In World Transport Policy & Practice___________________________________________________ 12 Volume 14. Number 2. July 2008
  13. 13. 1999, the African region had the highest Figure 1: Global distribution of road deathsroad traffic injury mortality rate in the in 1999world with 28.3 deaths per 100,000population (Jacobs and Thomas, 2000).The WHO forecasts that by 2020, roadcrashes will be the third most commoncause of premature death globally.According to some estimates, morechildren died in Africa in 1998 from roadcrashes than from the HIV/AIDS virus(Dhliwayo, 2007). More recent WHOestimates place road deaths second toAIDS related deaths. Equally compelling isthe statistic that more young adults inAfrica aged 15 to 44 years died from roadaccidents in 1998 than malaria. AlthoughTogo’s roadways appear to be fairlyhazardous, where 132 fatalities per 10,000 Source: U.S. Dept. of Transportation,vehicles were estimated in 1999 (Jacobs Federal Highway Administration (Jacobsand Thomas, 2000), Nigerian roads have a and Thomas, 2000)reputation for being some of the mostdangerous in the West African region. Results from the Africa Road Safety Review by the U.S. Dept. ofIn many Asian, African, and Middle Transportation, Federal HighwayEastern countries between 40 and 50 Administration (supported by the Worldpercent of people killed as a result of a Bank) reveal that the total 35,394 deathsroad accidents are pedestrians. Data from in 42 Sub-Saharan African countries and1995 indicate that 33 percent of all road the individual national totals arecasualties in Kumasi, Ghana involved “significant underestimates of the truepedestrians (Jacobs and Thomas, 2000). totals” due to under-reporting and non-In 2002, it was estimated that between 40 reporting of fatalities in transportationand 45 percent of road user fatalities in related accidents (Jacobs and ThomasGhana were pedestrians (Dhliwayo, 2007) 2000, Section 3.1). Figure 1 shows that(see Figure 2). Although unsubstantiated Sub-Saharan Africa accounts for 10here, it would stand to reason that if the percent of the estimated total 750,000trend in pedestrian fatalities exhibits an road deaths world wide in 1999 (low endoverall increase, bicycle related fatalities of 750,000-880,000 fatalities). Of the 10would likely be increasing in Ghana and percent, a fraction were bicyclists, butneighbouring West African countries as data were not available for all Sub-well. The validity of road safety statistics is Saharan countries, therefore nosubject to question. meaningful estimates of the numbers or percentages of road deaths that were cyclists could be extrapolated. However, some statistics for selected countries can give an indication of how hazardous the roads are for bicyclists in Africa.World Transport Policy & Practice___________________________________________________ 13Volume 14. Number 2. July 2008
  14. 14. In 1998, there were 303 reported bicyclist Figure 2: Percentages of Bicyclist,deaths in Uganda accounting for 19 Pedestrian and Motorist Fatalities inpercent of total road casualties, while Selected Countries (2002)Kenya reported285 bicyclistfatalities (14percent of totalcasualties) forthe same year.For the sameyear in Ghana,an estimated 4percent of roaduser fatalitieswere cyclists(Jacobs andThomas, 2000).As one mightexpect, thosecountries thathave a largerurbanpopulation andmore vehicles/10,000 population tend to Source: World Health Organisation (Krug,have greater risks for bicycle riders. 2007)Similarly, countries where a significantpercentage of the population relies on In an effort to address road safety issues,bicycle transport as the primary form of the World Health Organisation (WHO) andmobility, such as the Netherlands, tend to the United Nations (UN) Economichave higher bicyclist fatalities than Commission for Africa held the Africanpedestrian fatalities. Figure 2 shows the Road Safety Conference during Februarynumber of bicyclist fatalities in selected 2007 in Accra, Ghana. The overallcountries indicating that relative to other conference theme was road safety and thecountries, cyclist fatality rates may not be millennium development goals, one ofabnormally high in West Africa. The high which is to reduce road traffic fatalities inpercentages of pedestrian fatalities in Africa by half by 2015. Among the keyAfrican countries are a clear indication that strategies identified by the WHO towalking is still the most widely used improve safety for cyclists in Africa is themethod of mobility. One might expect to encouragement of helmet use. Althoughfind that less urbanised areas in Africa the use of helmets may receive policywould have lower pedestrian and bicyclist support, the very real economic challengerisks, but again, the research is of distributing affordable helmets forinconclusive on this point. purchase is certain to be an issue in most Sub-Saharan African countries. The use of reflectors and headlights is another safety concern. Traffic calming measures such asWorld Transport Policy & Practice___________________________________________________ 14Volume 14. Number 2. July 2008
  15. 15. the installation of speed bumps or even hence the need to discuss subsistence, orreduced speed limits can have an impact smallholder farming communities and theiron cyclist safety as well. Speed bumps dependence on bicycles and, possibly, bikereduced crashes by 35 percent at a high- trailers. Smallholder farmers in Sub-risk accident site in Ghana (Krug 2007). Saharan Africa experience uniqueAnother factor indirectly related to bicycle challenges in the only world region wheresafety is the development and per capita food production has declinedenforcement of laws that prohibit driving over the past fifteen years. As farmlandwhile under the influence of alcohol or holdings are subdivided into smaller plotsdrugs (currently, South Africa is the only to accommodate new family members orAfrican country with such law land sales, and fallow periods decline inenforcement). closer proximity to village settlements, household members and labourers areIdeally, cities such as Sokodé in Togo will forced to travel greater distances betweensome day designate bicycle routes, but the the village and farm, then village to urbanrecent addition of a traffic light at the market. Transport of goods between themain intersection in the centre of town farm and village is most often made onalong the exceedingly dangerous Route single-track by foot, or occasionally byNationale, or main national highway bicycle (when affordable and available).bisecting the country, marks a step in the Smallholders with farms of four or moreright direction. If distribution and sales of acres that are more diversified in terms ofbicycles in Togo and other Sub-Saharan their production, compared to farms ofcountries are to be promoted, bicycle (and one to two acres, are often morevehicular) safety must be addressed. financially successful (Dorsey, 1999), andWhile the UN and WHO policy goal to may therefore be more likely to have thereduce traffic fatalities by half may be an financial flexibility to purchase a bicycleimportant first step, there must be and, though less likely, an accompanyingassurance that bicycle safety issues will be integral part of any policy framework.An equally important solution to the A similar demographic pattern related tobicycle safety issue is to work through economies of scale occurs among urbanlocal civic groups, village organisations, households. Commuters living in Africanschools, sports and fitness and other primary cities often have better access togroups to educate motorists and cyclists, public transit, typically diesel buses, thanparticularly younger people, regarding those in secondary cities. While ITDProad safety concerns noted above. estimates that less than 2% of commuting trips in major African cities are made by Demographic- and gender-based barriers bicycle, it is possible that the figure forClosely related to economic growth factors secondary cities is as high as 45 percentand market conditions are demographic (Gauthier and Hook, 2005). ITDPdifferences between urban, peri-urban and members also note that many of the bikesrural areas that are essential to take into sold in major cities are actually used inconsideration when evaluating bicycle secondary cities and, or villages (op. cit.).transport systems. Most secondary cities Based on the previous discussion of tariffsin Africa are tightly bound to surrounding on imported bicycles, it would also berural settlements and marketing activity, expected that bikes bought in major citiesWorld Transport Policy & Practice___________________________________________________ 15Volume 14. Number 2. July 2008
  16. 16. where there are low to no tariffs may be tariffs in Togo, ITDP staff members aretaken into adjoining countries with higher working to establish a bicycletariffs, e.g., from Ghana to Togo. manufacturing facility in Senegal. Since Togo is a member of the West AfricanPerhaps one of the more challenging Economic and Monetary Union, or Uniondemographic issues influencing the use of économique et monétaire oust-africainebicycles is that of gender. Women in (UEMOA), bicycles manufactured inmany Sub-Saharan communities often Senegal could then be imported in Togocarry the majority of goods (farm without exceedingly high tariffs. ITDPproducts, tools, fuel wood, etc.) between representatives have been working with anthe village and larger urban markets independent bike dealer in the secondary(Leinbach, 2000; Mozer, 2000). city of Tamale (Ghana’s third largest city)Development of policy and support for to distribute the CA Bike in Ghana.bicycles and bike trailers offers a critically Tamale has a population of about 270,000needed alternative to the burden of people, and is centrally located in a regionmoving loads by foot, yet few African where bicycle use is expected to be highwomen ride bikes. Due to cultural (Figure 4). If the CA Bike were to bedifferences in western Burkina Faso, introduced in a secondary city in Togo, thewomen’s bicycles do not sell as well as in most closely comparable city would beother regions (Sifa, 2001). One aspect of Sokodé, Togo’s second largest city with aculture and gender effecting bicycle use is population of roughly 75,000. Althoughdress. The traditional wrap of cloth, a Tamale is larger than Sokodé, both cities“pagne” or sarong, worn by women in are characterised by large, sprawlingmuch of Africa is not conducive to bicycle residential areas with relatively weakriding. An effort to break down the norm infrastructural development. Connectionsof gender division in cycling was pursued to surrounding rural villages are strong,during the “Tour des Femmes” in Senegal and intensified agricultural production isand the “HIV/AIDS Education Bike Ride” in prevalent in both peri-urban areas.Ghana. Both bike tours, initiated in 2002,were bold attempts by local men andmostly women, US Peace Corpsvolunteers, and various non-governmentorganizations to raise awareness aboutgirl’s education and health issues. Thetours lasted from four days to three weeksand inspired many Africa women to trycycling even where traditionally only mencycle (ITDP, 2005). Bicycles and trailers in Ghana and TogoSo as not to put the cart before the bike, itshould be noted that Togo has little to noinvestment or policy support for bicycletransport, nor has ITDP attempted tointroduce the California Bike (CA Bike) inTogo (Figure 3). Discouraged by highWorld Transport Policy & Practice___________________________________________________ 16Volume 14. Number 2. July 2008
  17. 17. will be cooperation with localized groups such as “Jeunesse et Sports,” or Youth and Sports advocates, as well as local retailers who may wish to sell the CA Bike. Such decentralised cooperation is now viewed as essential to the success of transportation projects. Indeed, the development of bicycle and bike trailer projects in Africa has been largely the result of cooperation between organisations such as ITDP, the Swedish International Development Agency (SIDA) and others.Figure 3: ITDPs California Bike (shown bycaptain of the Ghanaian cycling team) Bicycle trailers are being used in manyPhoto by author world regions to broaden the mode share for transit. Trailer use may be most pragmatically applied in secondary urban areas and the urban periphery where access to transportation is more constrained than in capitol cities such as Accra or Lomé. The development of prototype bicycle trailers has a history of mixed success in Ghana, but the extent to which trailers are being used in Togo is unclear. Development of trailers as bicycle powered “ambulances” has been undertaken to assist in urgent medical care where vehicle access is limited. Bike ambulances produced in Ghana are now being used in Uganda with considerableFigure 4: Locator map for Tamale and success (Gauthier 2005).Sokodé One of the few critiques of bicycle trailerGiven the obvious need to have quality, projects comes from Ghanaian researcher,affordable bikes, with readily available M. Salifu (1994). In the Transportreplacement parts and repairs, before Rehabilitation pilot project in Northernbicycle trailers can be used, the first stage Ghana, supported by the World Bank, theof the bicycle sale and support project bicycle and trailer combination was oftendiscussed below was to survey the market unaffordable. Salifu concludes thatfor quality bikes. Cooperation with the although the trailer was a reasonableTogo Ministries of Transportation and technology, it was inappropriate given theAgriculture will be requisite to the long failures of the pilot project. Several keyterm promotion of intermediate transport findings are worthy of note: the trailerspolicy, but more immediately importantWorld Transport Policy & Practice___________________________________________________ 17Volume 14. Number 2. July 2008
  18. 18. lacked the structural integrity to haul if travel is made by bike, cost ofheavy loads, the trailers were cost bicycles remains an impedimentprohibitive, and as noted above, generally, for subsistence farmers, butwomen do not use bikes (Salifu, 1994). remains affordable for many urbanFurthermore, the trailers did not perform residents;well on village footpaths (Starkey et al, availability of “quality” bicycle2001). As a result, the Togo pilot project repairs, parts and accessories,discussed below will target two different including trailers remainstrailers. First, an affordable, locally problematic in the region;produced, two-wheel trailer will be average income of bicycle ownersdesigned for urban areas, and at a later varies from low to high (widedate, a single-wheel trailer designed for range, but fewer in upper incomeuse on single-track footpaths will be range use bicycles as primary formdeveloped in hopes that diversified of transport).smallholders will be able to afford a light-weight, simply constructed vehicle. Currently, avenues for funding of a shipment of the California Bikes is beingWhere bicycle use is high, and/or markets explored jointly with ITDP and otheroffer promise, it would follow that the use interested organisations. It isand sales of bicycle trailers might also hypothesised that not only urbanaccelerate. Hence, the objectives of the household members, but also diversifiedpilot project are two-fold: to identify urban smallholder farmers with more than 3and urban periphery transport challenges hectares under production could benefitrelated to the movement of goods from affordable bikes and trailers. Asbetween villages, farms and urban prototype two-wheel and single-wheelcenters; and to develop a sustainable bicycle trailers are developed, tested andbicycle trailer construction program for introduced in Central Togo, attention canimproved rural to urban transport. be shifted to larger scale production andAdministration of a brief survey of urban marketing of trailers in Togo, or perhapshousehold members, smallholders, and Senegal where CA Bikes are produced. Bycyclists in the Sokodé urban and peri- surveying women and developing a trailerurban area of Central Togo in June of 2006 that may be converted to a hand cart, theidentified the following: Bicycle Trailer Pilot Project (BTPP) in Togo will take issues of gender into the most common means by which consideration in the testing and household members travel to work introduction of this appropriate technology is still by foot, followed by program. The BTPP will attempt to form a bicycles, but use of motor scooters partnership between researchers, is increasing rapidly; smallholders, the Togo Ministries of average distances and travel times Agriculture and Transportation, ITDP, and for travel to work vary widely perhaps the U.S. Agency for International according to rural versus urban or Development. peri-urban settings; average cost of travel to work Conclusions remains low, while increasing fuel Urban transportation planning is often prices favor cycling; focused on mass transit and roadwayWorld Transport Policy & Practice___________________________________________________ 18Volume 14. Number 2. July 2008
  19. 19. improvements that inadvertently bicycle use in countries such as Togo couldperpetuate dependence on the singly also expand rapidly. If bicycleoccupied vehicle. As transit oriented manufacturing within the West Africandevelopment becomes more widespread, Monetary and Economic Union can bealternative modes of walking and bicycling achieved, perhaps marketing of quality,may be revitalised despite trends in affordable bike trailers will follow.primary cities such as Shanghai where Transportation survey data from Westbicycle ridership has begun to decline. African urban and peri-urban areas, asDirect cooperation between organisations well as the development of trailersuch as the Institute for Transportation prototypes will undoubtedly assist us inand Development (ITDP) and cycling meeting the demand for improvedadvocates in secondary cities, such as movement of both agricultural goods andTamale in north central Ghana and Sokodé central Togo, are critical to providingreliable intermediate transport to those Referenceswho cannot afford private vehicles. While American Public Transportationsome transportation specialists and many Administration. 2002. APTA Transitelite in Sub-Saharan Africa anxiously Ridership Report. www.apta.orgembrace motorised vehicles, some are Author unknown. 2003. Is the wakeningrealizing the value of less polluting, more giant a monster? Economist magazine Feb.dependable, efficient, quality bicycles for 13, 2003commuting to work, school, the market Bureau of African Affairs, US Departmentplace or other destinations. of State. 2005. Togo Profile. www.state.govDevelopment of the market for quality Dhliwayo, M. E. 2007. Road Safetybicycles such as the California Bike may be Development in Africa. Presentation at thekey to promoting non-motorised transport Economic Commission for Africa, Africanin Africa. Currently, those countries with Road Safety Conference in Accra, Ghana,growing GDP per capita are being targeted February 5-7, 2007.for bicycle sales, yet countries plagued by Dorsey, B. 1999. Agriculturalpolitical unrest and faltering economies intensification, diversification, andmay miss these marketing opportunities. commercial production among smallholderIn fact, countries such as Togo may be in coffee growers in central Kenya. Economicgreatest need of bicycles and trailers to Geography 75(2): 178-195.perpetuate food security in the urban European Commission. 2003. Clean Urbanperiphery. As less expensive, but adequate Transport. bicycles from Chinese European Union Council. 2005. Recordsmanufacturers saturate the market outside from the Council of the EU. Official Journalof China, countries with low import tariffs of the European Union no. L 183.will capitalise on intermediate transport. Gauthier, A. & Hook, W. 2005. TappingGhana’s recent per capita bicycle sales the market for quality bicycles in Africa.that exceed those of China indicate that Sustainable Transport 19: 8-11, 30.the African market should not be ignored Gauthier, A. 2005. Scaling up foras it has been for decades. However, given healthcare mobility in Africa. Sustainablethat actual bicycle ridership is not Transport 19: 20-23, 25.necessarily correlated with bike sales, Hook, W. 1995. Economic Importance ofWorld Transport Policy & Practice___________________________________________________ 19Volume 14. Number 2. July 2008
  20. 20. Non-motorised Transportation. Sifa (Société Industrielle du Faso). 2001.Transportation Research Record #1487. Production et distribution de vélos enWashington, DC: Transportation Research milieu rural cas de la Sifa (SociétéBoard, National Research Council. Industrielle du Faso). Paper presented at aHowe, J. 1997. Transport for the Poor or Group Forum National du Burkina seminarPoor Transport? A General Review of Rural on gender and transport, 3-5 April,Transport Policy in Developing Countries Ouagadougou, Burkina Faso.with Emphasis on Low-Income Areas. Summerville, P. 2005. A Bicycle Built forGeneva: International Labour Office. ISBN Two. Report on Canadian International92-2-110473-7. Trade Tribunal.Institute for Transportation and Policy (ITDP). 2005. Non- Starkey, P., Ellis, S., Hine, J., & Ternell, A.motorised transport in Africa. 2002. Improving Rural Mobility: for Developing Motorised and Non-International Bicycle Fund. 2005. Bicycle motorised Transport in Rural Areas. WorldStatistics. Bank Technical Paper No. 525.Jacobs, G. and A. Aeron-Thomas. 2000. US Census Bureau. 2000. Census.Africa Road Safety Review Final Report. www.census.govPublished by the U.S. Department of World Bank. 2002. Cities on the Move: ATransportation / Federal Highway World Bank Urban Transport StrategyAdministration. Review. Washington D.C.: World BankKrug, E. 2007. Preventing Road Traffic Publications.Injuries. World Health Organization, World Watch Institute. 2001. State of theDepartment of Injuries and Violence World 2001. World Watch Institute:Prevention. Washington DC.Leinbach, T. R. 2000. Mobility indevelopment context: changing Author contact informationperspectives, new interpretations, and the Bryan Dorsey, Associate Professor ofreal issues. Journal of Transport Geography, Department of Geography,Geography 8: 1-9. Weber State University, 1401 UniversityLitman, T., Blair, R., Demopolous, W., Circle, Ogden, UT 84408-1401 USAEddy, N., Fritzel, A., Laidlaw, D., & Tel.: +1 801 626 6944Maddox, H. 2000. Pedestrian and Bicycle Fax.: +1 801 626 7130Planning: A Guide to Best Practices. E-mail: BDorsey@weber.eduVictoria Transportation Policy Institute,Victoria BC, Canada. <>. Bryan Dorsey holds a Ph.D. in geographyMozer, D. 2000. Transportation, Bicycles from the University of Colorado-Boulderand Development in Africa. International (1996). He specialises in environment andBicycle Fund: Seattle, WA. society interactions, specifically, land useSalifu, M. 1994. The cycle trailer in Ghana: planning and rural development. HeA reasonable but inappropriate coordinates the interdisciplinary Urban andtechnology. African Technology Forum Regional Planning Program at Weber State7(3): 37-40. University in Ogden, Utah.World Transport Policy & Practice___________________________________________________ 20Volume 14. Number 2. July 2008
  21. 21. An International Review of The Significance of Rail inDeveloping More Sustainable Urban Transport Systems inHigher Income CitiesJeff Kenworthy Introduction income cities where there appears to beWith growing attention being paid to less financial capacity to afford the extrasustainability issues, most cities are capital costs of rail systems (Badamimaking efforts to restrain the growth in 2005). Others argue that rail systems inautomobile dependence. Many avenues general have greater intrinsic passengerare available to cities in the pursuit of this appeal and that they compete better withgoal. Physical planning policies can aim to cars (Newman and Kenworthy 1991).make development more compact with Hass-Klau et al (2003) have mademixed land uses, thus building in less extensive studies of European cities withauto-dependence at the start (Cervero and without light rail systems and have1998, Newman and Kenworthy 1999a). concluded strongly that those cities thatEconomic policies towards the automobile develop LRT systems consistentlycan seek to minimise car ownership and outperform, across many criteria, thoseuse through higher prices that perhaps cities that attempt to run their publicbetter reflect the car’s true social cost, as transport systems only using buses.has happened in Singapore for example(Ang 1990, 1993). Amongst these efforts, Likewise, a report from Litman (2004) ofthere is a general recognition that the role the Victoria Transport Policy Instituteof public transport needs to be enhanced, called ‘Rail Transit In America:along with its companion modes, walking Comprehensive Evaluation of Benefits’and cycling, and the latter for reasons of evaluates rail’s benefits in terms ofhealth, not just transport (Pucher 2002, transport system performance in 130 U.S.Pucher and Dijkstra 2003). cities. It finds that cities with large, well- established rail systems have a wideWithin this general recognition that public range of system-wide benefits relative totransport can play a much greater role in those that have no urban rail (see later).most cities, arguments exist about themost appropriate modes to install to It is further argued that rail stations areachieve enhanced public transport use natural sites for dense residential andand other desirable qualities, such as mixed-use development which can help tocost-effectiveness, integration with land reshape the city into a more sustainableuses and ability to shift people out of urban form (Cervero 1995, Kenworthycars. In particular, there is considerable 1995, Cervero 1998, Newman anddebate about buses versus rail (e.g. Kenworthy 1999a, Hass-Klau, et al 2004).Henry 1989, Pickrell 1990). Some arguethat rail is very capital intensive and that In order to contribute a morewell-conceived bus systems can achieve international perspective on the issue ofthe same results at a fraction of the cost the merits of rail in cities, this paper will(Bonsall 1985, Kain and Liu 1999). This explore a wide range of transport,argument is strongly used in lower economic and environmental features inWorld Transport Policy & Practice___________________________________________________ 21Volume 14. Number 2. July 2008
  22. 22. 60 higher income metropolitan areas that Rail in this study is defined as thehave strong urban rail systems compared combined modes of trams, LRT, metroto those that have weak rail systems or and suburban rail. The strong rail citiesno rail systems at all. The term ‘cities’ in (SRCs) have been defined using threerelation to data in this paper refers criteria:generally to whole metropolitan regions,not the smaller administrative unit at the • To be classed as a SRC, cities wereheart of the region, which often bears the required to have more than 50% ofsame name (e.g. City of New York etc.). their total public transport task (publicHigher income cities were defined for the transport passenger travel measuredpurposes of this research as those with as passenger kilometres) on rail, theannual GDPs per capita of $US10 000 or weak rail cities (WRCs) have railmore (i.e. it embraced those cities that systems that account for less thanare generally perceived as being part of 50% of their total public transportthe ‘developed world’, as opposed to cities passenger kilometres and no rail citiesthat are clearly in developing nations). It (NRCs) have either no rail systems orwill examine the evidence for whether rail systems that are so negligible inurban rail in a city’s public transport terms of extent and usage as to besystem appears to make any observable, tantamount to having no rail. Cities instatistically significant difference to the table 1 that fulfill this last criterion arebroad patterns of transport and related Tel Aviv, Denver, Los Angeles andfactors at a metropolitan scale. Taipei where rail usage in 1995 is negligible due to the existence of only Method very small rail systems.This paper draws upon the MillenniumCities Database for Sustainable Transport • SRCs also had to have no less thandeveloped by Kenworthy and Laube 40% of total public transport(2001), which in turn built on and boardings by rail modes.extended earlier work by Newman andKenworthy (1989) and Kenworthy and • Finally, for classification as a SRC,Laube (1999). Some details about items cities were required to have railin the Millennium database, including systems that are competitive with thedefinitions of indicators and car in speed terms. The overallmethodologies behind the research can be average speed of all rail modes infound in Kenworthy and Laube (1999), each city was calculated, weighted byKenworthy and Laube et al (1999) and passenger hours, and expressed as aNewman and Kenworthy (1999a). More ratio of the average road trafficspecific details about other variables in speed. Only those cities with anthe Millennium database are available average rail speed that was equal tofrom the author. or greater than 0.90 of the road speed were classed as SRCs. MostThe list of 24 ‘strong rail’, 28 ‘weak rail’ SRCs exceeded this criterion, often byand 8 ‘no rail’ cities involved in the a considerable margin.research in this paper, together with their1995/6 populations, appears in table 1.World Transport Policy & Practice___________________________________________________ 22Volume 14. Number 2. July 2008
  23. 23. STRONG POPULATION WEAK RAIL POPULATION NO POPULATIONRAIL (1995/6) CITIES (1995/6) RAIL (1995/6)CITIES CITIESWashington 3,739,330 Calgary 767,059 Ottawa 972,456New York 19,227,361 Atlanta 2,897,178 Denver 1,984,578Brisbane 1,488,883 Chicago 7,523,328 Houston 3,918,061Sydney 3,741,290 S. Francisco 3,837,896 L. Angeles 9,077,853Wellington 366,411 Montreal 3,224,130 Phoenix 2,526,113Barcelona 2,780,342 San Diego 2,626,714 Bologna 448,744Berlin 3,471,418 Toronto 4,628,883 Taipei 5,960,673Berne 295,837 Vancouver 1,898,687 Tel Aviv 2,458,155Brussels 948,122 Melbourne 3,138,147Frankfurt 653,241 Perth 1,244,320Hamburg 1,707,901 Amsterdam 831,499London 7,007,100 Athens 3,464,866Madrid 5,181,659 Copenhagen 1,739,458Munich 1,324,208 Dusseldorf 571,064Oslo 917,852 Graz 240,066Paris 11,004,254 Helsinki 891,056Ruhr 7,356,500 Lyon 1,152,259Stockholm 1,725,756 Marseille 798,430Stuttgart 585,604 Nantes 534,000Vienna 1,592,596 Rome 2,654,187Zürich 785,655 Geneva 399,081Osaka 16,828,737 Glasgow 2,177,400Sapporo 1,757,025 Newcastle 1,131,000Tokyo 32,342,698 Manchester 2,578,300 Milan 2,460,000 Hong Kong 6,311,000 Singapore 2,986,500 Seoul 20,576,272Table 1: Strong rail, weak rail and no rail cities in the studyThe Millennium Cities Database contains are clearly located in ‘developing nations’.complete data for 84 metropolitan areas However, Eastern European cities such asworldwide, of which 24 can be considered Prague in 1995 had low GDPs per capitaas lower income (i.e. with a GDP per but cannot be considered as ‘developingcapita of less than $US10 000 per cities’, whilst South African cities presentannum). All of these cities, apart from a starkly mixed picture whose GDPs perthose in Eastern Europe and South Africa, capita are low because of the hugeWorld Transport Policy & Practice___________________________________________________ 23Volume 14. Number 2. July 2008
  24. 24. majority poorer populations. Attempts these are shown in the last column ofwere made to conduct the analysis of the each table, with significant results markedrole of urban rail in all these lower income with an asterisk*.cities as well, but by the criteria justdescribed, only three of these 24 cities Urban form and GDPcould be considered as having strong rail Table 2 shows the differences in urbansystems. A larger sample of lower income form between the groups of cities, ascities worldwide for which comprehensive reflected by density and centralisation ofand reliable data were available would jobs, as well as economic differences inyield more SRCs so that the analysis the cities expressed through the GDP percould be meaningfully conducted, but this capita of the urban regions.was not possible for this paper. The focusof this paper is therefore on cities in the Although urban densities are‘developed world’, as shown in table 1 systematically higher in the cities with railwhose GDPs per capita range from $US10 and lowest in the no rail cities, the result305 up to $US54 692 per annum. is not statistically significant. Since density is a powerful determinant ofTables 2 to 7 systematically examine how transport patterns, especially private carthe strong rail, weak rail and no rail cities use (e.g. Kenworthy and Laube et alperform on a wide range of factors using 1999, Newman and Kenworthy 1999), it1995/6 data. The values for each variable is useful for the purpose of this researchin the tables are the medians for the that differences in densities between thethree groups of cities, since the data in three groups of cities are not significant.each case are generally skewed On the other hand centralisation of thedistributions where the median value is a city, as measured by the proportion ofbetter representation than the mean. In metropolitan jobs in the CBD, is clearlyorder to test the statistical significance of highest in the SRCs (18.2%) and lowestthe difference amongst the medians, the in the NRCs (10.2%) and the differencesnonparametric Kruskal-Wallis test from are statistically significant. This might beSPSS was used. The Kruskal-Wallis test is expected, given the link between radialused for simultaneously testing multiple urban rail systems and the developmentcases and eliminates the increased of strong city centres, through rail’sprobability of significant results that capacity to deliver large numbers ofoccurs where, in this case, three separate people into small areas (Thomson 1978).pair-wise tests could have beenundertaken for each variable. Since the Amongst these high-income cities, thesamples are relatively small and the SRCs are clearly wealthier than both otherasymptotic significance value is not groups of cities in a statistically significantaccurate enough, the Monte Carlo way, and as the next section shows, theysimulation of the Kruskal-Wallis test was are also more public transport-oriented.employed using 100 000 iterations, which This undermines the idea that citiesgives a 99% confidence level for the p- inevitably become more auto-dependentvalue (significance of the difference in the and move inexorably away from publicmedians for each variable). P-values of transport as they become wealthier. In0.05 or less (95% confidence level) were this significant international sample ofconsidered statistically significant and higher income cities, the reverse wouldWorld Transport Policy & Practice___________________________________________________ 24Volume 14. Number 2. July 2008
  25. 25. appear to be true. We have argued The additional relevance of some of theseelsewhere that excessive automobile data to the arguments made in this paperdependence drains the economy of cities will become more apparent in laterand there is some tacit support for this in discussions.the results in table 2 (e.g. see Kenworthyet al 1997). Strong Weak Rail No p-Urban form and GDP Rail Cities Cities Rail Cities valueUrban density (persons per ha) 47.6 36.6 27.7 0.453Job density (jobs per ha) 27.4 16.1 13.4 0.293Proportion of jobs in the CBD (%) 18.2% 14.6% 10.2% 0.008*Metropolitan GDP per capita (US$1995) $35,747 $26,151 $27,247 0.014*Table 2: Median values and statistical significance for urban form and GDP in strong, weakand no rail cities (1995) Operational performance of public three to four times higher in the SRCs transport than in the NRCs, depending on theTable 3 examines differences in public measurement used. This is especiallytransport operational performance interesting in the light of the urban(service and use). The first item reveals a density data in table 2, which show thatkey basis for the formation of the groups there is no statistically significantof cities. It shows how the SRCs clearly difference in the median population andrely much more heavily on rail systems to job densities between the three groups ofdeliver public transport mobility, with a cities.median value of 74% of passengerkilometers on rail modes, compared to Interestingly, however, despite these big43% and 0.4% respectively for the other differences in the supply and use of publictwo groups of cities. transport, per capita use of public transport energy is only some 1.6 timesLooking more broadly at the public higher in the SRCs than in the NRCs,transport operational measures, table 3 though the difference amongst theshows that the supply of public transport medians on this factor is statisticallyservice rises systematically from NRCs to significant. This demonstrates theSRCs for both vehicle and seat kilometres intrinsically high energy efficiency ofof service per capita. SRCs have over four public transport systems in providingtimes higher seat kilometres of service mobility (i.e. service and use are fourper capita than the NRCs. In usage, there times higher in the SRCs compared to theis the same ascending pattern from NRCs NRCs, while energy use to run theto SRCs for boardings, passenger systems is only 1.6 times higher).kilometres and the proportion of totalmotorised passenger kilometres on publictransport. Public transport use is someWorld Transport Policy & Practice___________________________________________________ 25Volume 14. Number 2. July 2008