Cedelft.eu

Oplossingen voor
milieu, economie
en technologie
Although most of us are aware of the benefits of cycling many people arenot. When we want to develop and implement a bicycle plan it is useful whenwe can show them the economic benefits of cycling. It will enhance thechance that we find support and financial contributions for our plan. This pa-per is meant to give you some support on the methodology when setting upa cost-benefit analysis.
In the highly developed western economies cycling is often considered to bean odd and old-fashioned mode of transport with low status. With the excep-tion of a few countries that are known for their relatively high share of cycling(in Europe: Denmark, Germany, the Netherlands) the cyclist is a rare spe-cies on the road even when bicycle possession rates are high. Recently, theattitude towards cycling has started to change somewhat. As the negativeeffects of motorised transport become increasingly visible (poor air quality,congestion, noise nuisance) the benefits of the alternatives also becomemore prominent and the benefits of the bicycle are rediscovered. Especiallyin the urban areas cycling is considered to be a good alternative for the car.
In relatively rich developing countries cycling generally has a low status andis considered to be a transport mode for the poor, or inappropriate forwomen and people of high status. In this respect there is a danger that thetransport system develops along the same lines as it has done in the west-ern countries and that the benefits of cycling, but also of walking, as a modeof transport are not used to the fullest. An example of this is Bangkok, Thai-land that suffers from congestion problems and air pollution as a result of atransport investment policy that was focussed on the car.
In poor developing countries for many people cycling has another statusthan in the Western world: it would be an aide for their economic prosperity ifthey could afford one. Here, the bicycle is an alternative for walking andmatters of environmental pollution or congestion are less important.
It is this contrast that we return to at the end of this paper to illustrate theeconomic benefits of cycling. In both situations (as an alternative to the carand for walking) the bicycle offers economic benefits to society, althoughthese benefits are different in kind. This paper will try to give handles for asystematic analysis of the benefits (and costs) of cycling.
When we want to make a cost benefit analysis of a bicycle project the firststep we need to take is to estimate the effects on mobility.
This contains: • estimating the normal traffic, i.e. the traffic that would have taken place on the facility in any case, even without the new investment; • estimating the amount of newly generated traffic; • estimating the amount of diverted traffic between transport modes.
It is important that these three categories are determined and compared to aproper reference scenario. In order to appreciate the exact effects of theproject this reference scenario should also take into account the expectedmobility developments were the project not been carried out at all.
4.641.1/The benefits of cycling and how to assess them For the objective of this paper we assume that the transport effects of theproject are known, i.e. the changes in bicycle use, car use, walking et ceteraare ex-ante determined. We can than continue with the second step: estab-lishing the costs and benefits arising from these changes in mobility.
We approach the cost benefit analysis from a socio-economic point of view:we consider the costs and benefits of cycling to society as a whole. We dothis because we assume that the main use of this paper will be to providesome support setting up a cost benefit analysis for transport (infrastructure)projects that are meant for the public benefit. It is such a project that wehave in the back of our mind when we write this paper.
In this section we briefly discuss the various costs and benefits of cycling. Atthe end of this section we will focus on the costs and benefits from two per-spectives and identify the most important costs and benefits that one shoulddetermine in order to estimate the benefits of cycling:1 From the perspective of the bicycle as an alternative for walking (most likely in poor developing countries), and 2 The bicycle as an alternative for the car (most common situation in western countries and richer developing countries).
We will see that from both perspectives cycling offers significant benefits toboth the private users and society as a whole. In the following sections wewill further address these benefits and costs.
Under private costs and benefits we consider the costs and benefits that theprivate user experiences. These comprise:• costs of purchasing, driving and maintaining a bicycle (or car); • costs of travel time and increased economic opportunities; • benefits of an improved health condition.
Hereafter we address these issues in more detail.
&RVWV RI SXUFKDVLQJ GULYLQJ DQG PDLQWDLQLQJ D ELF\FOH RU FDUIn general the costs of purchasing a bicycle or car are relatively easy to de-termine by comparing market prices. As a rough estimate one can assumethat the purchasing and operational costs (maintenance and fuel) per kilo-metre of a bicycle are about 10 times cheaper than that of a car, but largedeviations exist (see the table below).
4.641.1/The benefits of cycling and how to assess them Purchase costs and average yearly operational costs (use and maintenance)in relation to income per capita in 1992 (in US$) 6s‡r…)ÃWIBÃ!Ãh††ˆ€vtÃ&$Ã$Ãhqà Ãxvy‚€r‡…r†Ãƒr…Ã’rh…Ãs‚…Ãivp’pyrÀ‚‡‚…p’pyrhqÃph…Ã…r†ƒrp‡v‰ry’ The private costs of walking are very low. Probably only the wear and tear ofshoes could be considered in a cost-benefit analysis. For the case of Moro-goro, Tanzania, the travel costs of the various modes are summarised be-low.
Travel costs per kilometre in US$ for the year 2000 When we want to compare the purchasing and operational costs it is impor-tant that they are calculated on the same basis. Most commonly, these costsare expressed as yearly costs by using depreciation functions that calculatethe average yearly costs on the basis of purchase price, residual value, life-time and interest rate.
&RVWV RI WUDIILF DFFLGHQWVRoad accident rates in developing countries are typically an order of magni-tude higher than that in industrialised countries (Adler, 1987). This is due toa number of reasons among which:• poor road design and pavement quality; • motorised vehicles and slow vehicles, pedestrians and animals are not • poor condition and overloading of vehicles.
Traffic accidents generally impose costs on both the involved victims and thesociety. The costs for the user comprise primarily the material damage andmedical costs. These costs can in principle be derived from the costs of in-surance. In the Netherlands the costs of accident insurance are on averageabout 500 Euro per year. Insurance for damage to bicycles is not common,as the owner can generally raise these costs with relative ease. Although indeveloping countries many vehicles are not covered by insurance we canderive the 4.641.1/The benefits of cycling and how to assess them The costs of medical insurance are generally not dependent on the transportmode that is used and is therefore not important for our cost-benefit analy-sis.
&RVW RI WUDYHO WLPHIn the industrialised world the phrase “time is money” certainly applies intransport. As people have a remarkable similar daily time budget for travel-ling (about 1 hour), the faster one can travel the more destinations you canreach (see below).
In order to establish the costs and benefits of (saved) travel time we need tostart with some theoretical background on their determination. If a bicycleproject leads to changes in travel time, generally three categories of userswill benefit (as mentioned in chapter 2):1 The people that continue to use the bicycle.
The people that switch from another mode to the bicycle.
Suppose the building of a free-lying dedicated bicycle path leads to reduc-tions in travel time for cyclists on that route. The people that continue to usethe route will experience the full benefits of the travel time reduction (whichwe can express in financial terms, as we will see later).
The new users however, will experience a lower benefit: The first newcomerthat in the situation without the bicycle path was almost willing to make thetrip by bicycle will experience benefits that are almost as large as that for thecyclists who continue to use the route. On the other end there will be newusers that are only just willing (and thus gaining just more than zero benefits)to use the new bicycle path. On average the benefits for these new usersare therefore half that of the existing users. This principle is called the rule ofhalf in cost-benefit analyses.
Finally, the people that switch from another mode (e.g. the car) to the bicyclewill gain at least the benefits of that of the people that continue to use thecar. But, because they switch from the car to the bicycle there must be extrabenefits or they would not do that. On average these extra benefits equalhalf of the difference between the level of benefits of the bicycle and that ofthe car in the situation with the new bicycle path1.
In summary, the total travel time benefits % of this project equal D)(3D−3D)+ 0.5 (4 −4 3D = costs of travel time for cycling 3E = costs of travel time for carWithout (0) and with (1) the new bicycle path.
If the new bicycle path does not influence the travel by car, the benefits of the car with or without the bicycle path will be equal. If the new path e.g. alleviates the congestion on the road there will be benefits for the car as well. The extra benefits for those who switch from the car the bicycle will then be somewhat smaller. If, on the other hand the bicycle path is combined with a narrowing or closure of the road the benefits of the car will decrease. The extra benefits for those who switch from the car the bicycle will then be larger.
4.641.1/The benefits of cycling and how to assess them In the example above the demand and costs can be calculated on the basisof the number of people or vehicle-kilometres etc. It is of course importantthat all costs and benefits share the same basis so that they can be addedand compared. For the purpose of this paper we will not worry about the ba-sis when we identify the various costs and benefits of cycling and indicatehow we can determine and value them.
The valuation of travel time depends on the motive of travel. Business tripsare generally valued highest and the valuation is mostly derived from wagerates. The travel time value of commuting trips and leisure or holiday trips isgenerally lower. An example of these differences for the value of time (VOT)is given in the table below. A value of time for cycling of 4 euro (9 Dutchguilders) is mentioned for a case study for Amsterdam in VNG, 2000. Thisvalue is in line with that for leisure trips with motorised modes.
Values of time (in euro of 1998 per hour) for Europe, based on state of theart studies Values of time in the developing countries are significantly lower. An exam-ple is the case study for Morogoro, Tanzania from VNG, 2000. In this casethe time costs per hour are estimated to be somewhere between 0.17 US$and 0.68 US$2. However, with a purchasing power parity per capita that isabout 40 times lower than that of the EU (CIA, 2003) the valuations of timein real terms are comparable.
As a last remark we note that the value of time can be very dependent onthe local circumstances. In many poor developing countries the rate of un-employment is high. Therefore wages do not properly reflect the economiccosts of labour. A derivation of the value of time from wage rates shouldtherefore be carried out with care (Adler, 1987).
As one of the benefits of a faster mode of transport is that it makes moredestinations accessible, it is likely that the economic opportunities of theuser will increase. On average, a person that walks will cover about 4 kilo-metres per hour. In rural areas of development countries this will allow himto get a round his own village on maybe reach some neighbouring settle-ments. If however such a person would use a bicycle his radius would in-crease to about 10 kilometres. In that case he will be able to reach manymore destinations and economic activities. In our example he would be ableto reach the market in a regional town in 1.5 hours and sell his products in-stead of in over 4 hours (and probably not making the journey at all).
In principle the valuation of time is independent on the transport mode, as it is the one unique value of time that can make a person choose between one mode or the other.
4.641.1/The benefits of cycling and how to assess them Two examples of these increased economic opportunities are given in VNG,2000:1 In Midrand, South Africa bicycles are used to collect used paper for re- cycling in a paper mill. This raises an income for the collector that is 3times that of his colleagues who collect and transport paper by foot.
2 Near Accra, Ghana women sell their oranges for very low prices at the local market because local supply exceeds demand. However, only 15km down the road in Accra supply is relatively low and prices are high.
Had these women possessed bicycles they could have offered their or-anges in Accra and raise their income substantially.
There is great discussion about these increased economic benefits. Manyargue that they represent no new benefits other than those of travel timegains and should therefore not be added to the other benefits.
In addition to an increase in economic opportunities, new economic activitiescan arise. A flourishing bicycle culture will bring about bicycle rental, saleand repair shops. When attributing these benefits one should be very carefuland take into account the economic development that would have takenplace without the bicycle project as well as the economic activities that arelost in other sectors as a consequence of the project.
%HQHILWV RI DQ LPSURYHG KHDOWK FRQGLWLRQIn the developed countries cycling is often seen as a means of physical ex-ercise. Where many people suffer from health conditions due to lack ofphysical labour and a lavish diet, cycling can indeed contribute to health im-provements. However, in most developing countries conditions are different.
The benefits of extra physical exercise from cycling are likely to be negligi-ble.
After we have addressed the costs for the private user we will here discussthe costs that are not carried directly by the users but are covered by thegovernment or by society as a whole. These costs comprise: • costs of infrastructure building and maintenance; With the exception of the costs of infrastructure all the above costs can bereferred to as external costs of transport. External costs are the external ef-fects of mobility that are judged to be financially negative. External effectsare taken to mean those which the originator does not taken into considera-tion in his or her decision on mobility (CE, 1999). Hereafter we addressthese issues in more detail. At the end we shortly discuss the benefits ofwelfare sharing for which the bicycle can act as a facilitator.
&RVWV RI LQIUDVWUXFWXUH EXLOGLQJ DQG PDLQWHQDQFHThe costs for construction and maintenance of infrastructure are relativelyeasy to assess. If we assume the same high quality of asphalt to be used aspavement the costs of a square metre of bicycle lane does not differ sub-stantially from that of a road for cars. And although wear of the pavement ismuch smaller for bicycle lanes as for car lanes, cyclists are more sensitive tothe quality of the pavement than motorists, meaning more regular mainte-nance.
4.641.1/The benefits of cycling and how to assess them However, the cost of bicycle infrastructure will be lower per user since itscapacity per metre is larger: bicycles need less space, so the lane can ac-commodate more cyclists per metre. In the Netherlands an urban road (with1 lane of 3-4 metres wide for each direction) has a capacity of 2000 cars perhour at most for each direction. With an average occupancy rate of 1.6 thatequals 3200 persons per hour. A one-way bicycle lane of 2.5 metres widecan accommodate 6500 persons per hour, which is double that of the car.
In reality the costs of infrastructure will greatly depend on its design and theaddition of facilities as tunnels, bridges and traffic lights. As an illustration ofthe cost of bicycle infrastructure serve that of Morogoro, Tanzania where onaverage the costs amount 35,000 US$ per km and Bogotá, Colombia wherecosts amount 360,000 US$ per km. When we know that the ratio of the pur-chasing power parity per capita equals 10, the costs become similar3.
(PLVVLRQV DQG QRLVH QXLVDQFHIn contrast to the car a bicycle does not produce noise, polluting emissionsor contribute to global warming. As an illustration, the emissions of a carsand buses in the Netherlands are given in the table below.
Emission factors (in gram per km) for urban and rural trips of the averagevehicle built in 1993 and 2002 The polluting emissions of vehicles in developing countries are typically andorder of magnitude larger because they are older (newer vehicles are gen-erally cleaner due to ongoing tighter emission standards in the industrialisedcountries, see the table) and their condition is worse due to lack of propermaintenance. As a result the air quality in many cities in developing coun-tries is poor.
Monetary valuation of environmental effects is relatively well establishedduring the last decade and will be based ideally on an estimate of the dam-age (reduction in welfare) caused by the environmental effect. This methodis to be preferred for environmental problems that are closely related toquality of life, because a good estimate of the loss of welfare can be made inthis type of problem.
In the case of noise nuisance, a satisfactory estimate of the loss can be ef-fected by means of the loss of house values as a result of noise. Allowances This is a rough approach as the design of the bicycle lanes and the facilities that are in- cluded in its costs per km are not exactly the same.
4.641.1/The benefits of cycling and how to assess them must be made in this case for the fact that such a loss represents a totalsum, which must be subsequently allocated to the various types of vehicle.
However, for more complex environmental problems and those related tosustainability, it is often impossible to make an appropriate damage cost es-timate. A good example is CO2 emissions in connection with climate change;moreover direct valuation of damages soon proves to be inadequate foracidification, too. Society has nevertheless established implicit damagevaluation for these emissions, usually on the basis of the precautionary prin-ciple, having decided that it wants to achieve certain objectives for reduc-tions. By analysing the costs of measures that are necessary to achieve ob-jectives, it can be ascertained how much society implicitly wants to reducethe environmental effect. In this case the evaluation is based on what areknown as prevention costs. Prevention costs represent a kind of social will-ingness to pay, in which all uncertainties regarding the actual environmentaleffect are included. Valuation of environmental effects relating tosustainability, based on the prevention cost method, also results in a muchsmaller spread than evaluation based on estimation of direct damage. Be-cause the prevention cost method is based on marginal prevention costs(the costs of the most expensive measures required to meet the objectives),the costs of prevention represent the marginal costs of emissions. Since theprevention costs are expressed in ε per kg of emissions and emission fac-tors in gram per km are known, the external costs of emissions can be cal-culated directly (‘bottom up’). This is in contrast with for instance to the costsof infrastructure and noise nuisance where an indirect method (‘top down’with allocation factors) must be applied.
From the valuation efforts a set of so called shadow prices is established forthe major pollutants (see the table below).
Environmental effects and their financial valuation (in euros of 2001 per kg)for the EU These shadow prices however are not readily applicable to developingcountries as the valuation depends on the national welfare level. A good ap-proximation of the shadow prices in a particular country is therefore to cor-rect the prices in the table with ratio of the purchasing power parity (PPP) of 4.641.1/The benefits of cycling and how to assess them that country with that of the EU. An exemption is in place for the shadowprice of CO2. As this is a pollutant with global effects its shadow price is de-rived from the expected level when a global CO2-emissions trade schemewere in place.
7UDIILF DFFLGHQWVThe external costs of road traffic accidents can be divided into four catego-ries:1 Processing and prevention costs: these are police, fire brigade, court, insurer, investigation, information and congestion costs (congestion dueto accidents).
2 The costs of medical care, convalescence and replacement, if applica- ble: the proportion not covered by insurance is external.
The costs of productivity losses due to accident victims being off work.
The costs of human suffering. This last item is very significant in the totalexternal effects of road traffic accidents.
The allocation of external costs to vehicle categories in one-party accidents(e.g. a car driving into a tree) is simple. With these, 100% of external costsare allocated to the party concerned.
For multi-party accidents allocation is more complex. In their driving behav-iour, every road user attends to his or her own safety. A good measure ofthe external costs of road accidents is therefore the extent to which the vari-ous road users expose others to danger.
Cyclists are vulnerable but harmless. For motorised vehicles the heavier andthe faster a vehicle, the more likely it is that they impose risk to others. Thisintrinsic risk is reflected in road accident statistics. In a crash between a cy-clist and a lorry it is less likely for the cyclist to survive than it is for the otherparty. In a certain sense, then, the mere presence of the lorry is also re-sponsible for the occurrence and seriousness of the accident, even if thedrivers of these vehicles are not to blame for the cause of the accident. Theso-called ‘conflict tables’ used in accident statistics show how casualties inmulti-party accidents are divided over the various means of transport.
As a consequence of allocating responsibility for multi-party accidents on thebasis of LQWULQVLF ULVN only an insignificant part of the external costs of roadtraffic accidents is to be allocated to cyclists (and pedestrians). Moreover, onthe basis of study results it is believed that when the number of pedestriansand cyclists increases the risk for all vulnerable road users will fall (HLG,1999). The expert advisors to the high level group on infrastructure charging(an advisory board to the European Commission) even mark in their finalreport on the calculation of transport accident costs that “&RQVHTXHQWO\ WKH UHPDLQLQJ H[WHUQDO FRPSRQHQW RI DFFLGHQWV FRVWV LV QHJDWLYH LH WKH\ VKRXOG EH VXEVLGLVHG EHFDXVH PRUH SHGHVWULDQV DQG F\FOLVW UHGXFH WKH ULVN IRU H[LVWLQJ YXOQHUDEOH URDG XVHUV”.
Of all the costs from accidents that of human suffering are generally mostimportant. A common quantity in which to express these costs is that of thevalue of a statistical life (VOSL). This is the value of a ‘random’ human lifethat is derived from the expressed willingness to pay for risk reduction. INEurope a value of 1.5 million euro is commonly used. For use in developingcountries this value should be adjusted according to the ratio of PPP’s. Thecosts of processing and prevention, medical care and the costs of productiv-ity losses can, as a rough estimate, be valued as 10% of that of a VOSL(Unite, 2001).
4.641.1/The benefits of cycling and how to assess them &RQJHVWLRQA bicycle takes less space than a car. On (urban) roads intensive car usewill lead to congestion.
As opposed to the costs of infrastructure, noise and emissions, in the caseof congestion costs we take ‘external’ to mean not so much that the trans-port system DV D ZKROH generates delay costs for society, more that roadusers generate delay costs for RQH DQRWKHU. A road user has it within hispower to decide to travel at a time which is susceptible to queues, allowingin any case for his own (anticipated) delay (as a result of which it is internal-ised), but not for the delay he causes other road users through this decision.
Congestion costs can amount to great sums as the valuation of lost produc-tive time is relatively high. It is estimated that Bangkok, Thailand misses outon a third of its economic growth due to congestion. In this city the conges-tion due to a policy of facilitating the car has lead to an average speed forcars of 8 km per hour. As a contrast, in Singapore where space has beengiven to public transport and the bicycle the average speed of cars is about30 km per hour (VNG, 2000).
%HQHILWV IURP ZHOIDUH VKDULQJ WKH ELF\FOH DV D EULGJHNow that we have addressed the most important external costs and benefitswe would like to add one more benefit of cycling that is often forgotten. Inmany countries the welfare distribution over the population is not even. Forexample only the small rich class experiences the benefits of the car. Thebicycle could help to change this situation, and offer opportunities to manypeople who now have very little chances. In the long term this will contributeto a larger economic participation and stability of the country as a whole.
Although it is hard to quantify this effect it is believed that benefits can bepotentially large.
&RQFOXVLRQ WKH PDLQ EHQHILWV RI F\FOLQJ As an alternative for walking the major benefit of the bicycle in developmentcountries for the user is the reduction in travel time that allows for an enor-mous increase in the socio-economic opportunities that the owner has. Nowhe or she can reach more jobs and markets than before and existing desti-nations can be reached in less time.
As an alternative for the car the bicycle’s main benefits for the user are:• its much lower purchasing and operational costs; • in congested urban areas: travel time savings; • a better physical condition (health effects).
• smaller safety risks, especially to vulnerable groups (pedestrians and The size of the benefits of the bicycle over the car will depend on the specificlocal situation (whether or not a city suffers from congestion), but as timeprogresses small benefits may grow.
4.641.1/The benefits of cycling and how to assess them As a conclusion we may therefore say that the bicycle is a relatively fastmeans of transport with major benefits over walking and without the draw-backs of the car.
4.641.1/The benefits of cycling and how to assess them CIA, 2003Website of the US Central Intelligence Agency:http://www.cia.gov/cia/publications/factbook/fields/2004.html CBS, 2003Website of the Dutch national statistics office:http://www.cbs.nl/nl/cijfers/statline/index.htm Adler, 1987(FRQRPLF DSSUDLVDO RI WUDQVSRUW SURMHFWV D PDQXDO ZLWK FDVH VWXGLHVRevised and expanded Edition, published for The World Bank.
Hans A. Adler1987 CE, 1999(IILFLHQW SULFHV IRU WUDQVSRUW (VWLPDWLQJ WKH VRFLDO FRVWV RI YHKLFOH XVHDings, J.M.W et al.
Delft, October 1999.
CE, 2001%HQ]LQH GLHVHO HQ /3* EDODQFHUHQ WXVVHQ PLOLHX HQ HFRQRPLH(Petrol, diesel and LPG: balancing between environment and economy)Kampman, Bettina et al.
Delft, August 2001 HLG, 1999&DOFXODWLQJ WUDQVSRUW DFFLGHQW FRVWVFinal report of the expert advisors to the high level group on infrastructurechargingApril 1999 Unite, 20019DOXDWLRQ FRQYHQWLRQV IRU 81,7( 81,7( 81,ILFDWLRQ RI DFFRXQWV DQG PDU JLQDO FRVWV IRU 7UDQVSRUW (IILFLHQF\Nellthorp, J et al.
April 2001 VNG, 2000'H HFRQRPLVFKH EHWHNHQLV YDQ KHW ILHWVHQ HHQ RQGHU]RHN RP NRVWHQ HQ EDWHQ YDQ ILHWVEHOHLG WH LO XVWUHUHQ (The economic significance of cycling: astudy to illustrate costs and benefits of cycling policy), including the annex:7KH UHVXOWV RI IRXU FRVWEHQHILW FDOFXODWLRQV $PVWHUGDP %RJRWD 'HOKL 0RURJRURStichting Habitat Platform, Interface for Cycling Expertise I-ce , Verenigingvan Nederlandse GemeentenVNG Uitgeverij, Den Haag, 2000.
4.641.1/The benefits of cycling and how to assess them

Source: http://www.cedelft.eu/art/uploads/file/03_4641_38.pdf