Carsharing: Evolution, Challenges and Opportunities

22th ACEA

SEPTEMBER 2014

Dr Scott Le Vine, Dr Alireza Zolfaghari, Professor John Polak

Centre for Transport Studies, Imperial College London

22 Content

Introduction 3

What is carsharing? 3

Variations on a theme 5

The carsharing ecosystem 7

New types of interactions with the public sector 9

Who uses carsharing and how? 10

Policy and governance issues 13

Carsharing’s novel system-level properties 14

Acknowledgments The authors would like to thank Arnd Bätzner, Stephan Herbst, Barbara Lenz, Fuensanta Martinez Sans, Joseph Seal-Driver, and Ivo Wengraf for comments on earlier drafts. Any remaining errors are the authors’ responsibility. cusses the set of key industry-level issues. keyindustry-level of set the cusses rather it and identifies dis chronicle activity; carsharing societies. loped less-deve in operations of examples of number a growing is heavily concentrated in countries, there industrialised are of thousandsof vehicles several million deliver improvements service (see to technologies new of use making are that but viable, cially information and communication technology to be commer older, forms thatof services do mobility not require modern alternative, are hire car traditional Taxis and sector. public the and neworganisational structures, ways of with interacting new propositions, value new requires this users, end to cars selling of format traditional the to contrast In (a car). asset cal car-based the without consumer mobility owning the physi to access enable to technology modern on draw that vices’ rationale and main objectives Introduction: 1.

This briefing paper is not intended to comprehensively comprehensively to intended not is paper briefing This The market carsharing worldwide today encompasses the within emerging sits classCarsharing of ser ‘mobility Figure 1

customers. The fleet consists of some tens tens some of consists fleet The customers. source 2006-2014 Vehicles, and Members of Number Market, Carsharing

Frost and Sullivan (2014) Strategic Insight of the Global Carsharing Market. Report #ND90-18, June 2014. 2014. June #ND90-18, Report Market. Carsharing Global the of Insight Strategic (2014) Sullivan and Frost 2

. Though carsharing activity today . Though activity carsharing

TEXT BOX ‘ BOX TEXT A

’ ). - - - - - The • the with following services general characteristics: mobility to refer to world the of most throughout used term the is ring residents. its than rather visitors ajurisdiction’s tax to adesire by ted motiva were cases many in which rental, car traditional for intended taxes to subject be to deemed been occasionally hasspecific form oftaxation.For instance,carsharing activity a to subject is service agiven whether determine to used is definition atechnical when as such however, important, be professionals and canend instancesIn users. clarity certain industry both for confusion of source ongoing an is it dised; isWhat carsharing? 2. The • verification of and identity once, driving-record and isthen hire (ie not a paid chauffeur, as in a taxi). The end user may may user end The taxi). in a as chauffeur, apaid not (ie hire require the user to manually exchange the car’s keys. (see services carsharing the section) next peer-to-peer vehicle that telematics this requires. For instance, many typically, though not have all services carsharing the in- is access Keyless time. each staff of amember with ting interac without future in cars service’s the access to able Thoughcarsha definition, there is no uniquely-correct The terminology of hascarsharing never been standar vehicle user

must is driven go through by the a end

pre-qualification user as in traditional process car for - - - -

3 22th ACEA Scientific Advisory Group Report 4 Carsharing: Evolution, Challenges and Opportunities

TEXT BOX A

SHARING VS ACCESSING

The term carsharing is used for historical reasons, but it is debatable whether ‘sharing’ accurately describes the behaviour. Carsharing generally involves be making use of the vehicle on a personal basis, or on accessing a car owned by another person or entity in behalf of an employer (sometimes called corporate carsha- exchange for an agreed monetary payment. During the ring). The vehicles tend to be models that are uncomplica- period of time when a person has access to a carsha- ted for users to operate, as with standard rental cars. ring car, they are responsible for it and its use is for their • Usage is billed in time increments of minutes or hours, exclusive benefit. Rather than carsharing cars being and sometimes also on the basis of distance travelled. shared between consumers, it is the authors’ view that Many operators allow multi-day usage at discounted rates, the behaviour is more accurately described as sequen- though daily rates are typically higher than for traditional tial short-term car access. Even in instances in which car hire. consumers collectively own a carsharing operator • There may be a one-time sign-up fee or an annual subscrip- through a cooperative structure (eg Mobility Carsha- tion fee, in addition to time-based and/or distance-based ring in Switzerland), they continue to use the fleet via charges. the sequential-access-in-exchange-for-monetary-pay- • Usage is in some cases spontaneous and in others reser- ment model. ved in advance (this point is discussed further below). • The vehicles are typically available from distributed loca- Many analysts position carsharing within the tions across a service area, in contrast to traditional car hire increasing array of other connectivity-enabled access- in which vehicles are accessible only from a small number based consumer services, such as cloud computing of storefront or airport locations. and peer-to-peer accommodation rental (AirBnB 4 ). • Servicing/cleaning is done by the operator’s staff on an There are counter-examples, however, where the domi- occasional basis, rather than after each usage. In many nant paradigm has shifted in the opposite direction cases collaboration with users facilitates fleet logistics; for over time, from access to private-ownership. Today, instance users may be incentivised to re-fuel a carsharing for instance, nearly all adults in high-income countries vehicle through a modest benefit such as an additional in- carry their own personal mobile phone, and the traditio- crement of time to use the car. nal shared form of phone-access (the phone booth) has become irrelevant. In the UK carsharing refers to multiple people travelling together in a car at the same time; the term car clubs is used in IS ‘UBER’ A CARSHARING SERVICE? the way that carsharing is elsewhere. Media for general audiences frequently use the The next section describes the diversity amongst dif- term carsharing more loosely than it is defined here. ferent types of carsharing services. For instance, companies that connect car-drivers with fare-paying customers wishing to be driven from point ‘A’ to ‘B’, such as Uber and Lyft, are regularly described as ‘carsharing services’ in the mass media (eg, Bloom- berg, Forbes, Wall Street Journal). ‘Transport Network Companies’ (TNCs) is the term that now seems to be emerging amongst professionals to describe chauffeur-driven (as opposed to the end user driving himself) technology-enabled mobility services. Such services are becoming increasingly prominent, with Uber being valued at $18 billion (€13 billion) when it raised capital in early June 2014, as compared to a valuation of $3.5 billion in a previous capital-raising round in August 2013.

In June 2014 the taxi trade in a number of European cities staged public demonstrations against smart- Scientific Advisory Group Report phone-app-based TNCs. The question is how such services will be regulated by the public sector; the taxi industry’s overriding concern is that app-based systems

ACEA will be treated preferentially relative to incumbents in th the industry. 22 on-street (which requires permission from the street network network street the from permission requires (which on-street allocated dedicated parking spaces, which in some cases are a professional entity. operating carsharing The vehicles are The fleet ofcarsiscarsharing ownedcentrally (orleased) by reservation. their of end the at it return they when and car the to access gain they when between time entire the for pay and accessed, was it that place same the to car the return tions) Usage duration. its and reservation their begin to wish they which at time the both specify must user the all, not but cases, most In one, in apps or general a via smartphone dedicated website. Users researchers. by extensively most studied been has and cially, consequences. wider the in and cases, use typical because there is great in variability the customer experience, types is of different important carsharing services between carsharing a single carsharing operator delivers more than of one type Variations on a theme 3. ROUND-TRIP CARSHARING

This type ofThis is carsharing commer type the established best cases some in found; be can carsharing of forms Several generally Figure 2 is ‘round-trip’

service

reserve $20 + Million Opportunity (Avis Budget Group Presentation to Investors, May 2014) May Investors, to Presentation Group Budget (Avis is units complementary, and hasand therefore round-trip-carsharing begun to pool their fleets Zipcar’s parent company Avis Budget Group indicates that fleetutilisation traditionalbetweenits car hire

model. as a the car

customer Understanding ahead of when must

(with they the

distinctions wish few excep to use - -

users of peer-to-peer carsharing, due carsharing, users of to peer-to-peer the fleet notbeing to available typically is vehicles of selection diverse a more carsharing, of forms other to Compared episode. usage the of thetransfer car’s keys to the at vehicle-renter the beginning whereas physically must in otherthe vehicle-owner systems remote with provide vehicle-renters access via smartcard, to devices telematics with equipped are vehicles the cases some In out. rented is it when payments receive others by use rator. People choosing to make their private car available for ope a central by owned – not individuals private by owned cribed above is that– the fleet carsharing is de-centralised des model round-trip the with keydistinction The episodes. largest worldwide. provider ofservices carsharing round-trip manager) and in are others Zipcaris the located off-street. their online marketplace. through transacted rental each of a percentage collects and bespoke the vehicle-owner, insurance that protects product a provides typically operator the model, business the of part As vehicle-renters. prospective with vehicle-owners connect operatorcarsharing is to provide an online marketplace to The principal rolecentrally-managed. of the peer-to-peer PEER-TO-PEER CARSHARING This model is also characterised by round-trip usage usage round-trip by characterised also is model This - -

5 22th ACEA Scientific Advisory Group Report 6 Carsharing: Evolution, Challenges and Opportunities

TEXT BOX B

THE CONSUMER PERSPECTIVE

In a memorable exchange (in 2009), a focus group par- ticipant described to the authors the view that selling her car to join a carsharing service was akin to “stepping off of a diving board without knowing whether there’s any water POINT-TO-POINT FREE-FLOATING CARSHARING in the pool”.

Point-to-point free-floating carsharing (often referred It is well-established that carsharing services thorou- to as flexible carsharing) enables one-way journeys within ghly restructure the costs of car-based mobility, away from a specified geographic zone, in contrast to round-trip cars- high fixed costs / low marginal costs and towards average haring. Usage is typically spontaneous – ie not reserved at costs. The other ways that carsharing differs from owning a all, or only reserved several minutes in advance. The fleets car are less well-understood but nevertheless essential to are centrally-owned by the system operator. A contractual ar- understanding consumers’ attraction to carsharing. rangement with the entity that manages on-street parking is generally required; typical agreements involve the payment Dealing with damage is a drawback for consumers. In of an agreed sum in exchange for the right for customers principle, users are obligated to check a vehicle for damage to park in any (or nearly any) legal on-street parking space. prior to each time they use a carsharing car; failing to report Though this type of carsharing allows one-way journeys, it could mean that they are liable for damage caused by a customers may also use cars for round-trip excursions. The previous user. Any damage deemed to have been caused largest operator of point-to-point free-floating carsharing by a user is subsequently repaired by the operator and services worldwide is car2go. billed to the customer, whereas when a person owns a private car they have the choice of whether or not to pay to fix POINT-TO-POINT STATION-BASED CARSHARING cosmetic damage.

Some point-to-point carsharing services are station- Carsharing addresses some of the classical problems based, meaning that the user picks up a car from one parking of owning a car – such as arranging permanent parking, station and returns it to another. Fixed infrastructure can be vehicle inspection, maintaining insurance cover, and repair located at the parking stations, such as charging points for costs that can be both uncertain and large. But the consu- electric vehicles and kiosks for customer service. France’s mer must bear new risks. Many carsharing services require Autolib’ is the largest point-to-point station-based carsharing that a user indicate the duration of their rental in advance; system (and has plans to expand internationally); the point- the customer must therefore either pay for reservation- to-point system currently being piloted by Zipcar in Boston time that they do not end up using, or run the risk of paying (USA) is also station-based. The logistics of a point-to-point large penalties for not returning the vehicle at the agreed station-based system are less challenging to manage (in end of the reservation. Also, a private car is (nearly) always comparison to a point-to-point free-floating system). The available to its owner, whereas carsharing operators make trade-off, however, is that users are provided a lower degree no promises to users that a car will definitely be available of flexibility, and as with round-trip carsharing the stations when and where desired. Cambio carsharing, for instance, require the allocation of dedicated space. reports that 93% of customers’ requests for reservations are accommodated to the customer’s satisfaction – this Both types of point-to-point carsharing (free-floating and means that 1 in every 15 requests is not satisfactorily accom- station-based) are subject to tidal flows. Clustering of vehi- modated. cles is undesirable from the perspective of both users and the service provider. The Autolib’ system addresses this pro- In future, larger fleet sizes and fleet-management tech- blem collaboratively with its customers, by offering free ren- niques to better predict patterns of fleet utilisation will help tals to users willing to re-position vehicles on an ad-hoc basis. carsharing operators design offers that balance between degree-of-certainty and price-charged in more sophisticated ways. This will allow consumers to better match, on a journey-by-journey basis, between their appetite-for-risk and willingness-to-pay.

In order for carsharing operators to unlock new consumer Scientific Advisory Group Report segments, the operational challenge will be to retain the benefits of the access-based mobility, while in future providing customers a very high degree of assuredness that they

ACEA will have access to a car when and where they wish (as is the th case with private car ownership). 22 DriveNow service). service). DriveNow Sixt’s and BMW and service, car2go Daimler’s and Europcar are cooperating to (eg deliver jointly services carsharing one another. In other instances carmakers and car hire firms with competition in themselves find so and services, sharing car deliver both now firms hire car and carmakers However, relationship continues today.rers, and this well-established nally major purchasers of new from cars vehicle manufactu ness relationships. Car hire firms, for instance, aretraditio busi of types new to leading is sector carsharing the in firms 2013). in Group Budget Avis company parent its of revenue the world’s largest operator,carsharing accounted for only 3% of the Zipcar, (eg organisation larger amuch of part are others forpendent entities whom iscarsharing the core business; smaller operators) are(typically not-for-profit.Some are inde others whilst entities, for-profit are operators Many vices. ecosystem carsharing The 4.

The growing interest of incumbent automotive sector sector automotive incumbent of interest growing The Various of organisations deliver types ser carsharing Figure 3

1995-2013 Total number of users and vehicles ------

(approximately annually of as €165,000) social part its 200,000 CHF of aloss at pods these operates it that locations do not cover reports costs; the co-operative its of 16% municipalities. Swiss of 65% in service electric). (fully vehicles electric battery are cars) (27 fleet its of discounted rates and in governance. taking its part 1% investing a small amount in in the service exchange for of users are its members of co-operative, the Mobility 47% vehicles; 2,650 with Switzerland across customers 112,000 serve staff 180 Today its 1980s. late the to back to-peer offeringsto-peer (see discussion later inthis section). beginning to and peer- experiment point-to-point with fortrip services many carsharing years and are now set of first-waveoperatorsthat haveoperated round- andboth Mobility Communauto are among the small (see 2010s early the in time first the for size fleet its in stabilisation of evidence bited exhi also that Canada) Quebec, in (located operator carsharing independent to, another well-established Communau with aparallel is There 2013). to (2012 0.8% by decreased operations core from revenue and 2012) to 2011 from 2% only by grew (and unchanged remained size fleet the though 2013, to 2012 from 7% by increased priordrivers to acquiring their full licence. driving learner by use for available cars its makes that scheme innovative an 2013, in launched, also It platform. ring carsha peer-to-peer Sharoo the in invested it 2014 in Earlier 2014. September for planned is service floating free- apoint-to-point of atrial though model, service profit margin of approximately 5%. a at operates awhole as business The commitment. TEXT BOX TEXT SWITZERLAND’S MOBILITY CARSHARING Mobility operates 1,400 vehicleMobility locations, with traces originsSwitzerland’s Carsharing its Mobility Mobility reports that their number reports ofMobility customers currently operatesMobility a carsharing round-trip

C FIGURE

3 ). 5 Interestingly, - - - 7 22th ACEA Scientific Advisory Group Report 8 Carsharing: Evolution, Challenges and Opportunities

TEXT BOX D Whilst the overall carsharing sector is growing rapidly, there is churn among individual operators. In Britain, for in stance, three significant carsharing operators have with- CARSHARING AND INSURANCE drawn from the marketplace since 2013: • Whipcar operated a peer-to-peer carsharing service Carsharing operators that cannot self-insure must between 2010 and 2013. instead purchase insurance in the marketplace. Accor- • The Dutch firm GreenWheels withdrew its round-trip car- ding to Guy Fraker, CEO of risk-mitigation software sharing service in 2013 (it continues to provide service in company get2kno, insurance charges for carsharing continental Europe). services are typically 3-4 times what a comparable pri- • Car2go withdrew its point-to-point free-floating service in vate car owner would pay. Innovation in the insurance spring 2014 (its operations continue to grow in continen- sector is limited by its high degree of regulation, and tal Europe and North America). It was reported that “coor- designing customised products to address the chal- dinating 32 separate [local] authorities across different lenges of a small sector such as carsharing is in general [London] boroughs proved to be more difficult than anti- not a priority for insurers. cipated”. A car2go spokesman stated that the business was not financially sustainable in Britain because demand There is cause for optimism, however. Fraker points from the fewer than 10,000 customers was not sufficient to out that carsharing services have a relatively rich set of achieve the required usage level of five to eight bookings evidence for each of their customers (in comparison to per car per day. 6 traditional car hire) to contribute to insurers’ underwri- ting decisions. The data streams from in-vehicle tele- Different organisational forms for providing carsharing matics can in principle be exploited in order to better services each have distinct strengths and weaknesses. Cars- assess risk – encouraging safer drivers, if not actively haring is not a traditional core competency for vehicle manu- discouraging unsafe drivers. facturers, and therefore requires allocating resources to set up dedicated teams with the specialised skillsets as well as Carsharing operators today price insurance to their capital investments in information-technology systems. Car- customers in quite crude ways, in comparison to the makers are, however, well-positioned in several respects to sophisticated pricing models used for private motor deliver carsharing services. They have the financial depth insurance. The price operators charge for insurance to bear risks such as residual values and to self-insure, whe- frequently does not vary among different drivers, but is reas smaller independent carsharing operators frequently rather embedded in the overall car-usage charges and struggle with insurance issues (see TEXT BOX D ). Vertical therefore invisible to the end user. integration between vehicle manufacturing and service- provision also allows telematics equipment to be efficiently Insurance issues have proven particularly trouble- designed and fitted into a carsharing service’s vehicles in the some for peer-to-peer carsharing services, as vehicle factory, rather than as after-market add-ons. Finally, carma- owners who rent their cars out may in some cases kers are able to leverage existing organisational strengths expose themselves to financial liability. (eg back-end IT systems, market research capabilities, brand recognition, optimal vehicle maintenance regimes) in ways A new paper by Dixit and Rashidi 7 , available online that competitors cannot. now and to be published in print in September 2014, appears to be the first in the public literature to quantify Operators of round-trip carsharing services have taken the impact of various factors associated with carsha- notice of the rapid growth of competing point-to-point car- ring members’ crash risk. Also later in 2014, a global sharing services. At the time of writing, at least three major actuarial study of carsharing crash risk will be publicly ‘legacy’ round-trip carsharing operators are experimenting disseminated; the research is being undertaken by with point-to-point services to complement their existing UC-Berkeley’s Transportation Research Sustainabi- service offers (Communauto pilot-tested a one-way system lity Center, get2kno, Metavera, and sponsored by the Scientific Advisory Group Report in Montreal, Canada in summer 2013, with a larger launch in insurance company Assurant. Additional information is autumn 2013; Zipcar began a pilot project in Boston, USA in available by contacting Susan Shaheen at TSRC or Guy late spring 2014, with plans for wider rollout later this year; Fraker at get2kno.

ACEA and Mobility Carsharing plans a pilot programme in Basel, th Switzerland, to be called Catch-a-Car, in September 2014). 22 and larger than any one operator. carsharing individual How (forring the organisations is reasons structural above) outlined had on strong the impacts taxi trade. havesame actions way historically that local-government the in operators, for important be always will government take • are • are • are • are • general: in also, They space. on-street on a monopoly have ments bodies. sector public national and Europe-wide whichindustry, is accustomed to with primarily interacting automotive the for concept anovel is user end and supplier operating. The intermediation of local government between spaceprevent canfrom effectively carsharing a service on-street to access required the attain to inability the cases nicipalities. This can be a criticalvulnerability, as in certain parking space, managedon-street by which is mu typically for carmakers. the with publicin interact sector must ways that are unfamiliar however,Carsharing services, car manufacturers. affect directly that safety, and emissions as such areas in decisions, the public sector New types of interactions with 5. influenced. positively be to is loyalty brand if important be will way asustained in expect customers is not a foregone conclusion. Delivering the level of service this future, in cars purchase they when and if loyalty brand encourage may services carsharing via 6) Section (see mers the marketplace. carsharing younger custo While serving graphic area, ring operators, and ment for commerce), nue generation, social inclusion, and a environ supportive high of quality life for residents, emissions reve reductions, a (eg another one with conflict may which of some comes, elections), The asymmetry between local government between andThe carsha asymmetry The and relationshipoperators carsharing local between In contrast to higher levels of government, local govern requireMany access services carsharing privileged to National governments and European-level bodies make in carmakers for opportunities as well as risks are There under fragmented, subject under actions no pressure at to obligation a changes slower each to representing pace to deliver in reach policy than a agreements private-sector range direction a relatively of desirable

(eg with small entities. following carsha geo out

------

3. 2. 1. level bodies Industry shouldments. undertake tasks such as: vely mediate individual operators between and local govern effecti can third-party, a trusted as that, bodies level industry ever , there are ways manage tothis relationship. effectively terms-of-market-access. than rather provision, service of quality on) another one with compete (and towards energies their devote to operators for it operator; carsharing individual is for better all concerned each than rather level industry the at addressed best are cess take on this role. principle is that The terms-of-ac important should one(s), new or body(ies), existing which on aview take www.acea.be ly (eg European Automobile Association, Manufacturers www.bvrla.co.uk the organisations that represent the sector(eg vehicle-rental tential of mobility”. shared-use policies, programs and standards that demonstrate the po craft to agencies governmental other and cities industry, sharedusemobilitycenter.org), withambitions to “work with www. 2014, June (10 launched recently very was Center lity rica, uk www.carsharing.de sentation of include operators carsharing BCS(in Germany, 5. 4.

), and the Carsharing Association (primarily in North Ame ), in and (primarily North Association the Carsharing Providing a secure, moderated online forum (perhaps an Maintaining an online archive of documents contract vernment. and operators carsharing local between go contracts Preparing (and online) posting sample standardised fair. as them see sides all on people reasonable that away such in drafted be must they time, over sustained and impactful be to order in voluntary; be Thefor principles would information-sharing. of necessity standards as well as terms-of-access, and payments ting relationship, including transparent formulae for calcula both sides of the local government/carsharing-operator guide to principles of set simple and a clear Proposing terms-of-on-street-parking-access in differentplaces. terms-of-on-street-parking-access to information than local government staff regardingthe access better have currently cities multiple in active tors wherebywould opera carsharing address the asymmetry places to exchange information each with other. This for local government listserv) from staff e-mail different ormisunderstandings disagreementsfrom escalating. Offering a formaltomediation service prevent small-scale local governmentbetween and carsharing-operators. British Carsharing services must be must represented services Carsharing by competent Examples of current industry bodies with partial repre partial with bodies industry current of Examples www.carsharing.org Vehicle ). The authors wish to be clear that they do not not do they that clear be to wish authors The ). ) and the automotive sector more broad more sector automotive the ) and Rental ), CarPlus ). and In the Leasing (in

8 USA, There are also established the Association UK, the Shared www.carplus.org. Use in the Mobi UK, ------

9 22th ACEA Scientific Advisory Group Report 10 Carsharing: Evolution, Challenges and Opportunities

Even with the best of intentions among all parties, the Information about the socio-demographics of peer-to- structural nature of the relationship between carsharing- peer carsharing customers is particularly scarce. In one stu- operators and local-government creates inefficiencies. It in- dy, a 2012 survey of customers of the Whipcar peer-to-peer jects a layer of uncertainty that constrains the ability of opera- service found that large majorities were under age 45 (73%), tors to invest, and leads all parties to mis-allocate resources. and lived in a carless household (58%). 15 A major research Respected, independent industry level body(ies) that can project is currently underway in Portland, Oregon; early re- undertake tasks such as those listed above would provide sults show that people who list their cars for rent through a net benefits to both carsharing operators and local govern- peer-to-peer service are also relatively young, well-educa- ments, and ultimately to end users. ted, and in moderate/upper income households 21 . A majo- rity (58%) indicated that the car they make available for rent through the service is the only car their household owns. 6. A recent study 19 (May 2014) examined the user profile of Who uses carsharing and how? the point-to-point station-based service Autolib'. Like customers of round-trip carsharing services, Autolib' users were Carsharing activity has been studied quite extensively by found to be relatively well-educated and to live in households both academic and private sector researchers. This section with above-average income levels. This study also found that describes what is known about carsharing customers and customers of both Autolib' and the Mobizen round-trip car their travel patterns. sharing service are motivated primarily by economic consi- Though the publicly available evidence is fragmentary, it derations and convenience, rather than environmental sen- is clear that carsharing users have a distinctive socio-econo- sitivity. Interestingly, this last point is somewhat in contrast to mic profile. This section briefly highlights patterns and rela- earlier findings (2006) from Switzerland, where environmen- tionships that are typically observed; readers interested in tal consciousness was found to be a distinguishing feature of further particulars are directed to more extensive reviews of round-trip carsharing users. 18 the literature. 9, 10, 11, 12, 13 The evidence base regarding carsharing users and their Whereas the discussion of user demographics up to now behaviour is most robust for round-trip carsharing services, has focused on personal use of carsharing, the services are though a number of research studies are presently underway also used for commercial purposes, and some operators spe- investigating other types of carsharing services. cifically target the business-to-business market (eg Alpha- Users of round-trip carsharing services tend to be: 14, 15, 16, city and Ubeeqo). A reported 8% of round-trip carsharing 17, 18, 19, 20 journeys in London (which accounts for 83% of Britain’s car- • well-educated, sharing users) are for business purposes, with higher rates • predominantly males (Loose 20 reports a range between elsewhere in Britain (17% in Scotland and 22% in England and 53% and 69% male among various services in Europe, but Wales, excluding London). 15 Loose 20 reports that “at least Martin and Shaheen 17 reported a 57% to 43% split in favour 16% of authorised users [in Europe] are business customers”. of females in North America), • young adults, predominantly between ages 25 and 45 For carsharing operators, increasing the business travel (70% of British carsharing users are in this age range), as a share of overall usage can help increase fleet utilisa- • living as single-person or childless-couple households, tion during periods when demand for personal use of cars- • living in middle or middle/upper income households, haring services is low (eg daytime hours during weekdays). • living in carless or single-car households , Employers may also find staff use of carsharing attractive as • living in urban neighbourhoods, a substitute for providing company cars or pool cars (for cost • relatively heavy users of non-car forms of urban transport and sustainability reasons) or conversely staff use of their (eg public transport, walking and cycling). own private cars for business travel (due to insurance/liability issues). In some cases an employer provides its staff with By way of contrast, comparatively little reliable data is access to a carsharing service for their own personal use as Scientific Advisory Group Report available regarding the user profiles of other types of car an employment benefit (eg the employer may pay their em- sharing services. What evidence exists suggests broad simi- ployees’ subscription fees) as well as their business use. larities in the users’ socio-economic profiles, though there

ACEA appear to be more substantial differences in the ways the th services are used and hence their wider impacts (discussed 22 later in this section).

TEXT BOX TEXT changing technologies and other external constraints. contribution of changing preferences as opposed to relative the quantify to yet have researchers and place, taken have consumption adults’ young on constraints the in changes structural time same the At cars. to relating in youngshowing adults’ shifts consumption preferences but are the authors unaware of unambiguous evidence access carsharing cars when needed. This may be true, to prefer instead and cars owning in interested less are Millennials) as to referred (sometimes adults young of rall Peak Car phenomenon. ove the of component amajor is group demographic this downwardstructural amongst trend in ‘auto-mobility’ decreases in young adults’ car ownership and use. This inthe has carsharing growth taken place alongside sharp ‘PEAK CAR’ AND CARSHARING

One interpretation is that the current generation Carsharing customers tend to be young and adults, Figure 4

Time-trend in personal income of British adults aged 20-29 and by 50-59, gender (Authors’ analysis of HMRC’s Survey of Personal Incomes) Personal of Survey HMRC’s of analysis (Authors’

22 - constraints) to explain young adults'changing car-re native hypotheses (changing preferences vs changing moremade services carsharing practical. private at cars, the same time as new technologies have incomes may have increasingly constrained their use of incomes. ral downward trend in young German adults’ real Kuhnimhof and colleagues a report similar structu the concurrent in growth older adults’ income levels. to contrast in is standards living falling of trend This beginning well in advance of the 2008 recession. trended adults British downwards through the 2000s, industry more broadly.industry automotive the and sector, carsharing the of trajectory commercialimportant consequences for the future their relative power explanatory Quantifying will have issue. asettled yet not is patterns consumption lated In summary, these between alter distinguishing FIGURE

23 4 For young in adults these countries, falling below shows that real incomes for young young for incomes real that shows below - - - 11 22th ACEA Scientific Advisory Group Report 12 Carsharing: Evolution, Challenges and Opportunities

The impacts of carsharing are complex to calculate, indicated that they drove less since taking part (3% provided though many efforts have been made. Firnkorn 24 points out no response). These early findings (suggest that the net effect that correctly-interpreting the impacts of carsharing hinges was an increase in driving mileage associated with this peer- on the methodology used to elicit the responses. A particu- to-peer carsharing service, though to establish this with any larly powerful technique is to enquire about the counterfac- certainty will require the accumulation of further evidence. tual; Martin and Shaheen did so by asking how carsharing customers’ behaviour would change if carsharing were to A body of literature is now beginning to take shape regar- disappear suddenly. 16 (Shaheen and colleagues refer to im- ding the early impacts of point-to-point carsharing. Car2go in pacts elicited from this question-wording as the ‘full impacts’ Seattle (USA) repor ts that 39% of their customers have recon- of carsharing; see Table 1). sidered the need for a personal car, and that on average their customers are using private cars, taking public transport, There is a consensus that the impacts vary quite stron- cycling and walking less than they did prior to using car2go. 26 gly between different carsharing service models. It is in These findings are consistent with earlier findings by Firnkorn appropriate, for instance, to apply the established impacts of and colleagues who studied the car2go system in its early round-trip carsharing to predict the prospective impacts of stages in Ulm, Germany. 21, 27 , Likewise, 2013 survey results peer-to-peer and point-to-point carsharing systems. from Munich, Germany show that 30% of users of the Drive- Now point-to-point free-floating system defer, reduce or plan Table 1 below (reproduced from Shaheen and Cohen to reduce their private car ownership due to their participa- [2013]) summarises reported carsharing impacts from North tion in the service. America and Europe, showing evidence of reductions in both private car ownership and greenhouse gas emissions (ari- In Paris, recent (2014) findings suggest that both round- sing from a combination of reduced average driving distance trip and point-to-point carsharing encourage reductions in and driving in lower-emitting cars). It should be noted that car ownership, with the effect being stronger on a per-cus- the average net reduction in driving distance by round-trip tomer basis for round-trip carsharing (a reduction of 67% carsharing users comes about from an increase in driving by for round-trip carsharing, as compared to 23% for point-to- some (eg people who otherwise would not own a car) and a point). 18 Both types of services in Paris are likewise asso- decrease in driving by others (eg those who otherwise would ciated with decreased driving distance, again with a larger be car owners). per-user impact (-127 kilometres per user per month) for customers of round-trip carsharing than point-to-point carsha- The 2012 survey of British peer-to-peer carsharing custo- ring users (-43 kilometres per user per month). In Paris it was mers provides some limited evidence regarding the effect of also found that point-to-point carsharing customers use the this service model on the amount of driving. Of the 207 res- service on average more frequently than round-trip carsha- pondents (all of whom were vehicle-renters, as opposed to ring users, with 57% doing so more than twice per week whe- people listing their car for rent), 29% reported driving more (in reas 80% of round-trip carsharing customers reported using terms of distance) than they did before taking part in the ser- it less than 3 times per month. vice, while 68% said there was no change. No respondents

Table 1 Reported impacts associated with round-trip carsharing source reproduced from Shaheen and Cohen [2013] 13

IMPACT EUROPE NORTH AMERICA Scientific Advisory Group Report Carbon dioxide emission reduction 39 to 54% 27% (observed impact) | 56% (full impact) 25

Number of private cars a carsharing vehicle replaces (sold/forgone purchase) 4 to 10 cars 9 to 13 cars

ACEA Sold vehicle due to carsharing 16 to 34% 25% th

22 Forgone vehicle purchase due to carsharing N/A 25% cles in requires fleets carsharing ofa clearthe understanding vehi electric of use the encourages that policy public fore, of operator. the service carsharing the perspective There are more complex to operate and overall less economic from In general,electrically-propelled. however, vehicles electric is fleet entire operator’s an where Amsterdam) in car2go and Paris, in Autolib’ (eg instances some and fleet, operator’s an of part asmall forming vehicles apure-electric of examples many however are There diesel). or (petrol fuels fossil by red agenda. research the for item tant impor an therefore is It commercially. mature to yet have that forms forof dressed, newly-emerging carsharing particularly ad comprehensively been yet not has it ‘yes’, but be to likely quite is question this to answer The impacts). environmental includingthe andfull widerfits/costs social impacts, of costs bene user both of account (taking value of creators net are be asked regarding is also services carsharing whether they to question the Likewise, criterion. money’ for ‘value the by is guided decision-making investment, transport blic-sector a pu considering When assessment. a wider of part as but account, into taken be must emissions and levels traffic on suchdeliver through as Impacts thosepublic actions. policy of life, and encouraging social inclusion) that they seek to economicporting development, improving residents’ quality sup (eg outcomes desirable of anumber of one be must tions have and negative emissions that emissions impacts, reduc makers must keep in mind that few very consumer products of Whatever carsharing. the answer to thispolicy question, forms other for robust as presently not is base evidence the though as carsharing, notedround-trip earlier in this paper for decreases’ ‘net be to clearly quite appears answer The to net increases or decreases in greenhouse gas emissions. rators offer and services carsharing jointticketing products. marketplace. A related ope issue is whether public transport the enter can entrants new which under terms the publicly discourage competition between operators and make clear enough is competition much how of question the to answer simple no is There levels. service lower-than-otherwise and prices high inefficiently in result can that power pricing of a degree places one with operator, carsharing provider the has service in that means This transport. urban of forms other for stitute discussed in 5. Section Policy and governance issues 7. Carsharing services predominantly use predominantly vehicles services Carsharing powe A frequent research is question whetherleads carsharing is, sub inCarsharing economics terms, an imperfect those beyond issues policy of anumber raises Carsharing

28 , but it is important that policymakers do nothing to to nothing do policymakers that important is it , but ------young adults to use their services (see (see services their use to adults young [carsharing provider] for a nil or nominal charge”. “consider offeringthe members option totransfer to another to also and practical”, is this if notice months” “two members that take in accreditation its scheme part to provide their market; in Carplus for Britain, instance, requires operators a exit to needs operator an that situation the handle to how among operators. There are therefore questions of policy churn continued be to likely also is there market, this in tion evolu rapid the to adapt to well-positioned is sector private the While organisations. sector private by mainly provided than otherwise. sector local government will generate less revenue for the public that the thearrangementsbetween trade-off operator and Belgium) and (Germany Cambio UK) (in Zipcar UK) (in 24/7 Hertz 2014) May in market from withdrawing to prior UK, (in Car2go (USA) RelayRides Germany) (in DriveNow (France) Mobizen Netherlands) (in Car2go Buzzcar (France) Autolib’ (France) (Switzerland) Carsharing Mobility the of part important an becomes and commercially up scales to But this is insurance less restrictions. viable as carsharing

OPERATOR In many providers do cases not service carsharing allow be to continue will services carsharing that likely is It Table 2 Table

carsharing services services carsharing Minimum age to use selected 21 18 18 MINIMUM CUSTOMER AGE CUSTOMER MINIMUM two years) two least at for licence a driving held have if younger, 25 (or business or university) affiliated an through joining if (19 21 an affiliated university) through joining if (19 21 21 course) driving asafe taken have if (18 20 year one least at for licence adriving held have and 20 year one least at for licence adriving held have and 18 18

Table 2 Table

), due primarily

29 - 13 22th ACEA Scientific Advisory Group Report 14 Carsharing: Evolution, Challenges and Opportunities

transport network. An alternative regime would be for cars- train operating companies, ongoing subsidies for individual haring services to price insurance to their users on a more routes disbursed by the Department for Transport, and a individualised basis than they currently do, and to charge separate public sector regulator that oversee the industry. young drivers insurance prices that are actuarially fair (ie Drawing on lessons from railways and other regulated mar- neither subsidise other users nor require subsidy from other kets (eg electricity and water) will be valuable as the carsha- users). A similar argument applies to drivers with poor driving ring sector matures. histories, who in some cases are not allowed to use carsharing services at all. The policy question is whether these characteristics of today’s carsharing systems are acceptable as 8. the sector increases in importance. Carsharing’s novel One of the great strengths of carsharing is that unlike system-level properties many forms of urban transport it can be profitable, generating a revenue stream for the public purse rather than requiring on- This section concludes the paper with a brief discussion going subsidy. It is therefore an ideally suited product for the of how the properties of large-scale carsharing systems differ current era of austerity; private-sector carsharing services’ from other forms of urban transport, particularly private car main request from the public sector is to be able to operate ownership. in a stable investment environment. This places relatively few demands on public sector entities. Another strength is that, Connectivity is essential to the operation of all contem- from the public sector’s perspective, facilitating carsharing is porary types of carsharing systems. Carsharing services with an action that can be modified in due course or even rever- centrally-owned fleets rely on uninterrupted wireless com- sed if deemed necessary. This flexibility is in sharp contrast to munication to reliably provide customers with access to vehi- other types of public sector decisions, such as whether to pro- cles at the agreed time. However, even services that do not vide planning permission for a land development or whether use in-vehicle telematics (eg many peer-to-peer carsharing to build a new road. Such cases require major one-off deci- services) could not operate without buyers and sellers being sions that cannot be easily re-visited, rather than the opportu- able to arrange car rentals through an online portal. Carsha- nity to learn and adapt over time, as in the case of carsharing. ring systems are therefore vulnerable to unexpected disrup- tion to communications networks in ways that the traditional What, if any, actions are appropriate for national govern- regime of private car ownership and use is not. ments to take is an important decision. National government departments often have large fleets of cars that in principle Urban travellers are quite accustomed to the first-in / can be reduced through providing staff with access to a cars- first-out nature of congested road traffic. With few exceptions haring service. National governments may also directly spon- (eg emergency vehicles), all private cars move in relatively- sor new forms of carsharing; an example of this is Britain’s sha- orderly queues, meaning that all travellers experience similar red-autonomous-vehicle project in Milton Keynes. 30 As with level of service. Carsharing operates differently, however – any public funding of industrial activity, there must be very rather than all drivers of private vehicles getting to their des- clear public benefits in order to justify this type of participa- tination but at a slow speed, carsharing systems experience tion. Such benefits are more likely to be found in demonstra- congestion at the point of vehicle-access. When demand ex- tion projects than applications of mature technology. Finally, ceeds capacity, some users are simply unable to gain access national governments frequently give policy guidance to local to a carsharing vehicle at all, whilst others who have gained government; it would be appropriate to provide overarching access to a carsharing-system car experience no delays due policy guidance that local government should have a pres- to the congestion in the vehicle-access system. Carsharing umption to facilitate innovative forms of carsharing services systems today allocate vehicle-access on a first-come / first- in the first instance, and later revisit the arrangements to opti- served basis through fairly blunt tariff structures, though mise the terms once impacts are better understood. other mechanisms may emerge in future (eg pricing that is dynamic to take spatial and temporal context into account; or Scientific Advisory Group Report The relationship between public and private sector enti- mechanisms that enable customers to trade vehicle reserva- ties requires careful management, but there are parallels tions in secondary markets 31 ). Service operators are likely in to other industrial sectors. The rail industry, for instance, is future to differentiate their service offers to better align with

ACEA characterised by similarly complex functional relationships individual customers’ willingness-to-pay, such as offering th between different actors. In the UK, for instance, there is a ‘premium’ services that provide varying degree of priority in 22 quasi-public owner of rail track, a number of private-sector the vehicle-access system. card is stored in the vehicle; carsharing it swiping when paying REWE. retailer grocery German and DriveNow stores). A example between real-world is the partnership larger (presumably stores food different at shop to journeys into less-frequent shopping occasions, but to make longer areharing service likely to consolidate their shopping grocery that gain accessthat non-car-owners cars to a point-to-point authors the by research exploratory Recent neys than do other shoppers during each of their store visits. spend more on each of shopping their jour less-frequent they that mean may car) private own one’s owning of typical occasionallyservice (rather than the car-usage more-frequent get consumers for many retailers.Further, using a carsharing profile ofthey users, carsharing cio-economic areprime tar so the Given car). a telematics-equipped in driving accepts their with in-vehicle user and interfaces, interacts explicitly vehicles, service’s the in rides user (the user the with tionship retailers, and have the advantage of an ongoing physical rela with customers connecting for vectors are similarly services turn retains a percentage of the value exchanged. Car in and amerchant, and acustomer between a transaction facilitates issuer card Acredit strategy. business their of pect leveraging their customer relationship as will be an important ness. For to operators carsharing avoid such an outcome, than a corresponding private car owner would experience. penalizes travel in congested conditions much more strongly that signal price direct and astrong is This charges. additional of €5 roughly very for liable be would instance, for 20 minutes, tion user carsharing pricing; stuck in a trafficfor point-to-point ofstructure carsharing is a pure particularly form of conges tariff the that out point can operators Service demand. travel and as operators network a desirable mechanism to manage 2014. spring in problems pollution air acute to response in Paris in days several for implemented suchof as restrictions, driving number-plate-based those peer could, carsharing for instance, reduce the effectiveness Peer-to- management. network road of task the complicate own In cars. other ways, however, may services carsharing table way, than blanket on using people their restrictions done accep in and a politically much more straightforward, be could This time. of period some during parked vehicles their keep to operator sector private the paying effect in fleet, system’s acarsharing of all or some block-book simply could manager network road the event), sporting amajor hosts city a when (eg event specific for a volumes traffic restrict to tant sing network conditions. For instance, if it is deemed impor sector road network managers with new options for optimi Car rental is traditionally a low-margin commodity busi commodity alow-margin traditionally is rental Car Road pricing is increasingly seenplanners by transport at systems largeCarsharing scale will also provide public 32 34 indeed found 35 A discount sharing ------and their customers. operators andrelationshipcarsharing between persistent that areyears – enabled by partnerships the relatively deep coming in expected be can retailers and providers service carsharing between partnerships increasingly-sophisticated during opening hours. For the reasons mentioned earlier, location store aparticipating at is customer the that rifying ve system telematics vehicle’s the with parking, of minutes free 10 available make to receipt store their on printed code aunique enter they vehicle the to returns customer the When purchase. in-store the on a5%discount unlocks groceries for that engage braking in emergency situations). (egfull-automation collision-avoidance technologies technologies driving highly-automated that are of short with experiment to manufacturers for test-beds ideal be so mayharing aretightly-managed, prove fleets very to time. any at system automation the from away control taking of capable be driver ahuman that require automation vehicle rage encou to aim that countries) European many by (ratified the to 2014) (April changes recently-agreed the Even 2014. steering controls announced publicly by Googlein May manual without test-vehicles bespoke of fleet the ding notwithstan proposition, ashort-term however, not, is positioning a on without human public streets inside them. re-position to drivers human ding sen of expense the without high, be to likely is demand carsharing vehiclesre-position to where and when to ability the as identified frequently is carsharing to the main unique advantage mobility; vate-car-based to operators. carsharing provide would cars fully-automated that opportunities hauser Korn Alain Technology, of Institute Massachusetts the of prominent researchers (eg Adriano Alessandrini of singly available in the new car market, and a number TEXT BOX TEXT VEHICLE AUTOMATION AND CARSHARING United Sapienza Highly-automated driving technology driving isHighly-automated increa In addition to the prospect of full-automation, cars full-automation, of prospect the to addition In of toFull-automation cars carsharing enable re- Automation many offers new possibilities for pri

E Princeton Nation’s University 33 1968

University) of Convention Rome,

have Emilio

on highlighted Frazzoli Road Traffic of

the the ------

- 15 22th ACEA Scientific Advisory Group Report 16 Carsharing: Evolution, Challenges and Opportunities

References

6. Taylor, E. (2014) Daimler's car sharing business car2go 1. Estimates of the scale of carsharing participation are to quit UK, London a challenge. Reuters. Available prepared by several respected industry analysts. The at: uk.reuters.com/article/2014/05/28/uk-daimler- University of California, Berkeley’s Transportation europcar-carsharing-idUKKBN0E81ZN20140528. Sustainability Research Center’s (TSRC) most recent survey of operators (October 2012) showed 7. Dixit, vs and Rashidi, T.H. (2014) Modelling crash 1.8 million carsharing customers worldwide and propensity of carshare members. Accident Analysis 700,000 in Europe (Shaheen, S., Cohen, A. 2012, and Prevention. Vol.70, p.140-147. DOI: 10.1016/j. tsrc.berkeley.edu/node/701 ). Navigant Research aap.2014.03.005. estimates 2.3 million users worldwide as of 2013 (www. navigantresearch.com/research/carsharing-programs). 8. Shared-Use Mobility Center Formed to Accelerate Frost and Sullivan estimates 3.5 million customers Availability, Equity and Impact of Innovative New Urban worldwide in 2014, with 1.5 million located in Europe Transportation Models. Press release dated 10 June (Briggs, M. 2014, www.slideshare.net/FrostandSullivan/ 2014, retrieved from: www.marketwired.com/press- corporate-carsharing-3-1814). release/-1919053.htm.

2. TSRC: 45,000 carsharing vehicles worldwide as 9. www.communauto.com/biblio.html. of October 2012, a 38% increase in the two years from October 2010; Frost and Sullivan: 90,000 10. Millard-Ball, A., Murray, G., ter Schure, J., Fox, C., vehicles (2014), a 33% increase over the fleet size in and Burkhardt, J. (2005) Car-sharing: When and how 2013. Readers should take note that operators are it succeeds. Transit Cooperative Research Program incentivised to report large and growing customer Report #108. Transportation Research Board. numbers; car numbers, while also based on self- reported data from operators, may provide a more 11. Le Vine, S. (2011) Strategies for personal mobility: A study accurate indication of market scope and growth. of consumer acceptance of subscription drive-it-yourself Also, when customer numbers are aggregated across car services. Chapter 2: Background. Doctoral thesis, operators in places with multiple carsharing services, a Centre for Transport Studies, Imperial College London. person that is a customer of more than one service can (https://workspace.imperial.ac.uk/people/Public/s.le- be counted multiple times. vine/S%20Le%20Vine%20PhD%20Thesis.pdf).

3. Bieszczat, A. and Schwieterman, J. (2012) Carsharing: 12. Jorge, D., and Correia, G. (2013) Carsharing systems Review of its public benefits and level of taxation. demand estimation and defined operations: a Transportation Research Record #2319. P.105-112. DOI: literature review. European Journal of Transport and 10.3141/2319-12. Infrastructure Research. Vol. 13(3), p.201-220.

4. Zervas and colleagues recently reported an 13. Shaheen, S. and Cohen, A. (2013) Carsharing econometric analysis of the effect of AirBnB on the and Personal Vehicle Services: Worldwide Market hotel industry. The authors report that a 1% increase in Developments and Emerging Trends. International AirBnB listings is associated with a 0.05% decrease in Journal of Sustainable Transportation. Vol. 7(1), p.5-34. hotel revenues, with ‘lower-end hotels and hotels not DOI: 10.1080/15568318.2012.660103. catering to business travelers’ most affected. Zervas, G., Prosperio, D., Byers, J. (2013) The rise of 14. de Lorimier, A. (2011) Factors affecting vehicle usage the sharing economy: Estimating the impact of AirBnB and availability in the Communauto carsharing network. on the hotel industry. Boston University School of MSc thesis, McGill University, School of Urban Planning. Management Research Paper Series, #2013-16. papers. Scientific Advisory Group Report ssrn.com/sol3/papers.cfm?abstract_id=2366898. 15. Carplus (2014) Annual Survey of Car Clubs 2013/14: London. 5. Data sources: www.mobility.ch/fr/a-propos-de-mobility/ 16. Cervero, R., Golub, A., and Nee, B. (2007) City

ACEA mobility-societe-cooperative/a-propos-de-nous/ CarShare: Longer-Term Travel Demand and Car th rapports-de-gestion and www.communauto.com/ Ownership Impacts. Transportation Research Record 22 images/usagers.html” #1992. p.70-80. DOI: 10.3141/1992-09. 26. 25. 24. 23. 22. 21. 20. 19. 18. 17.

Published March 2014. Report. Program Pilot Share Car Free-Floating Seattle of Transportation Department Seattle exist. not did service the versuscarsharing emissions their level hypothetical if based on people’s emissions in while participating is impact’ ‘full The emissions. in change and-after Vol. 46, p.1654-1672. DOI: 10.1016/j.tra.2012.08.003 Germany. in system carsharing measuring of the a impacts free-floating Firnkorn, J. p.443-450. DOI: 10.1016/j.jtrangeo.2012.04.018. men. for use andGermany: declining increasing car multimodality D. Kuhnimhof, T., Buehler, and M., Kalinowska, R., Wirtz, DiscussionPapers/DP201213.pdf at: Available #13. Paper Discussion Forum Transport and a Research Agenda. Implications, Policy Issues, Unresolved Evidence, Mobility: of Future Goodwin, P. Washington 2014. January DC, Annual of Meeting the Transportation Research Board, Objectives Meet Policy to Potential the and Oregon, Portland, in Owners of Vehicle Analysis A Carsharing: Preliminary to-Peer N. MacNeil, S. Howland, Jennifer, Dill, SI2.499387. European Commission Grant Agreement # IEE/07/696/ D2.4, Report Final study Momo W. Loose, d’Autolib’ Paris). cas (le directe trace en autopartage service d’un l’impact sur enquête grande première la de Recherche de Bureau 6-t Car-Sharing. Swiss Federal Office Energy of Mineta Transportation #09-11. Report Institute America. in of North Carsharing Impacts Emission S. Shaheen, and E., Martin, The ‘observed impact’ is based on a simple before- asimple on based is impact’ ‘observed The (2012) (2012) http://www.internationaltransportforum.org/jtrc/ Travel trends among young adults in in adults young among Travel trends (2010) (2010) (2012) (2012) (2012) (2012) Journal of Transport Geography. Vol.24, The state of European carsharing. carsharing. European of state The Triangulation of two methods methods two of Triangulation Peak Travel, Peak Car and the Transportation Research Part A, A, Part Research Transportation . Paper presented at the 93rd 93rd the at presented . Paper (2014) Sommaire: (2010) (2010) . Work Package #2. (2006) Evaluation Greenhouse Gas Gas Greenhouse International International (2014) 2013 2013 (2014) (2014) Résultats Résultats Peer-

29. 28. 27. 35. 34. 33. 32. 31. 30.

the addressed Commission Competition UK’s The Secondary markets inSecondary which restaurantreservations 14, April 2014. Available at: at: Available 2014. April 14, Provisional, and Basic Standard Accreditation. Version Carplus zipcar_streetcar/pdf/final_report.pdf competitioncommission/docs/pdf/inquiry/ref2010/ http://www.competition-commission.org.uk/assets/ goVSuk/20140402141250 2010: in transport and and carsharing between other forms of issue of competition between carsharing operators Economics. DOI: 10.1016/j.ecolecon.2011.03.014 systems? car-sharing free-floating new of effects environmental M. Muller, and J., Firnkorn, parken/ zu-rewe-parken-einkaufen-5-sparen-und-verguenstigt- http://blog.drive-now.de/2014/05/05/mit-drivenow- Transport Policy. DOI: 10.1016/j.tranpol.2013.12.008 shopping. grocery of study Case method: response stated- aneeds-based through services shared-mobility by enabled patterns activity/mobility of forms new J.W. Polak, O., Adamou, S., Vine, Le TRANS-WP1-145e.pdf www.unece.org/fileadmin/DAM/trans/doc/2014/wp1/ECE- com/node/67874 Paris? S. Vine, Le a-good-table-by-flicking-an-app-not-greasing-a-palm.html cf. trading, debateindustry’s over the of merits reservations is conceptually similar. Regarding the restaurant time agiven at vehicle acarsharing uses customer which of outcome the with willingness-to-pay of carsharingto reservations align better customers’ are exist traded exchange today; the mutually-willing build-prototype-self-driving-vehicles view/-/blogs/transport-systems-catapult-selects-rdm-to- www.innovateuk.org/web/intelligent-mobility/article- Criteria-provisional-full-basic-standard-April-2014-1-1.pdf wp-content/uploads/2014/04/Car-Club-Accreditation- There will be an app for that. that. for app an be will There (2014) Car Club Accreditation Criteria: Full, Full, Criteria: Accreditation Club Car (2014) Ecological Ulm. in car2go of case The (2014) (2014) www.nytimes.com/2014/06/14/dining/getting- . The road to better air quality in in quality air better to road The http://webarchive.nationalarchives.

http://www.carplus.org.uk/ (2011) What will be the the be will What www.planetizen. (2014) (2014) Predicting Predicting

17 22th ACEA Scientific Advisory Group Report 18 22th ACEA Scientific Advisory Group Report

Association des Constructeurs Européens d’Automobiles

European Automobile Manufacturers Association

Avenue des Nerviens 85 B – 1040 Bruxelles Belgium t +32 2 732 55 50 f +32 2 738 73 10

@ACEA_eu www.acea.be