Carsharing and Personal Vehicle Services: Worldwide Market Developments and Emerging Trends

Abstract

Carsharing (or short-term auto use) provides a flexible alternative that meets diverse transportation needs across the globe, while reducing the negative impacts of private vehicle ownership. More than 65 years ago, carsharing began appearing in Europe. It has expanded to approximately 1,100 cities worldwide, in 26 nations on five continents. This article provides a global perspective of carsharing growth and future developments from 2006 through 2015, employing data from three surveys conducted in 2006, 2008, and 2010. The authors feature the interview findings of 25 carsharing experts worldwide representing 25 of 26 nations, collected in 2010. As of October 2010, approximately 1.25 million members were sharing nearly 32,000 vehicles as part of an organized carsharing service. Another eight countries were exploring carsharing.

Full text

Personal vehicle sharing services in North America
Susan A. Shaheen
a,
⁎
, Mark A. Mallery
a,1
, Karla J. Kingsley
b,2
a
Transportation Sustainability Research Center (TSRC), University of California, Berkeley, 1301 S. 46th Street, Bldg 190, Richmond, CA 94804‐4648, United States
b
Department of City and Regional Planning (DCRP), University of California, Berkeley, 228 Wurster Hall, #1850, Berkeley, CA 94720‐1850, United States
abstractarticle info
Article history:
Received 1 November 2011
Received in revised form 26 March 2012
Accepted 27 April 2012
Available online xxxx
Keywords:
Personal vehicle sharing
Carsharing
Peer-to-peer (P2P) carsharing
Collaborative consumption
Over the past three decades, carsharing has grown from a collection of local grassroots organizations into a
worldwide industry. Traditional carsharing, though expanding, has a limited network of vehicles and loca-
tions. The next generation of shared-use vehicle services could overcome such expansion barriers as capital
costs and land use by incorporating new concepts like personal vehicle sharing.
Personal vehicle sharing provides short-term access to privately-owned vehicles. As of May 2012, there were
33 personal vehicle sharing operators worldwide, with 10 active or in pilot phase, three planned, and four
defunct in North America. Due to operator non-disclosure, personal vehicle sharing member numbers are
currently unknown. The authors investigated personal vehicle sharing in North America by conducting 34
expert interviews. This research explores the development of personal vehicle sharing including business
models, market opportunities, and service barriers to assess its early viability as a sustainable transportation
mode and to provide a foundation for future research on the topic. Personal vehicle sharing has the potential
to impact the transportation sector by increasing the availability and interconnectivity among modes and
providing greater alternatives to vehicle ownership in more geographic locations.
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Although the personal automobile remains the primary transporta-
tion mode in North America, recent research supports the view that pri-
vate vehicle use is in decline in numerous countries (Millard-Ball &
Schipper, 2011; Newman & Kenworthy, 2011). Indeed, the U.S. Depart-
ment of Energy recorded a drop in ownership of four million vehicles in
2009—the first significant decline since it began recordkeeping in 1960
(Mittelstaedt, 2010). This decline coincides with a growing prevalence
of alternative modes such as traditional carsharing and the develop-
ment of new modes such as personal vehicle sharing (short-term access
to privately-owned vehicles).
Traditional carsharing provides members access to a vehicle for
short-term daily use. Automobiles owned or leased by a carsharing
operator are distributed throughout a network; members access the
vehicles with a reservation and are charged per time and often per
mile. They benefit by obtaining personal automobility without the
need to own a private vehicle; this can result in considerable monetary
savings and environmental benefits.
Traditional carsharing is intended for short trips and as a supplement
public transit. Initial market entry in North America focused on the
neighborhood carsharing model, characterized by a fleet of shared-use
vehicles parked in designated areas throughout a neighbo rhood or
municipality. In recent years, business models have advanced and di-
versified. Variations on the neighborhood model developed in North
America include: business; college/university; government/institutional
fleet; and public transit (carsharing provided at public transit stations or
multi-modal nodes). Despite differences in target markets, these models
share a similar organizational structure, capital ownership, and revenue
stream.
The next generation of shared-use vehicle services, which provide
access to a fleet of shared-use vehicles, incorporates new concepts,
technologies, and operational methods.These models representinnova-
tive solutions and notable advances. They include one-way carsharing
and personal vehicle sharing. One-way carsharing, also known as
“free-floating” carsharing, frees users from the restriction of having to
return a vehicle to the same location from which it was accessed. In-
stead, users leave vehicles parked at any spot within the organization's
operating area, allowing for the possibility of one-way trips. The one-
way model resembles more traditional forms of carsharing—except for
the logistics of vehicle redistribution and the need for expanded vehicle
parking.
Personal vehicle sharing, which is the focus of this paper, represents
a more distinct model due to differences in organizational structure,
capital stock, and liability. Personal vehicle sharing involves short-term
access to privately-owned vehicles, enabling a lower operating cost
and a wider vehicle distribution. While two versions of personal vehicle
sharing first occurred in North America beginning in 2001, the personal
Research in Transportation Business & Management xxx (2012) xxx–xxx
⁎Corresponding author. Tel.: +1 510 665 3483; fax: +1 510 665 2183.
E-mail addresses: sashaheen@tsrc.berkeley.edu (S.A. Shaheen),
markmallery@tsrc.berkeley.edu (M.A. Mallery), kkingsley@berkeley.edu (K.J. Kingsley).
1
Tel.: +1 510 665 3461; fax: +1 510 665 2183.
2
Tel.: +1 510 485 9292; fax: +1 510 642 1641.
RTBM-00034; No of Pages 11
2210-5395/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.rtbm.2012.04.005
Contents lists available at SciVerse ScienceDirect
Research in Transportation Business & Management
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

vehicle sharing model did not begin significant expansion until 2010
when companies began to emerge across the globe. Since then, personal
vehicle sharingmodels have evolved rapidly. As of May2012, there were
33 personal vehicle sharing operators worldwide, with 10 active or in
pilot phase, three planned, and four defunct in North America.
Personal vehicle sharing remains an extremely new concept within
the shared-use vehicle spectrum, the potential of which is unknown.
This research explores the development of personal vehicle sharing in-
cluding business models, market opportunities, and service barriers to
assess its early viability as a sustainable transportation mode, and to
provide a foundation for future research on the topic.
This paper is organized into six sections. First, it presents a back-
ground section, which provides an overview of the emer gence of
personal vehicle sharing. Next, it provides the study methodology.
This is followed by the results of 34 expert interviews on personal
vehicle sharing—exploring business models, barriers, and opportu-
nities for this market. Next, a continuum framework for understand-
ing shared-use models is presented, f ollowed by a brief discussion of
the implications of personal vehicle sharing for managerial practice.
Finally, it concludes with a summary of key find i ng s a n d re c om me n -
dations for future research.
2. Background
This section focuses on the emergence of personal vehicle sharing as
an extension of traditional carsharing, in part resulting from recent
shifts in consumption patterns referred to as “collaborative consump-
tion.” Collaborative consumption is an economic model that empha-
sizes “access” or “sharing” instead of owne rship. The discussion of
traditional carsharing incl udes its evolution over time in North
America and a revi ew of its social and e nvironmental impacts—key
drivers to service growth.
2.1. Collaborative consumption
According to Rachel Botsman, author of What's Mine is Yours: The
Rise of Collaborative Consumption, the phenomenon of a “sharing
economy” has become more prevalent in recent years due to a number
of factors: online connectivity, which makes shared access networks
ubiquitously accessible; technology; the “living local” movement , which
facilitates community focused lifestyle; cost consciousness due to the
economic downturn that began in 2008; and environmental conscious-
ness. The sharing of information, photos, and music is widespread and
mainstream; thus, it is not surprising that the digital sharing template
has been applied to physical goods (N Gorenflo, 2011, pers. comm.,
28 July).
Online social networking, such as LinkedIn, Facebook, and Twitter,
has allowed people to connect, influence, converse, and share infor-
mation about new products, companies, and ideas more easily and
rapidly than ever before. While people have always exchanged such
information, the Internet makes it easier, more wide reaching, and
faster. It has also enabled an offline experience, where people can
connect with their communities face to face through peer-to-peer
or person-to-person (P2P) sharing networks (L Anderson, 2011, pers.
comm., 8 August). Social media, both offline and via the web, is an im-
portant facet of marketing for most consumer-facing businesses today
and perhaps even more so for P2P sharing businesses, including person-
al vehicle sharing, which must overcome a lack of familiarity with the
concept and fear of sharing among users.
The public understanding of “access to shared goods” advanced
by the t raditional carsharing industry has likely increased consumer
acceptance of sharing high-value assets, such as vehicles. Carsharing
mar keting, education, and expe rimenta tion h ave i mpacted the way
many consumers view their car and perhaps are d iminish ing the im-
portance of the private vehicle as a status symbol (S Savoure, 2011,
pers.comm.,9June;andMWilliams,2011,pers.comm.,21June).
The next se ction provides a discussion of traditio nal carsharing's
evolution in N orth Amer ica.
2.2. Personal vehicle sharing: evolution and growth
Traditional carsharing was first introduced in North America by way
of two experiments: Purdue University's Mobility Enterprise (1983–86)
and Short-Term Auto Rental (STAR) in San Francisco (1983–85). Car-
sharing later reemerged in 1994 in Quebec City, Canada, with the
founding of the cooperative Auto-Com, which operates today as the
for-profit Communauto in Montreal and the Province of Quebec. In
1998, CarSharing Portland in Portland, Oregon was the first successful
launch of modern carsharing in the United States (U.S.) (Shaheen,
Cohen, & Chung, 2009). In 2000, Zipcar was started in Cambridge,
Massachusetts, and Flexc ar was established in Seattle, Washington.
These organizations expanded rapidly with the help of outside capital
and eventually merged to dominate several large cities in the Northeast
and Pacific Northwest by the end of the decade.
As of July 2011, 26 U.S. carsharing programs claimed 560,572
members and 10,019 vehicles. In Canada, 78,840 members shared
2605 vehicles among 20 carsharing organizations. At present, car-
sharing organizations operate in 20 m etropolitan regions in the
U.S. and Canada. Despite sustained member ship growth, traditional
carsharing remains geographically limited. Sullivan and Magid (2005)
estimated that the profitability of a carsharing business is contingent
upon gaining approximately 25 active members living within 0.40 km/
0.25 miles of each point of departure (POD) to ensure sufficient use.
This estimation is supported by the findings of Cervero et al. (2007),
which indicated that 80% of all City CarShare members lived within
0.80 km/0.50 miles of the nearest POD, and more than 50% lived within
0.40 km/0.25 miles of the nearest POD.
As early a s 2001, eGO CarShare ( formerly Boulder Carshare) of
Boulder, Colorado implemented the first person al vehicle sharing
model by i ncorporating private vehicles into their commercial fleet
through the transfer of personal vehicle titles to the carsharing organi-
zation. In this model, the carsharing operator becomes the temporary
owner and maintainer of the transferred vehicles. In the same year,
RentMyCar was launched using a marketplace approach (similar to
eBay®), whereby vehicle owners pay a small fee to list their automobile
on the website. B eginning in 2007, a variety of companies offering
per sonal vehicle sharing were established internationally i ncluding:
Wombat Car Club (England), Zilok (France), and Drive My Car Rentals
(Australia). Since then, the personal vehicle sharing market has contin-
ued to expand domestically and internationally.
The next section provides a summary of the social and environmental
impacts of traditional carsharing, which may be common to personal
vehicle sharing, and have been associated with user adoption and
public- and private-sector su pport.
2.3. Social and environmental impacts
An increasing body of empirical evidence indicates that traditional
carsharing can provide numerous transportation, land use, environmen-
tal, and social benefits (Dallaire, Lafond, Lanoix, & Viviani, 2007; Econsult
Corporation, 2010; Shaheen et al., 2009). Public benefits are generated
through reduced vehicle ownership, vehicle miles/kilometers traveled
(VMT/VKT), and greenhouse gas (GHG) emissions, as well as through
the provision of short-term auto access. Table 1 provides an overview
of the changes in auto ownersh ip impac ts and VMT/VKT due to car-
sharing from a range of studies condu cted in North America over
the last decade.
One of carsharing's notable impacts is vehicle ownership reduction.
A 2008 survey of more than 6281 carsharingmembers in North America
found car ownership among the survey population dropped by approx-
imately 50% due to carsharing participation (Martin & Shaheen, 2011a).
Data from the same survey indicate that carsharing removed between 9
2 S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

and 13 automobiles for each carsharing vehicle in North America
(Martin, Shaheen, & Lidicker, 2010). While only 40% of all households
joining carsharing own one or more vehicles in North America, these
households exhibited a dramatic shift toward a carless lifestyle (Lane,
2005). Carsharing promotes alternative travel modes, such as public
transit, biking, and walking. These lifestyle shifts not only improve
health but also lead to decreased traffic, congestion, and parking de-
mand in urban areas. A study by Martin and Shaheen (2011b) reported
the net impacts of carsharing on public transit, walking, and bicycling.
The net change in: all public transit was − 1%; traditional public transit
(rail and bus) was − 3%; walking was +3%; and bicycling was +6%
(Martin & Shaheen, 2011b).
North American studies also document overall reductions in VMT/
VKT and emissions. Carsharing has facilitated significant reductions in
some households' annual emissions, which compensate for the collec-
tive small increases of other households. Studies of North American car-
sharing operators found reductions in VMT/VKT ranging from 7.6 to 80%
(City CarShare, 2004; Martin & Shaheen, 2011b; Zipcar, 2006).The most
recent study demonstrated a reduction of the average VMT/VKT of 68%
(Martin & Shaheen, 2011a; Martin et al., 2010). Along with decreased
VMT/VKT and vehicle ownership, low-emission fleets contribute to
lower GHG emissions.
Carsharing's GHG emission impacts have recently been measured
through two distinct metrics in North America: an “observed impact”
describing the emission change that actually occurred and the “full
impact,” which includes the observed impact but also an additional
component of avoided emissionsthat would have occurred, if carsharing
were not available. On balance, net carsharing emissions are negative
(i.e., reducing total emissions). Further, in the 2008 North American
carsharing survey, the average emission change over 2000 respon-
dents was − 0.58 t GHG per household per year for the observed im-
pact, and − 0.84 t GHG per household per year for the full impact
(Martin et al., 2010). From the same survey data, Martin and Shaheen
also e stimated a reduction of 27% in GHGs (observed impact) a nd a
reduction of 43% (full impact) (Martin & Shaheen, 2011a).
Finally, carsharing offers an array of individual benefits by providing
“pay-as-you-go” automobility without the need for personal vehicle own-
ership (Martin & Shaheen, 2010; Shaheen, Mollyanne, & Wipyewski,
2003). Due to carsharing membership, average monthly transportation
costs decreased by a range of US $154 to $435 for American members
(Cervero, Golub, & Nee, 2007; Price, DeMaio, & Hamilton, 2006)andUS
$375 to $471 for Canadian members (Martin & Shaheen, 2011a; Robert,
2000). The next section provides the research methodology.
3. Methodology
Between May and September 2011, researchers conducted 34 ex-
pert interviews. Semi-structured questionnaires were selected as the
method of data collection.The intent was to enable multiple researchers
to conduct interviews and provide comparable qualitative data while
maintaining the flexibility to probe respondents on relevant topics. A
review of the literature and Internet resources were used to identify po-
tential interview topics and subjects. Subject populations were divided
according to respondent expertiseso that the perspectives of variousin-
dustry stakeholders were represented. Interview questionnaires were
developed for each subject population and pre-tested with one expert
in each field. Sub ject populations included personal vehicle sharing
ope rators, traditional carsharing operators, insurance providers, and
public policy authorities.
Researchers developed questionnaires to address key topics includ-
ing: business models, market opportunities, barriers to adoption and
expansion, technology, insurance, an d public policy. Questionnaires
were provided prior to administration so that experts could formulate
answers before being interviewed. Interviews lasted for approximately
1 to 2 h and were recorded and transcrib ed for analysis purposes.
Subjects reserved the righ t to non-disclosure of any given topic.
Researchers interviewed five active, two pilot, one planned, and one
defunct personal vehicle sharing operators in North America and three
active organizations in France. In addition, 17 traditional carsharing
operators, one carsharing consultant, three insurance experts, and one
public policy expert were interviewed. Questions regarding industry
technology were integrated into each interview type.
4. Personal vehicle sharing expert interview results
This section provides an overview of the personal vehicle sharing
models identified as part of this research; a discussion of market
Table 1
North American carsharing impacts.
North American studies (year) Location Sample size Participants selling
owned vehicle (%)
Participants avoiding
vehicle purchase (%)
# of privately-owned
vehicles removed per
carsharing vehicle
Change in avera ge
VMT/VKT (%)
Martin and Shaheen (2010) North America N/A 25 25 9–13 N/A
Econsult (2010) Philadelphia, PA 300 25 7 15.3 N/A
Cervero et al. (2007) San Francisco, CA N/A 24.2 N/A N/A − 33
Cervero et al. (2007) San Francisco, CA N/A N/A N/A N/A − 67
c
/24
d
Dallaire et al. 2007 Quebec Province, Canada N/A 24 53 4.6 N/A
Price et al. (2006) Arlington, VA 369 29
a
71
b
N/A − 43
Zipcar (2006) United States N/A 32 39 20 − 79.8
Millard-Ball et al. (2005) North America 1340 55.2
a
70.5
b
14.9 − 37
Lane (2005) Philadelphia, PA 502 24.5 29.1 10.8
e
− 42
Price and Hamilton (2005) Arlington, VA 403 25 68 N/A − 40
Cervero and Tsai (2004) San Francisco, CA 516 29.1 67.5 6.8 − 47
c
/− 73
d
Autoshare (2003) Toronto, Canada N/A 15 25 6–8 N/A
Katzev (2003) Portland, OR 64 26 53 N/A N/A
Cervero et al. (2002) San Francisco, CA 404 2.5 60 N/A − 3
c
/− 58
d
Jensen (2001) Quebec Province, Canada N/A 21–29 55–61 9.1 N/A
Cooper et al. (2000) Portland, OR N/A 23 25 N/A − 7.6
Katzev (1999) Portland, OR 110 26 53 N/A N/A
Walb and Loudon (1986) San Francisco, CA N/A 15.4 43.1 N/A N/A
Note: N/A denotes data not provided.
a
Percentage that strongly agreed or agreed that they were able to sell one or more cars due to carsharing.
b
Percentage that strongl y agreed or agreed that they were able to postpone buying a car due to carsharing.
c
Reflects existing members' reduction in VMT/VKT.
d
Reflects only trial members' reduction in VMT/VKT.
e
Reflects vehicles removed by members who gave up a car.
3S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

opportunities; and a synopsis of barriers to personal vehicle sharing
adoption and expansion, as identified by the study's expert interviews.
4.1. Personal vehicle sharing models
Shared-use vehicle services are entering a new phase of develop-
ment characterized by short-term access to privately-owned vehicles,
referred to as personal vehicle sharing. Broadly speaking, personal
vehicle sharing companies broker transactions among car owners and
renters by providing the organizational resources needed to make the
exchange possible (i.e., online platform, customer support, auto insur-
ance, and technology). As of May 2012, there were 33 personal vehicle
sharing operators worldwide, with 10 active or in pilot phase, three
planned, and four defunct in North America. Researchers identified four
distinct models through the business model portion of the personal
vehicle sharing operator interviews. Each model is described below.
Fractional ownership
In the fractional ownership model, individuals sub-lease or subscribe
to a vehicle owned by a third party. These individuals have “rights” to
theshared-usevehicleserviceinexchangefortakingonaportionof
the expense. This could be facilitated through a dealership and a part-
nership with a carsharing operator, where the car is purchased and
managed by the carsharing operator. This enables access to vehicles
that individuals might otherwise be unable to afford (e.g., higher-
end models) and results in income sharing when the vehicle is
rented to non-owners.
Hybrid P2P-traditional carsharing model
In the hybrid P2P-traditional carsharing model, individuals access
vehicles by joining an organization that maintains its own fleet of
vehicles, but also includes private autos, throughout a network of
locations. Insurance is provided by the carsharing organization
during the access period for both carsharing and P2P vehicles.
Members can access vehicles through a direct key transfer from
the vehicle owner or through operator installed in-vehicle technol-
ogy that enables “unattended access.” In exchange for providing
the P2P service, operators keep a portion of the vehicle usage fee.
An example of this approach is Communauto's carsharing service,
which started to pilot P2P vehicles in 2011.
This model could also incorporate private vehicles into a
commercially-managed fleet through indefinite transfer of title from
a private car owner to a carsharing organization. In this case, the pri-
vate owner receives carsharing access at a reduced cost, while trans-
ferring the costs of private vehicle ownership to the carsharing
organization. This model was pioneered by eGo CarShare of Boulder
Colorado (formerly Boulder Carshare) in 2001, and was adopted in
2011 by Go-Op of Pittsburgh, Pennsylvania (now defunct), during its
pilot phase.
P2P carsharing
P2P carsharing employs privately-owned vehicles made temporarily
available for shared use by an individual or members of a P2P
com pany. It allows short-term (hourly or daily) or longer-term
(multi-day) vehicle access; members pay only for the time they
use the auto and the mileage driven. Insurance is generally pro-
vided by the P2P carsharing organization during the access period.
In exchange for providing the service, operators keep a portion of
the usage fee. Members can access vehicles through a direct key
transfer from the vehicle owner or through operator installed in-
vehicle technology that enables “unattended access.” The P2P key
exchange method represents a point of debate among carsharing
experts; some feel that an in-person exchange excludes it from car-
sharing's definition, which emphasizes unattended access (i.e., a key
box or smartcard access). Others contend that any service that facil-
itates vehicle sharing can be classified as carsharing. RelayRides,
launched in June 2010, and Getaround, which launched in May
2011, represent the first two examples of P2P carsharing in North
America.
P2P marketplace
P2P marketplac e enables direct exchanges between individuals via
the Internet. This model can focus exclusively on the shared use of pri-
vate vehicles or on the buying, selling, and trading of a broader set of
products and services, such as autos, tools, and party rentals. Terms
are generally decided among parties of a transaction and disputes
are subject to private resolution. RentMyCar, which launched in May
2001, represents the first example of a P2P marketplace exclusively
offering personal vehicle sharing. Other P2P marketplaces, such as
Zilok, focus on the exchange of a wider range of goods and services.
This model is distinct from P2P carsharing in that transactions are
made between parties versus a third-party provider, which offers in-
surance coverage and technology, for instance, as part of the service.
Table 2 lists the North American personal vehicle sharing operators
chronologically by launch date and operational status; it also includes
model type and location. Table 3 lists the personal vehicle sharing oper-
ators outside of North America chronologically according to launch date
and includes location. Model classifications for operators located outside
of North America are beyond the scope of this study.
4.2. Market opportunities
Traditiona l carsharing is most suited to walkable, high-density,
mixed-use urban areas with convenient public transit, allowi ng for
Table 2
North American personal vehicle sharing operators.
Organization Launch
date
Model Launch location
Active
RentMyCar 2001 P2P marketplace New York, NY; Germany
eGO CarShare 2001 Hybrid P2P-traditional
carsharing
Boulder, CO
Jolly Wheels 2009 P2P marketplace New York, NY
RelayRides 2010 P2P carsharing Cambridge, MA
Getaround 2011 P2P carsharing San Francisco, CA
Wheelz 2011 P2P carsharing Palo Alto, CA
PATS Carshare 2011 Hybrid P2P-traditional
carsharing
San Jose, CA
JustShareIt 2011 P2P carsharing San Francisco Bay Area,
Los Angeles, CA
Rent2Buy
(formerly HiGear)
2011 P2P Carsharing Unknown
Pilot phase
Communauto 2011 Hybrid P2P-traditional
carsharing
Quebec, Canada
Planned
Spride Unknown Unknown San Francisco, CA
Koolicar Unknown P2P carsharing Unknown
Mesh Motors Unknown Fractional ownership San Francisco, CA
Defunct
Divvy 2009 P2P marketplace Seattle, WA
Spagg Network 2009 P2P carsharing Los Angeles, CA
HiGear 2011 P2P carsharing San Francisco, CA
Go-op 2011 Hybrid P2P-traditional
carsharing
Pittsburgh, PA
4 S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

high vehicle usage rates (Sullivan & Magid, 2005; Cohen et al., 2008).
Personal vehicle sh aring services have the potential to expand the
geographic range of vehicle sharing s ervices by renting underused
autos and thus lowering vehicle usage r equirements. As of February
2012, there are approximately 1500 autos operatingin personal vehicle
sharing services in North America. Due to operator non-disclosure,
per sonal vehicle sharing member numbers are currently unknown.
This section discusses the geographic viability and market adoption
of personal vehicle sharing.
4.2.1. Geographic viability
Hampshire and Gaites (2011) assessed the financial feasibility of
personal vehicle sharing relative to supply and demand across the
U.S. by adapting Sullivan and Magid's (2005) market penetration
threshold for the model. They found that a 10:1 user-to-vehicle ratio
would be sufficient to ensure market viability (Hampshire & Gaites,
2011). This ratio is much lower than Sullivan and Magid's 25:1 ratio
for traditional carsharing because the organization is not responsible
for the vehicle's capital costs.
“Private” vehicle use in personal vehicle sharing enables the poten-
tial to expand the range of geographic environments (e.g., suburbs)
for shared-use vehicle services. One of the P2P carsharing operators
interviewe d in this study, RelayRides, referred to their service as
“neighbor-to-neighbor carsharing” because it is especially focus ed
on neighborhoods. Private vehicles used in this service are typically
parked at or near an owner's residence, creating “walkable” access
and the potential for increased community member use. Eight per-
sonal vehicle sharing and 13 traditional carsharing expert interview
respondent s identified the potential for personal vehicle sharing to
expand into areas previously considered less viable for tr aditional
carsharing—due to the capital costs and usa ge requirements needed
to support a traditional carsharing model.
4.2.2. Market adoption
Not surprisingly, personal vehicle sharing services benefit from
the operational history and marketing of traditional North American
carsharing companies, particularly when targeting markets with a deep
understanding of shared-use vehicle services. Traditional carsharing op-
erators cited population density related to land use as a primary distinc-
tion between early adopters of personal vehicle sharing and traditional
carsharing members. The potential to deploy personal vehicle sharing
services in low- to mid-density areas may enable the incorporation of
new member demographics. For instance, Getaround stated that their
members in San Jose, California were older on average than those
in San Francisco. Despite the potential for personal vehicle sharing
to expand into lower-d ensity areas, the initial target markets have
been in primarily dense urban centers where t raditional carsharing
companies currently operate, resulting in analogous user populations.
On the vehicle renter side, the demographic and psychographic
characteristics of early adopters varied by company. Members of
Getaround and RelayRides in the San Francisco Bay Area were repre-
sentative of an urban population and tended to be cost-conscious,
technologically savvy individuals. Hybrid P2P-traditional carsharing
services, such as eGo and Communauto, had personal vehicle sharing
renters that show similar demographics to members of their traditional
carsharing service. Communauto reported that its personal vehicle
sharing members were more likely to be partnered than single and
had an average age of 41 years.
On the vehicle owner side eight personal vehicle sharing and tradi-
tional carsharing experts identified economic incentives and perceived
environmental benefits as the primary motivations for early adoption.
Vehicle owners receive 60 to 65% of the rental fee. Income from renting
a personal vehicle is taxable; however, automobile expenses and depre-
ciation can be deducted from revenue. One expert believed that envi-
ronmental concern was more important among affluent populations
to whom the financial incentive may be less significant (W Knapp,
2011, pers. comm., 27 June). Another noted that households with
multiple automobiles might be more likely to dedicate a vehicle to
P2P than households relying on a single vehicle (e.g., a household that
requires multiple cars during the week but only one on weekends)
(K Mclaughlin, 2011, pers. comm., 16 May). Further market research
is needed to establish the market potential for personal vehicle sharing
services.
Revenues from personal vehicle sharing provide an incentive to
vehicle owners to market their own vehicles through diverse mecha-
nisms (e.g., friends and family). One operator, Jolly Wheels, noted that
it is critical for vehicle “providers” (i.e., vehicle owners) to operate as
a small business, publicizing the availability of their vehicles to potential
customers (A Maus, 2011, pers. comm., 12 August). Getaround, another
P2P carsharing operator, is a proponent of “grassroots” marketing—a
strategy that relies primarily on vehicle owners and less on centralized
marketing. Personal vehicle sharing companies often rely heavily on the
“social graph,” relationships among individuals that lack the physical
advertisements common to commercial enterprises, such as vehicle
and parking signage used in traditional carsh aring. However, many
successful companies have grown their customer base using word-of-
mouth and online social networks. The founders of Zimride, a dynamic
ridesharing service, have cited connections with Facebook as the reason
their company was able to scale successfully (J Zimmer, 2010, pers.
comm., 4 November).
The next section discusses the barriers to the adoption and expansion
of personal vehicle sharing including: insurance; public policy; fear of
sharing; pricing/revenue and other key issues, such as technology and
safety.
4.3. Barriers to adoption and expansion
Researchers asked both personal vehicle sharing and traditional
carsharing operators (n=29) to identify the top three barriers to
widespread adoption of personal vehicle sharing (with “one” being
the most important); the results are shown in Fig. 1.Althoughitwas
noted that various barriers were inter-related, the two most common
barriers identified i ncluded insuran ce coverage (23/29) and fear of
sharing personal vehicles (21/29). Th e next most prevalent barriers
included balancing revenue and pricing (10/29), public policy (9/
29), and technology (8/29). Th ese and oth er barriers are discussed
below. Public policy is discussed f ollowin g insurance due to their
close interrelation.
Table 3
Personal vehicle sharing operators outside North America.
Organization Launch date Location
WOMBAT Car Club 2007 United Kingdom
Zilok 2007 France
DriveMyCar Rentals 2008 Australia
CaFoRe 2009 Japan
E-loue 2009 France
DEways 2010 France
LivOp 2010 France
Tamyca 2010 Germany
Voiturelib' 2010 France
WhipCar 2010 United Kingdom
Autonetzer 2011 Germany
Buzzcar 2011 France
CityZen Car 2011 France
FlexiDrive 2011 Sweden
MyWheels 2011 Netherlands
Nachbarschaftsauto 2011 Germany
Rent-n-roll.de 2012 Germany
Snappcar 2011 Netherlands
Social Car 2011 Spain
Une Voiture a Louer 2011 France
WeGo 2011 Amsterdam
PosodiAvto 2012 Slovania
Cartribe Unknown Switzerland
5S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

4.3.1. Insurance
As in the early phase of traditional carsharing, cost and availability
of auto insurance in personal vehicle sharing represents one of the
most notable barriers to the adoption, implementation, and profit-
ability of this service.
Automotive insurance policies are historically designed to provide
a separate coverage of commercial and personal vehicles. Insurance
coverage of personal vehicle sharing requires policies designed to
provide umbrella coverage in order to cover personal vehicles under
high usage rates for any number of vehicles and users. The commer-
cial umbrella coverage developed for traditional carsharing was an
enabling factor for the development of insurance policies for personal
vehicle sharing. However, significant modifications were required for
umbrella coverage of personal vehicles. The creation of a new policy
or modification of existing policies is encumbered by development
costs and setup/legal fees for filing policies with state-level insurance
departments.
A common coverage limi t is US$1 million f or general liability;
however, such policies sometimes provide only secondary coverage
to a vehicle owners' primary policy. In February 2012, a serious colli-
sion occurred in Boston, Massachusetts involving a vehicle rented
through a P2P carsharing company. The accident caused irreparable
damage to the vehicle, debilitating injuries to four individuals, and one
fatality. If the liabilities in this case exceed insurance coverage limits for
the P2P carsharing company, the resulting settlement will likely set a
precedent for future cases (Leiber, 2012).
Based on interview response, insurance costs currently represent
20 to 25% of personal vehicle sharing operators' overall costs; one op-
erator cited it as their largest marginal cost. Member and vehicle
screening is essential to maintaining manageable premiums. At the
time of this writing, the full-time equivalent cost to insure personal
vehicle sharing automobiles in the U.S. was US$2000 to $2500 per ve-
hicle per year (W Curtis, 2011, pers. comm., 10 June). The cost of phy-
sical damage insurance is assessed on a per hour basis. William P.
Curtis, Jr. of Porter & Curtis (Media, PA) proposed an alternative
model whereby liability is established on a per mile basis to reflect
vehicle exposure rates, since vehicles are rarely driven for the full du-
ration of the rental period. Communauto uses a per-mile and per-hour
basis to price their insurance. In the province of Quebec, La Capitale
charges US$910 per vehicle per year. The relatively low cost of insurance
is a product of Communauto's strict screening requirements and univer-
sal health care in the province of Quebec (remov ing the necessity for
per sonal injury coverage).
While insurance costs reflect one of the largest barriers to profit-
ability for personal vehicle sharing operators, all insurance experts
interviewed agree that improvements in risk assessment could signif-
icantly reduce insurance costs. Personal vehicle sharing introduces
unique exposure circumstances and a lack of operational experience
hinders liability assessment. Accurate insurance pricing requires as
much as 10,000 vehicle-years of operating data to determine liability
costs and realize industry-standard profit margins (insurance expert
on carsharing, 2011, pers. comm., 4 August). The most immediate
means of reducing insurance coverage costs, while expandin g the
pro viders' interest, is for personal vehicle sharing operators t o ag-
gregate non-identif ying operational data so that insurance compa-
nies can make coverage profitable. One personal vehicle sharing
ope rators' insuran ce premiums decr eased by 30% in the first year
alone due to operational data collection (W Curtis, 2011, pers.
com m., 10 June). Although dispute d among a number of the experts
interviewe d in th is study, technology may affect th e development
and cost of insurance policies through data collection, vehicle securi-
ty, and P2P operator control. The next section discusses public policy
developments for personal vehicle sharing services.
4.3.2. Public policy
Insurance laws vary state-to-state in the U.S. and province-to-
province in Canada. Indeed, insurance coverage may vary somewhat
among locations even within a single policy due to state or province
requirements. Key elements of personal vehicle sharing legislation in-
clude the classification of personal vehicle sharing as non-commercial
vehicle use (which allows them to be shared in exchange for a rental
rate, if less than the annual vehicle expense), liability determination,
the use of information to determine price and coverage, and economic
incentives or taxation.
Lobbying in California, Oregon, and Washington has facilitated the
operation of personal vehicle sharing services through the passage of
AB 1871, HB 3149, and HB 2384 respectively. The enactment of AB
1871 has been the key for personal vehicle sharing in California and
serves as the model legislation for other states across the U.S. These
bills classify personal vehicle sharing as non-commercial use and
limit “the circumstances under which the vehicle owner's automobile
liability insurance can be subject to liability” in order to prevent can-
celation of primary automobile insurance policies (AB, 1871, 2011).
Thus, personal vehicle sharing programs assume liability when the
vehicle is rented in a shared-use capacity, and the owner's insurance
policy resumes coverage once it is returned. Vehicle owners are indem-
nified for any loss or injury that occurs through shared-use not resulting
from their negligence. The date, time, initial and final locations of a ve-
hicle must be clearly delineated through “verifiable electronic records
identifying” when it is being used as part of a personal vehicle sharing
program (AB, 1871, 2011). This prevents premium spikes for primary
insurance policies resulting from unverified shared use. Vehicle owners
that share their autos in states lacking personal vehicle sharing legisla-
tion risk non-renewal of primary insurance policies, as well as premium
spikes resulting from increased use.
Finally, legal classifications of shared-use vehicle services a ffect
the establishment of taxation rates or financial incentives. In many
states, excise taxes are applied to traditional carsharing services.
New York Stat e, for instance, classifies car re ntal and traditional car-
sharing services as commercial ente rprises, which are taxed 6% on
transactions. While many carsharing operators perceive this as a fi-
nan cial disadvantage, this classification benefited Zipcar (the largest
traditional carsharing oper ator in the world) in the New York court
case, Minto v. Zipcar. A p ersonal injury action arose whereby Leslie
Minto, the plaintiff, alleg ed tha t his vehicle was rear-ended while
stopped at a red light by a ve hicle owned by Zipcar, which was driven
by the defendant. The defen dant asserted the claim against Zipcar in
an attempt to attach vicarious liability to the company as the vehicle
owner. Zipc ar successfully refuted the action “precl uded by section
14 of the Federal Transportation Equity Act of 2005, better known
as the Graves Amendment,” which protects commercial vehicle
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Percent of Respondents, n=22
Obstacles
#1 Obstacle #2 Obstacle #3 Obstacle
Fig. 1. Barriers to adoption and expansion of personal vehicle sharing.
6 S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

owners from vicarious liabili ty in cases void of negligence or criminal
wrongdoing (Zipcar, 2010). The next section discusses “fear of shar-
ing” as a potential obstacle to the growth and expansion of personal
vehicle sharing services.
4.3.3. Fear of sharing
In this study, fear of sharing personal assets was cited as one of the
primary barriers to the adoption of P2P sharing services. A combined
21 personal vehicle sharing and carsharing respondents (out of 29)
identified this as one of the top three barriers, second only to insur-
ance barriers. Indeed, the private vehicle is among an individual's
most valued possessions, and a person may be more likely to rent
out their home than their automobile (G Kohli, 2011, pers. comm.,
17 June).
In 2010, University of California, Berkeley graduate students con-
ducted an intercept survey of people at the Department of Motor Ve-
hicles to test user response to personal vehicle sharing. This study
found that more than half of all participants indicated lack of trust
as their main reason for not converting their personal vehicles to
shared vehicles. Survey respondents most commonly cited operator
screening, user rating, and feedback systems as mechanisms for P2P
ope rators to address trust issues among renter s and owners. This re-
search supports the survey findings, with variations attributed to
subject population and methodology. Fig. 2 indicates how traditional
car sharing and personal veh icle sharing operators believe that the
personal vehicle sharing industry can address the trust issue.
When traditional carsharing and personal vehicle sharing experts
(n=28, one personal vehicle sharing operator abstained) were asked
to identify all the mechanisms that they thought would help to address
the trust issue among individuals in personal vehicle sharing, the top
three choices included: user rating and feedback system (22/28), oper-
ator screening and selection (18/28), and social networking (13/28).
Other mechanisms identified included vehicle quality assurance and
responsive customer service.
In personal vehicle sharing, user rating and feedback systems pro-
vide a critical medium for establishing trust and credibility among
members. Feedback systems provide a mechanism for accountability
and a way to “blacklist” users in the event of misuse or vehicle damage.
There may be instances where damage liability is difficult to determine,
which may require operators to investigate. So me personal vehicle
sharing operators incorporate vehicle owner control over who can rent.
User rental criteria may be established through user feedback and ratings,
behavioral analysis of driving data, and social networking (whereby only
certain “communities” or “groups” are granted access). Two traditional
carsharing operators expressed concern regarding potential liabilities
to service operators and vehicle owners resulting from practices that
may be perceived as discriminatory; however, four carsharing opera-
tors agreed that the vehicle owner's choice of renters would help to ad-
dress the trust issue. One respondent believed that vehicle renters
would tend to take better care of a personal auto than a traditional car-
sharing fleet vehicle (A Quirk, 2011, pers. comm., 20 June). Face-to-face
interaction of the vehicle owner with the renter could also play a major
role in establishing trust (G Kohli, 2011, pers. comm., 17 June; and J
Scorpio, 2011, pers. comm., 24 June).
Screening and selection criteria must be met for individuals to
gain membership to shared-use vehicle organizations in exchange
for service use, “unattended access” or “skip the counter” services,
and/or member rewards incentive programs. Companies commonly
screen for and verify age, identity, and driving record, which enable
ope rators to satisfy insurance policy guidelines. In the future, this
could also include criminal checks. Some personal vehicle sharing
organiz ati ons screen vehi cles for maintena nce issues, age, fuel effi-
ciency, and model specifica tions.
Seven of 12 personal vehicle sharing operato rs indicated that in-
creased legitima cy of personal vehicle sharing could be achieved
through marketing and social media. Three personal vehicle sharing
ope rators and 10 traditional carsharing experts agreed that media
coverage p rovides education about the service and helps to establish
legitimacy. Howeve r, it also has the potential to affect the industry
negatively through reporting of a major incident.
Airbnb, a P2P lodging service that enables people to rent a room or
bed, experienced a widespread media blow in the summer of 2011
after a case involving property damage (Mills, 2011). One expert thought
that personal vehicle sharing companies would be less vulnerable to this
type of incident for two reasons. First, the majority of personal vehicle
sharing services require member screening to meet mandatory auto
insurance requirements. Home- or room-sharing, on the other hand,
doe s not require in surance; thus, thos e companies are not forced to
confront this issue up-front. Second, personal vehi cle sharing may
occur most often within one's own community or local region, thus
encouraging a positive connection with neighbors (L Anderson,
2011, pers. comm., 8 August). Conversely, P2P lodging generally oc-
curs when someone is traveling away from home—it could be a
one-time transaction, which can eliminate community connection
and accountability.
One approach advocated by two operators is to focus exclusively
on sharing between affinity groups–peoples' pre-established commu-
nity networks–instead of the community at large (A Freed, 2012, pers.
comm., 20 March). This enables personal vehicle sharing only among
trusted community members, as well as increased accountability
0%
10%
20%
30%
40%
50%
60%
70%
80%
Number o Respondents, n=22
Trust Issues
Fig. 2. Personal vehicle sharing trust issues.
7S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

through personal interaction. The next section discusses barriers to
balancing revenue and pricing for personal vehicle sharing operators.
4.3.4. Balancing revenue and pricing
Barriers to balancing revenue and pricing result from a trade-off
between generating operator profit and vehicle owner revenue. Person-
al vehicle sharing companies must find renters and offer incentives to
owners, such as 60 to 65% of the rental fee to match the demand. Too
manyP2P vehicles could result in owners not receiving enough reserva-
tions to make their participation worthwhile (K McLaughlin, 2011, pers.
comm., 16 May). In addition, there may be a lag period between when
vehicle owners put their auto into a personal vehicle sharing service
and begin generating rentals (A. Freed, 2012, pers. comm., 20 March).
In personal vehicle sharing, pricing is dependent on membership
fees, technology and insurance costs, hourly or daily use, and mileage.
Competitive pricing is critical in areas where traditional carsharing is
available to incentivize personal vehicle sharing use. Although P2P
services require less capital investment, they also generate a lower re-
turn per transaction in contrast to traditional carsharing due to revenue
sharing between the operator and vehicle owner. Thus, profitability
is contingent on rental volume. Among the personal vehic le sharing
ope rators in North America, mem bership fees are charged primarily
by hybrid P2P-traditional carsharing providers and range from US
$50 to $345 annually. Technology installation costs are often waived
during the start-up period, but they range from US$250 to $1000. As of
April 2012, the typical hourly fee set by vehicle owners renting their
vehicles through Getaround ranged between US$5 and $12, with daily
rates between US$25 and $65-not including fuel. By comparison, Zipcar
charges US$8 to $14.25 per hour with discounts available through
monthly plans, overnight rates of US$39, and daily rates between US
$72 and $102 Brook, 2012.
At present, three methods exist to determine pricing and operator
revenue. One method, referred to as the “owner-pricing model,” allows
owners to set the rental price of their vehicle within the bounds
established by the operator. This method enables market mechanisms
to dictate the price, from which the operator takes a percentage of typ-
ically 35 to 40%. In the second approach, the “vehicle-model pricing
model,” operators determine rates on a per-vehicle basis. Go-op used
this system and commented that balancing renter fees, car owner reve-
nue, and P2P operator return was difficulttoachieve,butitwaskey
to supporting a profi table business model (R Hampshire, 2011,
per s. comm., 11 May). In the third method, referred to as the “P2P
marketplace pricing model,” operators charge a fee for car owners
to post their vehicles to an online marketplace where owners set
the price and/or renters b id against each other, setting the usage
rate. Since a fraction al ownership model has yet to be demonstrated,
this model is n ot described in terms of pricing. Nevertheless, it is pre-
sum ed that a fractional ownersh ip model could operate much like a
traditional carsharing service with respect to renter pricing, and it
would most likely involve discounted pricing or inclusive service to
individuals that invest in vehicles owned by the operator.
4.3.5. Technology
Tec hnology development costs an d per unit expense represent
major barriers to profitability, especially during the start -up phase.
The development and implementation of a personal vehicle sharing
system that addresses software, hardware, and security needs in a
universal solution have proven resource-intensive and costly. Stand-
alone packages range from US$500 to $1000 per unit. Most personal
vehicle sharing companies currently waive technology inst allation
fee s to boost membership and plan to absorb a portion of future in-
stallation costs. Some auto owners prefer not to install aftermarket
technology in their vehicles because it may void manuf acturer war-
ranties or hurt resale values. Compatibility issues may arise with in-
vehicle technology in certain vehicle makes and models, especially for
advanced vehicle types (e.g., electric vehicles). Through partnership
with General Motors (GM), RelayRides replaced their aftermarket tech-
nological package for use of the OnStar© technology, available in GM
vehicles, and a key exchange for non-GM autos. JustShareIt plans to
use a single technology platform for multiple vehicle types, including
boats and all-terrain vehicles, further complicating technological devel-
opment. It is important to note that oversight and maintenance costs, as
well as user ease-of-use, vary by technology.
The importance of technological applications in personal vehicle
sharing schemes was highly contended among experts—ranging from
the need for no technology to advanced systems. The majority of
interview subjects agreed that rapid technological advances and digital
communications could play a key role in establishing a market conducive
to personal vehicle sharing, but they disagreed regarding the need for
in-vehicle technologies to provide operator control and vehicle security.
For instance, some operators claimed that an in-person key exchange
between owner and renter facilitates access to personal vehicle sharing
and trust among users. Others asserted that unattended access pro-
motes user safety and reduces potential access issues caused by coordi-
nation among individuals. The failure of the high-end personal vehicle
sharing operator, HiGear, due to vehicle theft reinforces the need
for vehicle security, as well as careful member screening to prevent
fraud and identity theft.
4.3.6. Vehicle availability
While personal vehicle sharing has the potential to greatly expand
the number of shared-use vehicles, nine of the 18 traditional carsharing
operators interviewed expressed concern regarding vehicle supply (or
availability) meeting demand. In traditional carsharing, vehicles are
made available for reservation during all allotted times in contrast to
personal vehicle sharing models, which reflect more limited availability.
Potential operational issues include: short and limited usage periods,
non-availability during peak-use hours, and emergency situations when
the owner needs a vehicle unexpectedly. Vehicle availability represents
a larger issue for personal vehicle sharing transactions that involve a
key exchange due to the lack of operator control over vehicle access.
Personal vehicle sharing services focus highly on customer service
to provide fast and adequate response to issues that could arise. For
instance, operators plan to address emergencies by honoring pre-
existing reservations and assisting the owner in acquiring an alternative
vehicle. However, P2P carsharing and P2P marketplace models may
prove less able to provide a vehicle in some cases.
4.3.7. Safety
Vehicle safety is a critical issue for personal vehicle sharing services.
Traditiona l carsharing operators actively m aintain fleet vehicles to
ens ure standards of safety and quali ty. P2P carsharing and P2P mar-
ketplace models rely on vehicle owners to upkeep their autos and
address maintenance issues. Lack of operator control over vehicle care
may result in variable auto conditions vs. condition of autos and create
risk for vehicle renters in instances of negligent maintenance practices.
P2P carsharing and P2P marketplace operators are advised to conduct
regular vehicle maintenance inspections and require vehicle owners
to sign agreements that their vehicle is properly maintained in order
to reduce safety and liability concerns.
Table 4 identifies the most impactful barriers to the spread of per-
sonal vehicle sharing, as well as opportunities for addressing many of
them.
5. Shared-use vehicle services: a continuum
This section proposes a continuum framework for understanding
the current state of shared-us e vehicle services, which includes per-
sonal vehicle sharing—spanning from commercial car rental to the
P2P marketplace model. This framework ha s been created to guid e
local and regional governments, policymakers, researchers, and busi-
ness practitioners in understanding key differences among the models
8 S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

bas ed on: capital ow nership, technology, social and environmenta l
impacts, an d industry competition. Fig. 3 illustrates the placement
of the models along the shared-use vehicle continuum.
5.1. Capital
Automobile capital is a central determinant of organizational struc-
ture and thus a key distinguishing factor along the continuum. Models
rangingfromcommercial car rental to hybrid P2P-traditional carsharing
involve company-owned vehicle capital. The costs of purchasing or
leasing and maintaining a vehicle fleet are among the highest opera-
tional costs for models with company-owned capital. Vehicle capital
represents approximately 69% of the total operating expenses of car-
sharing companies (Sullivan & Magid, 2005). Company-owned vehicles
require reserved parking, which can be an asset for marketing and ser-
vice users, but this typically requires additional expenditures. The costs
of vehicle maintenance and parking reflect geographic location, with
costs tending to increase in dense urban areas. Personal ownership of
vehicle capital is a central distinguishing factor of personal vehicle shar-
ing, which can greatly reduce operator capital requirements.
5.2. Technology
Technologysolutions are rapidly evolving among shared-use vehicle
services and vary widely even within a single model. Key elements of
shared-use vehicle technology include reservations and billing, access,
vehicle control and security, and data recording and transmission. Ad-
vances in information and communication technology have enabled
reservations and payment through multiple platforms including auto-
mated phone systems, web-based user interfaces, and smart phone
applications. The use of a manual reservation process has steadily di-
minished since the mid-1990s, although they a re still available
through customer service representatives (Shaheen et a l., 2009).
Vehicle access may be provided through an in-person key exchange
(i.e., no technology), referred to as attended access, or through
unattended access mechanisms such as lockboxes, key fobs, smart
cards, or smart phone applications. Shared-use vehicle service opera-
tors may provide attended access, unattended access, or both. The dis-
tinction between attended access and unattended access is considered
critical for security and safety by three of the personal vehicle sharing
and three of the traditional carsharing operators interviewed. At
Table 4
Opportunities and barriers to personal vehicle sharing in North America.
Personal vehicle sharing Barriers Opportunities
Market adoption • Competition among shared-use vehicle models
• Availability of personal vehicles for shared-use
• Expansion of shared-use vehicle services to
lower-density areas
•“Walkable” community access to shared-use vehicles
• Increased shared-use vehicle supply
Insurance • Cost and availability of auto insurance
• Lack of operational experience to accurately
assess risk
• Circumstances with indeterminate liability
(operator provided insurance liability vs.
vehicle owner insurance liability)
• Indefinite number of users and vehicles
• Strict screening requirements
• Aggregation of operational data to improve
risk assessment
• In-vehicle security and control mechanisms that could
reduce risk and lower insurance costs
• Dynamic insurance pricing based on user demographics
and usage rates
Public policy • Lack of state-level legislation in the U.S.
• Clear delineation of liability
• Rental tax
• Lobbying
• Classification of personal vehicle sharing as non-commercial
vehicle use
• Economic incentives (e.g., tax credits)
Fear of sharing • Value of private vehicles
• Personal attachment to private vehicles
• Lack of trust among members
• Personal vehicle sharing legitimacy needs to be
established (i.e., reliable business)
• User rating and feedback system
• Operator screening of members
• Social media and marketing
• Vehicle access limited to affinity groups
• Insurance to cover all parties in the case of an accident
(e.g., driver, owner, and operator)
• Vehicle owner screens renters
• Critical mass of users
• In-vehicle technology facilitates vehicle security
Balancing revenue and pricing • Trade-off between operator profitandvehicle
owner revenue
• Competitive pricing in areas with competitive
shared-use vehicle services
• Reaching a profitable rental volume
• Operator selection of pricing model
Technology • Development of a universal software, hardware,
and security suite
• Cost per unit
• Vehicle theft
• Compatibility among makes, models, and
vehicle types
• In-vehicle technology facilitates operator control and
vehicle security
• Technology fee waivers boost membership
• Fraud and identity theft protection
Safety • Lack of operator control over maintenance • Regular vehicle maintenance inspections
• Vehicle owner agreement to properly maintain vehicle
Commercial Hourly Traditional Fractional Hybrid P2P P2P
Car Rental Car Rental Carsharing Ownership P2P & Carsharing Carsharing Marketplace
Fig. 3. Continuum of shared-use vehicle services.
9S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

present, seven out of ten personal vehicle sharing operators in North
America provide unattended access.
Three pre-existing classifications of shared-use vehicle technology
have been adapted to reflect the current systems: manual processes,
partially-automated systems, and advanced technologies (Shaheen
et al., 2009). These classifications are based on reservations and billing
as well as access mechanisms, which are common to al l shared-use
vehicle services. Manual processes involve reservations and billing
in-person or via telephone with attended access via key exchange
or unattended access using a lockbox. Partially-automated systems
include reservations and billing via automated telephone system, In-
ternet, s mart card, or smart phone, with attended access via key ex-
change or unat tended ac cess using a lockbox. Advanced technologies
involve reservati ons and billing via automated telephon e system, In-
ternet, sma rt card, or smart phone, and unattended vehicle access
using key fobs, smart cards, or smart phone applications.
Shared-use vehicle providers are increasingly integrating additional
technologies for vehicle control and security, as well as data recording
and transmission. These systems are most commonly used in personal
vehicle sharing systems to provide operator control over shared assets
and to reduce the risks associated with P2P transactions. Table 5 iden-
tifies the mechanisms and purpose of additional technologies used for
vehicle control, security, data recording, and transmission.
5.3. Social and environmental impacts
With respect to social and environmental impacts along the con-
tinuum, traditional carsharing is the only model that is documented
in the literature, at present. While the benefits of traditional carsharing
are widely accepted among experts, the social and environmental im-
pacts of personal vehicle sharing are still debated. Most respondents be-
lieved that personal vehicle sharing would encourage lifestyle changes
that foster modal shifts and ultimately provide environmental benefits
analogous to thoseof traditional carsharing. About half of the traditional
carsharing operators expressed concern that personal vehicle sharing
has the potential for negative environmental impacts. They argued that
personal vehicle sharing would increase VMT/VKT through increased ve-
hicle usage rates and encourage households to keep a personal vehicle
they otherwise might sell or even purchase new vehicles to support
rentals. Further research is needed to document and understand the
social and environmental impacts of personal vehicle sharing.
5.4. Industry competition
Shared-use vehicle services within the continuum have the
potential to provide both complementary and competitive services
depending on geographic location, density, and company strategy.
Some experts cited the private automobile as the primary competitor
to mobility alternatives as a whole. However, market competition in
the San Francisco Bay Area illustrates how a wide variety of similar
services can compete in the same marketplace. In general, market
com petitio n tends to increase in high-density areas with widely
available public transit, and shared-use vehicle services tend to
com plement each other more in low-density area s with f ewer alter-
native transportation modes.
6. Implications for managerial practice
Today, the P2P carsharing business model is receiving a great deal
of attention and is the most widely available personal vehicle sharing
approach in North America. Traditional carsharing operators that in-
corporate personal vehicle sharing into their services (i.e., hybrid
P2P-traditional carsharing) can have a financial advantage–especially
during the initial stages of market development–since they do not de-
pend on venture capital funding or their personal vehicle sharing ser-
vices to be self-sustaining.
Expert respondents were asked to extrapolate the potential for
profitability of personal vehicle sharing over the next 10 years. Five
personal vehicle sharing and four carsharing operators thought that
personal vehicle sharing would likely become profitable in the next
5 years depending upon: reduced insurance and technology costs, a
sustainable balance of rental fees with owner revenues, and well
matched supply and demand. Some operators recognized the trade-
off between expansion and profitability, although others identified
expansion as key to profitability.
In the early phases of personal vehicle sharing market development,
organizations offering these services could mutu ally benefitfrom
cooperation. Communication and information sharing could aid in
technological development, reduced insurance costs, supportive
public policies, and an expande d marketplace. The establishment
of a set of best practices that could provide operational guidelines,
which also occurred in traditional carsharing, could facilitate a bet-
ter understanding of the social and environmental benefits of t his
approach. In the future, other transportation-related shared access busi-
ness models–such as ridesharing (Avego, Zimride), public bikesharing
(BIXI, Capital Bikeshare), and parking services (parkatmyhouse.com)–
will likely interconnect with personal vehicle sharing schemes.
7. Conclusion
In recent years, the concept of personal vehicle sharing has gained
momentum in Europe and North America, largely as an outgrowth of
traditional carsharing. This evolution has coincided with reduced auto
ownership trends and shifting consumption patterns, often referred to
as “collaborative consumption” (or a sharing economy). This approach
has facilitated a range of new business models that create access to shared
resources as an alternative to ownership. This has resulted in privately-
owned vehicles being incorporated into shared-use vehicle fleets.
Table 5
Mechanisms and purpose of additional shared-use vehicle technologies.
Technologies Mechanism Purpose
Vehicle control and security Door lock controls Manage access
Remotely controlled security system Prevent theft
Engine disable Prevent theft
In-vehicle sensor network Detect accidents or parts theft
Data recording and transmission Global positioning system (GPS) tracking that combines satellite
navigation and vehicle tracking (e.g., time, distance traveled, speed)
Record vehicle use to satisfy personal vehicle sharing legislation
requirements (e.g., records identifyi ng the date, time, initial and
final vehicle locations, and kilometers/miles driven); vehicle tracking
to prevent theft; and operationa l data collection and analysis
On-board diagnostic parameters (e.g., fuel used, emissions, engine load) Veh icle operational data collection and analysis (e.g., environmental
impact assessment)
Wireless data transmission Real-time data recording
In-vehicle communication Emergency response and reservation changes
Note: In-vehicle data recording and transmission devices are commonly referred to as telematics (combined computer and wireless communication systems).
10 S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005

As of May 2012, there were 33 personal vehicle sharing operators
worldwide, with 10 active or in pilot phase, three planned, and four de-
funct in North America. Personal vehicle sharing could provide a model
that overcomes some of the financial constraints and geographic limita-
tions of fleet ownership and distribution, as in traditional carsharing. In-
terestingly, all personal vehicle sharing and traditional carsharing
experts interviewed in this study agreed that personal vehicle sharing
holds the potential to notably expand the shared-use vehicle market.
However, a range of issues related to adoption and expansion need to
be addressed including: insurance and liability, technology, vehicle avail-
ability, maintenance, and trust among auto owners and renters. Experts
identified a number of opportunities to address barriers including: lobby-
ing and legislation, screening and data aggregation, user rating and feed-
back, operator screening and selection, rental volume, pricing model, and
technology .
Moving forward, more research into personal vehicle sharing is need-
ed to assess market potential and its social and environmental impacts.
Future research opportunities include: understanding geographic and
land use opportunities, demographics and psychographics, economic vi-
ability, business model evolution, and market growth. While this paper
assesses industry developments in its nascent stages, continued tracking
and assessment of future developments is recommended.
Acknowledgments
The authors express their gratitude to the 34 specialists who provided
valuable expertise to this study through expert interviews. They would
like to thank Guy Allen, Eric Murugneux, and David Rolnitzky for
their help in initial data collection and questionnaire development, as
well as results from their class project on P2P carsharing conducted at
UC Berkeley. They also acknowledge Shelby Clark, Dave Brook, Kevin
McLaughlin, and William Curtis for their assistance in questionnaire
pre-testing. The contents of this report reflect the views of the authors,
who are responsible for the facts and the accuracy of the data presented
herein.
References
Autoshare (2003). About us: In the news/PR. (online) Available at:. www.autoshare.
com/aboutus_news.html (Accessed July 23, 2008).
Brook, D. (2012). Portland Carsharing Choices. [online] Available at http://www.carsharing.us
[Accessed April 20, 2012].
California Assembly Bill 1871 (2011).An act to addSection 11580.24to the Insurance Code,
relating to motor vehicle insurance coverage. (Online) Available at: http:/ /info.s en.ca.
gov/pub/09-10/bill/asm/ab_1851-1900/ab_1871_bill_20100212_introduced.html
(Accessed February 2, 2011).
Cervero, R., Creedman, N., Pohan, M., Pai, M., & Tsai, Y. (2002). City carshare: Assessment
of intermediate-term travel-behavior impacts. Working Paper 2002‐02. Berkeley:
Institute of Urban and Regional Development, University of California.
Cervero, R., Golub, A., & Nee, B. (2007). City carshare: Longer-term travel demand and
car ownership impacts. Transportation Research Record: Journal of the Transportation
Research Board, 1992,70–80.
Cervero, R., & Tsai, Y. (2004). City carshare in San Francisco, California: Second-year
travel demand and car ownership impacts. Transportation Research Record: Journal
of the Transportation Research Board, 1887, 117–127.
City CarShare (2004). First-ever study of car-sharing. (Press release), 12 January 2004,
Available at: http://www.citycarshare.org/pressrelease_01-12-04.do (Accessed 18
August 2011).
Cohen, A., Shaheen, S., & McKenzie, R. (2008). Carsharing: A Guide for Local Planners.
[online] American Planning Association. Available at: http://76.12.4.249/artman2/
uploads/1/pasmemo0508.pdf [Accessed 15 August, 2011].
Cooper, G., Howe, D., & Mye, P. (2000). The missing link: An evaluation of CarSharing
Portland Inc. Portland, OR: Oregon Department of Environmental Quality.
Dallaire, Y., Lafond, N., Lanoix, C., & Viviani, M. (2007). Le projet auto+bus: Évaluation
d'initiatives de mobilité combineé dans les villes canadiennes. (Online) Communauto.
Available at www.communauto.com/abonn es/PT-CS_RapportFinal_ju l06.pdf (Accessed
17 August 2011).
Econsult Corporation (2010). The economic and environmental impact of PhillyCarShare
in the Philadelphia region. (Online) Available at: http://www.phillycarshare.org/wp-
content/uploads/2010/07/pcs-impact-study.pdf (Accessed 17 August 2011).
Hampshire, R. C., & Gaites, C. (2011). Peer-to-peer carsharing: Market analysis and
potential growth. Transportation Rese arch Record: Journal of the Tran sporta tion
Research Bo ard, 2217,119–126.
Jenson, N. (2000–2001). The co-operative auto network social and environmental report.
Vancouver, British Columbia, Canada: Co-operative Auto Network.
Katzev, R. (1999). Carsharing Portland: Review and analysis of its first year. (Online). Portland,
OR: Department of Environmental Quality Available at: bwww.publicpolicyresearch.
net/documents/ CSP_first_year_eval.PDF> (Accessed 18 August, 2011).
Katzev, R. (2003). Car sharing: A new approach to urban transportation problems.
Analysis of Social Issues and Public Policy, 3(1), 65–86.
Leiber, R. (2012). Fatal Collision Makes Car-Sharing Worries No Longer Theoretical.
[Online] The New York Times Available at: bhttp://www.nytimes.com/2012/04/
14/your-money/relayrides-accident-raises-questions-on-liabilities-of-car-sharing.
html?pagewanted=all> [Accessed 14 April, 2012].
Lane, C. (2005). PhillyCarShare: First-year social and mobility impacts of carsharing in
Philadelphia, Pennsylvania. Transportation Research Record: Journal of the Transportation
Research Board, 1927,158–166.
Martin,E.,&Shaheen,S.A.(2010).Greenhouse gas emission impacts of carsharing in
North America. (Final report). San Jose, CA: Mineta Transportation Institute
Available at: bhttp://76.12.4.249/artman2/uploads/1/Greenhouse_Gas_Emission_
Impacts_of_Carsharing_in_North_America.pdf> (Accessed 18 August, 2011).
Martin,E.,&Shaheen,S.A.(2011).Greenhouse gas emiss ion imp act s of car sh arin g
in North America. IEEE Transactions on Intelligent Transportation Systems, 12(4),
1074–1086.
Martin, E., & Shaheen, S. A. (2011). The impact of carsharing on public transit and non-
motorized travel: An exploration of North American carsharing survey data. Energies,
4,2094–2114, http://dx.doi.org/10.3390/en4112094.
Martin,E.,Shaheen,S.,&Lidicker,J.(2010).Theimpactofcarsharingonhouseholdvehicle
holdings: Results from a North American shared-use vehicle survey. Transportation Re-
search Board: Journal of the Transportation Research Board, 2143, 150–158.
Millard-Ball, A., Murray, G., ter Schure, J., Fox, C., & Burkhardt, J. (2005). TCRP report
108: Car-sharing: Where and how it succeeds. [Online]. Washington, D.C.: Transpor-
tation Research Board of the National Academies Available at: b http://onlinepubs.
trb.org/onlinepubs/tcrp/tcrp_rpt_108.pdf > [Accessed 17 August, 2011].
Millard-Ball, A., & Schipper, L. (2011). Are we reaching peak travel? Trends in passen-
ger transport in eight industrialized countries. Transport Reviews, 31, 357–378.
Mills, E. (2011). SF police arrest suspect in trashing of Airbnb Rental. CNET News. (On-
line), 29 July, Available at: http://news.cnet.com/8301-27080_3-20085741-245/sf-
police-arrest-suspect-in-trashing-of-airbnb-rental/ (Accessed 1 August, 2011).
Mittelstaedt, M. (2010). U.S. car ownership shifts into reverse, The Globe and Mail. (online)
Available at: http://www.theglobeandmail.com/news/world/us-car-ownersh ip-shifts-
into-reverse/article1418860/ (Accessed 17 August, 2011).
Newman, P., & Kenworthy, J. (2011). Peak car use: Understanding the demise of auto-
mobile dependence. World Transport, Policy & Practice, 17,31–39.
Price, J., DeMaio, P., & Hamilton, C. (2006). Arlington Carshare Progr am: 2006 Re-
port. (Online) Available at http://www.commuterpage .com/pdfdocs/ArlC o_
CarshareReport_2006.pdf (Accessed 17 August, 2011).
Price, J., & Hamilton, C. (2005). Arlington Pilot Carshare Program: First-year report.
(Online) Available at http://www.commuterpage.com/pdfdocs/ArlingtonCarshare
Program.pdf (Accessed 17 August, 2011).
Robert, B. (2000). Potentiel de L'Auto-Partage Dans Le Cadre d'Une Politique de Gestion
de La Demande en Transport. Presented at Forum de L'AQTR, Gaz à Effet de Serre:
Transport et Développement, Kyoto: Une Opportunité d'Affaire? Montreal, Canada.
Shaheen, S. A., Cohen, A. P., & Chung, M. S. (2009). North American carsharing: 10 year
retrospective.
Transportation Research Record: Journal of the Transportation Research
Board, 2110,35–44.
Shaheen,S.A.,Mollyanne,M.,&Wipyewski,K.(2003).USshared-usevehiclesurvey
findings on carshar ing and statio n c ar growth: Ob st acle s and opport unit ies.
Tra nsport ati on Research Record: Journal of the Transportation Research Board, 1841,
90–98.
Sullivan, E., & Magid, L. (2005). Bringing carsharing to your community. (Online) City
CarShare. Available at: http://www.citycarshare.org/download/CCS_BCCtYC_Long.
pdf (Accessed 17 August, 2011).
Walb, C., & Loudon, W. (1986). Evaluation of the short-term auto rental service in San
Francisco, California. [Online]. Cambridge, MA: Cambridge Systematics Available at:
b http://www.communauto.com/images/0 3.coupures_de_presse/WalbLoudon 1986_
B.pdf > [Accessed 18 August, 2011].
Zipcar (2006). Zipcar customer survey shows car-sharing leads to car shedding. (Press
release), 16 February 2009, Available at: http://zipcar.mediaroom.com/index.php?
s=43&item=108 (Accessed 17 August, 2011).
Zipcar Granted Graves Protection (2010). Auto rental news: For the car and truck rental in-
dustry. (Online) Available at http://www.autorentalnews.com/Channel/Legislative/
News/Story/2010/06/Zipcar-Granted-Graves-Protection.aspx (Accessed March, 2011).
11S.A. Shaheen et al. / Research in Transportation Business & Management xxx (2012) xxx–xxx
Please cite this article as: Shaheen, S.A., et al., Personal vehicle sharing services in North America, Research in Transportation Business & Man-
agement (2012), doi:10.1016/j.rtbm.2012.04.005