The

De­mystification
of Car Sharing
“What it is, what it’s not, and what it could be”
An in-depth analysis of customer perspective,
underlying economics, and secondary effects
August 2019

The Demystification of Car Sharing 1

 Key Messages
Only a small group of members uses car sharing on a regular

1
basis—the substitution effect between car sharing and private
car use is limited—it is primarily seen as a complementary
service, at best substituting public transportation.

People still value the inherent advantages of private car

2
ownership, such as permanent availability, privacy, and a feeling
of personal freedom, which will make it challenging to increase
car sharing’s penetration rate.

At the same time, current car-sharing members primarily choose

3 car sharing for economic and convenience reasons, resulting in
an economically-challenging business model.

The lack of major urban agglomerations and the relatively low

population density of many major European cities prove to be key
4 limitations to increasing the vehicle usage and financial success
of car sharing—in Germany, only 5 percent of the theoretical
market can be served in economically viable terms.

Recommendations for Action

Collaborate closely with city administrators to integrate car
1 sharing in the public transportation network and gain exclusive
benefits, such as reserved parking.

2 Select car-sharing cities based on population density, availability

of interconnected hot spot areas, and local government support.

Proactively use positive secondary effects of car sharing such as

3
brand image building, customer relationship management, and
data collection to balance likely financial losses (especially in the
start-up phase).

Target potential younger customers by marketing efforts, since

4 they’re more open to alternative value propositions based on
vehicle offerings compared to potential older customers.

1
Introduction
The purpose of this publication is to shed light on a publicly hyped mode of transportation: car
sharing. It’s gained significant presence in major cities around the world and has received
heavy media attention over the past few years. The end of personal car ownership is frequently
touted, with car sharing taking over and replacing this old-fashioned ownership model.
Additionally, it is high on the agenda of many automotive OEMs, who perceive it as a potential
cornerstone of their future business models, geared toward becoming more of a “mobility
provider” instead of a manufacturer and seller of vehicles.

However, recent developments in the industry speak both for and against a rapid advancement
of this mobility solution. While the global number of car-sharing members increases signifi-
cantly, major car-sharing players have removed their services from cities like London and
Copenhagen, which at first sight seem like obvious locations for car sharing.

In this publication, we will explore the parameters

of the car-sharing business model on three levels:
The estimated number
of global car-sharing 1. Customer Perspective

members has 2. Car-Sharing Economics

increased from 3. Secondary Effects for OEMs

7 million Our in-depth analysis aims to demonstrate to what

degree the service will reshape mobility, and what it
to 27 million will take for car sharing to become financially
between 2015 and 2018 successful going forward. We will focus on Western
markets (Europe and North America), where
car-sharing services are established in most major
cities. An emerging-markets perspective on car sharing, where accelerated urbanization and
growing megacities shape the mobility of tomorrow, will follow in a future publication.

A World of Change?
Multiple socioeconomic trends are currently at play and can potentially help the car-sharing
business model:

• Governments are taking action to increase sustainability and reduce emissions via changes in
their mobility-related policies.

• Digital technology is increasingly becoming part of our daily lives and is inherent in the way
we use many products and services, including how we organize our individual mobility.

• The sharing economy is gaining acceptance, particularly with millennials and younger gener-
ations, and thanks to connectivity through smartphones, it’s never been easier to gain access
to assets owned by corporations or other private individuals.

These trends have fueled car sharing offerings in recent years—this new business model has
sparked enthusiastic hopes for multiple parties, from policy makers to OEMs and new mobility
players.

The Demystification of Car Sharing 2

However, recent M&A activity in the industry raises questions about the financial performance
of car sharing. Moreover, consumers appear reluctant to let go of the traditional ownership
model. Will car sharing change the mobility industry as we know it today, or will it only grow to
become a complementary addition to the existing portfolio of mobility options?

Several withdrawals of car-sharing providers from

major cities and recent M&A activity highlight the
difficult path to profitability for the business model

Consumer and media attention to car sharing is a global phenomenon. Headlines like “City car
owners moving to car sharing schemes” (Sunday Times) and “Owning a car will soon be a thing
of the past” (The Guardian) clearly show the media’s interest. Consumers are showing similar
excitement as the global membership base of car-sharing platforms has grown from 3 million
members in 2013 to close to 27 million in 2018. Although our sources are aligned historically on
these numbers, forecasts of future trends show wide discrepancies for the anticipated future
member numbers.

In this section, we investigate the the reasons behind consumers’ increased interest in this
relatively new mobility option. Do consumers use car sharing as a mobility option to increase
convenience and flexibility, or is the demand purely financial? Will consumers take on car
sharing as a complementary addition to owning a car or will it become a complete substitute to
car ownership, thus reshaping the industry and mobility as we know it?

Figure 1
Global Development of Membership Base of Car-Sharing Platforms
and Fleet Size of Car Sharing

# of Car-Sharing Members (millions) Car-Sharing Fleet Size (’000 cars)

27 268
+32%
+58%

+29%
+42% 118
7 71
3

2013 2015 2018 2013 2015 2018

Members Fleet size CAGR = +xx%

Source: Frost & Sullivan, Berg Insight, A.T. Kearney

The Demystification of Car Sharing 3

With membership rates rising, the global fleet size of car-sharing vehicles reached 268,000 in
2018. As members grow faster than fleet size, the member-to-car ratio rises, thus improving
unit economics. Yet, implementing car sharing in major cities like London, Copenhagen, and
Stockholm has proven difficult for some players—we will explore what’s needed to foster a
profitable environment for car sharing.

Multiple OEMs, large car rental players, and new-to-the industry entrants are participating in
the car-sharing industry in a variety of setups. While their agendas may be financial, additional
reasons influence the choices made by these players. We will now investigate how OEMs can
benefit from offering a car-sharing service as it may increase brand awareness and improve
relationships with existing and new customers. Can these secondary effects justify operating a
car sharing service from the OEM perspective, even if the service itself is not financially
sustainable (yet)?

Customer Perspective
A first look at the customer side of the car-sharing ecosystem seems to suggest a positive
outlook for the business model—the number of car sharing members is increasing rapidly and
providers are proudly announcing the milestone of 27 million customers globally. To allow for a
more differentiated evaluation, A.T. Kearney conducted a survey with more than 1000
car-sharing members and non-members in Germany, the United States, and the United
Kingdom to investigate customer motivations, service satisfaction, and improvement potential.

User Behavior: What Do We Learn from Actual Usage?

The results of usage frequency are a first indicator that membership numbers do not equal
active customers. For the United States and United Kingdom, more than 50 percent of
car-sharing members use the service never or less than once per month. Germany has a slightly
better outlook with only 33 percent of car sharing members using the service never or less than
once per month. The higher share of frequent users in Germany can be explained with a more
mature market offering, which has been especially shaped by large car-sharing providers like
car2go, DriveNow, and Flinkster.

In terms of satisfaction with the car-sharing service, our survey signals significant differences
across the three analyzed markets. While the United Kingdom (-12) and in particular the United
States (-35) show a negative Net Promoter Score (NPS), car sharing members in Germany (+23)
are more likely to recommend their provider(s). Frequent users show a better NPS compared to
occasional users. Based on the NPS results, providers in the United States and United Kingdom
might look to Germany to identify best practices such as vehicle offering, availability, and
pricing.

To investigate potential substitution effects, we asked members in Germany to list the mobility
options they use in addition to car sharing. The results show that the share of private car use is
unaffected by an increasing frequency of car-sharing use. Thus, we can assume a limited
substi­tution between the private and shared car. This finding of limited substitution, based on
our customer survey, is also supported by an analysis of new car registrations in Hamburg and
Berlin. To assess whether the introduction of car sharing on a larger scale has affected

The Demystification of Car Sharing 4

Survey Results

How frequently do you use Would you recommend the car-sharing

the car-sharing service? service(s) you use to a friend or acquaintance?

47%
Net promoter score

31% 30%
27% All users -12
22% 22% 23%
17% -35
16% 13%
8% 12% 10% 23
6% 6% 6% 4%
1%
Frequent users 0
Never, Less 1-2x per 3-4x per Several Daily (>1x week) -22
even though than month month times 52
I have a 1x per per week
membership month
Occassional users -13
(<1x week)
UK US GER -39
18

Occassional users Frequent users

(<1x week) (>1x week) UK
US
26%
50% 50% 49% 51% 74% 51% 49% 52% 48% GER

female male female male female male female male female male

50% of US car-sharing members use the service never or Car-sharing members in Germany are more satisfied
less than once per month. gender difference in UK with the service offering

Which other mobility options do you use besides Which factors do you miss in a car-sharing
car sharing? (Example: Germany) service compared to a private car?

Share of users using Permanent availability 33%

23%
mobility option (%) of vehicle 39%

Privacy e.g., keep personal 30%

50% 24%
Public transportation belonging in the car 38%
40%
Feeling of personal 14%
Private 28%
30%
freedom/independence 31%
car

Lower costs in case 23%

20% Bicycle 30%
of regular use 31%
Rental car
10%
Special brand/ 20%
Private Motorbike 22%
model selection 18%
Never Less than 1-2x per 3-4x per Several UK
1x per month month month times per 16%
Personalization of the US
week or 19%
vehicle (e.g., colour) 13%
daily GER
Usage frequency of car sharing

With increasing usage frequency, car-sharing users in Germany Key arguments for owning a car can not be achieved
use less public transportation—the use of private cars and by a car-sharing service
motorbikes remains unchanged, signaling limited substitution

Source: A.T. Kearney analysis

customer demand for privately-owned cars, we compared the share of new car registrations for
each of the two cities to the total number of new car registrations in Germany. By looking at the
relative share instead of the absolute number of new car registrations, we can cancel out the
effect of potential overall market fluctuations, which are caused by macroeconomic forces. The
analyzed time periods are 2006–2011 and 2012–2017, respectively. This period was chosen as
from mid-2011, car2go and DriveNow started to roll out their services in Hamburg and Berlin. As

The Demystification of Car Sharing 5

shown in Figure 2, there has been no significant change in the share of new car registrations
since the introduction of these two large car-sharing service providers. Thus, market analysis
mirrors the results of our customer survey. Even without the possibility of car sharing, it’s
remarkable that, despite presumably growing traffic jams and ever more densely-populated
cities, city dwellers are still, as much as ever, following the traditional car ownership model.

Figure 2
Impact of Car Sharing on New Vehicle Registrations

Germany
(ø%-share of new
vehicle registrations for
total German market)
Berlin Hamburg
(3.5 million/6.0 million)1 (2.0 million/5.2 million)1

4.3 4.2

2.5 2.5 Car demand

stable in big cities
despite car-sharing
introduction

06-11 12-17 06-11 12-17

Pre CS Post CS Pre CS Post CS

1) Population (City district/metropolitan area) in millions

Source: Kraftfahrtbundesamt (KBA), A.T. Kearney

For public transportation, a different result becomes obvious—with increasing car-sharing

frequency, the share of public transportation use decreases, thus signaling a substitution
effect. This observation supports critics who question the sustainability advantages of car
sharing based on predominant substitution with public transportation. One opportunity to limit
this substitution is the integration of car sharing into the public transportation network. This
approach was followed for example by DriveNow in Copenhagen, where the company
partnered with Arriva, a local public transportation operator.

To estimate the long-term potential of car sharing’s position in the mobility landscape, we
asked members to list factors they miss in a shared car compared to a private one. Responses
indicate that, based on the current mood, car sharing will remain only one of multiple mobility
options next to owning a private car. This is because car sharing members rank aspects such as
permanent availability and storing personal belongings as key advantages of a private car.
Naturally, these are inherently difficult for a car-sharing service to provide.

Based on the current mindset of customers

regarding key advantages of owning a private car,
car sharing will remain only one of multiple
mobility options.

The Demystification of Car Sharing 6

Member Motivations: What Are “True” Drivers
of Members and Non-members?
In addition to analyzing actual user behavior, we were particularly interested in user and
non-user motivations for using or not using car sharing. What makes a user a satisfied and loyal
customer? And what hinders non-members to become car-sharing members?

Methodology Excerpt: Understanding “True” Relevance of

Drivers and Discovering “Hidden” Motivators
To decipher the “true” drivers behind user and non-user behavioral patterns, we applied
advanced analytics. We created two distinct analytical models for members and non-members,
each designed to identify the “true” relevance of drivers and motivational patterns. To also
understand potentially “hidden” motivations, we applied Structural Equation Modeling (SEM).
This methodology allows us to accurately and indirectly assess member and non-member
motivations, as we do not directly ask how important certain drivers and levers related to car
sharing are. Instead, we ask members and non-members to assess the validity of statements
related to the use of car sharing and use their answers feed a Partial Least Squares (PLS)
algorithm. Most customer investigations ask directly about the relevance of certain drivers, thus
frequently falling into the trap of, for example, socially desired answers. Our methodology is
thus designed to understand what motivates member and non-member behavior regarding car
sharing and also uncover otherwise hidden aspects of their decision-making.

Figure 3
Analytical Models

Car-Sharing Non-members Car-Sharing Members

Appreciation by friends
Expression of progressiveness
Distance to next vehicle Distance to next vehicle
Expression of environmentalism
Size of area of operations Size of area of operations
Social
Conve- Conve-
recogni-
nience nience
tion
Easy vehicle booking Usage Easy vehicle booking Usage
Easy usage of vehicles and Ope- Easy usage of vehicles and Ope- Recommendation
Availability of service contact rations Attractiveness of Availability of service contact rations to friends and
car sharing acquaintances
Attrac- Attrac-
Eco-friendly vehicles tiveness Probability of Eco-friendly vehicles tiveness Loyalty to
Attractive brands becoming a member Attractive brands car sharing
Vehicle Vehicle provider
Attractive models Attractive models
offering offering
Diversity of vehicle models Diversity of vehicle models
Vehicle cleanliness Vehicle cleanliness
Eco- Personal Eco-
nomics freedom nomics
Value for money Value for money
Personal freedom and independence
Price transparency Price transparency
Permanent availability of vehicles

• Sample size: 250 car-sharing members and 250 car-sharing • The Performance in the driver categories were used as independent
non-members in Germany variables used to explain the dependent variable, which is either the
• Rating different aspects of car sharing concerning performance service recommendation (for members) or the probability of
becoming a member (for non-members)
• Different aspects clustered into economics, vehicle offering, and
usage & operations
• Additional categories social recognition and personal freedom for
car sharing non-members

Source: A.T. Kearney analysis

The Demystification of Car Sharing 7

What Drives Car-Sharing Members to Become Satisfied and
Loyal Customers?
In a nutshell, the answer is favorable economics. This accounts for a full 38 percent of a
member’s satisfaction and loyalty—price transparency and good value for money specifically
are key levers for members to recommend car sharing. This is followed by the convenience
factor, predominantly explained by a desired short distance to the next vehicle and a large
operating area, which accounts for another 25 percent.

Together, favorable economics and short distances to the next vehicle in a large covered area
make up the recipe for a razor-thin margin business. Customers predominantly looking for the
cheapest offer and a vehicle around the next corner, thus driving up costs of the car-sharing
offering, significantly lower achievable margins for car sharing providers. The “true” preference
structures of car-sharing members mean making money with this service is a challenging task.

When we differentiate members into younger (18 to 29 years) and older segments (40 years+),
we see similarities, yet also striking differences. For the older group, rational economic reasons
(47 percent) are the most relevant. This is also true for the younger target group of members,
but to a lesser extent (34 percent). Older members are thus primarily “economically forced” to
use car sharing, which explains the high relevance of price for this target group. In contrast,
younger members care very much about the vehicle offering, which becomes their main
satisfaction and loyalty driver (38 percent). This high relevance of concrete vehicle offerings for
younger members allows for (profitable) differentiation of car-sharing providers. In combination
with a relatively low relevance of costly driver convenience (2 percent), this makes younger
car-sharing members a more promising customer group (compared to the older demographic
of members) from a profitability perspective.

What Is Needed to Turn Non-members into Advocates?

We now turn to people who are not yet members of a car-sharing offering. It becomes apparent
that a powerful bundle of levers for converting them to members can be found in the “social
recognition” aspect of car sharing (accounting for 37 percent of what could increase
non-members’ likelihood of becoming car-sharing members)—expressions of progressiveness
and environmentalism and appreciation by friends are the main factors that increase the appeal
of car sharing for non-members. Nevertheless, non-members don’t feel that car sharing
currently benefits their social recognition. This indicates that marketing efforts of providers may
want to focus on promoting a progressive and socially valuable image of car sharing in the
future.

The second most important factor is the apparently missing feeling of “personal freedom”
(relevance score of 22 percent)—providing personal freedom and independence, which is also
expressed by permanent vehicle availability, is an almost unachievable task for car-sharing
providers when compared to the user benefits of having a private car. Simply making it cheaper
(9 percent relevance) and providing better brands and models (21 percent) will hardly turn
non-members into members. The same holds true for increasing convenience (7 percent). In
sum, non-members clearly don’t shy away from car sharing due to economic or convenience
aspects. Rather, today, it is due to the perceived lack of freedom and independence,
non-permanent availability of cars, and insufficient social recognition.

The Demystification of Car Sharing 8

For non-members, we also differentiated our analysis for younger and older demographics.
Overall, older non-members are significantly more skeptical about car sharing than younger
ones, indicated by a much lower attractiveness score (on average, three out of ten for older
non-members compared to a similarly low five for younger non-members). Going forward, this
exceptionally low attractiveness score for older non-members will prove hard to overcome.

The biggest gap in demographics regarding the relevance of individual driver categories comes
from “personal freedom,” where it seems like younger non-members do perceive this as signifi-
cantly more relevant than older ones. For older non-members, it seems like the “traditional”
lever of what vehicles (brands, models) are being offered (30 percent) could, at least theoreti-
cally, convert them—this is significantly less important for younger non-members (1 percent).
Differences can also be found in the importance of economics: while this category ranks high
for young non-members (20 percent), it is of almost no importance for the older group
(4 percent).

Figure 4
Drivers of Satisfaction and Loyalty (Members) and Offering Attractiveness
and Likelihood to Participate (Non-members)

Car-Sharing Non-members Car-Sharing Members

Relevance Economics 9% Economics 38%

of drivers
Convenience 7% Convenience 25%
Vehicle
V
Vehicle offering 21% Ve
V hicle offering
Vehicle 21%
Usage and operations 5% Usage and operations 16%
Social recognition 37%
Personal freedom 22%

Age group Young Old Young Old

differentiation
Economics -16% Economics +12%
Convenience -7% Convenience +16%
Vehicle offering +30% Vehicle offering -22%
Usage and operations +-3% Usage and operations -7%
Social recognition +11%
Personal freedom -15%

Levers for
improvement Personal freedom Price transparency
and independence Others
22% 20%
Others 32%
37%
Expression of Value
15% progressiveness 18% for money
Availability
5%
6% of eco-friendly
Value 11% vehicles 8%
10% Appreciation
17%
for money
by friends Size of area Distance to
Expression of
of operations next vehicle
environmentalism

Source: A.T. Kearney

The Demystification of Car Sharing 9

Comparing Motivational Patterns of Members and Non-members
When comparing members and non-members and differentiating between the different
demographics, a few aspects become strikingly apparent and need to be emphasized:

• While it’s predominantly an economics and convenience game for members, non-members
can’t be persuaded to join the club by these arguments.

• Favorable economics are the predominant motivator for older members and younger
non-members to turn to car sharing.

• Conversely, the car sharing offering (specifically the brands and models offered) is of high
relevance predominantly for younger members and older non-members.

Thus, a differentiated approach to target groups is recommended—and the differences

between segments do warrant a differentiated perspective on which potential customers can,
and which potential customers can’t, be served profitably.

The Demystification of Car Sharing 10

Car-Sharing Economics
From an economic perspective, we need to consider costs for operating a specific number of
cars in a city and revenues generated from time-based rentals of these cars. Both the number of
required cars on the cost side and the number of potential revenue-generating users depend
significantly on various parameters of the city or region in which the car sharing provider
operates.

Revenue and Costs of Operation

Based on a forsa study, a key customer criterion for considering car sharing as a mobility option
is finding a car within a walking distance of less than one kilometer.1 A minimum number of cars
per city is thus required to fulfill this condition, assuming cars are equally distributed in the
operation area. On a single-car level, the purchase price of the car, fuel (or electricity), mainte-
nance, and overhead are the most important cost categories. The specific costs per car for the
provider depend on several parameters:

• Car fleet: the type of car and depreciation period mainly influence the hardware cost of
the cars. Car-sharing services operated by OEMs will benefit from larger discounts on the
purchase price compared to independent providers. Smaller cars result in lower fuel costs
and lower depreciation and maintenance costs.

• Operating model: free-floating car-sharing services face additional costs for relocating cars.
Furthermore, they incur higher costs for more advanced hardware (for example, equipment
for car unlocking by smartphone, Internet connectivity). In comparison, stationary services
face additional costs for reserved parking lots.

On the revenue side, the most important factors are car usage and the pricing model of the
car-sharing provider. Based on studies and expert interviews, average car usage is below
10 percent per day. Parameters influencing car usage are:

• Number of users of the car-sharing service: besides population size, other factors like
population age and income level influence the size of the user group.

• Usage frequency: the usage frequency of the service by a user is determined mostly by the
attractiveness of the service. Car sharing is in competition with public transportation, taxis,
and private cars in terms of convenience and price. By increasing the density of cars in the
operation area, the average distance to the next available car is decreased, which increases
usage frequency.

• Ride duration: the duration of a single ride depends on driving speed and distance traveled.
The duration of a ride is generally longer in large cities, because the distance between points
of interests increases as driving decreases due to higher traffic congestion. On the other
hand, car sharing becomes less attractive in cities with slower-moving traffic, since most
providers offer a time-based payment model.

1 Study “Car-Sharing in Großstädten,” forsa, 2016

The Demystification of Car Sharing 11

Figure 5
Factors Influencing Revenue and Cost of a Car-Sharing Provider

Cost Calculation

Number of cars/area Daily costs per car (in €)

(walking distance <1 km) (purchase, fuel, maintenance, overhead)

10.8
9.0
3.1 8.1
Overhead 2.6 1.9
2.7 1.2
Maintenance 2.7
1.8 1.7
Fuel/Energy 1.7
Car purchase 2.0 3.2 3.3
Free- Free- Local
floating floating operator
(small) (medium)

Revenue Calculation
Potential users in city Usage frequency Duration of ride
(based on current data) (based on current data) (based on current data and city size)

Daily profit (€/car)

Berlin Cologne Essen

15.8
13.3
10.8 10.8 10.8
9.0 8.9 8.1 8.7 9.0 10.2 8.1 9.0 8.1
5.8 3.9 4.7
2.6

Free- Free- Local Free- Free- Local Free- Free- Local

floating floating operator floating floating operator floating floating operator
(small) (medium) (small) (medium) (small) (medium)

Revenue Costs
Source: A.T. Kearney analysis

The Demystification of Car Sharing 12

The Effect of Population Density
Based on these parameters, it becomes clear that population density has a significant impact
on car-sharing profitability in a given city. Cities with a high population density will have a larger
number of users per car and hour, leading to advantages in terms of revenue. Detailed infor-
mation on theoretical profitability for major German and several international cities can be
found in the table below. Theoretical profitability is calculated based on:

• The number of cars needed to cover the hot-spot areas of the city (with the assumption that
the distance to the next available car is less than one kilometer)

• The potential user group living in the hot spot areas of the city

• The average member rate of car sharing in the country

• The usage frequency of car sharing members and the average duration of a ride

Figure 6
Comparison of Demographic and Economic Parameters for Major Car-Sharing Cities

City Population Population in Area Population Share of Car-Sharing Car Salary

(in thousands) Hot-Spot Areas (km2) Density Population in Profit Margin Owner Index
(in thousands) (1/km2) Hot-Spot Areas Rate

Berlin 3,520 1,783 892 3,946 51% 32% 29% 95%

Munich 1,450 422 310 4,677 29% 32% 33% 117%
Hamburg 1,790 427 755 2,371 24% 19% 33% 105%
Cologne 1,060 216 405 2,617 20% -5% 34% 107%
Düsseldorf 610 110 217 2,811 18% -9% 35% 101%
Frankfurt 730 83 248 2,944 11% -80% 30% 119%
Dortmund 585 25 280 2,089 4% -1,056% 41% 101%
Essen 580 94 210 2,762 16% -131% 40% 105%
Bremen 555 30 327 1,697 5% -1,025% 35% 105%
Dresden 545 97 329 1,657 18% -250% 34% 97%
Hannover 530 71 204 2,598 13% -196% 32% 101%
Nuremberg 510 71 187 2,727 14% -172% 37% 101%
Milano 1,190 182 6,538 9%
Vienna 1,790 415 4,313 16%
Brooklyn 2,560 180 14,222 -46%
Miami 454 143 3,175 -154%
Prague 1,200 496 2,419 -55%
Not available Not available Not available
Paris 2,200 109 20,183 81%
Zurich 403 92 4,380 14%
Kopenhagen 780 88 8,864 58%
Helsinki 635 214 2,967 -26%
Brussels 1,175 161 7,298 49%

Source: Statistisches Bundesamt, city websites, A.T. Kearney analysis

Even though mean population density and absolute city population size can be indicators for
the financial success of car sharing, it’s more important to look at a city’s detailed population
distribution. Significant car use can only be achieved in areas with a high local population
density—so-called hot-spot areas. In these areas, the operator can afford a high density of cars,
which ensures a short distance to the next available car and thus makes car sharing a conve-
nient mobility option.

The Demystification of Car Sharing 13

The right threshold for local population density can be identified by looking at the development
of revenue and profit as a function of the area of operation with a certain cut-off in local
population density. Assumptions for this calculation are:

• Cities are divided into official districts (Berlin: 96, Cologne: 86, Munich: 26, Hamburg: 104) for
which area and population are known.

• Districts for the operation area are chosen based on different threshold levels of population
density (from 3,000 to 15,000 people per square kilometer).

• The total area and population of the operation area is calculated.

• The number of cars in the operation area is calculated from the area (with a maximum walking
time to a car of 5 minutes).

• Potential users are calculated from the total number of people living in the operation area.

• Revenue and costs are calculated from the number of potential users and number of cars
needed (as described above).

Figure 7
Profit, Revenue, and Profit Margin as a Function of the Area of Operation Based on
Local Population Density (sample: cities of Berlin, Munich, Hamburg, and Cologne)

Profit
Revenue
Profit margin

Source: A.T. Kearney

This figure shows the aggregated result for the four biggest cities in Germany (Berlin, Munich,
Hamburg, and Cologne). If the threshold for population density is lowered, the profit margin
decreases, as the ratio between potential users and necessary cars is lower. Although these
areas are less profitable, they contribute to total revenue, which therefore increases. Due to
both effects, a plateau can be found in the total profit for threshold values between 6,000 and
10,000 people per square kilometer . By choosing the operation area based on these conditions,
a car-sharing service can operate in the most profitable way.

The Demystification of Car Sharing 14

Replacement of the Car Ownership Model:
5 Percent Is Not a Revolution
Assuming a minimum population density of 6,000 people per square kilometer, only a small
part of the city population qualifies for the user group who can be served profitably (see Figure
7). By comparing different cities with a similar mean population density, significant differences
can be identified in the share of people living in hot-spot areas (for example, 24 percent in
Hamburg, 21 percent in Cologne, 11 percent in Frankfurt). In addition to the number of people in
hot-spot areas, it’s important that the areas are connected to each other and not dispersed
throughout the city (see Figure 8). This ensures a profitable area of operation and reduces
relocation costs.

Figure 8
Calculated Areas of Operation for Hamburg, Cologne, and Frankfurt

Hamburg Cologne Frankfurt

Hot-spot area (population density >8,000 people/km²) Extended area (population density >6,000 people/km²)
City center/tourist areas Population density < 6,000 people/km²)

Source: Statistisches Bundesamt, city websites, A.T. Kearney analysis

Based on the analysis of hot-spot area inhabitants in all German cities with a population of more
than 500,000, only 4 million people qualify as potential users. This translates to a mere
5 percent of the German population. Assuming these 5 percent would switch from owning a car
to exclusively using car-sharing services (considering an average car owner rate of 57 percent in
Germany), this would in the long-term lead to a car park reduction of around 2 million cars in
Germany, or 5 percent of the total number of cars. For the rest of the population, car sharing can
only become a further mobility option in addition to owning a car or using public transportation
for the duration of a hot-spot area visit. These figures—and the full switch of the theoretical
5 percent of the population to car ownership is already a serious assumption—clearly indicate
that car sharing will not replace the ownership model.

The Demystification of Car Sharing 15

Figure 9
Share of the German Population Able to Participate in an Economically-Successful
Car-Sharing Service

Hamburg Hamburg

Bremen Bremen
Berlin Berlin
Hannover Hannover
Essen Essen
Dortmund Dortmund
Dusseldorf Leipzig Dusseldorf Leipzig

Cologne Dresden Cologne Dresden

Frankfurt Frankfurt

Nuremberg Nuremberg

Stuttgart Stuttgart

Munich Munich

83 million people 13 million people 4 million people

(Germany) (large cities with (hot spots in large cities with
population >0.5 mn) population density >8,000 people/km²)

Source: Statistisches Bundesamt, city websites, A.T. Kearney

Other Factors Influencing Economic Success

We can observe the significance of population density in the financial success of car sharing in
a given city by looking at Miami, where car2go ceased operations due to low car usage.
However, it’s not only population density that affects the success of car sharing in a city. Cities
like Stockholm are good candidates from the perspective of population density, but in 2018,
DriveNow had to discontinue its service after three years in operation because city-specific
congestion charges and parking fees decreased profitability. Furthermore, administrative
complexity can be responsible for the failure of car sharing. In London, car2go initially received
an operation permit for only three of the 32 London boroughs. The company failed to convince
the other borough administrations, which individually decide on parking permits and charging
stations. As a result, the area of operation remained too small and car2go had to discontinue its
service after 18 months.

The Demystification of Car Sharing 16

On the other hand, big car-sharing providers operate in cities with negative calculated profit
margins—so clearly, secondary effects need to be considered when starting a car-sharing
service in a city. The merger between car2go and DriveNow will have a significant impact on the
profitability of the cities in which they are operating. By merging the two car fleets, the
companies will be able to either reduce the number of cars in each city (leading to advantages
on the cost side) or attract a larger number of users because of higher availability and shorter
walking distances to cars.

Only cities with a high population density and

connected hot-spot areas are suitable for financially
successful car-sharing operations

Figure 10
Crucial Success Factors in the Economic Success of Car Sharing for a City Sharing Provider

City size (>0,5mn people) Local population density (> 6,000 people/km2)

Necessary amount of users and districts with Districts to operate car sharing service profitabily
high local population density

Agreement with administration Moving traffic (average speed >15 km/h)

Permit for operation, parking, and placement Otherwise, slower than public transportation and
of infrastructure (e.g., charging stations) not much cheaper than taxis

Parking Situation Population Demographics

Short times for finding parking lot, otherwise Enough young people accepting
car sharing too expensive car sharing as a mobility option

Source: A.T. Kearney analysis

The Demystification of Car Sharing 17

Secondary Effects for OEMs
Following the profitability assessment of car sharing, there are additional aspects that can
justify operating a proprietary service from the OEM perspective. While car sharing may not
currently be regarded as a financial investment, it can certainly be considered a strategic one.
OEMs can generate so-called “secondary effects” that can balance the financial losses their
services cause if fleet usage is still insufficient to break even.

Brand Image
As car sharing is one of the more frequently discussed topics regarding new business models in
the automotive industry, getting involved with a proprietary service has a direct effect on an
OEM’s brand image. The effect can be differentiated in two ways: the perception of an OEM’s
sustainability and the perception of an OEM’s progressiveness.

One of the key pillars of the sharing economy is sustainability. Sharing assets results in the
consumption of fewer resources. To a certain extent, this also holds true for car sharing. If two
people share one car, instead of buying two new ones, less raw material and energy are
consumed for production and assembly. Similarly, only half of the physical space is needed, so
cities could, for example, prioritize green spaces over parking areas. However, a reduction in
emissions compared to private car ownership can only be achieved if car sharing extends to
ride-sharing (with multiple passengers) or if the shared car has a lower fuel consumption than
the private car.

Since car sharing is still a novel business model for major OEMs rather than an industry
standard, running a proprietary service signals progressiveness and the ability to adopt new
trends. Furthermore, it showcases that the OEM recognizes changes in the mobility landscape
and is willing to adjust its offering to serve new client needs. In a time of continuous brand
rejuvenation facilitated through design language, technology development, and marketing
campaigns, car sharing presents an authentic way for OEMs to communicate the zeitgeist to
existing and potential new brand customers.

Customer Relationship Management

Getting involved with a new customer can be a challenging and expensive process for car
brands. The broad spectrum of competitors combined with the high financial burden
associated with car ownership make new customers a very valuable currency for brands. The
annual report of the United States’ National Automobile Dealers Association (NADA) measured
that the advertising expense per new vehicle retailed at $624 in 2018. This amount has proven to
be stable with only a slight decrease from the respective 2017 amount of $629.

Car sharing offers OEMs the opportunity to present their products to a wider audience in a
cost-efficient way. In many cases, this audience is young and, outside of car sharing, would not
experience the brand’s products at their age. Car sharing thus enables OEMs to establish
relationships early in the customer life cycle. This can positively influence brand loyalty and

The Demystification of Car Sharing 18

increases the likelihood of brand interactions beyond using the car-sharing service. A customer
can, for example, promote the brand’s products in front of his or her peers. In case of a
purchasing decision later in the customer life cycle, the brand is automatically part of the
customer’s awareness set. Since the customer is already familiar with the brand and its
products, the purchase involves less uncertainty for him or her.

Due to this valuable opportunity, OEMs try to present themselves and their products in the best
way possible. This is one reason why car-sharing fleets are closely monitored in terms of age,
mileage, and mechanical and visual condition.

Furthermore, car sharing can lower the entry barriers for potential customers to try products.
An official dealership test drive usually comes with several inconveniences. The customer must
convince the salesperson that he or she is interested and in a financial position to purchase the
car. From here, the customer and salesperson’s schedules need to align, and the customer
needs to drive to the dealership, which is often located outside the city center. Finally, after the
test drive, the customer usually needs to evaluate the car, which may entail potential follow-up
calls from the salesperson. In a car-sharing scenario, the customer can integrate the test drive in
his or her daily routine without interacting with the dealership at all. This makes the process
leaner, less binding, and more spontaneous for the customer. The dealership also benefits, as
the customer pays for the test drive and does not consume any resources at this early stage of
the potential purchasing process. Based on expert interviews, we estimate the average cost of a
one-hour test drive at roughly €40, including human resources and vehicle depreciation.

The lowered entry barriers brought on by car sharing can be particularly well-observed for more
exotic models. BMW has chosen to integrate the electric i3 in its DriveNow fleet and Daimler’s
car2go added ten Mercedes-AMG CLA 45 models to their Munich and Hamburg fleets between
November 2017 and October 2018. Both models have a high potential to excite customers and
strengthen their relationship with the brand. However, due to their exclusivity, only a limited
number of customers can experience them outside of the car-sharing context.

Data Collection
Another positive secondary effect that OEMs can profit through their car-sharing service is the
opportunity to collect data, both on customers and cars.

On the customer level, the OEM receives contact data. If the customer agrees to extended user
conditions, he or she can be contacted regarding offers from the car-sharing service itself or
even from the OEM brand. Furthermore, the OEM can track customer movements within the
operating area of its service, preferences for specific models depending on customer
demographics, and the usage of specific vehicle features (like the infotainment system). Finally,
the car-sharing customer base offers an additional channel for direct product feedback.

On the car level, the OEM can analyze the durability of its products as input for future R&D
improvement measures. Short-distance city driving with constantly changing drivers presents
one of the toughest tests in terms of material stress. Analyzing wear and tear patterns of
car-sharing vehicles can therefore serve as an early warning system, which complements the
OEM’s own test drives for continuous product improvements.

The Demystification of Car Sharing 19

Fleet Aspects
Car sharing’s positive secondary effects for an OEM’s fleet can be differentiated between
emission penalties and demand volatility.

From 2021 on, the fleet of car manufacturers in Europe must not exceed 95 g of carbon dioxide
emissions per kilometer. For every gram above this threshold, a penalty of €95 per sold car will
be charged. A recent study from PA Consulting2 estimates that six out of eleven manufacturers
active in Europe will not achieve this target and face annual penalty charges between €0.2 and
€1.7 billion, depending on their average fleet emissions and number of cars sold.

An OEM’s car-sharing fleet, equipped with Bonus/Penalty

electric vehicles or low-emission economy (in €)
cars, could counteract these penalties by
lowering the fleet’s average carbon dioxide 9,025
emissions. Naturally, the impact of an OEM’s
car-sharing service on fleet emissions can
only become relevant if the service’s fleet
665 -5,130
size is significantly increased. However, on a
single-car level, the planned carbon dioxide
penalty could significantly improve the
business case for an electric vehicle in a Electric Economy Premium
car-sharing scheme. If electric vehicles vehicle vehicle vehicle
0 g/km 88 g/km 149 g/km
continue to be certified with an emission of CO2 of CO2 of CO2
standard of 0 grams per kilometer of carbon
dioxide, as BMW’s i3 and smart electric drive Bonus Penalty

currently are, each vehicle would carry an Source: A.T. Kearney analysis

artificial bonus of €9,025.

Similar to balancing fleet emissions, a car-sharing fleet could be used by an OEM to balance
demand volatility in its model portfolio. In the event that sales for a specific model drop and
factory production cannot be shifted to other models, the car-sharing fleet can function as a
temporary production capacity buffer. Certainly, current fleet sizes are not sufficient to serve
this purpose but increasing use of car-sharing offerings and the resulting growth of fleets could
make these aspects another positive secondary effect of car sharing in the future.

Car sharing presents an authentic way for OEMs to

communicate the zeitgeist to existing and potential
new brand customers

2 Study “The CO2 Emissions Challenge,” PA Consulting, 2017

The Demystification of Car Sharing 20

Outlook
A more differentiated look at car sharing has shown that there are many parameters to consider
beyond the growth figures of memberships and fleet size. While customers acknowledge the
benefits of the business model, they are aware of its current limitations and in many cases
regard car sharing as an additional mobility option rather than a fully-fledged substitute to
private car ownership. Current members who predominantly seek to combine the lowest cost
with the highest convenience will continue to challenge providers’ business models. Increasing
network density is thus an important way for providers to make the service more convenient
and boost usage. To achieve this, a close collaboration between providers and city adminis-
trators is required. This can facilitate a better integration of car sharing in public transportation
and ensure advantages over private vehicles such as reserved parking. From a financial
perspective, a boost in usage will decide whether car sharing can break even as an independent
service or whether it remains dependent on positive secondary effects for OEMs.

We will continue to observe the market and the impact of technological trends such as drive-
train electrification and autonomous driving. A future A.T. Kearney publication will analyze the
business environment for car sharing in emerging markets. Higher population density
combined with rapidly-growing metropolitan areas and a lower car-ownership rate will change
important fundamentals of the business model in these markets.

Authors

Dr. Wulf O. Stolle, partner, Berlin Wilhelm Steinmann, consultant,

[email protected] Düsseldorf
[email protected]

Vincent Rodewyk, consultant, Munich Angel Rodriguez Gil, consultant, Madrid

[email protected] [email protected]

Astrid Peine, senior research expert,

Düsseldorf
[email protected]

The Demystification of Car Sharing 21

A.T. Kearney is a leading global management consulting firm with offices in more
than 40 countries. Since 1926, we have been trusted advisors to the world’s foremost
organizations. A.T. Kearney is a partner-owned firm, committed to helping clients
achieve immediate impact and growing advantage on their most mission-critical
issues. For more information, visit www.atkearney.com.

Americas Atlanta Dallas San Francisco

Bogotá Detroit São Paulo
Boston Houston Toronto
Calgary Mexico City Washington, D.C.
Chicago New York

Asia Pacific Bangkok Kuala Lumpur Seoul

Beijing Melbourne Shanghai
Brisbane Mumbai Singapore
Hong Kong New Delhi Sydney
Jakarta Perth Tokyo

Europe Amsterdam Ljubljana Paris

Berlin London Prague
Brussels Madrid Rome
Bucharest Milan Stockholm
Copenhagen Moscow Vienna
Düsseldorf Munich Warsaw
Lisbon Oslo Zurich

Middle East Abu Dhabi Dubai Johannesburg

and Africa Doha Istanbul Riyadh

For more information, permission to reprint or translate this work, and all other
correspondence, please email: [email protected].

The signature of our namesake and founder, Andrew Thomas Kearney, on the cover
of this document represents our pledge to live the values he instilled in our firm and
uphold his commitment to ensuring “essential rightness” in all that we do.

A.T. Kearney Korea LLC is a separate and independent legal entity operating under the A.T. Kearney name in Korea.
A.T. Kearney operates in India as A.T. Kearney Limited (Branch Office), a branch office of A.T. Kearney Limited,
a company organized under the laws of England and Wales.
© 2019, A.T. Kearney, Inc. All rights reserved.