The Integration of Driverless Vehicles in Commercial Carsharing Schemes in Germany: A Prefeasibility Study

Table of Contents

Introduction
0.1 Overview
0.2 Definitions
0.3 Structure of the Study
0.4 Rationale
0.5 Case Study: Germany
0.6 Aim and Objectives
0.7 Methodology
0.8 Limitations

Part 1: Carsharing
1.1 Introduction
1.2 Background
1.3 Social and Business Change regarding Mobility
1.3.1 From “Generation Golf” to “Generation Hire Car”
1.3.2 From Carmaker to Mobility Provider
1.3.3 Sustainability Concerns
1.4 Carsharing as a Response to Social Change
1.4.1 Generation Hire Car
1.4.2 Carmaker or Mobility Provider
1.4.3 Sustainability Concerns
1.5 Limitations of Carsharing
1.5.1 Parking
1.5.2 Availability

Part 2: Driverless Vehicles
2.2 Background
2.3 Social Issue Traffic Accidents
2.3.1 Background
2.3.2 The Role of the Driver in Car-related Traffic Accidents
2.3.3 Traffic Accidents in Germany
2.3.4 Economic Costs of Traffic Accidents
2.4 Automation in Passenger Vehicles as a Response to Traffic Accidents
2.4.1 Driver Assistance Systems
2.4.2 Advanced Driver Assistance Systems
2.4.3 Car Drivers, Customers, and Advanced Driver Assistant Systems
2.5 Fully-Automated Vehicles

Part 3: Prefeasibility Study
3.1 Introduction
3.2 The Concept
3.3 Micro-Environmental Analysis: SWOT
3.3.1 Strengths
3.3.2 Weaknesses
3.3.3 Opportunities
3.3.4 Threats
3.4 Macro-Environmental Analysis: STEPLED
3.4.1 Society
3.4.2 Technology
3.4.3 Environment
3.4.4 Politics
3.4.5 Legislation
3.4.6 Ethics
3.4.7 Demographics

Conclusion

Recommendations

Bibliography

Abstract

The prefeasibility study at hand proposes to integrate driverless vehicles in commercial carsharing schemes in Germany. It finds indications that carsharing has the potential to become a significant mode of transport because it responds to social and business change. On the other hand, the study shows that it is the human component in traffic causing the majority of accidents, which is why driverless vehicles are expected to reduce traffic accidents and improve road safety. By making use of the business management tools SWOT analysis and an ad hoc STEPLED analysis, the study conducts a micro-environmental and macro-environmental analysis of the German carmakers BMW and Daimler in regard to the proposed concept. The main finding is that the technology is generally expected to be marketable within the next two decades, if not much earlier. It can be expected that customers will accept driverless vehicle technologies and that many new target groups could be reached by the realisation of the concept. Carmakers should benefit from this potential and might be able to offset possible losses in their core business. The analysis does also show, however, that the current legislation poses several obstacles and might, if unchanged, delay the introduction of driverless vehicle technologies. It is recommended for carmakers to become active and start a public debate similar to the one in the United States. Moreover, it is recommended to collaborate closely with other mobility providers in order to avoid lobbying against the idea.

The last five months have been a challenging time and I want to thank all those that supported me. William Sheward, my supervisor, who always took the time to discuss my ideas, not only for this study, but also for other assignments and on several other occasions. His feedback was consistently valuable and the directions he gave me helped to focus on the essential while not neglecting the complexity at the same time. I also want to thank Paul Sheeran, course leader and tutor of my MA programme Managing Contemporary Global Issues, for uncountable inspiring discussions and also for always emphasising the sheer complexity of our world. I want to thank Natalia Yakovleva, another tutor of my programme, whose interesting and challenging lectures were packed with many different study methods – one a little game which eventually led to the idea for this study. Thanks also to my fellow students and friends, Adeniyi Asifat and Robert Finegold, with who I developed the idea in the lecture.

I also want to thank Nachiket Pusalkar, a true friend, who was always there for me, always listened to my ideas and supported me throughout the year; Jean Paul Pollock and Dan Wilkinson, for lively debates and feedback on my topic; my parents Ramona and Jürgen, who did their best to support me in every way; as well as my brother Marco, who distracted me every now and then which was equally important.

The biggest thank-you goes to my amazing girlfriend Echo Yu, soon to be my wife Echo Kowalski. She is always there, listening to me crazy ideas, challenging my concepts and predictions, and supports me in just about every possible way. Love you honey.

List of Tables

Table 1.Comparison of traditional and commercial carsharing schemes

Table 2.Proposed definitions for different stages of the automation of the driving task

Table 3.Activities car drivers said they would engage in while driving

Table 4.Risk Rankings

Table 5.How many car drivers answer and make phone calls while driving

Table 6.How many car drivers set up and take a reading from their satellite navigation while driving

Table 7.Expected demographic change in Germany by 2050

List of Figures

Figure 1.How often cars caused accidents with other road users and how many people died comparatively

Figure 2.Share of seriously-injured car drivers in total population

Figure 3.Accidents per 1,000 population and erratic behaviour causing them

Figure 4.Distribution of accidents in which people were injured (within city limits) over a week

Introduction

0.1 Overview

The world population recently reached 7 billion people and is expected to rise to 10 billion by 2050 (UN, 2012a [online]). At the same time the share of people living in urban areas has been steadily increasing, from 29% in 1950 to nearly 52% in 2010 (UN, 2012b [online]). Because people move to cities in search of work and a better life (Firnkorn & Müller, 2012 [online]), this number is expected to rise further as well, to more than 67% by 2050 (UN, 2012b [online]). Growing cities are the consequent result of both trends (Firnkorn & Müller, 2012 [online]) and have led to various problems for respective local communities, ranging from poverty and the creation of slums, to water scarcity, health issues or pressure on the local and surrounding environment.

One problem that is shared by urban agglomerations all over the globe is the increasing traffic density. In general, road transportation provides benefits to individuals and societies as a whole, such as enabling economic markets (WHO, 2009 [online]). However, the aforementioned trends concerning population growth and increasing urban populations are accompanied by the process of the motorisation of the individual. In the political East and South this process has, from a historical point of view, just started, while demand for individual motorised mobility in Western and Northern societies has not yet reached full saturation levels (Firnkorn & Müller, 2012 [online]; Shell Deutschland Oil GmbH, 2009 [online]). The consequences of all these trends, however, are negative for the individual and the community. Traffic accidents and the resulting congestion in combination with the increasing traffic density cause mental stress, physical pain and economic inefficiencies (IMechE, 2012 [online]; OECD & ITF, 2008 [online]; Shankar & Singh, 2012 [online]; Straube, 2011 [online]); United Nations (n.d. [online]): people die or get injured in traffic accidents (UNECE, 2012 [online]), properties are damaged (Statistisches Bundesamt, 2012a [online]), noise and exhaust emissions pose a danger to the health of locals (Senatsverwaltung für Stadtentwicklung, 2006 [online]), time of individuals is wasted (Bratzel, 2011 [online]), and the environment is damaged (Union Investment, n.d. [online]). This result, depending on regions, countries and cities admittedly in vastly varying degrees, is expected to worsen further with a continuation of the abovementioned trends. The study at hand suggests a new mobility concept which would make use of carsharing and driverless vehicle technologies, and has the potential to tackle the abovementioned disadvantages of individualised transport while offering the same and even additional benefits.

0.2 Definitions

Different administrative approaches have been undertaken to tackle traffic-related issues, such as banning specific road users from certain areas (Firnkorn & Müller, 2012 [online]), strengthening public transport (Reidenbach et al, 2008 [online]) or encouraging residents to walk or make use of bicycles (UNECE, 2012 [online]). Private people, associations and companies, on the other hand, may also become a key in contributing to the solution.

One of these grassroots approaches is called carsharing, which is based on the idea that several people can share one car while at the same time enjoying fewer disadvantages compared to possessing an own car. The concept, which started within small groups of private persons, offers a new transport mode with the potential to reduce car ownership and consequently traffic density (Martin, Shaheen & Lidicker, 2010 [online]). These environmental reasons have strongly contributed to the early attractiveness of carsharing (BFE, 2006 [online]) and led to the first schemes in the 1980s, which were organised by associations and clubs (Autotipps.net, n.d. [online]; Schlesiger, 2011 [online]). In these usually small traditional carsharing schemes, customers can find available cars at specific stations in a city and rent them spontaneously or after reservation (Lawinczak & Heinrichs, 2008 [online]). Customers are charged with a monthly basic fee and a price based on rental time or distance, which includes all costs such as petrol, taxes, insurance, maintenance and repairs (Lawinczak & Heinrichs, 2008 [online]).

Increasing numbers of participant and schemes started to indicate a strong market potential (Zhao, 2010 [online]) and eventually attracted corporations who commercialised the idea (Daimler, 2008 [online]; DriveNow, 2011 [online]). In these commercial carsharing schemes, companies increased the flexibility for customers because cars can be parked on all public parking spaces as well as on specific stations (Lawinczak & Heinrichs, 2008 [online]), while customers can find the next available car by using smartphone or internet applications (Firnkorn & Müller, 2012 [online]). Moreover, commercial carsharing schemes distinguish themselves from traditional ones because of their minute-based charging and the absence of monthly fees (Firnkorn & Müller, 2012 [online]) and environmental commitments (Kramper, 2012a [online]).

These current commercial carsharing programmes, the study at hand proposes, should be combined with driverless vehicles, also called autonomous vehicles (Kalra, Anderson & Wachs, 2009 [online]) or robot cars (Davis, 2006 [online]). The technology is not yet sold to the public but recent developments indicate that the technology is likely to be released eventually. Two different approaches can be observed in the development of driverless vehicles.

The first approach is characterised by the development of so-called driver or advanced driver assistance systems (DAS or ADAS) by companies from the automotive industry. These systems like the anti-lock brake system (ABS), the electronic stability programme (ESP) or some more advanced ones like the adaptive cruise control (ACC) and the autonomous emergency braking assistant (AEB), address specific deficits of drivers and seek to make driving saver (Kubitzki, J., personal communication [email], 17/7/2012). With the continuous improvement of existing and the constant development of new systems, it is believed that technologies will gradually improve until fully autonomous driving is eventually achieved (Bohr, 2011 [online]).

The internet search provider Google, on the other hand, aims directly at fully-autonomous driving. The company does not manufacture the cars itself but attaches various sensors and cameras to cars and hopes to be able to improve algorithms, which make sense out of the data which is gathered by sensors and operate the car, until no more human intervention is needed for the driving task (Vanderbilt, 2012d [online]). That this approach could also deliver a marketable product indicates Google’s fleet of several driverless vehicles which had travelled almost 500,000 km in August 2012 (Urmson, 2012 [online]). Impressed by this success, United States’ Nevada was the first administration in the world to allow the official testing of autonomous vehicles on public roads, whereas Google was the first company to receive a licence under these new regulations (Nevada DMV, 2012a [online]).

While Google has not yet released any official statements on when its technology may be ready to go on sale, the automotive industry widely believes that its approach will lead to marketable solutions at latest within the next 20 years (Bohr, 2011 [online]). Either way, the successful realisation of such technologies is expected to improve the safety of car drivers and other road users such as pedestrians and bicyclists immediately (Urmson, 2012 [online]) and, with more cars making use of this technology, improve the traffic flow and fuel efficiency of the overall traffic in the long term (Kalra, Anderson & Wachs, 2009 [online]).

The study at hand proposes to combine both components by integrating such driverless vehicles in commercial carsharing schemes in the future. Customers could use these vehicles much like taxis which pick up customers where and when they want and bring them where they want to go, thereby driving fully autonomously and offering customers to use the travel time for other purposes. Some limitations of current carsharing schemes, such as limited parking spaces and the need to find available vehicles, would be solved as well, and the overall traffic safety could be improved.

0.3 Structure of the Study

The study, which will use Germany as a case study (see 0.5), will start with an analysis of the carsharing concept in the first part. It will be shown that carsharing offers a relatively new form of mobility and thereby responds to a changing mobility behaviour caused by social and business change. Moreover, the current limitations of carsharing will be explored to understand how the proposed concept could tackle those weaknesses.

The second part is concerned with traffic safety and how improving driver assistance systems benefit road users. It will be shown that the vast majority of car-related traffic accidents are caused by erratic behaviour of drivers, and that traffic accidents are a huge societal issue in terms of costs and physical and psychological pain. The part will continue to introduce current ADAS to show how carmakers and suppliers make driving saver, before it will close with a short analysis of the status quo of driverless vehicle development in order to understand the problems of the technology that have to be overcome before the introduction, the limitations it may encounter due to its system limitations, as well as to understand the potential such technologies have to improve traffic safety.

The third part constitutes a prefeasibility study. It starts by first describing the proposed concept and suggesting some components which would be required for the realisation. In order to analyse how the concept would impact on companies, the study makes use of two business management tools, the SWOT and an ad hoc STEPLED analysis (see 0.6 Methodology). This part will help to understand possible obstacles to the realisation but also benefits the proposed concept might bring to providers.

The conclusion will discuss the results before recommendations close the study.

0.4 Rationale

The study at hand proposes to integrate driverless vehicles, when available, in commercial carsharing schemes and will be evaluated from the perspective of carmakers for several reasons. These providers do already offer carsharing schemes and build up competences in the field. The proposed concept would thus suggest a mere logical and evolutionary step to make use of more technologies in carsharing vehicles, which are developed by carmakers for their regularly sold cars anyway. Moreover, other than small private associations or large government entities, carmakers have the potential to create an economic market with fewer inefficiencies and a higher customer penetration. On the other hand, with the possibility that the individual mobility behaviour is about to change, away from private car ownership towards other modes of transport such as carsharing, carmakers’ core business of manufacturing and selling cars may be significantly affected (BMW, 2008 [online]). The proposed concept would offer carmakers several opportunities in the future to reduce the reliance on the core business and take advantage of latest developments, three of which are described in the following paragraphs.

First, making use of the concept can potentially increase sales of driverless vehicles. The technology is widely regarded as the future of the automobile (HAVEit, 2011b [online]) and is thus expected to be offered by carmakers eventually (Bohr, 2011 [online]). Letting go off the wheel completely and putting one’s life at a car’s mercy, however, may pose a huge barrier to many customers for the adaption of such technologies (ADAC, personal communication [email], 17/7/2012). To overcome these fears it had been shown that customers become accustomed better with innovative products when they can test and associate positive experiences with those (Jansson, 2011 [online]). This effect is already used in carsharing schemes in order to make customers familiar with electric cars (Herz, 2011 [online]) and could thus be equally used for driverless vehicle technologies.

Second, the proposed concept has the potential to strengthen a carmaker’s competitive position and subsequently ensure an important share in the growing carsharing market. As will be shown in the first part of this study, carmakers are increasingly seeking to enlarge their business area by becoming mobility providers (BMW, 2008 [online]). While posing a threat to carmakers’ core business by negatively affecting private car ownership, carsharing and the proposed concept in particular may offer a new business field with the potential to partly or completely compensate those losses.

Not only are these opportunities advantageous for companies, but the concept should also be considered for a third reason. Analysts assume that carsharing is an attractive market with a huge customer potential (Zhao, 2010 [online]). Carmakers will have to retain a competitive advantage over other providers (Schlesiger, 2011 [online]) in order to survive in a changing industry (BMW, 2008 [online]). Keeping in mind that Google’s system could be attached to almost every car (Lavrinc, 2012a [online]), which would enable basically everyone to offer a similar mobility concept, potential future competition needs to be considered especially in terms of first-mover advantages.

0.5 Case Study: Germany

The study will use Germany as a case study. Concerning the three trends mentioned at the beginning, Germany fits almost each other society in regard to traffic. Its urban population is expected to increase from currently 73.8% to nearly 82% by 2050 (UN, 2012b [online]) and the figures for private car ownership are expected to increase further (Shell Deutschland Oil GmbH, 2009 [online]). Only the overall population is expected to decline (Statistisches Bundesamt, 2012b [online]) which should therefore positively impact on traffic density.

The study will use Germany as a case study, moreover, because it perfectly meets the criteria to introduce smart mobility solutions (Schlesiger, 2011 [online]). Global market leaders are located in Germany, such as carmakers and suppliers, or manufacturers of electrical components and providers of telematics (Schlesiger, 2011 [online]), and its automotive industry is regarded to be the most innovative (Fuß, 2011 [online]). Because of its importance for the country’s economic performance, the automotive industry is also comparatively well-supported by political parties and respective communities. On the other hand, the German market for automobiles is seen as one of the toughest worldwide due to high expectations and requirements of customers. This, in combination with supportive economic and political conditions, may not only allow introducing the proposed concept in Germany, but also to test and improve it there, before exporting it to other cities, countries and continents.

0.6 Aim and Objectives

The aim of the study is first, assessing the current social and business environment for the proposed concept as well as evaluating the potential to reduce traffic accidents by making use of driverless vehicles. Taking the perspective of carmakers, this study is subsequently going to analyse the feasibility of the proposed concept and will examine first, the strengths and weaknesses of carmakers in terms of their potential to introduce the proposed concept and also which opportunities and threats may result from the introduction of such a scheme. The study continues to analyse current states and possible future developments of the macro-environmental elements society, technology, environment, politics, legislation, ethics and demographics regarding the proposed concept.

0.7 Methodology

This report, because driverless vehicle technologies are in an early stage, will be conducted as a prefeasibility study. This type differs from a feasibility study insofar that it does not conduct specific calculations or surveys to determine specific price range or market potential, but rather seeks to find out whether the idea indicates to be profitable, for example, because of customer acceptance, technologies or costs (Lohmann, 2010 [online]). The study at hand makes use of components suggested specifically for prefeasibility studies by Carsten Lohmann (2012 [online]) from the economisation and innovation centre of the German economy association, such as conducting internet and literature reviews and talking with potential suppliers, cooperation partners and internal employees.

The study at hand contains three main parts. In the first two parts the study will concern itself with the two components of the proposed concept, carsharing and driverless vehicle technologies. After giving an overview of the general topic, both parts start by reflecting current scholarly debates concerning the respective topic. The theoretical debates are applied to the case study Germany by using primary data, such as [1] news releases and financial reports of companies; [2] legal texts, publications and statistics from governmental institutions, as well as secondary data, such as [1] book, journal and news articles; [2] reports and statistics from researchers and private organisations; and [3] statements of experts. In the next step the study shows how both concepts, carsharing and driverless vehicle technologies, have the potential to respond to the respective societal issues. By showing at the end of each part the current limitations of carsharing and issues that have to be overcome for the introduction of driverless vehicles respectively, the study seeks to indicate the potential for improvement in carsharing schemes as well as giving a realistic outlook on the challenges until driverless vehicles will be available to the public.

The third part makes use of two business management tools, the SWOT analysis and an ad hoc STEPLED analysis. The former synonym stands for strengths, weaknesses, opportunities and threats, and will be used to analyse micro-environmental factors. The first section will focus on strengths and weaknesses of carmakers concerning the proposed concept while the second section will explore possible opportunities and threats for carmakers. The second analysis tool STEPLED is an ad hoc tool created from STEEPLED, which will help to understand the current and possible future development of the macro-environmental factors society, technology, environment, politics, legislation, ethics and demographics. Because of the long-term outlook of the study at hand, the economic element (the second E in STEEPLED) will be left out and thus the analysis will be subsequently called STEPLED. The reason for this is that economic conditions such as interest rates or unemployment rates will hardly influence the overall development towards the proposed concept but also because such factors would rather require attention when preparing the actual introduction of a business proposal. This part conducts the analysis of the developments presented in the first two parts and uses statements of experts in their respective fields, either collected from publicly available interviews or used in private correspondence with the author of this study.

The advantage of the abovementioned approach is that no other study has tried to combine carsharing and driverless vehicles, despite being a popular theme in science fiction movies and future scenarios. The study does not only bring together two different trends but, from the knowledge of the author, for the first time proposes and analyses the potential concept in terms of the feasibility from the perspective of carmakers. This combination indicates possible future consequences but also necessary steps that require action.

0.8 Limitations

The proposed concept makes use of a technology which is neither available to the public yet nor is there a guarantee that it definitely will be in the future. The general view in the automotive industry and of Google is that the technology will be marketable eventually, however, the study could become obsolete if developers would encounter problems concerning driverless vehicle technologies which cannot be overcome.

The study will furthermore not analyse the possible advantages of increased fuel efficiency and improved traffic flow because both benefits would only be realised after a sufficiently large number of road users had switched to driverless vehicles (Kalra, Anderson & Wachs, 2009 [online]).

Moreover, similar to the proposed concept are systems called personal rapid transit systems (CityMobile, 2011 [online]). Although some systems are already in operation, such as the ULTra PRT system at Heathrow Airport in London (CityMobile, 2011 [online]), these systems are not included in this research because their autonomously driving vehicles require separate lanes or tunnels. The proposed concept, on the other hand, would make use of the existing road infrastructure, without the need to invest in new infrastructure.

Part 1: Carsharing

There is no human right to possess an own car.

Chandran Nair founder and CEO of the think tank Global Institute for Tomorrow

1.1 Introduction

This part is going to introduce the first of two components required for the proposed concept, carsharing. The background section gives an overview of recent developments in the carsharing market, which indicate that carsharing has a strong growth potential and could be established as an alternative mobility concept.

The following section investigates observable social and business change regarding mobility to show that [1] the automobile may further decline as a status symbol in the future with decreasing private car ownership levels, that [2] carmakers have started to change their traditional business models to become mobility providers, and that [3] sustainability considerations play a role for customers and carmakers and have the potential to influence purchase decisions.

The subsequent analysis focuses on how carsharing can respond to these changes. It will be shown that carsharing is currently especially used by the young generation who could do without an own car but demand environmentally-friendly mobility with high flexibility and low costs. This behaviour indicates that mobility patterns in Germany may be about to change sustainably. Carmakers, it will be shown, are seeking to respond to such trends by extending their core businesses in order to offer new mobility services. Whereas criticism revolves around the problem that an increased carsharing engagement may negatively influence sales figures, carmakers expect to bind young customers to their brands so that their brands would be considered as an option for a possible future purchase. In this context, the example of the electric car will be used to highlight how carmakers introduce new technologies in carsharing schemes to make their customers familiar with innovative technologies. The very same approach, the study suggests, can be used to integrate driverless vehicles in commercial carsharing schemes.

The last section of this part will focus on the two main limitations of current programmes, parking and the availability of cars. This overview will be used in the third part of this study to show how the second component of the proposed concept, driverless vehicles, could tackle most of carsharing’s current weaknesses.

1.2 Background

Carsharing is an organised form of joint car usage and offers, from the perspective of the customer, a mode of transport placed between car hire and private car ownership (Lawinczak & Heinrichs, 2008 [online]). The most significant carsharing models in terms of customer numbers are traditional carsharing schemes, organised by associations or clubs, and commercial schemes, offered by carmakers such as BMW and Daimler, the maker of Mercedes automobiles.

In general, both types work similarly. After having signed up, customers can rent available cars either after reservation or spontaneously, and drive to their desired destination. Opposed to traditional carsharing, customers of commercial schemes are limited to drive within certain areas, usually city centres, while they are not required to return the car to specific stations (Finanztest, 2012 [online]). In commercial schemes subsequent customers can find the randomly parked cars via customer hotlines, internet or smartphone applications (Firnkorn & Müller, 2012 [online]). Another difference is that customers of commercial schemes pay a fixed price per minute whereas carsharing associations usually require their customers to pay a monthly fee and offer different prices depending upon the vehicle customers choose and the duration or length of the trip (Lawinczak & Heinrichs, 2008 [online]). In both types, however, fees include all costs such as petrol, parking, insurance, tax, cleaning, maintenance or repairs (Grundhoff, 2008 [online]; Kramper, 2012a [online]; Lawinczak & Heinrichs, 2008 [online]; Finanztest, 2012 [online]). Table 1 summarizes the major differences of traditional and commercial carsharing schemes.

illustration not visible in this excerpt

Having started in Germany in 1988 (Autotipps.net, n.d. [online]), carsharing was made popular by independent organisations (Schlesiger, 2011 [online]). More than 110 of these organisations operated 4,600 cars for 158,000 customers in 285 German cities and communities in 2011, growing on average by 15.8% annually since 2005 (Loose, 2011 [online]). In October 2008 Daimler was the first company to offer its own carsharing scheme called car2go. Having started as a pilot project with 200 Smarts for its employees in Ulm (Daimler, 2008 [online]), Daimler extended the scheme to the public in March 2009 because of its huge success (Daimler, 2009 [online]). Offering carsharing that was as easy to use as mobile phones (Grundhoff, 2008 [online]), car2go attracted more than 10% of Ulm’s population within the first three months (Daimler, 2009 [online]). Daimler established a joint venture with the car hire provider Europcar (Daimler, 2011c [online]) so that the scheme was able to expand quickly to other cities and countries (car2go, n.d.b [online]). So far, car2go has acquired more than 120,000 customers in twelve cities (Daimler, 2012f [online]) who have travelled around 20 million kilometres (Daimler, 2012d [online]). In the near future, car2go is planned to be offered in forty to fifty cities (Daimler, 2011 [online]) and, according to car2go’s CEO Robert Henrich, Daimler wants to set itself at the peak of this trend to become the market leader (Bund, 2011 [online]). Despite the overall success, the scheme currently operates at a loss. Daimler, however, sees a huge potential in carsharing (Becker, 2010 [online]; Daimler, 2010 [online]) and expects profits by 2014 (Bund, 2011 [online]).

Other market entries of BMW, Volkswagen and Peugeot reflect that Daimler is not the only carmaker to believe in the success of carsharing (Firnkorn & Müller, 2012 [online]). Because Volkswagen’s Quicar (VW, 2011 [online]) and Peugeot’s Mu (Mu by Peugeot, n.d. [online]) rather resemble traditional carsharing and car hire respectively, however, this work will primarily focus on Daimler’s car2go and BMW’s carsharing scheme called DriveNow. The latter started almost three years after Daimler’s car2go, in June 2011 (DriveNow, 2011 [online]). According to BMW’s CEO Norbert Reithofer (2012 [online]), the scheme has gained more than 27,000 customers in Berlin, Munich and Dusseldorf in the first eleven months and is planning to expand its programme to Cologne in September 2012 (DriveNow, 2012c [online]). Similarly to Daimler, BMW opted for a joint venture with a car hire company, SIXT, but tries to distinguish itself from competition by offering premium cars in its scheme (SIXT, 2011 [online]).

Of the major German carmakers, only Audi, which is sceptical towards carsharing schemes (Magenheim, 2011 [online]), and Opel, which struggles with its own survival, are not engaged in carsharing (Hucko, 2012 [online]). This is surprising because many studies see a huge potential for carsharing. According to research from business consultancy Frost & Sullivan (Zhao, 2010 [online]), for example, “carsharing membership is expected to reach 4.4 million in North America and 5.5 million in Europe by 2016.” By comparing the advantages and disadvantages of carsharing with the other mass motorised transport modes [1] private car, [2] hire car, [3] taxi, and [4] public transport, the following paragraphs show reasons for this potential.

First, an own car provides, according to research project of the Bundesministerium für Verkehr, Bau und Stadtentwicklung (BMVBS, Federal Ministry for Transport, Construction and Urban Development) conducted by Jana Lawinczak & Eckhart Heinrichs (2008 [online]), unrestricted mobility and serves as a status symbol. Whereas the latter is intangible and varies for each person, the former can be measured in costs. When comparing the costs of an own car and those for carsharing, different calculations have been made available. Considering that these comparisons usually focus on traditional carsharing, one must be aware that the explanatory power of the following paragraph is limited in regard to commercial carsharing.

In general, the literature agrees that carsharing is cheaper for people living in urban areas who drive less than 12,000 to 15,000 kilometres annually (Finanztest, 2012 [online]; Lawinczak & Heinrichs, 2008 [online]; Zhao, 2010 [online]). Finanztest (2012 [online]), an independent foundation testing financial products, calculated costs of €0.49 per kilometre for a private car compared to €0.36 per kilometre with the same car of a traditional carsharing scheme respectively. Because DriveNow and car2go charge their customers by time (each programme charges €0.29 per minute), it is hard to compare these results with the above mentioned costs. For example, when using the vague statistics published by car2go, the average price per kilometre ranges between €0.69 and €1.39. This calculation neglects, however, that the above mentioned costs calculated by Finanztest contained travel times on rural roads and motorways. It can be assumed that the average speed on these roads is higher, or from another perspective, that the average speed within urban areas would be lower. The costs of traditional carsharing schemes and private cars may thus be higher if rural roads and motorways were excluded from the calculation. Other cost advantages of carsharing over private car ownership are highlighted by Lawinczak and Heinrichs (2008 [online]). They point out, for example, that customers of carsharing schemes have no need and also no costs for a parking lot, do not have to finance a car individually, and save time which would be required for the maintenance of the car.

Other authors such as Gernot Kramper (2012a [online]) and Michael Specht (2010 [online]) from Stern magazine highlight, however, that carsharing is neither suitable nor economical for daily commutes. Moreover, because carsharing cannot substitute the private car, costs for carsharing add up to other costs such as for public transport, which increases the overall price of mobility for the individual.

Second, Lawinczak & Heinrichs (2008 [online]) state two advantages of car hire over carsharing in their article from 2008, which are now offset by the schemes of BMW and Daimler: the availability of one-way journeys and the absence of monthly charges. Car hire, however, is still advantageous over these commercial schemes for all customers that plan longer or out-of-city journeys and it is also cheaper than carsharing for all those that only occasionally need a car (Finanztest, 2012 [online]).

Carsharing, on the other hand, has many advantages over car hire, for example, that each new trip does not require a new contract, that stations and cars are located within urban centres which can thus be accessed easier by the majority of customers, that cars can be rented for very short time periods or that cars do not have to be refuelled after each journey (Lawinczak & Heinrichs, 2008 [online]).

Third, being driven in a taxi has several advantages over driving a vehicle in carsharing schemes oneself. For example, customers do not have to locate an available car and walk to it before each journey but can easily use customer hotlines or smartphone applications to call a taxi, and will be picked up exactly where they are. Once in the taxi, customers can engage in other activities and are not required to pay attention to traffic or even to be sober. The taxi driver can take more than the two customers which fit in car2go’s Smarts and take the customer to the exact destination while the passenger is not required to find a parking lot (Kramper, 2012a [online]). Moreover, when the Allgemeiner Deutscher Automobil-Club (ADAC, German Automotive Society) tested taxis in several European cities, they found that 92.5% of drivers in German cities helped with luggage and 82.5% knew the destination right away (ADAC, 2011a,b,c,d,e [online]), which can also be advantageous over carsharing.

However, the test also emphasised some weaknesses of taxis compared to carsharing. Two drivers made costly detours, two others did not pay attention to traffic rules and three drove too fast, one of them as fast as 92 km/h were 50 km/h were allowed so that he brought the tyres to squeak in curves. One driver almost caused an accident with a bicyclist and another drove over two red lights and so aggressive that the tester got sick (ADAC, 2011a,b,c,d,e [online]). As opposed to taxis, customers in carsharing schemes can drive in a way so that they feel save. Moreover, Finanztest (2012 [online]) highlighted that carsharing users only pay 40% of the costs of a taxi.

Forth, public transport in urban areas is usually a service that contains bus lines, trams, light-rail traffic and, in some instances, a metro. The service generally covers a wide area of a city as opposed to the rather small business areas which are covered by commercial schemes so far. While also being usually cheaper than carsharing, customers have to orient their journeys on fixed time tables, may have to change between (different) transport modes, including the walking time to get from one station to another, and in the majority of cases are neither picked up where they are nor brought to where they want to go. Carsharing customers as well have to go (and even find) vehicles each time they want to make a travel. On the other hand, customers can rent a car flexibly whenever they want and can drive wherever they want to go faster because they do not have to stop at all stations or walk to change the mode of transport.

Some of the abovementioned disadvantages with traditional transport modes, like walking to a car, finding a parking space or not being able to use the travel time for other purposes than driving, would be offset by the proposed concept. This would increase the benefits for customers and potentially increase the market potential of carsharing significantly (see part three).

1.3 Social and Business Change regarding Mobility

The following section looks at three aspects of cultural change that can be observed in regard to mobility. The first covers the social change among the young generation which could indicate a broader change of the German population in the future. The second section concentrates on the changing business behaviour in the automotive industry and the third section covers sustainability concerns which, it can be argued, increasingly become a driver for customers’ purchase decisions and companies’ business orientations.

1.3.1 From “Generation Golf” to “Generation Hire Car”

Although some studies have shown that the vast majority of young people enjoys driving (Progrenium, 2010 [online]; Steiler, 2012 [online]), Robert Schönduwe (2011 [online]) from the Goethe University Frankfurt am Main claims that the young generation is fundamentally changing: the Generation Golf, a concept termed by Florian Illies to describe how young people gained mobility and flexibility due to cheap and affordable cars such as the VW Golf that had become status symbols at the turn of the millennium, may become the Generation Hire Car, which refers to the book by Michael Adler that was released eleven years after Florian Illies’ book in 2011 and describes how the young generation opts for mobility as opposed to cars as status symbols.

Indeed, Ballweg and Ebbing, 2011 [online]) suggest that the old maxim ‘one driving licence, one car’ does not seem to be true anymore. According to the youth study Timescout, 75% of young people have a driving licence but only 45% possess an own car (Lamparter, 2010 [online]). While a study of the Kraftfahrt-Bundesamt (KBA, Federal Office for Motor Traffic) (KBA, 2012a [online]) indicates that the numbers of young people acquiring a driving licence seems to have decreased, the Center of Automotive at the University of Bergisch Gladbach found that 31% without a car still enjoy access to one (Bratzel, 2011 [online]). The same study, however, also found that the number of 18-24 year-olds who are living in urban areas and have no access to a car, increased from 21% to 33% within five years (Bratzel, 2011 [online]). Another study of a German mobility panel (Das Deutsche Mobilitätspanel, 2011 [online]), shows that that the group of people aged 18-35, who hold a driving licence and have a car at their disposal, has declined from 84.2% in 2000 to 67.5% in 2010, while figures for those holding a driving licence but who possess no car increased from 7.1% to 19.3% in the same period. A report from the Deutsche Automobil Treuhand (DAT, German Automobile Trust) (DAT, 2012 [online]) which measures the purchase intention for cars, shows that the number of those not planning to buy a car anymore, has increased from 4% in 2004 to 12% in 2012. To describe the observed data, Schönduwe (2011 [online]) suggests analysing the following determinants.

First, possessing a car has become too expensive for many people, not only the young (Schönduwe, 2011 [online]). The business consultancy Progenium (2012 [online]) compared the costs of small, medium-sized and premium vehicles between 1980 and 2012. The result is that the costs for small vehicles increased by 9%, the costs for medium-sized vehicles went up by a third and those for premium cars almost doubled. The ADAC (2010 [online]) published an index in 2010 which depicts how various costs had developed between 1995 and 2010. This index shows that whereas the general living expenses increased by 24.6% in this period, expenses for cars increased by 41.6%. These developments hit the young generation especially hard because they only have an average amount of €500 a month at their disposal (Bratzel, 2011 [online]). Thus Stefan Bratzel (2011 [online]) from the Center of Automotive believes that costs have become the number one reason why cars increasingly fail to connect to young people on an emotional level.

Second, the value orientation is changing away from owning to using, which means that a car is not seen as a status symbol by the young generation anymore but used as a mean of transport (Schönduwe, 2011 [online]). This assumption is supported by several studies, such as by Puls (Steiler, 2012 [online]), which found that roughly half of all participants see the car as a transport mode only, or by Progenium (2010 [online]), which found that 41% consider the car as a mode of transport only. The latter study also asked participants to rate different things according to their status potential: holidays, iPhones, watches and exclusive mountain bikes, among others, were all rated higher than a Smart, Golf or Opel, with only 17% of participants regarding an automobile as a status symbol (Progenium, 2010 [online]). Another study of the Center of Automotive found that almost half of the interviewees said they prefer spending money on expensive holiday trips or leisure activities rather than on cars (Bratzel, 2011 [online]).

Third, biographies in Germany have changed (Schönduwe, 2011 [online]). The trend towards single and two-person households (Statistisches Bundesamt, 2012b [online]) combined with a reproduction rate below 1.4 (Statistisches Bundesamt, 2012c [online]) and the development that women are giving birth to their firstborn later, significantly influence private car ownership (Schönduwe, 2011 [online]). Schönduwe (2011 [online]) highlights in this regard that all families with children but 4% possess a car, compared to 18% of the average German household. Thus Schönduwe (2011 [online]) suggests that more families would require more cars and could thus influence car ownership rates positively.

Forth, new mobility services are pushed into the market and offer new flexibility (Schönduwe, 2011 [online]). New media offering forums and social networks give more information on mobility services and have led, among others, to car-pooling and the increased use of carsharing or bicycle hire (Schönduwe, 2011 [online]).

These determinants imply that car ownership among young people might decrease, however, data is ambiguous. Shell Deutschland Oil GmbH (2009 [online]) and the Deutsches Zentrum for Luft- und Raumfahrt (DLR, German Aerospace Center) (DLR, 2007 [online]) predict that car ownership will rise, and the business consultancy Ernst & Young concluded their survey that cars will remain an attractive product for customers (Fuß & Forst, 2012 [online]). Other studies, however, show that almost 80% of young people aged 18-25 could imagine a life without cars (Bratzel, 2011 [online]) and, again 80%, believe that they do not need an automobile in cities (Lamparter, 2010 [online]).

Whether the changing car ownership behaviour is a long-term trend or just the result of delayed purchase decisions, for example, because of the delayed start of having a family, cannot be answered clearly. Schönduwe (2011 [online]) believes that a new mobility behaviour is the reality in urban areas.

1.3.2 From Carmaker to Mobility Provider

For more than a century, selling cars to private owners has been a successful business strategy (Firnkorn & Müller, 2012 [online]). Even though readmissions are on the rise (KBA, 2012a [online]), studies have shown that more European automobile industry experts expect stagnation or even declining growth rates in the premium segment, where German carmakers are commonly seen as the global market leaders.

In the light of these developments carmakers have started to rethink their business models. According to Kramper (2012a [online]), carmakers are changing from being sellers of automobiles to becoming mobility service providers in order to, as Daimler (2011a [online]) puts it, “offer customers mobility services that match the requirement for traffic concepts in urban areas.” While this may sound at first as a threat to automakers’ traditional business model of selling cars, Jörg Firnkorn and Martin Müller (2012 [online]) from the University of Ulm present three reasons that could be the driver behind carmakers’ decisions to the introduction of new mobility services.

First, Firnkorn and Müller (2012 [online]) argue that in some countries the saturation levels of cars per inhabitants have already been reached. According to Shell’s forecast, Germany is a partly saturated market. While car ownership among men will only slightly increase from 700 to 715 cars per 1,000 men by 2030, women’s car ownership is expected to increase from 340 to 430 (Shell Deutschland Oil GmbH, 2009 [online]). However, compared to other countries, especially in Asia and Latin America, where the number of vehicles is expected to more than double by 2020 compared to 2000 (Sperling & Claussen, 2004 [online]), German figures are rather low and could be interpreted as describing a saturated market.

Second, carmakers seek new business opportunities in continuously growing cities (Firnkorn & Müller, 2012 [online]). Christian Schlesiger (2011 [online]), author for the news magazine Die Zeit, estimates that more than 80% of today’s economic value is created in cities, which means that the largest market lies within urban environments. Considering that the majority of Germany’s total population (UN, 2012a [online]) is currently living in urban areas (73.8%) and that this share is expected to increase to nearly 82% by 2050, this means that the main market for carmakers will further shift towards urban areas in Germany. However, considering the growth of cities and the emergence of megacities especially in developing and emerging countries, it can also be assumed that companies do not only see German cities as a new market but as a test field for the further expansion in global markets.

Third, Firnkorn and Müller (2012, p. 265 [online]) believe that automakers may also consider their “moral responsibility for the environmental externalities of their products,” such as greenhouse gas emissions or traffic jams, by addressing the pressure from their stakeholders who demand more corporate responsibility. A survey of Ernst & Young’s among companies of the automotive industry (Fuß, 2011 [online]) seems to prove this point because it has shown that companies expect that protecting the environment and reducing emission levels will become an important issue in the industry.

It is hardly possible to determine which of the aforementioned reasons, if any, led the German carmakers BMW and Daimler to offer new mobility services. However, both entered the market and seek to become the leaders. BMW has given up seeing the automotive market from the perspective of a supplier but puts customer orientation and service appreciation into the centre of its new business strategy, considering the complete area of individual mobility as their future business (BMW, 2008 [online]). BMW believes that carmakers are going to face the biggest challenge in the third millennium and only those that try to tackle this challenge will be able to survive (BMW, 2008 [online]). Daimler prepares itself for the future challenges and believes that those who offer the best mobility services will enjoy competitive advantages over many years (Lamparter, 2010 [online]). Thomas Weber, Head of Daimler’s research and development, states that it is better to be on top of a trend than being behind (Lamparter, 2010 [online]). This offers a fourth aspect, not mentioned by Firnkorn and Müller, which could be the fear to be left behind in a social environment that changes (see 1.3.1).

1.3.3 Sustainability Concerns

The concept of sustainable development is widely defined according to the Brundtland Commission (1987 [online]) which wrote that “Sustainable development seeks to meet the needs and aspirations of the present without compromising the ability to meet those of the future.” Today, sustainability considerations have become a driver of customers and corporations as a result of resource exploitation and climate change.

In terms of mobility, Shell (Shell Deutschland Oil GmbH, 2009 [online]) defines sustainable mobility as the capability of a society to move freely and exchange extensively without causing other important social or ecological disadvantages. Considering the following figures it can be argued that the current mobility regarding cars can hardly be described sustainable in Germany: the number of registered cars has increased from 11 million to around 42 million since 1952 (KBA, 2012a [online]). Considering that around 5,000 kilograms of raw materials for the production of a single car are needed (Union Investment, n.d. [online]), it becomes clear that this development is apparently not sustainable in terms of the definition of the Brundtland Commission and will thus very likely compromise the ability to meet the needs of the future. Recycling is seen as one possible solution to the problem. At the moment, section 5 of the Altfahrzeug-Verordnung (AltfahrzeugV, End-of-Life Vehicle Regulations) makes it mandatory to recycle 85% of a car’s weight (Bundesministerium der Justiz, 2012a, [online]). However, even recycling 100% will not provide more materials to the production process than what was returned to be recycled in the first place.

While many advocates concentrate their criticism on developing and emerging countries in Asia and Latin America and demand a need to acknowledge that those cannot afford to have the same amount of cars per capita as in the West (see, for example, Uken, 2011 [online]), the question should be allowed whether the present needs of Germany can only be fulfilled by having statistically more than one car for every two people (Statistisches Bundesamt, 2012d [online]; Statistisches Bundesamt, 2012e [online]). Weert Canzler, scientist at the Wissenschaftszentrum Berlin für Sozialforschung (WZB, Center of Social Sciences Berlin), estimates that the average private car is unused for 23 hours a day (Brenner, 2011a [online]). This means that, in theory, one car could be shared by several people without decreasing the value for the individual which would then reduce the need for resources to build more cars (Brenner, 2011a [online]).

Studies have shown that especially young buyers are willing to pay a premium of up to 10% to protect the environment (Bratzel, 2011 [online]). Moreover, the recently offered electric and thus more environmentally-friendly Smarts in some car2go cities are sought by customers and used above average (Daimler, 2012e [online]), which also indicates that sustainability considerations play a role for customers. As mentioned above, carmakers as well consider the environment as a top priority in the near future (Fuß, 2011 [online]).

1.4 Carsharing as a Response to Social Change

Considering the aforementioned social changes of customers and carmakers, this study will now analyse which role carsharing can take in responding to these changes.

1.4.1 Generation Hire Car

As the CEO of car2go Robert Henrich puts it, carsharing has become a lifestyle (Bund, 2011 [online]), which is especially true for the young generation. According to official figures of Daimler (2010 [online]) and a study among car2go users conducted by Firnkorn and Müller (2012 [online]), approximately 60% of the scheme’s customers are 18-35 years old; official figures of DriveNow were not available. This is a significant differently customer profile compared to traditional carsharing schemes, where two thirds of customers are between 30 and 50 years old (Finanztest, 2012 [online]). A study among young adults not older than 30 years by the marketing research institute Puls showed that 25% have already made use of carsharing (Steiler, 2012 [online]). Moreover, Ernst & Young’s study showed (Fuß & Forst, 2012 [online]) that people in general (50%), but those aged 18-35 in particular (58%), can imagine using carsharing if it was sufficiently available.

One reason why commercial schemes may attract especially young customers is the nature of these schemes. Cars can be parked anywhere within the business areas of the respective scheme and not only at fixed stations like in traditional carsharing programmes. This means that customers need to find a car first before they can start their journey and the easiest way to locate the next car is by using a smartphone application that leads customers to the vehicle (Daimler, 2012b [online]; DriveNow, 2012b [online]). However, to some extent this discriminates all those customers that do not possess a smartphone. Therefore Kramper (2012a [online]) and Finanztest (2012 [online]) note that the commercial carsharing schemes of BMW and Daimler are especially made for the internet generation. Taking Ernst & Young’s survey among car owners, 52% of all 18-35 year-olds possess a smartphone, compared to 29% of the total interviewees (Fuß & Forst, 2012 [online]). BMW even stated that 80% of its DriveNow customers possess a smartphone (DriveNow, 2012b [online]).

Besides offering the internet generation a native access to their schemes, BMW and Daimler offer additional degrees of flexibility to their customers. On the one hand, the new commercial schemes do not require any particular commitments to certain environmental behaviours (Kramper, 2012a [online]) and are characterised by the absence of monthly fees compared to traditional carsharing schemes (Firnkorn & Müller, 2012 [online]). The possibility to rent cars in a more flexible way, on the other hand, seems to be important to customers as well. 80% of the rentals have been made spontaneously (Solberg, 2009 [online]) and 90% were one-way journeys where the destination differed from the starting point (Daimler, 2010 [online]).

This is not only advantageous compared to traditional carsharing, where cars have to be returned to specific stations, but it also allows to use commercial carsharing schemes as an additional option to fulfil individual mobility demands. As Corinna Ballweg and Wiebke Ebbing (2011 [online]) from the news magazine Die Zeit note, new technologies are needed to reach the destination cheapest and fastest by using a mix of different systems and transport modes. This so-called multimodality, Kramper (2012c [online]) and Ballweg and Ebbing (2011 [online]) believe, is needed to break the dominance of private car ownership. In this regard Daimler (2012a [online]; 2012b [online]) believes that today’s urban populations combine different transport modes such as public transport, taxis, car rental and car2go to create a multimodal service that simplifies urban mobility and increases the quality of life in cities. Henrich, thus, advertises the car2go Smarts as the first ‘own’ public transport (Bund, 2011 [online]) and announced that car2go wants to become part of a connected traffic chain (Daimler, 2012a [online]). Whereas many transport modes are still to some degree island solutions where customers have to pay individually for each service (Ballweg & Ebbing, 2011 [online]; Kramper, 2012a [online]), politicians (Schlesiger, 2011 [online]), non-governmental organisations (Breitinger, 2010 [online]; Duhr, 2010 [online]) and companies (Daimler, 2011b [online]; Seiwert & Schlesiger, 2012 [online]) work on systems that allow to use all transport modes, for example, with a single customer card. It can thus be assumed that a functioning multimodal transport system will further reduce the need for private cars because customers can choose how to reach the destination cheapest and fastest (Ballweg & Ebbing, 2011 [online]).

1.4.2 Carmaker or Mobility Provider

The initial success and the continuous expansion of car2go and DriveNow makes it obvious that there is a market for commercial carsharing. Experts, however, worry that carsharing may sustainably damage carmakers core business, the selling of cars. David Zhao (2010 [online]), automotive expert for the business consultancy Frost and Sullivan believes that carsharing is a double-edged sword to carmakers, because “On [the] one hand, it means sales opportunities… On the other hand, as carsharing vehicles replace personal vehicles, a … decline in new vehicles sales is expected.” Zhao (2010 [online]) believes that this “WILL happen, and vehicle manufacturers need to carefully gauge the potential impact on their total sales.”

BMW and Daimler are confident that the impact on total sales will be positive. Henrich believes (Bund, 2011 [online]) that car2go is no competition to Daimler’s traditional business because, as he highlights, the average buyer of a new car is older than 50 years compared to car2og where 90% of customers are below 50 years. He rather hopes to attract and bind young people to the brand so when they move to rural areas and decide to have a family, for example, they will purchase a Mercedes. Ian Robertson, Chief Marketing Officer of BMW, also wants to lead potential customers to the brand (Magenheim, 2011 [online]). Only Tony Douglas, who markets newly developed mobility services for BMW, acknowledges that companies may not only offer carsharing by choice but also because they want to retain a competitive advantage. His statement reads like ‘If we don’t do it, someone else will do it. If it is not done by big companies, small will do it’ (Schlesiger, 2011 [online]).

1.4.3 Sustainability Concerns

Carsharing, traditional and commercial schemes, primarily responds to sustainability concerns in three ways.

First, by reducing car ownership. Several studies, which cover traditional carsharing schemes if not mentioned otherwise, have shown that on average several dozens of users share one carsharing vehicle in Germany (Der Blaue Engel, 2010 [online]; Loose, 2011 [online]). Frost & Sullivan estimates “that, on average, each shared vehicle replaced 15 personally owned vehicles” (Zhao, 2010 [online]). In a study among North American carsharing users, Elliot Martin, Susan A. Shaheen and Jeffrey Lidicker (2010, p. 15 [online]) from Berkeley University, California, calculated that each carsharing vehicle has removed 9 to 13 private vehicles and that “the vehicle holding population exhibited a dramatic shift towards a carless lifestyle.” In an evaluation on the impact of carsharing on Swiss customers, the study has shown that car ownership among carsharing users has declined from 40% before joining to 24% after joining a carsharing programme (BFE, 2006 [online]).

In Bremen, one of three city states in Germany, a study found that each carsharing vehicle of a traditional scheme substituted nine cars (Der Senator für Bau, Umwelt und Verkehr, 2005 [online]) and have led to a reduction of 1,000 vehicles up until now (Mobil.Punkt, n.d. [online]). In a rare study of commercial carsharing users, Firnkorn and Müller (2012 [online]) found that each car2go vehicle in Ulm had reduced car ownership by 2.3 to 10.3 cars, but has a potential to take 19.2 vehicles off the street in the long term.

Second, carsharing responds to sustainability concerns because customers reduce travelled kilometres in cars and increase the usage of public transport. A car owner, for example, will make use of the own car as often as possible because first, he or she already pays high fixed costs and other transport modes will only cause additional variable costs, and second, because drivers tend to neglect other variable costs than those for petrol (Loose, 2011 [online]). In contrast, customers of carsharing schemes have a strong incentive to drive as little as possible because they have to pay directly for each single journey and kilometre (Rodt et al, 2010 [online]). Willi Loose (2011 [online]), CEO of the Bundesverband Carsharing (BCS, Federal Association of Carsharing), speaks of the learning curve of carsharing users which leads, after some experience with this transport mode, to the bundling of several trips into fewer because costs are seen as variable costs which can be reduced, for example, by using public transport. Carsharing providers have recognized this connection and started offering discounts for customers of public transport (Daimler, 2012a [online]; DriveNow, 2012c [online]; Finanztest, 2012 [online]). In Switzerland, the above mentioned study has shown that the number of households taking part in traditional carsharing programmes and holding public transport passes, has increased by one quarter after one year (BFE, 2006 [online]).

Michael Specht (2010 [online]) from Stern reminds, however, that carsharing can only be beneficial for the environment when customers sell their cars but not when carsharing is used instead of a bike or public transport. And indeed, studies have shown that the availability of carsharing has led to journeys that would not have been made otherwise (BFE, 2006 [online]). Martin and Shaheen (2011 [online]) found the same effect in their study among carless households but note, however, that this increase is relatively small. The average travelled distance among formerly carless households reached the same level to which other formerly car owner households reduced theirs, so that the annually driven kilometres were reduced on average by 1,740 kilometres or 8% per customer (U.S. Department of Transport, 2011 [online]).

This reduction benefits the environment. Considering that the average German car drove 14,200 km in 2010 (Kuhnert & Radke, 2011 [online]) and emitted 144 g CO2 per kilometre (Rodt et al, 2010 [online]), the environmental CO2 emissions of each car are slightly more than 2 tonnes. Calculations of CO2 emission savings are rare, but in Switzerland a study concluded that each carsharing customer saved 290 kg of CO2 per year (BFE, 2006 [online]). Considering the Northern American and Swiss studies mentioned above, this would correspond to 8-14% CO2 savings in Germany or 1,140-2,013 kilometres travelled less per customer. Loose worries, however, that this effect may not be achieved in commercial schemes because the price calculation which is based on time units, gives, from his point of view, an incentive to use cars even for small trips (Lamparter, 2010 [online]).

The third way how carsharing responds to sustainability concerns is by providing cars that are more environmentally friendly than the average car in terms of their CO2 emissions. By emitting fewer CO2 emissions, carsharing can directly affect the climate positively and help to lower emission levels in German cities. The latter is a persistent problem (UBA, 2012 [online]) despite the fact that the general fuel consumption in Germany has decreased consistently, reaching 7.5 litres per 100 km in 2009 (BMVBS, 2011 [online]) and thus lowered the average CO2 emissions of a German car to 144g per kilometre (Rodt et al, 2010 [online]). Because carsharing programmes use small cars which they replace after as early as 18 months (Finanztest, 2012 [online]), the average car fleet of carsharing organisations emits only 132.4 g CO2/km (Loose, 2011 [online]). The fleet of Smarts in Daimler’s car2go programme, however, tops this by emitting a mere 97 g CO2/km (car2go, n.d. [online]); BMW has not released official figures of its fleet. While Daimler’s figures are certainly impressive, assessing the full CO2 impact of carsharing vehicles “require a complex analysis using … lifecycle assessments … and well-to-wheels balances“ (Firnkorn & Müller, 2012, p. 277 [online]), an assessment which has not been carried out yet (Firnkorn & Müller, 2012 [online]). There seems to be a consensus in the literature, though, that carsharing vehicles reduce CO2 emissions either by reducing the travelled kilometres or by increasing the usage of the even more environmentally-friendly public transport.

Further reductions, moreover, can be achieved by using electric vehicles (EVs) which use electricity produced from renewables, such as in Daimler’s case (Daimler, 2012e [online]). Daimler’s electric Smarts are offered at the same price as petrol Smarts despite that they cause increased costs for battery, charging times and maintenance (Daimler, 2012e [online]). One seventh of car2go’s fleet is already electric (Daimler, 2012e [online]) and Frost & Sullivan estimates that “EVs will be increasingly leveraged by … carsharing programs [so that] By 2016, one in five new shared vehicles … is expected to be an EV” (Zhao, 2010 [online]). Zhao (2010 [online]), among others, believes that this “means huge business opportunities for EV manufacturers.” Ferdinand Dudenhöffer, head of the CAR Center of the University Duisburg-Essen, expects that increased demand will not only come from carsharing programmes itself but that the use of electric vehicles in carsharing schemes may be a key factor to promote this new technology (Herz, 2011 [online]). In a study, where people tested electric vehicles and were asked about whether they liked the tested vehicles, 71% stated they would consider electric vehicles for their next purchase (Herz, 2011 [online]). And indeed, Johan Jansson (2011 [online]) from the Umeå School of Business, Sweden, supports this notion that the use of this new technology may subsequently lead to an increased demand. His study found that consumers in general become accustomed more easily to innovative products, when they can test them first and have a positive experience (Jansson, 2011 [online]). This effect is not only useful for introduction of electric vehicles, but as this study will suggest in part three, it could also be used for the introduction of driverless vehicle technologies as a way to benefit vehicle sales

Regarding the use of electric vehicles in carsharing schemes and its possible effects on sustainable mobility, however, another aspect needs to be highlighted. The current lack of sufficiently available charging stations is seen as one the biggest obstacles for the introduction of electric vehicles (Fuß, 2011 [online]). The increased use of electric vehicles in carsharing schemes will help to build up the necessary infrastructure and thereby accelerate, Daimler (2011b [online]) believes, the introduction of EVs in general.

1.5 Limitations of Carsharing

This section will look at two main limitations of current commercial carsharing business models. Both, it will be shown in part three, could partially be offset by the suggested concept of integrating driverless cars in carsharing schemes.

1.5.1 Parking

Customers of commercial carsharing schemes benefit from the possibility to park their cars not only at specific stations such as reserved spaces on private areas or in public streets as in traditional carsharing schemes, but also alongside public roads (Lawinczak & Heinrichs, 2008 [online]). Because carsharing is just one of several stakeholders with an interest in using public spaces (Firnkorn & Müller, 2012 [online]), communities have to decide how to increase the attractiveness of carsharing without risking displeasing local residents. This leaves them with two problems.

First, in order to offer their customers short waiting and seeking times at the individual destinations, carsharing schemes need an oversupply of parking spaces (Specht, 2010 [online]) which fulfil the requirements of customers, such as that the parking lot is close to their residence, directly accessible and easy to find (Lawinczak & Heinrichs, 2008 [online]). Moreover, the community is also interested in avoiding ‘parking-slot seeking traffic,’ which not only costs time and is stressful for the individual but also increases noise and exhaust emissions for the community (Senatsverwaltung für Stadtentwicklung, 2006 [online]). The problem is, according to Lawinczak and Heinrichs (2008 [online]) that there are at best few parking spaces available in densely populated city areas, which makes it often impossible for carsharing companies and their customers to rent and find, respectively, suitable parking spaces in close proximity to their customers.

Second, communities that want to create parking spaces for carsharing schemes cannot easily do so. A lacking uniform federal regulation makes currently every created carsharing parking lot legally contestable because according to sec. 41 in combination with sec. 283 of the Straßenverkehrsordnung (StVO, Highway Code), parking spaces can only be reserved for taxis, disabled people or residents (Bundesministerium der Justiz, 2010 [online]; Lawinczak & Heinrichs, 2008 [online]). Several attempts to pass a law which would equate carsharing with the abovementioned stakeholders, however, failed so far because of the fear, carsharing could be put at an advantage over hire car (Brenner, 2011b [online]).

Communities which do not want to wait for legal certainty found two ways to bypass existing legislation: granting, like Bremen, special usage (Sondernutzung) to carsharing companies, or partially confiscate public spaces (Teileinziehung), like Berlin did (Lawinczak & Heinrichs, 2008 [online]). Despite the fact that both solutions are legally contestable, Bremen and Berlin are the only communities that were able to increase carsharing parking lots significantly (Brenner, 2011b [online]). However, even in Berlin the proceedings to apply for parking spaces take between 6-9 months which is seen as too long by respective companies (Lawinczak & Heinrichs, 2008 [online]).

Another issue arises, once the parking spaces are established. A study by Lawinczak and Heinrichs (2008 [online]) observed unauthorised parking in 44% of all cases where no carsharing vehicle was parked. It comes thus to no surprise that customers desire more parking spaces (Solberg, 2009 [online]) and companies are in a competition for each slot that has been established by communities (Schlesiger, 2011 [online]). Henrich even stated in an interview that the refusal of local governments to establish parking lots is a major obstacle to the expansion of car2go (Specht, 2010 [online]).

1.5.2 Availability

The availability of vehicles in carsharing schemes is limited by three aspects.

First, the number of parking spaces is low which limits the number of vehicles. While at the moment the slowly increasing number of supplied carsharing vehicles meets the demand, it remains questionable whether carsharing providers can offer enough parking lots to meet future demand of their vehicles in the middle to long term. A corresponding law could have the potential to mitigate, but not entirely resolve this problem.

Second, the business area of carsharing programmes is limited. This means that several suburbs can be reached by cars but vehicles cannot be left there (Kramper, 2012a [online]). Customers that live in residential areas outside of urban centres can therefore not make use of carsharing schemes to drive home and leave the car in their respective neighbourhood.

Third, the willingness of people to walk to the next available car is limited. A study by BCS has shown that on average people do not want to walk further than 700 metres to find a car. This, however, can increase the costs for companies in terms of having to provide more cars and parking lots.

The current part has shown that traditional carsharing persistently grew over the past two decades and that commercial carsharing schemes grew especially fast in the past years. While beneficial for the environment, carsharing can be expected to grow even further in the future due to observable social and business change. Some of the limitations of current carsharing programmes might pose obstacles to further growth, but could be offset by the use of driverless vehicles (see below).

illustration not visible in this excerpt

Part 2: Driverless Vehicles

The fact that you’re still driving is a bug, not a feature.

Anthony Levandowski Business Lead on Google’s Self-Driving Car Project

2.1 Introduction

The part at hand is going to assess the second component which is required for the proposed concept, driverless vehicles. The background section gives an historical overview of past and current developments. It shows that this technology has seen major advances within the last eight years and indicates the feasibility of reaching marketability eventually.

The following section concerns itself with traffic accidents, a major societal issue that can be tackled by the widespread use of driverless vehicles, for example, in carsharing schemes. Because the vast majority of car-related traffic accidents are caused by respective drivers, this section will first look at the main causes for such accidents, which are [1] distractions, [2] deliberate non-compliance with traffic regulations, and [3] physical deficits in the elderly. Afterwards, the study continues to give an overview of traffic accidents in Germany, i.e. the consequences of the aforementioned deficits, and the economic costs caused by traffic accidents. Besides looking at the general population, this section will, as in part one, especially focus on young people because it is them who demand new mobility services and who make use of carsharing the most. However, because of a slow demographic change towards an aging society in Germany, which could eventually increase the use of carsharing among the elderly, the analysis will also include this age group.

The subsequent section analyses automation in passenger vehicles as a response to car-related traffic accidents. It starts by introducing first, driver assistance systems and second, advanced driver assistance systems, to show that these technologies were developed in order to tackle certain deficits of drivers and investigates how these systems are accepted by users and what, especially German customers, expect of these systems.

The last section of this part will investigate the current developmental state of driverless vehicle technologies in order to understand problems, which have to be overcome before the introduction, limitations it may encounter due to its system limitations, as well as its potential to improve traffic safety.

2.2 Background

Shortly after the introduction of the first motor car at the turn of the 20th century (Bird, 1960), driver assistance systems such as speed indicators were introduced to support drivers with their driving task (Becker, 2012 [online]). General Motors presented the idea of an autonomous vehicle for the first time to a wider public at the 1939 World’s Fair exhibition (Vanderbilt, 2012b [online]). In 1953, GM and the Radio Cooperation of America developed a scale model automated highway system (AHS) which was supposed to control cars by wires in roads (Wetmore, 2003 [online]). Since then “a few isolated systems have been built” (Wetmore, p. 18, 2003 [online]) but turned out to require extensive control and never worked reliably (Wetmore, 2003 [online]). In the 1960s and 1970s a team of Stanford University equipped a buggy with a video camera and image processing capabilities which in 1978 eventually “crossed a chair-filled room without human intervention in about five hours” (Vanderbilt, 2012b [online]). In the 1980s and 1990s Ernst Dickmanns achieved great distinction with his experiments in Germany and is since then regarded as pioneer of autonomous cars (Vanderbilt, 2012b [online]). His vehicles, usually transporters fully packed with computers and sensors, were capable of driving autonomously in real-world traffic on German motorways (Vanderbilt, 2012b [online]). Despite the success, limited processing power soon emerged as one of the biggest obstacles (Davis, 2006 [online]). Dickmanns referred to Moore’s Law in 1996, proclaiming that it would take another decade until sufficient processors were available and, as Joshua Davis (2006 [online])), author from Wired magazine notes, “He was right, and everyone knew it.”

In order to achieve the U.S. congressional goal that “by 2015, one-third of the operational ground combat vehicles are unmanned” (National Defence Authorization, 2000 [online]), the Defense Advanced Research Projects Agency (DARPA) decided in 2004 to spur innovation in the field of autonomous vehicle technologies by offering “$1 million to anyone who could build a self-driving vehicle capable of navigating 300 miles of desert” (Davis, 2006 [online]). The “prize [for the event dubbed Grand Challenge 2004] went unclaimed as no vehicles were able to complete the … route,” with the best team completing 7.3 miles (DARPA, 2007 [online]). One year later, DARPA doubled the prize money and 5 teams out of 23 completed the Grand Challenge 2005, a 132-mile race through the Mojave Desert (DARPA, 2005 [online]). In 2007, DARPA hold the last event of this kind, the so-called Urban Challenge which “represented the first full-scale demonstration of autonomous vehicles operating together in traffic” (DARPA, p. 4, 2008 [online]). Cars had to drive on unknown roads, consider traffic signs, park independently, filter into the traffic, overtake other cars (Bartels, Ruchatz & Brosig, n.d. [online]) and share the road with more than thirty professional human drivers (Stewart, 2007 [online]). Six out of eleven cars finished the competition successfully, with Stanford’s team finishing second (DARPA, 2008 [online]).

Whereas German carmakers such as Audi (Squatriglia, 2010a [online]), BMW (Boeriu, 2011 [online]) and Volkswagen (Bartels, Ruchatz & Brosig, n.d. [online]), suppliers such as Continental (2012 [online]) and research projects such as the European Highly Automated Vehicles for Intelligent Transport (HAVEit, 2011 [online]) are explicitly working on driverless vehicle technologies, it is the search engine provider Google who has announced the biggest advancements. In October 2010, Sebastian Thrun, formerly team leader of the Stanford Racing Team which won DARPA’s Grand Challenge 2005 and became second in DARPA’s Urban Grand Challenge, announced for his new employer Google that its fleet of autonomous vehicles had already travelled more than 225,000 km on public roads (Thrun, 2010 [online]). In August 2012, this number had more than doubled, while “there hasn’t been a single accident” (Urmson, 2012 [online]).

The different technologies of autonomous systems work very similar and have been improved since DARPA’s Grand Challenge competitions. In general, Nidhi Kalra, James Anderson & Martin Wachs (2009 [online]) from the RAND corporation note, autonomous driving is based on functions categorised as sensing, planning, and acting. After a system has gathered information about the external environment (sensing), it applies specific algorithms (planning) to “create a set of actions that bring the system closer to its objective” (Kalra, Anderson & Wachs, p. 7, 2009 [online]) and executes these plans subsequently (acting) (Kalra, Anderson & Wachs, 2009 [online]).

However, it is important to acknowledge that there is a continuum between a purely human-driven and completely autonomously-driven vehicle which has implications for the further study. Several, very similar, suggestions have been made to distinguish the different levels of automation, though no uniform definition, let alone any standard, exists. The project group Automation: Legal Consequences of Increasing Vehicle Automation of the Bundesanstalt für Straßenwesen (BASt, Federal Road Research Institute) proposes the stages [1] driver only, [2] assisted [3] semi-automated, [4] highly automated, and [5] fully automated (Rauch et al, 2009 [online]), with Bryant Walker Smith (2012e [online]) noting that these distinctions “are not absolute.” Table 2 shows the results of the project group.

Each stage, beginning from [2] assisted, is supposed to support the driver with the driving task to make driving saver, either by mitigating the worst outcomes of accidents or by preventing them at all. The next section will continue to analyse reasons for and results of traffic accidents.

2.3 Social Issue Traffic Accidents

2.3.1 Background

Road safety is seen as a major societal issue (European Commission, 2012c [online]). According to the United Nations Economic Commission for Europe (UNECE, 2012 [online]), 1.3 million people die and up to 50 million more are injured in traffic accidents each year. Despite the fact that over 90% of global fatalities (WHO, 2009 [online]) and around 80% of global injuries in traffic accidents occur in low and middle-income countries, which possess less than half of all registered vehicles (UNECE, 2012 [online]), high-income countries are not spared from the problem. For example, in 2010 1,473,443 people were injured and 31,029 lost their lives in 1,114,980 traffic accidents in the European Union (European Commission, 2012a [online]). With 49 road deaths per 1 million inhabitants, Germany had to suffer a considerably higher number of traffic deaths than the worldwide best-performing country Sweden (33 road deaths), but still less than the European average of 61 (European Commission, 2012b [online]). Before more detailed statistics for Germany will be presented, the following section will now turn to main causes of traffic accidents.

illustration not visible in this excerpt

2.3.2 The Role of the Driver in Car-related Traffic Accidents

Because technical defects and external factors such as defect roads play a minor role in causing traffic accidents (Statistisches Bundesamt, 2012a [online]), this section will concentrate on the role of drivers in car-related traffic accidents. In this regard, the literature distinguishes in general between three major sources of human-caused accidents: [1] distractions, [2] the deliberate non-compliance with traffic regulations such as driving at unsuitable velocities or under the influence of alcohol or drugs, and [3] physical deficits that increase with age and negatively impact the driving task (Kubitzki, 2011 [online]). Because distractions are widely believed to cause 90% of all accidents (Kubitzki, 2011 [online]), this study will now continue to investigate this aspect of traffic accidents before looking at the other two causes of traffic accidents.

2.3.2.1 Distractions

In general, every distraction is suboptimal for a driver because a new task shifts part of the attention away from the original driving task (Kubitzki, 2011 [online]). If several tasks compete for the attention of a person, however, one task will be put into the centre of attention and all others will receive fewer capacities due to the functioning of the human brain (Kubitzki, 2011 [online]). According to a study of the insurance company Allianz, the author Jörg Kubitzki (2011 [online]) summarises the literature and writes that one way to categorize distractions is by distinguishing their source, which can be localised [1] spatially (does the distraction lie inside the car, such as a conversation with a passenger, or outside of the car, such as a billboard at the roadside), [2] mentally (is the cause of the distraction internal, for example, is the driver angry, or is the distraction caused externally, for example, by an argument with another passenger) and [3] technologically (is the distraction device-related, such as using a mobile phone, or is it non-device related, such as smoking a cigarette).

These categorizations help to understand where, how and why traffic accidents are linked to distractions and what can be done to mitigate risks. For example, when considering the spatial occurrence of accidents, Mc Evoy et al. (2007, cited in Kubitzki, 2011 [online]) found that more than 90% of accidents were caused by distractions within the car, such as drinking, eating or smoking. A study of BASt (2012a [online]) found that 80% of drivers acknowledged that they would engage in non-driving related activities. The Allianz, too, asked drivers how often they would engage in non-driving related activities, the results of which can be found in the following table.

illustration not visible in this excerpt

Although these results need to be treated with caution because of their non-representative character and the subjective assessments of the interviewees, they indicate that distractions are ubiquitous. Despite their potential risk, however, only one quarter of drivers in the BASt study believed that their activities were distracting and less than 10% of drivers thought that distractions outside of the car are dangerous (BASt, 2012a [online]). Because these figures are subjective, traffic researchers create risk rankings to objectively evaluate which activities and distractions pose what risks. The following table shows the Risk Ranking New Zealand Focus Group Research, the Risk Ranking Regan, and the Allianz Risk Ranking.

These risk rankings have in common that distractions were often linked to mobile phones or smartphones, either caused by making or answering phone calls, sending text messages or emails, or using the internet. As Kubitzki (2011 [online]) notes, it may be easy for a person to handle a phone call and the driving task at the same time, but when the traffic situation changes suddenly, it may require the full attention of the driver. For this reason, German legislation prohibited phone calls without a hands-free set in 2001 (DEKRA, 2012 [online]) and also prohibited to use a smartphone or similar devices while driving, regardless of the function which was used (Kubitzki, 2011 [online]). Nonetheless, offences against these regulations almost doubled from 284,000 in 2005 to 450,000 between in 2011, with more

illustration not visible in this excerpt

than 40% of these offences committed by people younger than 35 years (KBA, 2012a [online]).

[...]