Supporting the adoption of electric vehicles in urban road freight transport – A multi-criteria analysis of policy measures in Germany

Transportation Research Part A 91 (2016) 61–79

Contents lists available at ScienceDirect

Transportation Research Part A journal homepage: www.elsevier.com/locate/tra

Supporting the adoption of electric vehicles in urban road freight transport – A multi-criteria analysis of policy measures in Germany Tessa T. Taefi a,⇑, Jochen Kreutzfeldt a, Tobias Held a, Andreas Fink b a b

Department of Mechanical Engineering and Production Management, Hamburg University of Applied Sciences, Berliner Tor 5, 20099 Hamburg, Germany Faculty of Economics and Social Sciences, Helmut-Schmidt-University, 22043 Hamburg, Germany

a r t i c l e

i n f o

Article history: Received 23 June 2015 Received in revised form 23 May 2016 Accepted 11 June 2016

Keywords: Electric vehicles Urban freight transport Transport policy evaluation Multi-criteria analysis

a b s t r a c t Policies in Germany to support electric vehicles, which are free of exhaust emissions, mostly focus on urban road passenger transport. However, road freight vehicles are a main source of the traffic air pollutants and noise emissions in cities. Available vehicle types, tour planning and purchase decisions in urban road freight transport differ from the passenger transport segment. The political and scientific literature lacks a comprehensive discussion of specific policy measures to support electric urban road freight vehicles. This article contributes to the existing body of knowledge, by undertaking a multi-criteria analysis of policy measures to support battery electric freight vehicles based on the rating by two stakeholder groups, ‘‘policymakers” and ‘‘freight electric vehicle users”. These stakeholders rate 23 policy measures as suggested in the literature or which are implemented in European countries. In comparing and ranking the rating results of the groups, we find that the discordance between the groups can be large and offers noticeable insight and room for future research and practice. Although financial support of electric vehicles is often named in the literature as the primary measure to overcome the total cost of ownership gap of freight electric vehicles, the current study shows that the effect of special legal measures and supporting the setup of company-charging infrastructure are underestimated by the policymakers. Recommendable policy options – beyond several fiscal measures – are to request emission-free vehicles in municipal tenders, to allow drivers with a class B license to drive freight EVs over 3.5 tons, or to implement a city toll on the long-term. The practicability of other policy measures depends on the local implementation goals of the municipality. Hence, a transparent debate on the aim of supporting electric freight mobility is as necessary as choosing measures targeted at the freight transportation segment. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction In order to achieving an emission-free urban transportation the European Commission (2011) suggests to take action in two segments of the urban road traffic: in passenger and freight transportation. While the European Commission’s framework aims to achieve emission-free urban passenger transportation by 2050, they suggest accomplishing an essentially ⇑ Corresponding author. E-mail address: [email protected] (T.T. Taefi). http://dx.doi.org/10.1016/j.tra.2016.06.003 0965-8564/Ó 2016 Elsevier Ltd. All rights reserved.

62

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

emission-free urban freight transportation already by 2030. In conjunction with the so-called ‘‘Energiewende”, the German federal government recognizes the necessity for emission-free transportation systems. At the same time the federal government stresses the importance of the transformation process for its automotive industry; here, 25% of the industry’s turnover and 20% of the country’s exports are generated; the large cars segment is a particular strength of the German car manufacturing industry (Merkel, 2013). As a logical consequence, the German federal government aims to become a lead supplier and lead market of electric mobility with one million electric vehicles on the streets by 2020 (NPE, 2014). The focus on the technical and economical chances and challenges of electric mobility for Germany lead to large investments in research, development and pilot projects. As a result, the car manufacturers will have launched 29 battery electric passenger car series models by the end of 2015 (NPE, 2014). International electric mobility benchmarks acknowledge Germany to be in a leading position, when ranking the suppliers of electric vehicles (EVs) (McKinsey, 2014; NPE, 2014). However, the same benchmarks conclude that Germany currently lags behind its self-set goal of becoming a lead market for electric mobility. By January 1st, 2015, only one out of 2434 registered passenger vehicles was an EV (KraftfahrtBundesamt, 2015). Since the German government aims to achieve electric mobility without permanent financial subsidies, an electric mobility law was adopted in Germany in March 2015 (EmoG, 2014) in order to strengthen the demand for EVs. The law foresees a labeling of electric vehicles on the vehicle registration plate, providing a legal basis for municipalities to grant privileges to electric passenger vehicles and light commercial vehicles. At the same time and while receiving a considerably lower attention and financial stimulus, the freight transportation market has outperformed the passenger car market. As an example, one out of 923 registered trucks between two and five tons had an electric drive-train by January 1st 2015 (Kraftfahrt-Bundesamt, 2015). Due to the unavailability of battery electric series freight vehicles, certain logistics companies interested in electric mobility became involved in importing, retrofitting, or producing freight EVs themselves (Taefi et al., 2015). Furthermore, freight EVs offer particular benefits in urban applications, since the vehicles are free of exhaust emissions such as nitrogen oxide and particulate matter and are more silent compared to conventional diesel trucks (Umweltbundesamt, 2013). Although only about five percent of the registered vehicles are trucks (Kraftfahrt-Bundesamt, 2015), they are responsible for more than ten percent of the driven kilometers in German cities (Wermuth, 2012) and are a main source of noise and air pollutants, such as particulate matter or nitrogen dioxide (Menge, 2013). As an example, trucks over 3.5 tons cause over 45% of the traffic’s NOx-emissions in Germany’s second largest city Hamburg (Böhm and Wahler, 2012). On a drive-cycle with frequent stops and low average speeds, medium duty delivery trucks emit 42–61% less greenhouse gas emissions compared to diesel vehicles (Lee et al., 2013). Duarte et al. (2016) found in a real-world case study that small electric urban delivery vehicles reduce the vehicle usage energy consumption by 76% (57% when considering the energy production stage). Despite the local environmental advantages of freight EVs, despite the interest of logistics companies in electric freight vehicles, and despite the results of research projects, which underline the potential of electric mobility in freight transportation (Tenkhoff et al., 2012), the German federal government excluded freight EVs over 4.25 tons from the electric mobility law. The public and scientific debates on policy options to support EVs so far misses to clearly differentiate between the two transport segments passenger and freight transportation, which have a fairly different structure and thus different requirements for support. Hence, this paper explores the question: Which policy measures are recommendable to support electric vehicles in urban road freight transport in Germany? Transport logistics companies indicate that policy measures are an important driver for the design of their logistics networks: 75% of the companies indicate that the political framework is an important or very important influencing factor (Fraunhofer IML, 2010). Thus, supporting electric freight vehicles through policy measures could increase the number of electric freight vehicles and abate freight transport-related emissions. In the next Section 2 we review the related literature, explicate the research questions and the contributions of this paper. In Section 3 the research methods used are described. This involves an exploration of the available literature and surveys considering two groups (policymakers and freight EV users). The results are presented in Section 4 and the differences between the ratings of the groups are discussed in detail in Section 5. This is followed by a conclusion and a discussion on the limitations of our work in Section 6.

2. Background and investigative questions Evaluations of urban freight transport policy measures often generally name electric or low emission vehicles as options to reduce freight transport emissions (Bozzoa et al., 2014; Lützenberger et al., 2014; Zanni and Bristow, 2010). Despite this, the literature does not yet provide an ample discussion of specific policy measures to support electric urban road freight vehicles. The high investment to purchase an EV is one of the main obstacles for commercial users (Amburg and Pitkanen, 2012; Ball and Wietschel, 2009; Kley et al., 2011; Taefi et al., 2015). Hence, some authors analyze fiscal policy options to bridge the gap between the total cost of ownership (TCO) between an EV and a vehicle with an internal combustion engine. Wietschel et al. (2013) discuss the efficiency of fiscal measures, such as a purchase price subsidy, tax abatement or depreciation model for the passenger vehicle market. They conclude that the segment of commercial passenger fleets offers a high replacement potential with electric vehicles and is sensitive to financial subsidies. Other authors who discuss fiscal tools in order to

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

63

compensate the higher TCO of EVs are Bozem et al. (2013) and Gass et al. (2012). The latter focus their discussion on possible fiscal measures to support the passenger EV market. In addition to describing fiscal measures, Bozem et al. (2013) briefly discuss one possible legal policy measure: to restrict the access into the city center for conventional vehicles. Their study evaluates the reaction of passenger vehicle drivers with regard to this measure. They find that the majority of private car owners would rather use public transportation instead of purchasing a vehicle with alternative technology – an option which is not realistic for the freight transport segment. The aforementioned fiscal measures that have been discussed in the literature include measures to reducing the total cost of ownership of electric vehicles or increasing the total cost of ownership for conventional vehicles. However, financial incentives for EVs or disadvantages for conventional vehicles are not the only policy options. Moreover, the market uptake of EVs cannot be explained by financial incentives alone and further research is needed to understand the impact of other policy measures, according to a global comparative study on passenger EV incentive policy (Mock and Yang, 2014). A general classification of the policy options of governments in the modern digital societies is provided by Hood and Margetts (2007). They suggest to cluster policy measures in a scheme with four categories: apart from fiscal measures, the government can adopt legal, communication and organizational measures. To adopt legal measures the government utilizes its authority to enable or prohibit certain activities, i.e., by changing the legislation or by issuing certificates. An electric freight mobility-related example is to allow zero-emissions freight vehicles to drive within the bus lanes. To implement communication measures the government uses its ‘‘nodality” to collect and dispense information. With regard to electric freight vehicles, this could be information on use cases, advantages, options and total cost of ownership of urban electric freight vehicles. Organizational measures include actions where the government uses its own capacity and capability, such as people, skills, land, buildings and infrastructure. The government could, for example, raise the number of electric freight vehicles in its own fleets. Salama et al. (2014) specifically discuss the legal policy option of providing loading bays for electric delivery vehicles. They argue that loading bays are an important legal privilege which can raise the efficiency of EVs and thus compensate the higher TCO for companies. However, the study does not compare the efficiency of the policy option to others, such as fiscal measures, legal privileges or communication measures and focuses on a single criterion – the efficiency of the measure. Since urban freight transport is a complex topic it requires including various criteria and the evaluation of various stakeholders in the decision-making process (Macharis, 2005; Suksri et al., 2012; Taniguchi and Tamagada, 2005). A multi-criteria analysis is a common methodology to evaluate various criteria and different stakeholder groups. An additional strength of this methodology is that it can take qualitative criteria into consideration, that are difficult to quantify (Suksri et al., 2012). Examples of multi-criteria analyzes of EV policy include the report of Grausam et al. (2014) and the manuscript of Bakker and Trip (2013). Both studies rate various measures by economic, ecologic and social criteria. However, both studies do not focus on freight transport, but instead give an overview about various policy options to support EVs in different segments, such as individual and public passenger transport as well as in commercial fleets. Thus, only a few discussed measures apply to the urban road freight transport segment. Furthermore, Bakker and Trip (2013) research the opinion of a single stakeholder group – policymakers – to rate policy options. Grausam et al. (2014) evaluate the opinion of a not clearly defined German expert group. Thus, none of the studies is specifically targeted at urban road freight transport, nor delivers a clear integration and comparison of the opinions of different stakeholder groups. The latter point is of specific interest when discussing the economical aspects and thus efficiency of policy options (Hood and Margetts, 2007). According to those authors, efficient policy options meet two criteria: (i) minimizing the effort for the government and (ii) limiting the burden on the general public, such as bureaucracy or undesired economic side-effects. Certain publications question the efficiency of discussed or implemented policy measures: Driscoll et al. (2013) simulate the market share for passenger EVs and conclude that only very high subsidies would achieve a general market penetration of 10% in Ireland, for example. Sprei and Brauener (2011) prove a positive effect of available EV incentives on the licensed EV stock in Europe, but question the cost effectiveness of the measures. Green et al. (2014) argue that current U.S. policies supporting EVs are neither efficient nor effective. They propose to shift incentives requiring major investment and government resources such as fiscal subsidies, research and development, and infrastructure and service equipment, away from the mass markets to the early adopters and niche markets, such as postal services, a sub-segment of urban freight transportation. As a conclusion, the available scientific literature focuses on policy options which are often not especially targeted at urban road freight transport. An overview about policy measures directed at emission-free urban road freight transportation is missing in the literature. The available assessments of the measures are predominantly discussing the effectiveness of fiscal measures and do not compare them to other prospective policy options. Multi-criteria ratings, which enhance ratings of the efficiency of measures by an assessment of their social feasibility, are sparse and again not especially targeted at the urban freight segment. Finally, available multi-criteria studies tend to focus on the opinion of policymakers, having the drawback of neglecting to indicate the efficiency of measures for freight EV users. This article strives to fill these gaps by breaking down the central research question, formulated in the above Section, into three investigative questions:
Which policy measures in support of freight EVs are currently discussed or implemented in European countries?
How do policymakers and freight EV users rate the ecological, economic and social feasibility of these measures?
Which policy measures are recommendable?

64

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

As a limitation, we will focus on policy options that aim to make an impact and influence behavior, and disregard measures to detect, take in and process information. Answering the above questions will contribute to the body of knowledge, by compiling an overview of single European policy measures suggested in the literature or implemented in European countries, to especially target electric freight vehicles. Furthermore, we will enhance the current method used in scientific literature to rate and rank electric vehicle policies: with a multi-criteria analysis we will compare the quantitative ratings of two stakeholder groups, ‘‘policymakers” and ‘‘electric freight vehicle users”, in order to yield insights about preferences and existing misconceptions between the two groups. In discussing explicit goals for adopting policy measures, we will differentiate policy recommendations based on the potential goals. 3. Methods We chose an inductive approach to explore the new field of policy adoption for electric mobility. Firstly, we enriched a review of available scientific literature on policy measures by reviewing research reports and current policy practices in European countries. The possible policy measures were clustered based on the categories of Hood and Margetts (2007). Secondly, two expert groups rated the possible policy measures in a Web-based survey. The two expert groups encompassed two of the most important stakeholder groups, on the one hand policymakers, and on the other hand companies which utilize EVs for road freight transport purposes. Thirdly, choosing a quantitative research design allowed statistical analysis to compare the answers of the two groups. In a final step, the data were interpreted to draw conclusions on recommendations and research opportunities. The example of Germany was chosen, since the country has the high aspiration of becoming a lead market for electric mobility, while still offering room to develop fiscal, legislative, communicative, or organizational freight EV-specific policies. 3.1. Proposed measures to support freight EVs Searched keywords in the literature review were: (‘‘Electric vehicle⁄” OR ‘‘Electric transport⁄” OR ‘‘Electric commercial vehicle⁄” OR ‘‘EV⁄”) AND (‘‘Policy” OR ‘‘Incentive⁄” OR ‘‘Stimulat⁄”). Databases and key journals were searched for publications from 2010 until spring 2014. The year 2010 was chosen as starting date, since the year was the start of large pilot projects in Western European countries, such as the Netherlands,1 Germany2 and the UK3 and marked the mass market introduction of modern EVs (Trigg, 2012). Current practices were gathered from the literature (E-Mobility NSR, 2013; Taefi et al., 2015; van der Steen et al., 2015) and by ‘‘snowballing” when looking up references, and following current news about electric mobility in Europe. The identified measures are clustered into to four categories based on the suggestion of Hood and Margetts (2007). 3.2. Composition of the online questionnaire After a pretest of the questionnaire, the survey was conducted by means of a standardized Online questionnaire based on the software LimeSurvey.4 The survey was available only by invitation, creating a selective sample of chosen experts of two distinct groups. An Online questionnaire fulfills test quality criteria, such as objectivity, reliability and efficiency as discussed by Bryman (2012). The possible measures were assessed on a Likert scale with five symmetric points from 1 (low) to 5 (high), with 2, 3 and 4 as numeric intervals without a description. For each possible policy measure three criteria were rated: Effort denotes the monetary and/or personnel effort required, to implement a measure in the respondents’ city (policymakers) or company (freight EV users). This criterion indicates the perceived economic impact of a measure for each group. Effect indicates the likelihood that the measure would lead to many electric freight vehicles in the respondents’ city (policymakers) or company (freight EV users). Hence this criterion is also an indication of the ecologic impact of the measure. Feasibility rates the feasibility to implement a measure, considering overall social, political and financial aspects in the respondents’ city. Additionally, the respondents were asked to estimate the time until the measure can be implemented. The choices were: the measure can be implemented in the near future (less than three years) or needs longer preparation (longer than three years). 1 Proeftuin elektrisch rijden. http://www.rvo.nl/onderwerpen/duurzaam-ondernemen/energie-en-milieu-innovaties/elektrisch-rijden/praktijkvoorbeelden/ proeftuinprojecten. 2 Electric mobility pilot regions. http://www.now-gmbh.de/de/mobilitaet/mobilitaet-von-morgen/modellregionen-elektromobilitaet.html. 3 Plugged in places. https://www.gov.uk/government/publications/plugged-in-places. 4 http://www.limesurvey.org.

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

65

The pre-test revealed that freight EV users were missing an option to rate whether the measure would be acceptable and useful for their company. In some cases they used the criterion ‘‘feasibility” and annotated ‘‘in my company”. In order to obtain a valid rating of the criterion ‘‘feasibility”, a fourth criterion ‘‘consent” was implemented in the survey of freight EV users: Consent describes the consent and acceptance of the freight transport company to the measure. 3.3. Choosing the sampling groups The sampling groups to respond to the questionnaire were purposively selected: in the first group ‘‘policymakers”, the respective civil servants responsible for electric mobility in the administration of the ten largest German cities by population were approached (Berlin, Hamburg, Munich, Cologne, Frankfurt, Stuttgart, Dusseldorf, Dortmund, Essen, and Bremen). This survey focused on local policymakers who were asked to answer the questions for their respective cities. This approach was chosen since the positive effects of deploying electric freight vehicles are predominately local, such as reduced noise and air pollutants. Federal policymakers have so far neglected freight EVs, since the indicators focused by the German federal government are mainly the number of electric vehicles and reductions in greenhouse gas emissions, both of which are comparably low for freight EVs, compare with Section 1. Nine out of ten policymakers answered the email request and filled in the questionnaire. Additionally, one representative of the German-wide coordination of the ‘‘electric mobility pilot regions” answered the questions (with respect to Berlin), raising the number of replies in this group to ten in the period June to November 2014. The sampling group ‘‘freight EV users” included those German companies, that utilize freight EVs in urban road freight transport, based on an assessment of existing projects (Taefi et al., 2015). Drawing on existing contacts or references by project leaders, 16 companies were identified. This number does not encompass all existing freight EV users in Germany, since not all freight EVs are purchased with the support of pilot projects and two project leaders did not respond to the request. Eight of the 16 invited companies answered the questionnaire during the period June 2014 to March 2015. The initially planned shorter time-span for the survey was extended repeatedly, in order to reach the initially planned number of ten participants in this group. Despite the efforts, only eight companies answered the questionnaire. The respondents were high ranking managers, responsible for electric mobility in the company; the owner or managing director in case of small companies; or in larger companies, the sustainability manger or site manager. The fleets of the participating companies ranged from one electric passenger car (in two cases) up to a large fleet of hundreds of electric vehicles with sizes between passenger cars and heavy electric trucks, with additionally several thousand electrically-supported freight cycles (in one case). The vehicle types used most frequently for freight transport purposes were electrically-supported freight cycles, electric trucks between 3.5 and 7.49 tons, and electric trucks between 2.3 and 3.49 tons. 3.4. Evaluation of the results The ratings of the two stakeholder groups were evaluated through a statistical analysis each. The symmetrically distributed items allow the scale to be treated as an interval scale, so that mean values can be calculated. The measures were ranked according to the independent criteria. The correlation of the non-parametric data was calculated using the Kendall correlation coefficients. 4. Results 4.1. Identified measures A total of 23 possible measures supporting freight EVs were gathered by reviewing the literature as well as European good practice examples. Ten measures fall into the category ‘‘legal”, one into ‘‘communication”, five into ‘‘organization” and seven into ‘‘fiscal”. However, some overlap between the categories, and thus the suggested clustering exists, as mentioned by Hood and Margetts (2007). Table 1 lists the possible measures and their abbreviations in columns 1 and 2, whether the measure has been discussed by decisionmakers in Germany (column 3) and in which European city or country the measure is implemented (column 4). 4.1.1. Communication measures This study summarizes the identified communication measures to support electric freight mobility in Europe into one policy option: to provide information on freight EV availability, their total cost of ownership, governmental grants or funding in pilot projects. Examples of these information centers are virtual (Websites), mobile (transportable booths) or stationary. The option to rate communication measures individually was dismissed for three reasons: first, in reality usually more than one aspect of electric mobility is communicated on Websites or in information centers. Second, the pretests yielded similar ratings of the communication measures. Third, the respondents of the pretest recommended to reduce the number of measures to be rated. Hence, in this survey we follow the practical example of grouping information measures.

66

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

Table 1 Overview of identified measures and their sources. Abbr.

Explanation

Communication Info Virtual or physical information on freight EVs, costs and state funding (incl. cargo cycles)

a

c

Examples

Grausam et al. (2014) and Aichinger (2014)

Amsterdam, London, Hoeje Taastrup

Legal Access Bays

Access privileges to areas restricted to heavy trucks above 7.5 tons Freight EVs can use privileged loading zones in inner city areas

BusLanes

Freight EVs are allowed to drive on bus lanes

Certificate DrLicense LEZ Lots

Certify transport companies with ecological fleets Allow EV drivers with class B license to transport a payload comparable to conventional 3.5 t diesel vehicles Low emission zone for freight vehicles of 3.5 tons and above Free and/or privileged parking for EVs

Noise

Access privileges for EVs into low noise zones or at nighttime

PedZones

Entry into pedestrian zones for EVs with city logistics approach

ZEZ

Access privileges for EVs into zero emission zones

Grausam et al. (2014)

Fiscal AFA

50% fiscal depreciation in year of purchase

Cash

Purchase price subsidy

NAPE (2014) and NPE (2014) Wietschel et al. (2013)

CityToll Diesel Projects

EV exemption from city toll Abolish subvention of diesel €0.219 Support of freight EV pilot projects

Grausam et al. (2014) Dudenhöffer (2013) NAPE (2014)

TaxIncent Tenders

Company tax incentive of € 100 per kW h (up to 20 kW h per vehicle) Demand freight EVs in tenders requiring goods transport

NPE (2012)

Organization Charging Setup of charging infrastructure on company compounds CycleLanes Cycle infrastructure improvement, facilitating freight transport with cargo cycles Hubs Space for micro-consolidation centers for freight EVs to operate in the near range and recharge the battery during loading of freight MunFleets Freight EVs in municipal fleet Repair Repair and service workshop b

Suggested

EmoG (2014) and NPE (2014) EmoG (2014)

NPE (2014)

EmoG (2014) and NPE (2014) EmoG (2014) and Grausam et al. (2014)

Amsterdam Oslo, Amsterdam, UK, Denmark Oslo, Newcastle-uponTyne Parma The Netherlands, Germany (since 31.12.2014) London, Amsterdam Oslo, Amsterdam, cities in UK, Denmark Paris Nuremberg, Hasselt, Gothenburg, Utrecht Venice

UK, The Netherlands, Denmark London, Oslo, Stockholm Germany, UK, Denmark and others Hamburg Bonna Copenhagen, Amsterdam

Aichinger (2014)

Brusselsb Amsterdam, Paris Dortmundc

Project: CO2-neutrale Zustellung in Bonn. http://www.eltis.org/discover/news/bonn-starts-co2-neutral-postal-delivery-germany-0. Project Straightsol. http://www.straightsol.eu/. Project Elmo. http://www.iml.fraunhofer.de/de/themengebiete/verkehrslogistik/themen_transportverkehrlogistik/Elmo.html.

4.1.2. Legal measures All ten identified legal measures are already implemented in European countries. Compared to other European countries, legal measures to support freight EVs are underdeveloped in Germany: of these measures, only one was adopted in Germany by the end of 2014 – the exemption to drive freight EVs up to 4.25 tons under certain conditions with a class B driver’s license. Further, four out of the ten identified measures are foreseen in the electric mobility law – privileged loading or parking zones; privileged entry into restricted areas due to noise restrictions; and the privileged use of bus lanes. By adopting the law, these possible measures are not yet implemented: with the marking of EVs, the federal government only provides the legal framework for federal states or municipalities to locally apply legal measures. The four measures discussed in the law could offer privileges to freight EV users in theory. However, in the electric mobility law it is stated that the directive targets to mitigate emissions apply to individual passenger transportation. Thus, the law only foresees the labeling of electric passenger cars, quadricycles, motorbikes and light commercial vehicles up to 3.5 tons (or 4.25 tons as exemptions). This means that freight EVs of higher weight classes are exempt from the possible benefits. Implementing more unpopular measures which would restrict drivers of conventional vehicles – such as access restrictions or zero emission zones – was more seldom discussed by experts and is not included in the law. A further identified measure is to certify fleets of companies that meet certain emission criteria. In Parma, Italy, only certified companies are allowed to deliver into the inner city area, making the certificate a legal measure. A less strict communication measure, suggested by companies in the pretest, is to issue a (federal) certificate as a token of appreciation of the companies’ efforts to achieve more sustainable transportation. Companies can use this certificate as a marketing tool in communications with their customers.

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

67

4.1.3. Fiscal measures Germany is one of the few countries which are actively fostering electric mobility, but has only implemented very limited financial subsidies. Direct purchase price subsidies as provided in many European countries are rejected by the federal government in the current phase (Merkel, 2013, p. 6). However, the purchase of electric freight vehicles can be subsidized in pilot projects. Implemented financial benefits for freight vehicles are the exemption from the vehicle tax. A fiscal depreciation of 50% in the first year for electric commercial vehicles was suggested by experts (NPE, 2014). A further suggestion of the expert group, counseling the German government, is to provide a company tax incentive of €100 per kW h of the EVs battery up to 20 kW h (NPE, 2012). This measure would offer a financial benefit, however freight EV users would not equally benefit as companies’ passenger car users, since the batteries of freight EVs are often larger than 20 kW h, currently up to 120 kW h. A further fiscal option is to prefer emissions-free vehicles in transportation tenders, for example, for the delivery of supplies or maintenance of technical equipment in offices. The city of Hamburg has implemented such a clause for transport tenders.5 A final measure suggested by Dudenhöffer (2013) is to discontinue the subvention of diesel fuel. The energy and environmental tax of diesel is about €0.18 (with is about €0.22 including VAT) lower than for petrol, although the nitrogen oxide and particulate matter emissions of diesel vehicles are problematic in many German cities. 4.1.4. Organizational measures The review found five organizational options for municipal administrations to use their own resources to support the adoption of electric freight mobility. The first option is to implement electric freight vehicles in own municipal fleets, or fleets sourced out to service providers. Examples are garbage removal, the maintenance of public parks or transporting books between subsidiaries of public libraries. Further options are to support the additional infrastructural needs of freight EVs as compared to conventional transport vehicles. Since EVs have a limited range, setting up a public charging infrastructure is commonly named to be one of the most important measures to increase the number of EVs (NPE, 2014). However, freight EVs are usually charged on company premises and thus require a charging infrastructure on the company premises instead. European examples highlight a further option to compensate the limited range: to offer space for micro-consolidation centers close to the city center. The hubs allow companies to operate their freight EVs within a near area. Moreover, the EVs can recharge their batteries during loading/unloading of freight at the consolidation center, increasing the usability and profitability of the freight EVs. An additional option is to support freight EVs in initiating repair centers. A fast and reliable maintenance and repair of their vehicles is essential for freight transporting companies (Taefi et al., 2016). The study found that abundant spare parts and trained mechanics for conventional vehicles were available, while non-series freight EVs need different spare parts and specially trained mechanics which were hard to come by in a timely manner. As a good-practice example, the City of Dortmund identified a repair shop which is able to provide spare parts and trained staff to repair all brands of electric freight vehicles in projects of the region. Finally, electric freight mobility generated new business models, such as utilizing electrically supported cargo cycles for urban road freight transport. Cargo cycles can replace small delivery vans in many urban transport applications (Reiter and Wrighton, 2014). The fast and bulky cargo cycles require an appropriate cycling infrastructure, to allow a safe operation in cities. The review fount that there is no strong focus on organizational measures, in discussions in Germany to date. 4.2. Rating of the measures Several criteria are rated by the two stakeholder groups for each of the identified measures. This section describes and discusses the results of the ratings for the criteria ‘‘time until implementation”, ‘‘consent” and ‘‘effect, effort and feasibility”. 4.2.1. Rating of the criterion ‘‘time until implementation” The participants of the study were asked to estimate how long it would take to implement a measure. The median of the answers is listed in Table 2 for each group and combined for all participants. ‘‘S” indicates measures that can be implemented within three years, ‘‘L” indicates measures that would take longer than three years. The rating of the two groups coincides for most measures. As discussed in Hood and Margetts (2007), communication and legal measures do not rely on depletable resources, and thus require in general relatively less effort (and hence time) to implement. The ratings of the respondents coincides with this theory. Information and legal measures are rated to require less time to implement. Exceptions are very restrictive measures, such as zero emission zones or access restrictions for conventional vehicles. Freight EV users are more pessimistic than policymakers about the duration before fiscal and organizational measures can be implemented. As an example, freight EV users estimate that those measures mainly driven by municipalities – ‘‘MunFleets”, ‘‘Tenders” and ‘‘Repair” – would take longer than three years. In general, both groups estimate that unpopular fiscal and organizational measures, such as abolishing the diesel tax subvention, establishing city tolls, setting up cycle lanes or consolidation hubs would require a longer period of time.

5

Section 3B(9) HmbGVBl. 2006, p. 57.

68

Group

Info Legal

Fiscal

Organizational

Access Bays BusLanes Certificate DrLicense LEZ Lots Noise PedZones ZEZ AFA Cash CityToll Diesel Projects TaxIncent Tenders Charging CycleLanes Hubs MunFleets Repair Policymakers S EV users S

L L

S S

S S

S S

S S

S L

S S

S S

S S

L L

S S

L L

L L

L L

S L

S S

S L

S S

L L

L L

S L

S L

All

L

S

S

S

S

S

S

S

S

L

S

L

L

L

S

S

S

S

L

L

S

S

S

S = within three years, L = longer than three years.

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

Table 2 Median rating of duration per measure and groups.

69

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

4.2.2. Rating of the criterion ‘‘consent” The group of freight EV users were asked to additionally rate the criterion ‘‘consent”, in order to offer a possibility to express their opinions concerning whether a measure would be acceptable to the company, see Section 3.2. The result of the rating is displayed in Fig. 1. The literature suggests that the TCO of electric freight vehicles is the predominant factor preventing companies from purchasing freight EVs, see Section 2. The rating of the criterion ‘‘consent” shows that freight EV users are widely, but not solely interested in financial subsidies: the average rating per category is the highest for informational measure (3.4), followed closely by legal and fiscal measures (both 3.3). The consent to organizational measures is lower on average (2.8). Among the top seven measures rated above 3.5 points on average, are four fiscal measures, see Fig. 1: a special fiscal depreciation (AFA), tax incentives (TaxIncent), purchase price subsidies (Cash) and that the government should use its financial resources to insist on freight EV usage in tenders (Tenders). Top-rated non-fiscal measures included the support for setting up charging infrastructure at the companies’ compounds (Charging), the privilege to drive freight EVs over 3.5 tons with a driver’s license class B (DrLicense), and free and privileged parking lots (Lots). The measure which the group of EV users rated to be the least desirable by far, is to discontinue the subsidy of the diesel taxation.

High

4.2.3. Rating of the criteria ‘‘effect, effort and feasibility” The ratings of the criteria ‘‘effect”, ‘‘effort”, and ‘‘feasibility” show a predominantly non-normal distribution of the data. Based on this indication, the methods for computing the correlation and further aggregations are chosen. No correlation was detected between the mean values of the criteria ‘‘effect”, ‘‘effort” and ‘‘feasibility” within the two groups. This indicates that the criteria are independent from each other and thus cover different fields. In the group ‘‘Freight EV Users” a correlation of 0.50 between the ‘‘consent” of a measure and the criterion ‘‘effect” indicates a moderate relationship: measures are more likely to be accepted by freight EV users if they are likely to raise the number of EVs in the companies’ fleets (or vice versa). However, due to the small number of respondents, the significance of the computed correlations is limited. The results give a first indication, but cannot be generalized. The average rating of the groups per measure and criterion can be found in Appendix in Table A.4. To provide a graphical overview of the results, the mean values of the ratings of the three criteria ‘‘effort”, ‘‘effect” and ‘‘feasibility” are combined in scatterplots. The dimensions are projected on the x-axis (‘‘effect”), y-axis (‘‘effort” – high and low were inverted for a better readability), and color coded markers (‘‘feasibility”). Measures that achieve a higher effect relative to the required effort are more efficient than others. Levels of similar efficiency are indicated by dotted gray lines in the plots. The Figs. 2 and 3 display the short-term measures, Fig. 4 the long-term measures for the two stakeholder groups. In order to derive suggestions for measures to support freight EVs, it is important to first define potential aims of introducing freight EVs: Ecological: A possible goal can be to focus on the reduction of the local road freight transport emissions, by introducing as many freight EVs as possible. This might be the case for areas or streets which do not comply with the local air pollution limitations. In this case the factor ‘‘effect” would be of main importance. Short-term measures that both groups rated as comparably effective are: to offer a fiscal depreciation of 50% in the first year after purchase (AFA, 4.22/3.88 (policymakers/freight EV users)); to allow freight EVs to enter pedestrian zones for delivery when being part of a city logistics approach (PedZones, 3.60/4.13); and to offer financial support in projects (Projects, 3.86/3.50). Regarding the effectiveness of the long-term measures, both groups agree that (in this order) a zero emission zone (ZEZ, 4.40/3.88), purchase price subsidies (Cash, 4.33/3.71) and a city toll (CityToll, 3.86/2.57) are the most effective measures. Economic: Fiscal and personal resources are often limited, hence a goal could be to implement the most efficient measures, by dividing the ratings of the criterion ‘‘effect” of a measure by the ‘‘effort”. However, the results of this study indicate that a measure which can be cost-efficiently implemented from a local policymaker’s perspective, might lead to higher costs in the companies which desire to deploy the vehicles, or vice versa, or might even lead to higher costs for the federal

Communication

Legal

Fiscal

Organizational

5 4.14

4

4

4

4

3.86 3.75 3.75 3.5 3.5 3.5 3.43 3.38

3

3.14 3.12

3

2.86 2.83

2.71 2.67

2.5 2.43 2.33

2

Diesel

Hubs

CycleLanes

Noise

Fig. 1. Mean rating of criterion ‘‘Consent” by freight EV users.

MunFleets

CityToll

NoAccess

LEZ

Repair

BusLanes

Certificate

Info

PedZones

ZEZ

Projects

Bays

Lots

Tenders

Cash

Charging

TaxIncent

AFA

DrLicense

Low

1.57

1

70

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

5

>> Higher Efficiency

1

High

Low

Feasibility

4

2 AFA

3

BusLanes

Bays

3

Charging

High

4

2

5

1

1

2

3

4

Low

Effort

Certificate TaxIncent Tenders DrLicense PedZones Repair Projects Noise LEZ Info Lots MunFleets

5 High

Low

Effect

Fig. 2. Scatterplot of policymakers’ short-term measures ratings.

2

Tenders

BusLanes Certificate

Effort

5

Repair

Info

3

Projects Bays TaxIncent

DrLicense Noise Lots LEZ

MunFleets

4 PedZones

AFA

3

Charging

4

High

High

1

2

5

1

1

2

3

Low Effect

4

Low

Low

Feasibility >> Higher Efficiency

5

High

Fig. 3. Scatterplot of freight EV users’ short-term measures ratings.

government. As an example, we regard the exemption to allow freight EVs to drive on bus lanes (BusLanes): to the federal government, granting this exemption is relatively effortless. In fact, the option has been adopted with the electric mobility law in mid-2015 in Germany. At the same time, this study finds that local policymakers evaluate the measure to be the least efficient by far. They assume that allowing freight EV to drive on bus lanes would be relatively ineffective and at the same time highly resource-intensive for the local government. On the contrary, to freight EV users the measure requires the second-least effort. The following three short-term measures are rated as comparably cost-efficient for both groups: to offer a fiscal depreciation of 50% in the first year after an EVs purchase (AFA, 1.81/1.50); to allow freight EVs to enter pedestrian zones for delivery (PedZones, 1.16/1.73); and to offer financial support in projects (Projects, 1.23/1.75). With regard to long-term measures, implementing a city toll is rated as the most efficient long-term measure by both groups (CityToll, 1.93/1.06). Social: To balance the social, ecologic and economic consequences of a measure ‘‘feasibility” might be utilized as a weighting criterion. Measures which are rated to be more feasible have a higher social acceptance. Both groups rate the following measures to be comparably socially acceptable: the exemption to drive EVs over 3.5 tons with a class B driver’s license (DrLicense, 3.40/3.43), the support of freight EV projects (Projects, 4.33/3.67) and to strengthen the information on the availability, costs and funding possibilities of freight EVs (Info).

71

CityToll

NoAccess Cash ZEZ

3

Hubs

1 Lo w

2

3

4

Effect

(a) Policymakers

5 High

Cash NoAccess

CycleLanes Hubs

High

High

CycleLanes

High

5

4

CityToll Diesel

4

5

>> Higher Efficiency

2

Diesel

3

1

3

ZEZ

4

2

5

1

1 Lo w

2

3

4

Low

>> Higher Efficiency

2

Effort

Low

Feasibility 1

Effort

Low

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

5 High

Effect

(b) Freight EV users

Fig. 4. Scatterplot of long-term measures ratings per group.

Timing: A further selection criterion can be the duration until a measure can be implemented. To achieve results as soon as possible, short-term measures may be preferred. The differences in the rating of the analyzed criteria indicate that it is important for municipal governments to clearly establish and communicate the aims, when supporting freight EVs. The city of Amsterdam in the Netherlands provides an elaborate example. The city aims to improve the local air quality and has identified the nitrogen oxide levels as most important indicator (Gemeente Amsterdam, 2011, 2013). After establishing the effects of potential measures in the most problematic areas, the cost-efficiency (for the municipality) of these potential measures were estimated. The resulting costabatement curve indicates measures that are preferable over others. For instance a subsidy of privately-owned electric passenger cars is considered not cost-efficient, while replacing company owned passenger cars by EVs is – since these vehicles are driving a higher average mileage. Further, offering a subsidy to replace medium-sized trucks by EVs is considered a costefficient measure, while the cost-abatement curve suggests that it is more favorable to subsidize the replacement of large trucks by Euro 6 vehicles in Amsterdam (Trip and Konings, 2014). 4.2.4. Recommendable policy options Different measures are preferable, depending on the defined local goal, as discussed in Section 4.2.3. However, a small number of policy options have been rated as comparably efficient and socially feasible by both groups. Hence, these measures can be interpreted as a minimal consensus and can be recommended independently from a specific goal: 1. 2. 3. 4. 5.

To support pilot projects which subsidize the purchase price of the freight EVs (Projects). To offer a fiscal depreciation of 50% in the first year after purchase (AFA). To subsidize the implementation of freight EVs by offering tax incentives (TaxIncent). To request emission-free freight vehicles in fleets that offer a transportation service to the municipality (Tenders). To allow drivers with a class B license to drive freight EVs over 3.5 tons (DrLicense), to transport a similar payload to diesel vehicles despite the heavy batteries. 6. In the long-term, to offer purchase price subsidies for freight EVs (Cash). 7. In the long-term, to implement a city toll (CityToll). Both groups rate all three criteria of the following two short-term measures rather low and thus rank these policy options as comparably not recommendable: the exemption for freight EVs to drive on bus lanes (BusLanes), since the public transportation with buses is an important priority in the cities; a low emission zone for freight transport (LEZ), since such a zone would not facilitate the use of EVs – diesel vehicles of Euro 4 emission standard and above would be used instead. In the long-term both groups rate the following two organizational measures as rather inefficient: setting up microconsolidation hubs to bundle goods and enable freight EVs to operate in the near-range (Hubs); as well as to improving the cycling infrastructure to allow the operation of wide and fast cargo bikes (CycleLanes). Both groups rate that the high effort does not result in an equally high effect. Local governments would need to invest into the infrastructure and redistribute the scarce and expensive public space. Freight EV companies would need to adapt their operations and planning. In comparison to improved cycle lanes or city hubs, both stakeholder groups rate areas or streets which are prohibited to entry with conventional vehicles (NoAccess) as more efficient. However, they feel that such a measure is not socially acceptable, even in the long-term. However, one has to keep in mind the relatively small number of respondents in this study (18 in total), which limits the generalization of this result.

72

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

5. Discussion This study aims to answering three investigative questions in order to analyze policy options in support of freight EVs in cities of Germany. Assuming that a city decided to support emission-free freight vehicles in order to reduce the negative impact of the urban road freight transport, we set out to screen the possible options. In answer to the first question – which are potential policy measures to support freight EVs – we collect policy options implemented in Europe, discussed by experts in Germany, or suggested by freight EV users. The compiled list of 23 potential measures is the first comprehensive collection of policy tools supporting freight EVs. So far, the literature has either discussed single measures; compared measures of a certain type, such as fiscal measures; or has not clearly differentiated between policy options to support individual or commercial passenger transportation and commercial freight transport, although the segments differ largely with regard to the operational profiles, requirements, as well as vehicle types. As an example, only six of the top ten measures discussed in the multi-criteria analysis of Bakker and Trip (2013) could directly support the uptake of freight EVs (support infrastructure build-up; show political leadership (e.g. EVs in fleet); provide information to businesses and citizens; reserve on-street parking spaces for EVs; exemption from toll for EVs; allow EVs to drive on bus/taxi lanes) while top measures identified by the current study, such as fiscal measures fro freight EVs, or allowing drivers with a class B license to drive freight EVs over 3.5 tons (DrLicense) are not included compare Section 4.2.4. In future discussions or when transferring the collection of possible measures to other countries, further or more detailed policy options supporting freight EVs might extend or adjust the measures derived in our study. The further two investigative questions answered by this study are to rate and recommend measures for the case of Germany. According to Hood and Margetts (2007), a quality criterion of a policy measure is its scalability, which means that a measure can be varied in its intensity. All measures researched and recommended in this study are scalable, either with regard to the economic value (for example for fiscal measures), the spatial dimension (for example for the area in which an exemption is granted) or the level requirement (for example the criteria which need to be fulfilled to issue a certificate or allow the entrance into a pedestrian zone as part of a city logistics approach). Hence, all discussed measures might be of interest to local governments, although some score better for their ecological, economic and social feasibility. Our study suggests that both freight EV users and policymakers agree on the applicability of certain policy options, see Section 4.2.3. These measures are recommendable independently from a more specific local goal. Beyond generally acceptable measures, choosing the most appropriate policy options depends on the goal of a city – whether it is desired to implement rather effective, efficient or socially acceptable measures, short-term or long-term options. When taking into account specific local goals, as discussed in Section 4.2.4, recommended freight transport-related measures differ from policy recommendations to foster electric mobility in general, i.e., discussed by Bakker and Trip (2013). The current research of policy options in electric mobility tends to focus on the rating of policymakers alone. In order to achieve a differentiated understanding of the estimated effect, efficiency and feasibility of the measures, the current study additionally includes the ratings of a group of freight EV users. The differences between the ratings of the groups offer an interesting potential for understanding conflicting opinions and their potential reasons. Hence, in the following section we will discuss and interpret differences between the groups in more detail. 5.1. Assessing differences of the ratings A graphical examination of the scatter plots in Figs. 2–4 indicates that for certain measures larger differences exist in the ratings of the two stakeholder groups. In order to analyze the differences systematically, we compare the ratings in two ways: the mean values of the criteria per category are aggregated in Fig. 5 and the difference per measure is assessed in Table 3. In Table 3, the single measures are ranked according to their rating. The lowest rank (1) indicates that the measure received the best mean rating of the respective group: The highest rating for the criteria ‘‘effect” and ‘‘feasibility”, the lowest rating for the criterion ‘‘effort”. Subsequently, the differences of the ranks are evaluated. In case the rank difference (r) is larger than or similar to half of the number of measures (n), the gap is considered to be large, indicated by a filled circle. In case the rank difference is smaller than a quarter of the number of measures, the difference is considered small. Otherwise, the gap is considered to be intermediate. As an example, there are n = 16 short-term measures. The ranking difference for the policy option ‘‘Bays” is 6 (intermediate: f) for the criterion ‘‘Effect”; 13 (large:
) for the criterion ‘‘Effort”; and 3 (small: ) for the criterion ‘‘Feasibility”. The scatter plots in Fig. 5 indicate the average opinion of the two groups regarding the rated policy options per category. Based on the classic assessment of Hood and Margetts (2007), we expected that the policymakers – representing the government – would rate communication and legal measures to require the least effort, since those measures are nondepletable; and communication and fiscal measures would impose less bureaucracy and hence effort to the public – represented by the freight EV users. 5.2. Rating differences of fiscal measures Freight EV users and policymakers agree that fiscal measures are on average the most efficient measures. Fiscal measures require a low effort and have a high effect, although they are rated to be only moderately socially feasible. The latter rating is

73

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

2

Effort

Fiscal

Effort

5

>> Higher Efficiency

4

2

3

Legal

Comm.

3

Organizational

5 1

2

3

Fiscal Legal

Comm.

3

Organizational

4 High

4 High

1

4

Low

2

5

5

1 1

2

3

4

Low

High

High

>> Higher Efficiency

Low

1

Low

Low

Feasibility

5 High

Effect

Effect

(a) Policymakers

(b) Freight EV users

Fig. 5. Average rating of categories per group.

Table 3 Ranking and ranking difference of the measures criteria. Criterion

Short-term measures MunFleets Bays BusLanes Projects Certificate Lots DrLicense Repair PedZones Tenders Charging TaxIncent LEZ AFA Noise Info

Difference

Freight EV users

Policymakers

(A)

(B)

(C)

(A)

(B)

(C)

(A)

(B)

(C)




f

   

13 6 12 6 14 8 4 16 1 9 2 5 11 2 9 14

16 2 2 1 7 13 8 10 6 4 15 4 14 8 10 12

9 13 14 2 11 4 3 11 9 7 5 7 14 5 14 1

2 12 16 4 7 9 10 12 5 3 8 11 12 1 6 15

12 14 14 10 2 12 5 7 6 4 14 3 7 1 9 11

6 10 16 1 2 12 7 4 15 9 5 10 14 8 13 2

5 6 2 4 7 3 1

3 7 1 4 5 1 6

1 7 2 5 4 3 6

4 6 2 5 7 3 1

2 6 4 3 6 1 5

6 2 4 7 1 3 5

f f



f 

f f f f f f

   

Longer-term measures Diesel  Hubs  Cash  NoAccess  CycleLanes  CityToll  ZEZ 






f       



f f f

 

      

 





   

 

f f

f

 ¼ Small ðr < n=4Þ f¼ Medium ðn=4 P r < n=2Þ (A) Effect (B) Effort (C) Feasibility.

f f f


¼ Large ðr P n=2Þ.

consistent, since allocating financial resources to emission free transportation would implicate a reduced availability of financial resources for certain other measures. Also the high rating of the effect of fiscal measures is not surprising, since the literature agrees that the higher costs of freight EVs are one of the main hurdles for the broader uptake of electric mobility. Furthermore, the low rating of the effort for freight EV users is in line with the expectations, based on the assessment of Hood and Margetts (2007): freight EV users would receive financial support when utilizing freight EVs, thus the low rating of their effort is obvious. However, why policymakers rated fiscal measures to require the lowest effort of all categories requires discussion: the factor ‘‘effort” is a combination of required personnel and monetary resources. Five of the six suggested fiscal measures undoubtedly require funding (measure six is a city toll which could, in sum, generate a financial surplus). An explanation for the rating is, that in Germany, as a federal state, responsibilities and budgets are strictly separated between the

74

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

national and the federal states governments. Presumably, the local policymakers expected from their experience, that the monetary component for fiscal measures would be mainly payed from federal budgets rather than out of their own municipal budgets. An inclusion of a third group ‘‘national policymakers” would be interesting for future studies, to support this theory and reveal the differences between the two policymakers groups opinions. Table 3 shows a large difference between the rating of the two groups with regards to two fiscal measures. The freight EV users rate receiving a subsidy for freight EVs in pilot projects (Projects) to require the lowest effort of all measures, whereas to policymakers estimate a rather high financial or personal effort for their organization. This rating is comprehensible, since a 25–50% subsidy of the purchase price of a freight EV is a large sum transferred from the governments funds to the freight EV users. In some cases, local budgets might have to supplement a national funding and the local governments are often responsible to coordinate activities in their cities, hence the effort for the policymakers is comparably higher than for the freight EV users. Surprisingly, this study finds that freight EV users rank higher taxes on diesel fuel (Diesel) as the most feasible long-term option – although to them the measure seems comparably ineffective, only moderately efficient and is the most disliked measure of all, when asked directly, see Fig. 1. Policymakers rate the general social acceptability of increasing the diesel fuel tax nearly as low as complete access restrictions for vehicles with combustion engines. A possible explanation of the relatively high score of the criterion feasibility in the group of freight EV users is, that companies which already utilize freight EVs are early adopters of electric mobility. One of their reasons to test electric freight vehicles is to achieve more sustainable means of freight transportation (Taefi et al., 2015). Hence, they might have a high level of awareness regarding the negative consequences of conventional diesel vehicles deployed in urban road freight transport. A hypothesis to be tested in further research is that the involvement with electric freight transport options, for example through information and freight EV usage, might raise the level of awareness of negative consequences of freight transport compared to non-freight EV users, and thus paves the way for the acceptance of a more sustainable legislation. 5.3. Rating differences of legal measures The importance of implementing legal measures to increase the market share of passenger EVs has been underlined by Figenbaum et al. (2015), who found that the exemption to drive on bus lanes was a crucial factor for the EV market development in Norway; as well as Langbroek et al. (2016) who showed in a stated-choice experiment that access to bus lanes or free parking are efficient policy options. Despite this, our study finds that privileges for electric or penalties for conventional freight vehicles are rated only moderately feasible by both stakeholder groups. A particular large gap is found in the ranking of the feasibility of the measure ‘‘Lots”. Freight EV users rate the granting of free and/or privileged parking lots for EVs as a socially feasible measure. Policymakers, on the contrary, estimate that the measure has a similar low socially acceptance as comparable measures. Presumably, the policymakers were influenced by a stormy debate in the media in 2014. Possible privileges for EVs, especially driving on bus lanes, as suggested in the electric mobility law, was publicly disputed and rejected by several stakeholders. The risk of becoming the target of such a debate might also explain the policymakers comparatively high rating for the effort of all measures that visibly offer advantages to electric vehicles in traffic. According to the prospect theory, people are systematically biased by the risk to ‘‘loose” more than they love to ‘‘win” (Mercer, 2005). This risk aversion of the policymakers might lead to the large differences between the groups in the ranking of the effort of the measures ‘‘Lots”, ‘‘Bays” and especially ‘‘BusLanes” in Table 3: utilizing privileged loading zones (Bays) or being allowed to drive on bus lanes in inner cities (Buslanes) requires comparatively low effort for the companies. Policymakers, on the contrary, rate these two legal measures to require the highest effort (together with one fiscal measure). Five of the policymakers used the comment field of the survey to explain that either their city does not have bus lanes, that the measure might lead to undesired obstruction of the public private transportation with buses, or that the measure is not enforceable, since the German Association of Cities and other important stakeholder groups have rejected the idea. Despite the apparent higher rating of the policymakers’ effort to implement legal measures, the rating of this criterion is still in line with our expectation based on Hood and Margetts (2007): policymakers rate the effort for communication and legal measures to be lower than organizational measures. In contrast to this, freight EV users rated the effort for legal measures lower than expected, even lower than communication measures, see Fig. 5. A possible reason for this rating is that legal measures are desired by freight EV users’, while only one legal measure has been implemented so far in Germany. This is underlined by the average ratings of the criteria consent and effect: the freight EV users’ average consent to legal measures is similar to fiscal measures, see Section 4.2.2. Furthermore, freight EV users rate the average effect of legal measures (3.1) to be nearly as high as fiscal measures (3.2), while policymakers estimate that fiscal measures are more effective (3.9 vs. 3.3). In particular, the effect of freight EV loadings zones (Bays) is underestimated by policymakers compared to freight EV users. At the same time, policymakers rank the effect of free or privileged parking lots (Lots) to be higher than for privileged freight loading zones. Except for short-term parking possibilities for postal and courier services, there is no indication how privileged parking lots could support the uptake of freight EVs, and even for the named segment loading bays could be equally interesting. This evaluation is underlined by the ranking of the freight EV users, who rank the effect of privileged parking lots lower than privileged loading zones. We assume that this (mis-) rating is symptomatic of the discussion of potential EV privileges: the German public discussion as well as the electric mobility law are focusing on the uptake of electric vehicles for passenger transportation. The electric mobility law adopted in mid-2015 provides a framework for

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

75

municipalities to implement legal benefits for electric vehicles, but deliberately excludes freight EVs above 4.25 tons from legal benefits. With this in mind, the policymakers’ answers might be biased towards the passenger transportation segment, even when being asked to rate measures for freight transportation. Recognizing the differences between the passenger and freight transportation segments and understanding the different needs are important when deciding on supportive legal policy options for freight EVs. 5.4. Rating differences of communication measures Both groups agree that communication measures are rather inefficient, require a mediocre effort compared to other measures, but are socially feasible, see Fig. 5. Obviously, the effort of preparing and distributing information is comparably higher for policymakers (3.4) than for freight EV users retrieving and processing them (2.9). Further, both groups estimate that the effect of communication measures would be the lowest of all measures, albeit the freight EV users are even more critical (2.1) than the policymakers (2.8). One company representative commented ‘‘An electric vehicle cannot be sugarcoated – it does not amortize”. However, it has to be considered that of the few freight EV users in Germany, presumably the companies most active in electric mobility answered the questionnaire. Those pioneer EV users possibly are not in need of more information about use cases, existing EV technology, advantages and challenges. Communication measures would rather aim at potential freight EV users. In consequence, it might be interesting to add the group of ‘‘potential freight EV users”, or even ‘‘prospective non-EV users” to the survey, in order to research their opinion regarding the effect of information measures. At the same time, one should stay aware that fragmenting groups within one actor can lead to a bias in multi-criteria analyzes (Macharis and Bernardini, 2015). Communication and acting as a role model are common and important administrative instruments to raise awareness and acceptance. Information on freight EVs, purchasing freight EVs for the governments own fleets and requesting emission-free freight transportation in tenders would mainly be implemented locally in municipalities and cities. However, centralized support might be preferable to prevent local actors from ‘‘reinventing the wheel”. A communication ‘‘toolbox” to promote electric freight mobility would be useful, but is not available yet in Germany. 5.5. Rating differences of organizational measures Both groups rate organizational measures to be rather inefficient, see Fig. 5. Organizational measures on average require the highest effort of all options, consistent with Hood and Margetts (2007). An interesting exemption is the rating of the option to set up charging infrastructure on the company premises: freight EV users rate the measure as one of the five most desirable measures when asked directly, see Section 4.2.2. The ranking in Table 3 also shows, that to freight EV users this measure is the second most effective. The potential positive effect of supporting charging infrastructure can be compared to the private car segment, where an adequate charging infrastructure is often named as one of the most important measures to enable electric mobility (Bakker and Trip, 2013; Grausam et al., 2014; NPE, 2014), albeit the reasons differ; the public charging infrastructure for private transportation is mainly necessary to overcome the range anxiety, although the majority of the EV users charge at home or at work (Morrissey et al., 2016). To companies, a charging infrastructure on their premisses is essential in order to ensure that the freight EVs are fully operational on the next day, or even in order to prolong the daily range by recharging (Taefi et al., 2015). While policymakers were slightly positive (3.3) about the feasibility of organizational measures in general, freight EV users had doubts (2.3). In particular, freight EV users rank the social acceptability of micro consolidation hubs (Hubs) as the least feasible option, while policymakers rank the measure to be the second most feasible in the long-term. A possible explanation of this result is that the freight EV users refer to previous experiences with pilot projects testing consolidation centers in Europe. Related city logistics approaches almost all ceased to operate due to a lack of profitability and coordination problems, once the project support finished (Wolpert, 2013). Policymakers, on the contrary, might be aware that city logistics approaches, such as consolidation centers, are envisioned in the long-run (European Commission, 2011). Despite the problems of these approaches in the past, further city logistics measures will be necessary to achieve a socially acceptable balance of livable cities, while ensuring the supply of the inhabitants of growing urban areas with goods and services. Taking into account the design of this study, the large difference of the rating of the effect of the organizational measure ‘‘MunFleets” in Table 3 is evident. Freight EV users indicate that raising the numbers of EVs in the governments fleets would not lead to an increase in their own fleets. Policymakers consistently rate the measure to be effective in raising the number of freight EV in their cities. However, one policymaker commented that due to the limited financial budget, the purchase of freight EVs for their own fleets ‘‘is not easy-going”. The differences between the groups were expected, regarding the setup of the questionnaire and is an indication that the groups have understood the questions and given conclusive answers. The measure to support the setup of a repair center (Repair) received an intermediate lower rating with regard to the effect and feasibility from the freight EV users than from policymakers. Similarly to communication measures, we suppose that the early adopters of electric freight mobility have successfully managed to implement their own service and repair structures for their EVs. It would be interesting to let a group of prospective EV users rate this measure and analyze the possible differences.

76

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

6. Conclusion This study aims to answer the question of which policy measures are recommendable in order to support electric vehicles in urban road freight transport in Germany. The aim of this study was to not only capture measures that support the purchase, but also the operation of freight EVs. In order do derive recommendations we followed the requirements for designing ‘‘intelligent” policies, as suggested by Hood and Margetts (2007): firstly we collected different possible policy options in order to allow a deliberate choice. Subsequently, our study assessed the potential policy measures based on the ratings of two expert groups, policymakers and freight EV users. The groups rated the measures’ environmental effectiveness, their social acceptability, and their economic efficiency for the government and the public. Finally we discussed the results, derived recommendations and come the following four conclusions. 6.1. Electric urban freight mobility requires dedicated policy measures Urban road transport is an important driver for the prosperity and economy in cities. At the same time governments strive to limit its negative environmental, social and economic effects. At a local level, trucks – which are nearly exclusively used for freight transport purposes (Wermuth, 2012) – are the main source of traffic-related air pollutants and noise emissions (Böhm and Wahler, 2012; Menge, 2013). Hence, increasing the number of EVs in urban road freight transport would mainly reduce the emissions of air pollutants and noise, as well as greenhouse gas to a smaller extent. Despite the local advantages and due to Germany’s national priorities of becoming a lead market and lead supplier of electric vehicles – both goals to which electric freight vehicles can only contribute sparsely – electric freight mobility is still in its infancy in Germany: (i) only a few companies utilize electric vehicles for urban road freight transporting purposes, albeit some of them in large quantities, see Section 3.1; (ii) though studies found that the vehicles are technically suitable, the EVs are often considered as too expensive to acquire outside of pilot projects (Amburg and Pitkanen, 2012; Taefi et al., 2015; Tenkhoff et al., 2012); (iii) purchase coalitions have formed to acquire electric passenger vehicles, but except for small delivery vans of about 2.3 tons, no series freight vehicles are available in Germany. Series production is often a prerequisite for the subsidy of purchase coalition projects; (iv) the German draft electric mobility law aims to support electric passenger transportation and foresees the labeling of electric vehicles up to 4.25 tons, in order to provide a legal framework for municipalities and cities for granting privileges to EVs. This excludes freight EVs over 4.25 tons from possible privileges. The scientific literature suggested already in 2001 that the marketing of electric vehicles should target early adopters (Gärling and Thogersen, 2001). More recently, Green et al. (2014) found that in order to increase EV policies’ effectiveness and efficiency, policy should foster niche markets such as parcel transport, instead of maintaining general undirected measures. The current study compiles the highest-scoring measures into generally recommendable policy measures. The identified measures differ from previously assessed general electric mobility recommendations, which are often directed at individual passenger transportation. Examples of recommended measures which do not target the individual passenger transportation segment are to support freight EVs in pilot projects focusing on freight transport; to offer companies a high fiscal depreciation in the year of purchasing an EV; to subsidize the implementation of freight EVs by offering tax incentives to companies; to demand for emission-free transportation in tenders; or to make exemptions regarding the driver’s license by subtracting the battery weight from the gross vehicle weight to enhance the utilization of a class B license (the latter measure has been implemented in Germany since the end of 2014); or, in the long-term, to implement a city toll. The results of our study indicate that the German policymakers who took part in the study were to a certain extent biased by a focus on passenger transportation, even when explicitly rating policy options for freight transportation. The preference of privileged parking lots over loading bays serves as an example. A clear differentiation between the segments ‘‘private passenger” and ‘‘freight” transportation is advisable when designing EV policies in the future. 6.2. Recommendable policy measures vary, depending on the local implementation goal This study identifies certain generally recommendable measures, but suggests that further policy measures depend on the local implementation goal. Because municipalities and cities face complex situations with individual challenges (Menge, 2013), we propose that administrations need to determine and clearly communicate their goals, when planning to reduce emissions by supporting freight EVs. Possible goals are to adapt economically efficient, environmentally effective, socially feasible or short- or long-term measures. Hence, a transparent debate on the aim of supporting electric freight mobility is as necessary as choosing measures targeted at the freight transportation segment. 6.3. Policymakers underestimate the effect of legal measures In contrast to existing research, this study includes ‘‘freight EV users” as a second stakeholder group to ‘‘policymakers” for rating proposed measures to support EV policy options. In comparing the results of the groups, we find that the consent and discordance between the groups can be large and offer noticeable insight and room for future research and practice.

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

77

Providing financial subsidies is often the primary subject of the discussion in the literature to overcome the TCO gap of freight EVs. In accordance to this, we find fiscal measures among the recommendable policy options to support freight EVs. However, this study also suggests that the effects of legal measures are underestimated by the policymakers. Examples of underestimated measures are privileged loading zones for freight EVs, exemptions regarding the driver’s license, or access rights to pedestrian zones as part of a city logistics approach. A prerequisite to benefit from the legal measures is to include EVs above 4.25 tons in the electric mobility law. A further measure underrated by the policymakers is to support a charging infrastructure at the premises of the respective companies. The option to provide a public charging infrastructure for individual EV users is well researched. Analyzing different options of supporting company-charging infrastructure is an open research opportunity, which can become especially interesting when larger fleets need to be charged. 6.4. Further stakeholder opinions are required for transport policy evaluation The differences identified by our study between the two stakeholder groups, policymakers and freight EV users, underlines the research of Ballantyne et al. (2013). The authors claim that the needs of freight operators are still not fully understood by political decisionmakers. Thus, we suggest to integrate the stakeholder group of freight EV users into future freight transport policy evaluation. Moreover, we find unexpected inconsistencies for certain ratings. These might be reduced by including further stakeholder groups in rating freight EV policies: the ratings of the criterion ‘‘effort” were unexpectedly low for the local policymakers. We hypothesize that including ratings of national policymakers would lead to different results. In case further research supports this claim, both levels of governments should be included when designing EV policy. Further, the ratings of measures that support a decision to become engaged in electric freight mobility, such as information, or a repair center, were unexpectedly low in the group of freight EV users. The early adopters who replied to the questionnaire in our study are electric mobility experts, who presumably already gather the necessary information and established the required structures. Including a group of ‘‘potential freight EV users” or even ‘‘prospective non-EV users” might relativize the ratings of the expert group for measures which support the uptake of freight EVs. However, fragmenting groups within one actor can lead to a bias in multi-criteria analyzes, which needs to be considered (Macharis and Bernardini, 2015). 6.5. Limitations, contributions and outlook As a limitation, the number of respondents rating the Web-based survey were small. Thus, this study can only provide a first indication of the ratings and the results are not fully generalizable. Receiving ten answers per group was the goal set at the start of the study. With eight responses in the group of ‘‘freight EV users”, the number of responses was lower than initially intended in this group. This low number is presumably caused by several factors: utilizing freight EVs is in its infancy in Germany – not many freight transporting companies deploy freight EVs at all; only limited data about companies which utilize freight EVs are publicly available (for instance, in project reports or pilot project Websites); project leaders of freight EV pilot projects did not respond to the request of forwarding the invitation to participating freight companies; thus only a small sampling group of 16 freight EV users could be identified; only half of those companies responded to the questionnaire. A possible explanation for a response rate of only 50% is that these early adopters might be approached too often with requests to answer questionnaires and interviews, and thus become less responsive. A future survey to question freight EV users could be integrated in the accompanying research of umbrella pilot projects, such as the electric mobility pilot regions or the showcase regions in Germany. This would make freight EV users more accessible and possibly raise their motivation to answer a questionnaire. However, future studies in the field of rating and ranking policy measures can build on several contributions of this study, which are adaptable for other countries: the compilation of possible measures to support freight EVs; the enhanced methodology of clustering the policy options; the list of rated criteria – which could be possibly extended. Furthermore we discussed the need of dedicated administrative goals, we applied a ranking methodology and analyzed the consent and discordance between the two sampling groups. For policymakers this study can hopefully highlight that freight EVs have special requirements and give a first indication of measures and activities to promote freight EVs and thus reduce urban transport related emissions. This study leads to formulating the hypothesis that experienced freight EV users are more willing to accept unfavorable policy measures in the long-term compared to the broader public, since EV users are aware of the negative impacts of freight transportation and are participating in establishing a solution. This hypothesis is based on the rating of the option to abolish the subvention of the diesel taxation and poses an opportunity for further research. Finally, our study has not analyzed which of the collected measures might be more sustainable than others in the longterm. The example of Norway suggests that legal and fiscal privileges for electric (private) vehicles are critical incentives (Bjerkan et al., 2016) and may kick-start the demand to support an early market. However, as the number of EVs increased, these measures lead to adverse effects (Aasness and Odeck, 2015) need to be scaled down or discontinued in the near future (Myklebust, 2013; van der Steen et al., 2015). One hypothesis to be explored is whether organizational and communication measures, as well as restrictive fiscal and legal measures can possibly lead to the ‘‘big leap” in taking up freight EV numbers that are necessary to meet given and future environmental standards.

78

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

Table A.4 Average rating of measures per group. Criterion

BusLanes Bays Lots PedZones LEZ NoAccess ZEZ Noise DrLicense Info Certificate MunFleets Tenders AFA TaxIncent Projects Diesel Cash CityToll Charging CycleLanes Hubs Repair a

Freight EV users

Policy makers

Effect

Effort

Efficienta

Feasible

Effect

Effort

Efficienta

Feasible

2.38 3.50 3.25 4.13 2.71 2.50 3.88 3.00 3.71 2.14 2.14 2.33 3.00 3.88 3.57 3.50 2.43 3.71 2.57 3.86 1.83 2.42 1.86

3.75 3.75 3.13 3.63 3.00 3.67 2.63 3.17 3.43 3.14 3.75 2.83 3.71 3.43 3.71 4.00 3.29 3.57 3.57 2.86 2.67 2.43 3.17

1.06 1.56 1.13 1.73 0.90 0.88 1.19 1.06 1.44 0.75 0.88 0.74 1.31 1.50 1.56 1.75 0.89 1.53 1.06 1.22 0.55 0.68 0.66

2.00 2.38 2.88 2.50 2.00 1.83 1.75 2.00 3.00 3.71 2.43 2.50 2.57 2.71 2.57 3.67 3.14 2.43 2.14 2.71 2.00 1.57 2.43

1.67 3.00 3.40 3.60 3.00 3.70 4.40 3.56 3.30 2.80 3.50 4.10 3.90 4.22 3.22 3.86 3.75 4.33 4.14 3.44 3.00 3.33 3.00

2.44 2.44 2.50 2.90 2.75 3.00 2.30 2.78 3.10 2.60 3.50 2.50 3.20 3.67 3.22 2.86 3.75 2.56 3.86 2.44 1.22 1.22 2.88

0.47 0.84 0.97 1.16 0.92 1.23 1.19 1.10 1.14 0.82 1.40 1.17 1.39 1.81 1.16 1.23 1.67 1.26 1.93 0.97 0.63 0.70 0.96

1.00 3.00 2.90 1.50 2.75 1.80 2.20 2.89 3.40 3.80 3.80 3.60 3.10 3.22 3.00 4.33 1.88 2.38 2.57 3.63 2.89 2.67 3.75

The efficiency was calculated by dividing the ratings of effect by effort.

Acknowledgments The research on which this paper has been based was carried out partly within the North Sea Region Interreg IVB project E-Mobility NSR. We would like to thank the anonymous reviewers for their comments and suggestions to improve this manuscript. Appendix A See Table A.4. References Aasness, M.A., Odeck, J., 2015. The increase of electric vehicle usage in Norway – incentives and adverse effects. Eur. Transp. Res. Rev. 7, 34. Aichinger, W., 2014. Elektromobilität im städtischen Wirtschaftsverkehr. Deutsches Institut für Urbanistik. Amburg, B.V., Pitkanen, W., 2012. Best fleet uses, key challenges and the early business case for e-trucks: findings and recommendations of the e-truck task force. In: EVS26 International Battery, Hybrid and Fuel Cell Electric Vehicle Symposium, Los Angeles, California, pp. 1–12. Bakker, S., Trip, J.J., 2013. Policy options to support the adoption of electric vehicles in the urban environment. Transport. Res. Part D 25, 18–23. Ball, M., Wietschel, M., 2009. The future of hydrogen – opportunities and challenges. Int. J. Hydrogen Energy 34, 615–627. Ballantyne, E.E., Lindholm, M., Whiteing, A., 2013. A comparative study of urban freight transport planning: addressing stakeholder needs. J. Transp. Geogr. 32, 93–101. Bjerkan, K.Y., Norbech, T.E., Nordtomme, M.E., 2016. Incentives for promoting Battery Electric Vehicle (BEV) adoption in Norway. Transport. Res. Part D 43, 169–180. Böhm, J., Wahler, G., 2012. Luftreinhalteplan für Hamburg, 1. Fortschreibung. Behörde für Stadt und Umwelt Hamburg. (last accessed on April, 15th 2015). Bozem, K., Nagl, A., Rath, V., Haubrock, A., 2013. Elektromobilität: Kundensicht, Strategien, Geschäftsmodelle. Springer Fachmedien Wiesbaden, iSBN: 9783-658-02627-1. Bozzoa, R., Concaa, A., Marangonb, F., 2014. Decision support system for city logistics: literature review, and guidelines for an ex-ante model. Transport. Res. Proc. 3, 518–527. Bryman, A., 2012. Social Research Methods. Oxford University Press, Oxford New York, iSBN 9783642200717. Driscoll, A., Lyons, S., Mariuzzo, F., Tol, R.S.J., 2013. Simulating demand for electric vehicles using revealed preference data. Energy Policy 62, 686–696. Duarte, G., Rolim, C., Baptista, P., 2016. How battery electric vehicles can contribute to sustainable urban logistics: a real-world application in Lisbon, Portugal. Sustain. Energy Technol. Assess. 15, 71–78. Dudenhöffer, F., 6, 2013. Dieselfahrzeuge und Stickoxide. In: Ifo Schnelldienst. Ifo Institut. (last accessed on February, 14th 2015) E-Mobility NSR, 2013. Comparative Analysis of European Examples of Schemes for Freight Electric Vehicles. Online. (last accessed on Febuary 10th, 2015). EmoG, 2014. Gesetz zur Bevorrechtigung der Verwendung elektrisch betriebener Fahrzeuge (Elektromobilitätsgesetz – EmoG). Drucksache 80/15. European Commission, 2011. White Paper: Roadmap to a Single European Transport Area – Towards a Competitive and Resource Efficient Transport System, Brussels. Figenbaum, E., Fearnley, N., Pfaffenbichler, P., Hjorthol, R., Kolbenstvedt, M., Jellinek, R., Emmerling, B., Bonnema, G.M., Vagane, F.R.L., Iversen, L.M., 2015. Increasing the competitiveness of e-vehicles in Europe. Eur. Transp. Res. Rev. 7, 28.

T.T. Taefi et al. / Transportation Research Part A 91 (2016) 61–79

79

Fraunhofer IML, 2010. Trendstudie Handelslogistik. EHI Retail Institute e. V. Köln. Gärling, A., Thogersen, J., 2001. Marketing of electric vehicles. Busin. Strategy Environ. 10, 53–65. Gass, V., Schmidt, J., Schmid, E., 2012. Analysis of alternative policy instruments to promote electric vehicles in Austria. Renew. Energy 61, 96–101. Gemeente Amsterdam, 2011. Schone lucht voor Amsterdam. Gemeente Amsterdam, 2013. Evaluatie amsterdamse maatregelen luchtkwaliteit. Grausam, M., Müller, U., Ried, W., Sperling, C., Wolpensinger, H., 2014. Elektromobilität in Kommunen. Handlungsleitfaden. Bundesministerium für Verkehr und digitale Infrastruktur (BMVI), Berlin. Green, E.H., Skerlos, S.J., Winebrake, J.J., 2014. Increasing electric vehicle policy efficiency and effectiveness by reducing mainstream market bias. Energy Policy 65, 562–566. Hood, C., Margetts, H., 2007. The Tools of Government in the Digital Age, second ed. Palgrave Macmillan, iSBN: 9780230001435. Kley, F., Lerch, C., Dallinger, D., 2011. New business models for electric cars – a holistic approach. Energy Policy 39, 3392–3403. Kraftfahrt-Bundesamt, 2015. Fahrzeugzulassungen (FZ) Bestand an Kraftfahrzeugen nach Umwelt-Merkmalen, 1 January 2015. Langbroek, J.M., Franklin, J.P., Susilo, Y.O., 2016. The effect of policy incentives on electric vehicle adoption. Energy Policy 94, 94–103. Lee, D.-Y., Thomas, V.M., Brown, M.A., 2013. Electric urban delivery trucks. Environ. Sci. Technol. 47, 8022–8030. Lützenberger, M., Masuch, N., Küster, T., Keiser, J., Freund, D., Voss, M., Hrabia, C.-E., Pozo, D., Fähndrich, J., Trollmann, F., Albayrak, S., 2014. Towards a holistic approach for problems in the energy and mobility domain. Proc. Comput. Sci. 32, 780–787. Macharis, C., 2005. The importance of stakeholder analysis in freight transport. Eur. Transp., 114–126 Macharis, C., Bernardini, A., 2015. Reviewing the use of multi-criteria decision analysis for the evaluation of transport projects: time for a multi-actor approach. Transp. Policy 37, 177–186. McKinsey, 2014. Electric Vehicle Index July. (last accessed September, 23th 2015). Menge, J., 2013. Relevance of commercial transport from a municipality’s perspective. In: Arndt, W.-H. (Ed.), Difu-Impulse: Städtischer Wirtschaftsverkehr. Dokumentation der Internationalen Konferenz 2012 in Berlin (ISBN: 9783881185189). Mercer, J., 2005. Prospect theory and political science. Ann. Rev. Polit. Sci. 8, 1–21. Merkel, A., 2013. Elektromobilität bewegt weltweit: Manuskript Angela Merkel. (last accessed on February 25th, 2015) Mock, P., Yang, Z., 2014. Driving Electricfication – A Global Comparison of Fiscal Incentive Policy for Electric Vehicles. Tech. Rep., The International Council on Clean Transportation. Morrissey, P., Weldon, P., O’Mahony, M., 2016. Future standard and fast charging infrastructure planning: an analysis of electric vehicle charging behaviour. Energy Policy 89, 257–270. Myklebust, B., 2013. EVs in bus lanes – controversial incentive. In: EVS27 International Battery, Hybrid and Fuel Cell Electric Vehicle Symposium, Barcelona, Spain. NAPE, 2014. Nationaler Aktionsplan Energieeffizienz, Federal Ministry for Economic Affairs and Energy. NPE, 2012. Fortschrittsbericht der Nationalen Plattform Elektromobilität. Dritter Bericht. (last accessed on January 7th, 2015). NPE, 2014. Fortschrittsbericht 2014 – Bilanz der Marktvorbereitung. (last accessed on January 7th, 2015). Reiter, K., Wrighton, S., 2014. Potential to Shift Goods Transport From Cars to Bicycles in European Cities. (last accessed on February 24th, 2015). Salama, P., Lubinsky, A., Roy, B., Ivanic, Z., 2014. Green Loading Zones to Suport Cost-Effective Zero-Emission Commercial Vehicle Operations in New York City. SAE Technical Paper 2014-01-2434. Sprei, F., Brauener, D., 2011. Incentives Impact on EV Markets. Victoria Institute. (last accessed on April 15th, 2015). Suksri, J., Raicu, R., Yue, W.L., 2012. Towards sustainable urban freight distribution – a proposed evaluation framework. In: Australasian Transport Research Forum 2012 Proceedings. Taefi, T.T., Kreutzfeldt, J., Held, T., Fink, A., 2015. Strategies to increase the profitability of electric vehicles in urban freight transport. In: Leal, W., Kotter, R. (Eds.), E-Mobility in Europe. Springer, Berlin, pp. 367–388, iSBN: 9783319131931. Taefi, T.T., Kreutzfeldt, J., Held, T., Konings, R., Kotter, R., Lilley, S., Baster, H., Green, N., Laugesen, M.S., Jacobsson, S., Borgqvist, M., Nyquist, C., 2016. Comparative analysis of European examples of freight electric vehicles schemes. In: Kotzab, H., Pannek, J., Thoben, K.-D. (Eds.), Dynamics in Logistics. Springer, iSBN 9783319235127. Taniguchi, E., Tamagada, D., 2005. Evaluating city logistics measures considering the behavior of several stakeholders. J. East. Asia Soc. Transport. Stud. 6, 3062–3076. Tenkhoff, C., Braune, O., Wilhelm, S., 2012. Ergebnisbericht der Modellregionen Elektromobilität 2009–2011. (last accessed on February 27th, 2015). Trigg, T., 2012. Third age of electric vehicles. J. Int. Energy Agency 2, 1–2. Trip, J.J., Konings, R., 2014. Supporting Electric Vehicles in Freight Transport in Amsterdam. Umweltbundesamt, 2013. Kurzfristig kaum Lärmminderung durch Elektroautos. (last accessed on February 10th, 2015). van der Steen, M., Schelven, R.V., Kotter, R., van Twist, M., van Deventer, P., 2015. EV policy compared: an international comparison of governments’ policy strategy towards e-mobility. In: Leal, W., Kotter, R. (Eds.), E-Mobility in Europe. Springer, Berlin, pp. 27–53, iSBN: 9783319131931. Wermuth, M., 2012. Kraftfahrzeugverkehr in Deutschland 2010 (KID 2010), Schlussbericht. Wietschel, M., Plötz, P., Gnann, T., Kühn, A., 2013. Markthochlaufszenarien für Elektrofahrzeuge. Fraunhofer ISI, Karlsruhe. (last accessed on April, 15th 2015). Wolpert, S., 2013. City-Logistik – Bestandsaufnahme relevanter Projekte des nachhaltigen Wirtschaftsverkehrs in Zentraleuropa. Tech. Rep., Fraunhofer SCS. Zanni, A.M., Bristow, A.L., 2010. Emissions of CO2 from road freight transport in London: trends and policies for long run reductions. Energy Policy 38, 1774– 1786.