ArticlePDF Available

Why is the market for hybrid electric vehicles (HEVs) moving slowly?

Abstract and Figures

Hybrid electric vehicles (HEVs) could be a good short term option to help achieve global targets regarding road transport greenhouse gas emissions. Several common and country-specific public policies based on price or tax rebates are established in order to encourage the adoption of HEVs. The present research empirically assesses market preferences for HEVs in Spain, looking at the role of subsidies. An interactive internet-based survey was conducted in a representative sample (N = 1,200) of Spanish drivers. Drivers are willing to pay an extra amount of €1,645 for a HEV model compared to a conventional vehicle, premium which is well below the price markup for these cars. Therefore, current levels of economic subsidies applied in isolation to promote these types of vehicles may have a quite limited effect in extending their use. Overall, it is found that drivers have clear misconceptions about HEVs, which affect their purchasing choices and perceptions. Therefore, a policy mix of various incentives (including informational campaigns) may be required in order to stimulate the demand for HEVs.
Content may be subject to copyright.
Why is the market for hybrid electric vehicles
(HEVs) moving slowly?
Djamel Rahmani
, Maria L. Loureiro
1CREDA-UPC-IRTA, Edifici ESAB, Parc Mediterrani de la Tecnologia, C/Esteve Terrades, Castelldefels,
Barcelona, Spain, 2Departamento de Fundamentos da Ana
´lise Econo
´mica, Facultade de C. Econo
´micas e
Empresariais, U. Santiago de Compostela, Santiago de Compostela, Spain
Hybrid electric vehicles (HEVs) could be a good short term option to help achieve global tar-
gets regarding road transport greenhouse gas emissions. Several common and country-
specific public policies based on price or tax rebates are established in order to encourage
the adoption of HEVs. The present research empirically assesses market preferences for
HEVs in Spain, looking at the role of subsidies. An interactive internet-based survey was
conducted in a representative sample (N = 1,200) of Spanish drivers. Drivers are willing to
pay an extra amount of 1,645 for a HEV model compared to a conventional vehicle, pre-
mium which is well below the price markup for these cars. Therefore, current levels of eco-
nomic subsidies applied in isolation to promote these types of vehicles may have a quite
limited effect in extending their use. Overall, it is found that drivers have clear misconcep-
tions about HEVs, which affect their purchasing choices and perceptions. Therefore, a pol-
icy mix of various incentives (including informational campaigns) may be required in order to
stimulate the demand for HEVs.
1. Introduction
In December 2015, a total of 195 countries ratified a universal agreement to combat climate
change at the Climate Summit in Paris (COP 21), where they expressed their willingness to
move together towards a low carbon economy. The European Union (EU) has announced its
plan to achieve its ambitious challenges, including the reduction of its greenhouse gas emis-
sions by 40% by 2030 from the 1990 level; improving energy efficiency by 40%; and increasing
the contribution of renewable energy in its energy consumption by 27% [1].
One of the priorities is the transport sector, due to its significant contribution to global
warming, and air pollution (Directive 2009/28/EC; 2009/30/EC; Directive 2009/33/EC). In par-
ticular, road transport is a major source of greenhouse gas emissions in European cities, being
responsible for one fifth of the EU’s total emissions of carbon dioxide (CO2), the main green-
house gas. Despite a slight decrease in the last few years, these emissions are still 20.5% higher
than in 1990 [2]. The consequences of air pollutants generated by the transport sector on
PLOS ONE | March 21, 2018 1 / 14
Citation: Rahmani D, Loureiro ML (2018) Why is
the market for hybrid electric vehicles (HEVs)
moving slowly? PLoS ONE 13(3): e0193777.
Editor: Xiaosong Hu, Chongqing University, CHINA
Received: October 23, 2017
Accepted: February 16, 2018
Published: March 21, 2018
Copyright: ©2018 Rahmani, Loureiro. This is an
open access article distributed under the terms of
the Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: Study data is
available at figshare (DOI:10.6084/m9.
Funding: This work was supported by Fundacio
´n Areces (,
Dr. Maria Loureiro; Secretarı
´a de Estado de
´n, Desarrollo e Innovacio
´n, ECO2016-
79446-R, Dr. Maria Loureiro.
Competing interests: The authors have declared
that no competing interests exist.
human health are an increasing cause for concern. According to the European Commission [3],
air pollution causes the premature death of more than 400,000 people in Europe every year.
However, and despite the “Dieselgate” scandal, diesel vehicles represented still 52% of all
Europe’s new registrations in 2015. While in United States, Chinese, and Japanese, the new
registrations in the same year were dominated by gasoline vehicles [4]. The registrations of
alternative fuel vehicles (AFVs) have increased in EU member states and European Free Trade
Association (EFTA) countries. In particular, Italy (14.1%), Norway (12.6%), and Poland
(8.1%) registered the highest growth in the AFV new registrations in the period from 2013 to
2015 [5]. However, the growth rate of the AFV new registrations was very low in most member
countries. The rise of the AFV diffusion is due to several factors, including the availability of a
wide range of models in the market, but also in part to different government incentives based
on direct rebates in purchase prices, tax reductions, free parking, and access to priority lanes,
among others [5].
The plug-in hybrid electric vehicles (PHEV), hybrid electric vehicles (HEVs), and electric
vehicles (EVs) accounted for only about 2.6% of 2015 new vehicle registrations in the EU.
Thanks to the CO2-based vehicle taxation scheme, Netherlands led the PHEVs and EVs sales
in 2015 with contributions of 8.8% and 0.9%, respectively. Among the EFTA countries, Nor-
way is a clear example in terms of fiscal incentives to promote the adoption of AFVs, especially
PHEVs and EVs. In fact, 22% of 2015 new vehicle sales in Norway were PHEVs and EVs [4].
Among the three models, HEVs are the most sold in EU, being the highest sales in 2015 those
registered in the Netherlands (3.3%) and France (2.2%). HEVs represented 25% of 2015
Toyota vehicles sold in the EU. The HEVs are also the bet of many countries like Japan and U.
S. where their 2015 market shares were around 22% and 5%, respectively [4].
Nevertheless, up to now, and despite the existence of various stimuli, the market penetra-
tion for EVs, HEVs and PHEVs is still quite low in most countries. This study explores the
potential reasons behind such a low adoption rate. In particular, it explores the relevant factors
that drive people’s vehicle choices, especially those that play a key role in preferring HEVs over
conventional vehicles, looking not only at economic incentives but also at perceptions and
knowledge about HEVs.
A HEV (non-plugin) is an AFV which uses internal combustion engines and electric batter-
ies. It uses braking energy, which is normally wasted, to recharge the battery. HEVs offer eco-
nomic and environmental advantages over conventional cars, ceteris paribus. The engineering
literature has produced very relevant references providing accurate and technical information
about the advantages in terms of current efficiency, environmental performance and future
possibilities of HEVs. [6,7,8,9]. In general, HEVs are cheaper than EVs and PHEVs; they do
not suffer from battery problems or lack of infrastructures, and benefit from public incentives
in many countries. Therefore, they should be very competitive with respect to diesel and gaso-
line vehicles.
It has not been clear why drivers have avoided switching to HEVs. Is it primarily related to
their price? And if so, what types of incentives are needed to encourage drivers to switch to
HEVs? Or is it based on other misconceptions and concerns associated with HEVs that may be
more important than the price markup? Based on a discrete choice experiment (DCE) included
in an extensive online questionnaire, the present paper aims to provide some insight into these
questions, as well as the type of incentives that are required in order to galvanize the HEV mar-
ket. Specifically, it explores preferences towards car attributes, including fuel consumption and
CO2 emissions (improved in HEVs). In addition, it tests whether drivers’ perceptions towards
subsidies may encourage demand for HEVs. Finally, individual heterogeneity in preferences for
car attributes, including the price–a factor that is often overlooked–is considered by specifying a
random parameter logit (RPL) model.
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 2 / 14
2. Literature review
Many contributions explored consumers’ preferences for different AFVs including EVs,
HEVs, PHEVs, liquefied petroleum gas, compressed natural gas, biofuel, and hydrogen pow-
ered vehicles [10,11,12,13,14]. The results detected heterogeneous individual perceptions for
different AFVs, with conventional cars remaining the most attractive option.
From the extensive existing literature on car choices, few studies [15,16,17,18,19] have
specifically investigated consumers’ preferences for hybrid cars (HEVs or PHEVs). Erdem
et al. [17] used a contingent valuation method to estimate the willingness to pay (WTP) for
HEVs in Turkey. The results showed that people were willing to pay an average premium of
US$ 858 to change to a HEV. Thatchenkery and Beresteanu [19] explored HEV demand in the
USA using the United States 2006 Polk New Vehicle Registration Cross-sectional Data. They
showed that people were sensitive to fuel efficiency, but were more sensitive towards horse-
power and weight. Axsen et al. [15] combined the revealed preferences (recent car purchases)
and stated preferences of Canadian and Californian car owners to explore how consumer pref-
erences for HEVs have shifted (specifically focusing on the neighbor effect) as HEV market
penetration increased. The results showed that the WTP for HEVs rose with the market share.
Chua et al. [16] employed scales and items to compare HEV and conventional car buyers in
Australia. The results from factor analysis showed that while preferences for conventional cars
were more sensitive to variations in quality and performance, and less sensitive to image and
social influence, HEV buyers placed a great amount of importance on their ‘green’ image and
social influence, and little importance on quality and appeal. Heffner et al. [18] used informal
face-to-face interviews to investigate whether the ‘green’ social image influenced United States
households to adopt HEVs. The results showed that all HEV owners placed some importance
on the ‘green’ image of their cars, although they did not adopt HEVs by only focusing on their
image. The present research joins this line of studies by adding a number of contributions.
First, it assesses the heterogeneity of preferences towards car attributes across drivers. Second,
this study investigates the importance of incentives when buying efficient cars, and explores
whether these incentives increase the demand for HEVs. Third, results are contextualized in
the current market conditions.
3. Case study and policy context
The present research was conducted in Spain. The current economic crisis has resulted in the
Spanish vehicle fleet being one of the oldest in Europe, currently with an average car age of 11.3
years, and with emissions that affect seriously the air quality in cities [20]. Vehicles older than 10
years account for 50% of all cars circulating in Spain [21]. Driving vehicles of this kind multiplies
the environmental damage caused by road transport. In this context, various strategies have
been promoted in Spain, including the Movele and the Pive public programs, both aimed at pro-
moting the market adoption of efficient cars. The Pive program is designed to encourage the
acceptance of HEVs, PHEVs, EVs, and extended range electric vehicles. Currently, the Pive pro-
gram [22] offers a discount of 1,500 on the purchase of a new vehicle, after turning in a private
car over 10 years of age, or any commercial vehicle over 7 years of age. The vehicle purchased
must be new and an efficient model (EVs, HEVs, PHEVs, or using alternative fossil fuels). As a
result of this program, the Institute for the Diversification and Energy Saving (IDAE) estimated
that from 2012 to 2014 the Pive plan has led to the replacement of 715,000 old vehicles, saved
248 million liters of fuel per year, and reduced greenhouse gas emissions by 513,000 tons of
CO2 per year [22]. The Spanish government also reformed the car registration tax (Law 34/2007
of 15th of November on air quality and protection of the atmosphere), making it inversely pro-
portional to the amount of CO2 emissions (0% for emissions lower than 120 g/km, 4.75% of the
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 3 / 14
value of the car for emissions between 120 and 160 g/km, 9.75% for emissions of between 160
and 200 g/km; and finally, 14.75% for emissions of 200 g/km or higher).
In spite of these current public policy efforts to encourage drivers to adopt HEVs, these are
still not particularly popular in the Spanish market. In 2016, HEVs only accounted for 2.70% of
new passenger car sales [23]. The Japanese Toyota brand led the sales of HEVs in Spain, with a
market share of more than 70%. Together with its premium brand Lexus, they accounted more
than 80% of the total HEVs sold in Spain in 2016. While the vast majority of HEVs sold were
gasoline, diesel hybrid cars only represented 6% of the total units sold [23].
4. Survey design
Data from drivers were collected using an online survey directed to a representative sample of
drivers over the age of eighteen. The survey was administered to 1,200 residents in Spain. The
number of fully completed and useful questionnaires was 1,016. The survey asked drivers to
provide information about several car related issues, including current car(s) ownership,
brand preferences, awareness of energy consumption issues, and their environmental atti-
tudes. Next, the survey provided information about HEVs, asking about their intentions and
plans for future car-purchases, including a DCE to elicit preferences to buy a future car. It con-
cluded with the socio-demographic characteristics of the driver.
Participants were asked about what size (small, medium or large) they would prefer to have
their next car? And thanks to the interactive aspect of the questionnaire, this information was
received immediately, and automatically, depending on their answer, they were assigned to
one of the two possible versions of the DCE survey. In particular, one was designed for drivers
interested in buying small or medium-sized cars and the other for those who were willing to
buy large cars. A total of 875 drivers (86.12% of the completed surveys) expressed their desire
to buy a small or midsize car in the future, while only 138 drivers (13.58% of the completed
survey) stated their wish to adopt a large size car in the future. The survey questions were com-
mon to the participants. The only difference between the two versions was the levels of the
attributes included in the DCE. In this paper, data from the survey completed by those drivers
willing to buy a small and medium-size car are analyzed.
4.1 Experimental design and DCEs
A DCE is used as it is the more appropriate way for measuring consumer welfare, and its results
are more consistent with the economic theory than a traditional conjoint analysis [24]. In addi-
tion, HEVs have a small market share, and revealed preference data sources are still scarce. The
DCE method is based on the assumptions of economic rationality and utility maximization
[24]. It consists of presenting drivers with several car alternatives, and asking them to choose
one of them based on their preferences. Each individual is expected to choose the alternative
that maximizes his/her utility. Moreover, the utility derived from an alternative is assumed to
depend on the marginal utilities associated with its attributes [25]. As a HEV is a quasi-public
good, both economic attributes and environmental (non-economic) attributes are included. In
the survey, and prior to the DCE exercise, participants were familiarized with HEVs and the
expected consumption and emissions for a mid-size car. They were also required to assume that
all non-specified attributes remained constant across alternatives. A DCE was then carried out,
in which the participants could select between a regular vehicle and a HEV, or just remain with
the status quo option (neither car).
Focus groups, pilot surveys and previous studies were used in order to identify the most
relevant attributes and suitable levels for our DCE exercise. Previous studies [13] summa-
rized the determinant factors of a car choice process mainly into economic attributes
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 4 / 14
(purchase price, fuel cost), non-economic attributes (refueling or recharging time, availabil-
ity of fuel or recharging opportunities, technological performance), and environmental
attributes (emissions). Besides the type of vehicle, two economic attributes have been
included: price and fuel consumption, factors that are highly and primarily valued by driv-
ers when considering the purchase of AFVs [26]. Apart from the monetary attributes, each
choice set included two non-monetary or environmental attributes. The environmental
attributes included were carbon dioxide (CO2) emissions, which were found to be signifi-
cant in earlier studies [13], and the option of biofuel adaptation (flex-fuel), which is a recent
trend in carmakers. In fact, European legislation (Directive 2003/30/EC) and national legis-
lation (Spain’s Royal Decree 61/2006) allow carmakers to incorporate bio-fuel directly into
conventional fuel without the need for specific labeling, unless the proportion exceeds 5%.
Some existing studies [10,27] have explored preferences for biofuel cars, although it has
never been investigated as an additional attribute to conventional and HEVs.
The attribute levels are based on information obtained from car suppliers in the Spanish
market for small and midsize cars. This information is used to determine 2 levels of vehicle type
(regular or HEV) and 3 levels of prices used in the analysis: a low price level (12,000), a medium
price level (16,000) and a high price level (20,000). The mid-price level considered corresponds
to the average price of new cars sold in Spain in 2012. From 2009 to 2013, most of the new cars
sold in Spain (80% of the total) were priced below 20,000, due in part to the decrease of purchas-
ing power of consumers caused by the economic crisis. For these reasons, and given the focus of
this work (analyzing the demand for small and medium HEVs), the upper price level is set at
20,000 and the lower price level at 12,000. The fuel consumption attribute was expressed as fuel
cost () per 100 kilometers [10,28]. This unit is used because drivers tend to remember how
much fuel their car consumes in terms of euros/kilometers. The fuel cost was computed as the
product between the numbers of liters of fuel the vehicle would require to travel 100 kilometers,
and the average fuel price in Spain (1.35 per liter at the time of the study). Similarly, the CO2
emissions were expressed as grams of CO2 per kilometer [10,28]. Again, for simplicity, and for
the purposes of this research, only two emission levels are included: a more efficient level (100gr
per kilometer) and an inefficient level (150gr per kilometer). Finally, the presence or absence of
the potential of biofuel adaptation corresponded with the two dichotomous levels specified for
the corresponding attribute.
The combination of these five attributes and their levels, using SPSS orthogonal main effects
design and then the procedure of Street and Burgess [29] (vector of differences = 12111), gen-
erated an optimal orthogonal design (OOD). The OOD is constructed so as to maximize the
differences in the attribute levels across alternatives, and therefore, maximize the information
from each respondent, forcing the tradeoffs of all attributes in the experiment [30]. It should
note that this design fits best choices where each alternative has the same number of attributes,
and each attribute has the same number of levels. The final design contained 8 choice cards
with a design efficiency of 98%.
Each respondent was presented with a total of 8 choice cards, a reasonable number that
does not affect data quality [31]. Fig 1 shows an example of a choice card. The no-choice alter-
native (neither car) was provided in order to make the choice decisions very similar to market
decisions (or more realistic).
4.2 Choice modeling specification
Assuming utility maximizing behavior, the empirical applications based on discrete choice
models make possible to estimate the probability that an individual chooses a given car alterna-
tive, among a set of available alternatives. The utility that an individual i derived from choosing
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 5 / 14
a car alternative j among a set of J alternatives (conventional car, HEV or no-choice option) in
each choice situation t may be expressed [32,33] as a sum of an observable utility component
ijtb) and unobservable component or error term (ε
ijtbþεijt ð1Þ
ijt: is a vector of specific car attributes and specific individual characteristics.
β: is a vector of parameters associated with the explanatory variables.
The multinomial logit model (MNL) [34] is derived assuming that the error terms (ε
) are
independently and identically extreme value type I distributed (IID). The MNL probability of
choosing an alternative a among a set of J alternatives is given by [32,33]:
The MNL is based on the assumption of the independence of irrelevant alternatives (IIA).
The MNL imposes homogeneity in tastes, inflexible substitution patterns in preferences
between different alternatives and independence in unobserved factors over time [35]. An
alternative model which is much more flexible and which overcomes the limitations of MNL is
the RPL. In addition to the fact of not complying with the IIA property, the RPL allows for: a)
random heterogeneous preferences across individuals, b) unrestricted substitution patterns,
and c) correlation in unobserved factors over time [35]. The RPL model probability (uncondi-
tional probability) is the integral of the conditional probability over all the random parameters
Fig 1. Choice experiment question and card example.
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 6 / 14
f(β|θ): is the density function of the parameters β. This density function may be assumed to
follow any closed-form distribution (normal, log-normal, triangular, uniform) [36,35]; θ: are
the parameters (mean and standard deviation) of the distribution.
In this application, RPL models are estimated assuming log-normal distribution for the
coefficients associated with price (PRICE), fuel consumption (FCONSUMPTION) and CO2
emissions (CO2) in order to force them to be negative (on one side of zero) for all individuals.
In the same line as previous studies [10,37,12], positive preferences for these attributes are
not allowed, as it is not expected that people would prefer higher prices, higher fuel consump-
tions or higher CO2 emissions. Several distributions (normal, log-normal, uniform, triangular,
etc.) are also tested for the coefficient associated with biofuel adaptation (BADAPTATION)
but its standard deviation was not statistically significant. Thus, it is considered as a nonran-
dom parameter. This valuation exercise also aims to predict respondents choices between the
two car-alternatives (conventional, and HEVs) and the no-car option (neither A or B), includ-
ing in the DCE models a no-choice-specific constant (ASC), denoting the election of the status
quo option. It is assumed that the no-choice-specific constant follow a normal distribution
because drivers may like or dislike staying or not with their current cars. In addition, it
explores how preferences for the no-car option, compared to the car options, shift with the fol-
lowing socioeconomic variables: gender (MALE), age (AGE), and monthly income under
1,800 (LHINC). It also analyzes the heterogeneity in preferences for the no-car option among
drivers who reported that incentives such as direct subsidies (SUBSIDY) would be important
factors when buying an efficient car. This incentive variable was created from the participants’
ratings, when they were asked to state how important (on a 5-point Likert scale: from 1 “not
important” to 5 “very important) this factor would be in their decision to select efficient cars.
Furthermore, and after the vector of parameters is obtained, the WTP welfare measures are
calculated in order to determine the monetary equivalent of the marginal utilities placed by
drivers in each car attribute improvement. This step may provide important information to
policy makers regarding the economic efforts that people are willing to make to acquire HEVs
and some improvements in car attributes.
WTP for a HEV compared to conventional vehicle is generally computed as the difference
– MU
) between the marginal utility obtained for HEV (MU
) and conventional
vehicles (MU
). Such values are obtained substituting in (1) the estimated parameters of our
empirical model, and then this difference in utilities is divided by the estimated price coeffi-
cient (β
) [33], as shown by the following formula:
In a RPL model, when the numerator and the denominator included in Eq 4 are random,
the expression of the WTP ratio shown in Eq 4 becomes a randomly distributed term. In
this case, Daly et al. [38] advised to ensure finite moments for the WTPs. In the present
application, WTP measures are constructed based on unconditional parameter estimates
[39], because they allow for prediction outside of the sample, unlike conditional parameter
estimates, which only predict within the sample [40]. Deriving WTP based on uncondi-
tional parameter estimates requires the population to be simulated [39]. Both, numerator
and denominator of (4) have been simulated employing random draws coming from the
log-normal distributions defined by the estimated parameters. Draws were generated from
both, the numerator and denominator, computing their respective ratio in each draw, as in
Hensher et al [39].
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 7 / 14
5. Data and results
Table 1 summarizes the drivers’ perceptions towards HEVs (prior to the information received
in the survey, given that participants could select several statements that they considered cor-
rect when thinking about HEVs). When compared to conventional cars, 62% of the drivers
perceived HEVs as being more expensive, although 28% stated that HEVs have low running
costs. In addition, 14% believed that HEVs are slower, and 18% considered that HEVs have
less power. These negative perceptions can be an obstacle to introduce HEVs in a country
where drivers are in general “speed lowers”. Furthermore, it is worrisome that 16% reported
that they did not know what HEVs are like. Finally, 17% reported that HEVs have limited
autonomy, showing a clear misunderstanding of the difference between HEVs and EVs.
Table 2 describes some drivers’ socio-demographic characteristics and the variables
included in the empirical models, containing basic information about the rated importance of
incentives including direct subsidies, registration tax exemption, free parking, access to prior-
ity lanes, and social image. In summary, public policies based on direct economic incentives,
such as subsidies or allowing free parking are perceived as the most important incentives for
drivers to buy an efficient car.
In terms of socio-demographics, the average age of participants in this sample is 46 years,
and 51% of the participants were male. One fifth of the participants were unemployed, and
about 50% of all households received a monthly income of less than 1,800. The participants
reported that on average they drive a car 4 days a week. The sample was representative of the
profile of a Spanish driver at least with respect to some important characteristics, such as age
and driving frequency. The Spanish Observatory of Drivers [41] defined, through a representa-
tive study, a typical Spanish driver as being a 44-year-old male, who uses a car an average of 5
days a week for work.
Table 3 summarizes the results of the estimated models. First, a MNL model is estimated
and the assumption of independent irrelevant alternatives (IIA) is tested using the Chi-squared
Hausman and McFadden test. The results of this test reject the IIA assumption [being the
omitted alternative the regular car: Chi-squared (5) = 156.808; with the omitted alternative
being the HEV: Chi-squared (5) = 160.883; the 99%; critical value: Chi-squared (5) = 15,086].
Then, to improve the performance, RPL models have been estimated, allowing for correlation
over time (but with uncorrelated parameters) and using NLOGIT.5 software with 2000 replica-
tion draws in the estimation processes. In particular, a baseline RPL and a RPL model with het-
erogeneity in the mean of the random parameter associated with the no-car specific constant
are specified.
In Table 3, Column 1 shows the results of the MNL; the RPL results are presented respec-
tively in Column 2 (baseline RPL), and Column 3 (extended RPL, with interaction terms with
Table 1. Perceptions for hybrid electric vehicles (HEVs).
Participants’ hybrid car Perceptions (1 = yes, 0 = no) Mean Std. Dev.
Compared to a conventional car:
A hybrid car is more expensive .620 .485
A hybrid car has lower running costs .281 .449
A hybrid car is slower .142 .349
A hybrid car has less autonomy .168 .374
A hybrid car is less powerful .176 .381
A hybrid car is less safe .013 .116
I do not know what a hybrid car is .158 .365
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 8 / 14
the constant). According to the values of the log-likelihood, adjusted pseudo-R
, Akaike infor-
mation criterion (AIC), the RPL improves the MNL model fit (which results are not directly
discussed). The moments of the coefficients associated with PRICE, FCONSUMPTION, and
CO2 which are calculated converting the log terms are presented in this table.
Table 2. Descriptive statistics of the variables included in the RPL model.
Variable Description Mean Std. Dev.
PRICE price of car-option divided by 10,000. 1.066 .805
FCONSUMPTION euros spent in fuel consumption per 100km. 4 2.943
CO2 grams of CO2 emitted per 1km. 83.333 62.362
BADAPTATION 1 if car-option is adaptable (flex-fuel) to run with biofuels and 0 otherwise. .333 .471
ASC no-car-option constant. .333 .471
MALE 1 for male and 0 otherwise. .513 .499
AGE age of participants (years). 45.972 13.546
LHINC 1 for monthly income under 1,800 and 0 otherwise. .505 .499
SUBSDY importance (score) attributed to the incentive “direct subsidies”. 4.199 .961
Table 3. Results of estimated MNL and RPL models.
MNL Baseline RPL RPL
Parameters in utility functions
Coeff. Std.
Prob. |z|>Z Coeff. Std.
Prob. |z|>Z Coeff. Std.
Prob. |z|>Z
PRICE -2.034 .055 .000 -2.450 .059 .000 -2.371 .051 .000
FCONSUMPTION -.289 .017 .000 -.336 .024 .000 -.348 .023 .000
CO2 -.009 .001 .000 -.013 .001 .000 -.014 .001 .000
BADAPTATION .157 .033 .000 .154 .041 .000 .148 .041 .000
ASC -6.017 .178 .000 -8.175 .220 .000 -9.487 .372 .000
Standard deviations of random parameters
LSPRICE . .832 .063 .000 .621 .035 .000
LSFCONSUMPTION . .150 .023 .000 .130 .019 .000
LSCO2 . .012 .002 .000 .012 .0004 .000
NSASC . 2.193 .109 .000 1.209 .080 .000
Heterogeneity in mean, Parameter Variable
ASC MALE . . -.620 .114 .000
ASC AGE . . .019 .004 .000
ASC LHINC . . .199 .111 .074
ASC SUBSDY . . .161 .049 .001
Measures of goodness of fit
N7,000 7,000 7,000
GROUPS 875 875 875
NB. OBS./GROUP 8 8 8
L.L. FUNCTION -6,655.621 -5,528.428 -5,284.316
K (factors number) 5 9 13
R-SQUARED .126 .281 .271
ADJ. R-SQUARED .125 .281 .270
AIC 13,321.2 11,074.9 10,594.6
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 9 / 14
The results of the baseline RPL model show that the mean of both nonrandom and random
parameters are significant and have the expected signs. The sign of the variables PRICE, FCON-
SUMPTION, and CO2 have been inversed and entered as negative in the RPL model specifica-
tions. The log-normally distributed coefficients are expressed as: β
= exp(b
) where, μ
IID standard normal deviate, b
and s
are the estimated mean and standard deviation for log-
normally distributed coefficient. The table above presents the moments of the coefficients asso-
ciated with PRICE, FCONSUMPTION, and CO2 which are calculated in the following way [42,
43]: Mean ¼expðbkþ ðs2
k=2ÞÞ;Standard deviation ¼mean ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
p. The mean of
PRICE, FCONSUMPTION, and CO2 were multiplied by -1 in order to reestablish the sign
changed a priori to the model estimation.
The results are similar to those provided by the MNL model. In line with previous findings
[26], the two monetary attributes that were included–price (PRICE) and fuel consumption
(FCONSUMPTION)–have negative effects on utility, indicating that drivers tend to prefer
cars with a lower price and lower fuel consumption. This implies that one of the motives that
may attract drivers to adopt HEVs is fuel economy. Similarly, and in agreement with findings
from previous studies [13], the coefficient associated with CO2 emissions (CO2) is negative,
implying that on average, drivers prefer cars with lower levels of CO2 emissions. Furthermore,
the fact that the car could be adapted to biofuels (BADAPTATION) carries a positive effect on
utility, implying that drivers prefer flexible-fuel cars to non-adapted cars. Regarding the stan-
dard deviations, it is found that the three random parameters (PRICE, FCONSUMPTION,
and CO2) have statistically significant standard deviations, implying that there are heteroge-
neous preferences for these attributes across drivers. Moreover, the no-car option constant is
negative and statistically significant at any critical level. This reveals that drivers prefer to
move from status quo to any of the two car type options. The standard deviation of the con-
stant (ASC) is also statistically significant suggesting that drivers preferences for their actual
cars are heterogeneous. In order to control for the potential wide heterogeneity of preferences
towards actual cars, it is considering the next extended RPL model.
The next RPL model explores whether the shift in the mean of the no-car-specific constant
due to drivers’ socio-demographic characteristics (MALE, AGE and LHINC) and drivers’ pref-
erences towards the incentives (SUBSIDY) when buying efficient cars, improves the fit of the
baseline RPL. The results show that MALE are less likely to choose the status quo no-car option
over an efficient car in terms of consumption (HEV or a conventional car) in the DCE, whereas
older drivers (AGE) prefer staying at the status quo option. Drivers with monthly income under
1,800 show significant preferences for their actual cars. Interaction term between the no-car-
option constant and the preferences towards direct incentives via subsidies is positive and statis-
tically significant, indicating that the more important subsidies (SUBSDY) are for drivers, the
more likely they are to stay with the no-car status quo option compared to enter the market of
HEVs or conventional vehicles. These may be caused by the fact that drivers who are more sen-
sitive to subsidies are also more sensitive to price in general, and as a consequence, they are not
willing to enter the HEV market, which is more expensive. This finding suggests that drivers do
not know that HEVs are supported with economic subsidies (Pive Plan) or they think that these
subsidies are not enough to afford a HEV.
Table 4 shows the sample mean WTP for the HEV compared to conventional car estimated
from the basic RPL empirical model. To estimate this WTP, we used the characteristics of a
concrete example of HEV and its conventional model of the Toyota brand. Table 4 shows
these characteristics for both models. Results show that drivers are willing to pay a statistically
significant premium of 1,645.032 to move from the conventional model to the HEV. There-
fore, participants are willing to make an economic effort to buy a HEV compared to
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 10 / 14
conventional car, although this is well below the price markup for these cars. Therefore, cur-
rent levels of economic subsidies (Pive Plan) applied in isolation to promote these types of
vehicles may have a limited effect in extending their use.
Once again, the results reiterate the need of informational mechanisms as to ensure the
understanding of HEVs attributes and the incentives supporting the use of HEVs. This is in
fact nowadays an obstacle to increasing the market share of hybrid technology.
6. Conclusions and implications
The present research explores the importance of monetary and non-monetary incentives, gen-
erally adopted by governments to boost sales of fuel-efficient cars. It tests whether drivers who
consider these incentives to be important are especially attracted to HEVs. Participants in a
national survey were asked to rate the importance that some public policies would have in
their decision to switch to efficient cars (HEVs or new medium cars). They rated policies
based on reductions in direct subsidies or allowing free parking to be the most important
incentives, followed by access to priority lanes and registration tax exemption, and finally by
the social image derived from driving an efficient car.
Nearly half of the sample perceives HEVs to be cleaner than gasoline, diesel, biofuels, and
liquefied petroleum gas (LPG) cars. Three out of ten drivers believe that HEVs have lower run-
ning costs than conventional cars. However, many drivers (62%) perceive them to be more
expensive, slower (14%), and less powerful than conventional cars (18%). In addition, some
drivers do not exactly know what HEVs are (15%), or clearly misunderstand the difference
between HEVs and EVs (16%).
The estimated stated preference models show that drivers prefer cheaper cars with low fuel
consumption, implying that fuel economy may be an attractive reason to buy HEVs. Similarly,
low CO2 emissions increase the utility derived from a car, and are another reason to encourage
drivers to buy HEVs. The subjects also expressed strong preferences for flexible-fuel cars, con-
cluding that offering HEVs adapted to run with biofuels could increase the demand for HEVs.
The results derived from the RPL model show that drivers are willing to pay an extra pre-
mium of 1,645 to change from conventional automobiles to HEVs; however this amount is
well below the price markup for these cars in the current market. Therefore, current levels of
economic subsidies applied in isolation to promote these types of vehicles may have a limited
effect in extending their use.
In this context, the measures implemented in countries such as Norway and France to
encourage adoption of EVs have already demonstrated their effectiveness and may be consid-
ered by third countries. In particular, Norway reached the target of 50,000 EVs in 2015 through
a mix of public monetary and non-monetary incentives in favor of EVs, including VAT and
registration tax exemption, access to priority lanes, free toll roads, free parking, free travel on
ferries, free municipal recharging, reductions in the annual road tax, and exemption from
Table 4. Mean WTP for HEV compared to conventional vehicle.
Attribute HEV Conventional model WTP
Price (PRICE) 20,200 18,550 Mean: 1,645.032
Std.Dev.: 5.319
Fuel/100Km.(FCONSUMPTION) 3.6l 5l Median: 1,646.7
1st.Qrtl: 1,609.1
3rd.Qrtl: 1,636.2
CO2 emissions per 1km (CO2) 75gr 112gr
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 11 / 14
company car tax [44]. The bonus-malus French system was able to reduce the emissions of
newly registered passenger vehicles by 19g/km in just three years after its introduction [45].
This system offers a bonus of 6,000 (up to 27% of the acquisition cost) for the purchase of a
new private car or van that emits up to 20 grams of CO2/km and a bonus of 2,500 for handing
over an old diesel vehicle put into circulation before January 1, 2006. These two financial aids
are cumulative, therefore the total aid may reach 8,500. Moreover, it imposes a tax up to
10,500 for the purchase of vehicles emitting more than 185 grams of CO2/km [46]. Further-
more, other restrictive measures (access to the city center, surroundings, or to parking) used by
some large European cities to deal with the emergency pollution situations could encourage
drivers to opt for AFVs if they were applied more frequently (a few days a week).
To conclude, and as earlier stated, the articulation of economic incentives may be crucial.
Further, designing information campaigns that provide accurate information on HEVs may
have as well a significant impact on sales. Another potential solution is to promote the use of
HEVs for taxis and public transport, and to encourage the public authorities to replace their
conventional cars with HEVs. This may help to further promote the image of HEVs, and to
reduce the current distrust towards this alternative fuel technology. Various governments
(Japan, EU, Canada, China, South Korea, Mexico, Brazil, and India) have fixed greenhouse gas
emission limits for passenger vehicles to meet short-term mitigation goals. In particular, EU
fleet target for 2021 is 95g/km [47], which is equivalent to the amount of CO2 emitted by vari-
ous HEVs currently available on the market. In this context, encouraging the adoption of
HEVs will not only reduce the compliance deadline but also the cost required. This set of find-
ings may be relevant in order to adopt appropriate and effective strategies in the future aimed
at reducing road transport, greenhouse gas emissions, and their contribution to climate
change. Future research should look deeper at the role of economic incentives under different
scenarios, characterized in many occasions by strong cultural differences, risk aversion, and
myopic time preferences.
Author Contributions
Data curation: Djamel Rahmani.
Formal analysis: Djamel Rahmani.
Investigation: Djamel Rahmani.
Methodology: Djamel Rahmani.
Writing – original draft: Djamel Rahmani.
1. European Commission. 2030 climate and energy framework. 2015. Available from:
clima/policies/strategies/2030/index_en.htm (accessed 12.07.16).
2. European Commission. Road transport: Reducing CO2 emissions from vehicles. 2012. Available from: (accessed 12.07.16).
3. European Commission. Proposal for a legislative instrument on control of emissions from Medium Com-
bustion Plants -Impact Assessment, commission staff working document (2013) 531. Available from: (accessed 12.07.16).
4. International Council on Clean Transportation. European vehicle market statistics, Pocketbook 2016/17.
2016. Available from:
(accessed 13.06.2017).
5. Eurostat. Passenger cars in the EU. 2017. Available from:
statisticsexplained/index.php/Passenger_cars_in_the_EU (accessed 13.06.2017).
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 12 / 14
6. Ehsami M., Gao Y., Emadi A. 2009. Modern Electric Vehicles hybrid electric, and fuel cell vehicles: fun-
damentals, theory, and design. CRC Press.
7. Liu J., Peng H. Modeling and Control of a Power-Split Hybrid Vehicle. IEEE Transactions on Control
Systems Technology. 2008, 16 (6):1242–1251.
8. Sun C. Hu X.,Moura S.J., Sun F. Velocity Predictors for Predictive Energy Management in Hybrid Elec-
tric Vehicles. IEEE Transactions on Control Systems Technology. 2015, 23(3):1197–1204.
9. Lv C., Zhang J., Li Y., Yan Y. Mechanism analysis and evaluation methodology of regenerative braking
contribution to energy efficiency improvement of electrified vehicles. Energy Conservation and Manage-
ment, 2015, 92:469–482.
10. Hidrue MK, Parsons GR, Kempton W, Gardner MP. Willingness to pay for electric vehicles and their
attributes. Resource and Energy Economics. 2011; 33 (3): 686–705.
11. Zhang Y, Yu Y, Zou B. Analyzing public awareness and acceptance of alternative fuel vehicles in China:
The case of EV. Energy Policy. 2011; 39 (11): 7015–7024.
12. Achtnicht M. German car buyers’ willingness to pay to reduce CO2 emissions. Climatic Change. 2012;
113(3–4): 679–697.
13. Caulfield B, Farrell S, McMahon B. Examining individuals’ preferences for hybrid electric and alterna-
tively fuelled vehicles. Transport Policy. 2010; 17 (6): 381–387.
14. Nixon H, Saphores JD. Understanding household preferences for alternatives-fuel vehicle technologies.
Report 10–11, Mineta Transportation Institute, San Jose
´, CA. 2011.
15. Potoglou D, Kanaroglou PS. Household demand and willingness to pay for clean vehicles. Transporta-
tion Research Part D. 2007; 12(4): 264–274.
16. Qian L, Soopramanien D. Heterogeneous consumer preferences for alternative fuel cars in China.
Transportation Research Part D. 2011; 16: 607–613.
17. Axsen J, Mountain DC, Jaccard M. Combining stated and revealed choice research to simulate the
neighbor effect: The case of hybrid-electric vehicles. Resource and Energy Economics. 2009; 31(3):
18. Chua WY, Lee A, Sadeque S. Why do people buy hybrid cars?, in Proceedings of Social Marketing
Forum, University of Western Australia, Perth, Western Australia, Edith Cowan University, Church-
lands, W.A. 2010: 1–13.
19. Erdemn C, Sentu¨rk I, Simsek T. Identifying the factors affecting the willingness to pay for fuel-efficient
vehicles in Turkey: A case of hybrids. Energy Policy. 2010; 38 (6): 3038–3043.
20. Heffner RR, Kurani KS, Turrentine TS. Symbolism in California’s early market for hybrid electric vehi-
cles. Transportation Research Part D: Transport and Environment. 2007; 12 (6): 396–413.
21. Thatchenkery SM, Beresteanu A. Determinants of Automobile Demand and Implications for Hybrid-
Electric Market Penetration. Honors thesis submitted in partial fulfilment of the requirements for Gradua-
tion with Distinction in Economics in Trinity College of Duke University. 2008.
22. Ecologists in Action. Informe: La calidad del aire en el Estado español durante 2011. EIA Report. 2011.
Available from:
(accessed 12.07.16).
23. Asociacio
´n española de fabricantes de automo
´viles y camiones (ANFAC). 2014. Available from: http:// (accessed 25.01.16).
24. Instituto para la diversificacio
´n y ahorro de la energı
´a (IDAE). Real Decreto 124/2015: programa de
incentivos al vehı
´culo eficiente (PIVE-7). 2015. Available from:
relcategoria.4030/id.838/relmenu.441/mod.pags/mem.detalle (accessed 12.07.16).
25. Asociacio
´n Española de Fabricantes de Automo
´viles y Camiones (ANFAC). 2017. Available from: (accessed 12.01.17).
26. Louviere J.J, Flynn TN, Carson RT. Discrete choice experiments are not conjoint analysis. Journal of
Choice Modelling. 2010; 3(3): 57–72.
27. Lancaster K. A new approach to consumer theory. The Journal of Political Economy. 1966; 74: 132–57.
28. Adler T, Wargelin L, Kostyniuk L, Kavalec C, Occhiuzzo G. Incentives for alternative Fuel Vehicles: A
Large-Scale Stated Preference Experiment, paper presented at the 10th International Conference on
Travel Behaviour Research, Lucerne. August 10–15, 2003.
29. Hackbarth A, Madlener R. Willingness-to-pay for alternative fuel vehicle characteristics: A stated choice
study for Germany. Transportation Research Part A: Policy and Practice. 2016; 85: 89–111.
30. Ziegler A. Individual characteristics and stated preferences for alternative energy sources and propul-
sion technologies in vehicles: A discrete choice analysis for Germany. Transportation Research Part A:
Policy and Practice. 2012; 46 (8): 1372–1385.
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 13 / 14
31. Street D, Burgess L. The construction of optimal stated choice experiments. New Jersey: John Wiley
and Sons Inc. 2007. Available from: (accessed
32. Street DJ, Burguess L, Louviere JJ. Quick and Easy Choice Sets: Constructing Optimal and Nearly Opti-
mal Stated Choice Experiments. International Journal of Research in Marketing. 2005; 22: 459–470.
33. Johnson RM, Orme BK. How many questions should you ask in choice-based conjoint studies?
Sequim, WA: Sawtooth Software. 1996.
34. Hole AR. Mixed logit modelling in stata: An overview. UK stata users group meeting. September 2013.
35. Train Kenneth E. Discrete Choice Models with Simulation. New York: Cambridge University Press.
2009; (2): 77–82.
36. McFadden D. Conditional logit analysis of qualitative choice behavior. In Frontiers in Econometrics, ed.
Zarembka P., New York: Academic Press. 1974: 105–42.
37. Daziano RA, Achtnicht M. Accounting for uncertainty in willingness to pay for environmental benefits.
Energy Economics. 2014; 44: 166–177.
38. Daly A, Hess S, Train K. Assuring finite moments for willingness to pay in random coefficient models.
Transportation. 2012; 39: 19–31.
39. Hensher DA, Rose JM, Greene WH. Applied choice analysis. Second edition, Cambridge University
Press. 2015.
40. Jones S, Hensher DA. Predicting financial distress: a mixed logit model. Accounting review. 2004; 79
(4): 1011–1038.
41. Spanish Observatory of Drivers. Estudio del conductor español: ha
´bitos en los desplazamientos y
viajes en vacaciones. 2014. Available from: (accessed 10.06.15).
42. Revelt D, Train K. Mixed logit with repeated choices: households’ choices of appliance efficiency level.
Review of Economics and Statistics. 1998; 80: 1–11.
43. Shimizu K, Crow EL. Lognormal distributions: Theory and applications. New York: Dekker. 1988.
44. Phillips L. Norway’s electric vehicle revolution: Lessons for British Columbia. Pacific Institute for Climate
Solutions (PICS), Briefing Note 2015–43.
45. Rubik F, Mityorn L. CO2-based Motor Vehicle Tax. Institute for Ecological Economy Research, The
SCP-Knowledge Hub. 30 September 2011.
46. Ministère de la Transition e
´cologique et Solidaire. Bonus-malus e
´cologique: de
´finitions et barèmes pour
2018. Available from:
baremes-2018 (accessed 08.01.2018)
47. International Council on Clean Transportation. Light-duty vehicle efficiency standards. 2014. Available
(accessed 13.06.2017).
The market for hybrid electric vehicles
PLOS ONE | March 21, 2018 14 / 14
... As a result, only a small number of electric cars are being sold and economies of scale are hard to achieve [3]. The examples of the countries analyzed show that the diffusion of EVs is hard to achieve without government subsidies [26] and incentives play a crucial role at the first stage of the diffusion of an innovation. This article shows what incentives have already been introduced in Norway, Germany and Spain and what results were achieved by each of these countries. ...
... The "Pive" program offers a subsidy of 1 500 euros on the purchase of a new model of such types, including AFVs, when such a vehicle replaces a private car of over 10 years of age or a commercial car of over 7 years of age 2 . The car registration tax was also reformed by the Spanish government (Act 34/2007 of the 15th of November on Air Quality and Protection of the Atmosphere), making it increasing in the amount of CO 2 emissions (0% for emissions lower than 120 g/km, 4.75% of the value of the car for emissions between 120 and 160 g/km, 9.75% for emissions of between 160 and 200 g/km; and finally, 14.75% for emissions of 200 g/km or higher) [26]. ...
... 14% of respondents believed that HEVs are slower, and 18% thought that HEVs have lower power, which suggests that the public perception of HEVs is rather negative. 17% of respondents also suggested that HEVs have limited flexibility, which indicates that respondents probably do not know the differences between HEVs and EVs [26]. ...
... [4,5]. To encourage strengthening R&D, some governments have provided strong R&D subsidy policies for NEV firms [6-10]. ...
... Proof of Proposition 4. When i qq > 0, (5) and (7) are compared. The results are shown in Table 2. ...
Full-text available
Different consumer groups accept new energy vehicles sequentially from the perspective of innovation diffusion theory, and the early adopter group has recently been identified. By assuming that the density of early adopters is increasing at minimum acceptable quality thresholds, this paper proposes a vertical quality differentiation model of product R&D with product subsidies. The impact of product subsidies on the R&D investment of new energy vehicle firms is discussed. We show that the early adopters’ characteristics may affect the stagnant marginal R&D investment of new energy vehicle firms by increasing sales, which determines the impact mechanism of product subsidies. For firms with decreasing marginal R&D investments, insufficient R&D investments result from financial constraints. If insufficient R&D resources deter firms from conducting R&D, substantial unit subsidies invariably incentivize firms to spend their entire R&D budget. Firms with increasing marginal R&D investments, insufficient R&D profits, or financial constraints are prevented from increasing R&D investment. Product subsidies generally have a crowding-in effect on firms not subject to financial constraints, and this effect increases with the unit subsidy. However, the existence of a crowding-in effect may require sufficiently large unit subsidies. In both situations, product subsidies cannot modulate financial constraints if the firm has spent its entire R&D budget. In the first situation, we also show that product subsidies should be replaced by a funding support policy. In contrast, the second situation shows that a funding support policy should be coordinated with product subsidies.
... Each respondent received a sequence of 8 choice sets, while they were asked to select their preferred alternative in each choice occasion. Each choice set was conformed by two automobiles and the no-choice alternative (neither alternative A nor B) (for more details see Rahmani & Loureiro, 2018). An example of the DCE card is shown in Fig. 2. ...
... In the empirical exercise, the assumption of IIA is tested using the Hausman and MacFadden test. Results from this test reject the null hypothesis (IIA assumption) (See specific details in (see Rahmani & Loureiro, 2018)), implying that the MNL model is not appropriate to fit our data. ...
With the aim of analyzing preferences for hybrid electric vehicles (HEVs), two stated preference methods (a contingent valuation exercise and a discrete choice experiment (DCE)) were used in a survey conducted in a representative sample of Spanish drivers. Overall, our findings show robustness between the willingness to pay (WTP) estimates elicited via a latent class model (LCM) and those from a payment card question. In both cases results show an average positive WTP, although insufficient to actually cover the extra cost of HEVs. The lack of interest for HEVs may be motivated by different reasons, including the low level of information related to this technology, and additional false believes about the autonomy of these vehicles. Furthermore, drivers who declare a willingness to buy HEVs do not always do so for environmental reasons, but rather for reputational issues related to their self-image. Thus, in order to increase the market share for HEV vehicles in the Spanish market, informative campaigns and additional economic incentives may be designed.
... According to the International Energy Agency, the number of EVs should continue to grow globally, because of changes in the preferences and attitudes of customers, technical progress in constructing cars and ways of charging, global economic and environmental trends, together with strong incentives from government subsidies, IEA [6]. Research shows that the diffusion of EVs is hard to achieve without government subsidies, Rahmani [7] and incentives play a crucial role in the first stage of the diffusion of an innovation, Davies et al. [8], Figenbaum et al. [9], Phillips [10], Praetorius [11] and Bakker et al. [12]. ...
Full-text available
Road transport causes one fifth of the EU’s total emissions of carbon dioxide (CO2), which are especially high in cities. A suggested solution to this situation is the introduction of electric vehicles (EV). However, evidence from European countries shows that, without any governmental support, the sales of EVs are low in comparison to other vehicles. Our pilot study, conducted in Wrocław (Poland), shows that car purchasers in Poland are aware of the difference between pure electric and hybrid vehicles (HEVs). As most car purchasers buy on the second-hand market, the potential for sales of EVs and HEVs still seems limited. Our study confirmed that consumers have a generally positive opinion about EVs. However, they expect that the purchase of EVs should be subsidized.
Plug-in Hybrid Electric Vehicles (PHEV) can be classified as low CO2 emissions vehicles in Europe if they emit no more than 50 g/km, and are viewed as favourable transitional technology for road transport electrification. Some recent studies challenge their effectiveness in reducing CO2 emissions in real-world conditions. This study tested four Euro 6 PHEVs, both in the laboratory and on the road, for CO2 emissions and energy consumption. The experimental results show that PHEVs in-use CO2 emissions can range from 0 to 6 times the official type-approval CO2 value. The results were analysed to derive relevant operation models and emission factors and benchmark their in-use performance with respect to the officially declared consumption values. A three-dimensional CO2 emissions model is proposed based on charge level and average trip speed or wheel energy. When considering different users' charging practices and representative real-world conditions, PHEVs in-use CO2 emissions are 1.5–2 times the official type-approval CO2 value. Although the real-world emissions of plug-in hybrids appear to increase compared to the official values, they generally remain lower than conventional powertrains. The single-modelling element approach presented offers a novel, robust, and simple to implement way to include PHEVs in planning exercises, emissions calculation models to support national and regional inventories, lifecycle emissions estimates, and fleet-wide emissions monitoring tools.
Full-text available
The performance and practicality of predictive energy management in hybrid electric vehicles (HEVs) are highly dependent on the forecast of future vehicular velocities, both in terms of accuracy and computational efficiency. In this brief, we provide a comprehensive comparative analysis of three velocity prediction strategies, applied within a model predictive control framework. The prediction process is performed over each receding horizon, and the predicted velocities are utilized for fuel economy optimization of a power-split HEV. We assume that no telemetry or on-board sensor information is available for the controller, and the actual future driving profile is completely unknown. Basic principles of exponentially varying, stochastic Markov chain, and neural network-based velocity prediction approaches are described. Their sensitivity to tuning parameters is analyzed, and the prediction precision, computational cost, and resultant vehicular fuel economy are compared.
Air quality is deteriorating, the globe is warming, and petroleum resources are decreasing. The most promising solutions for the future involve the development of effective and efficient drive train technologies. This comprehensive volume meets this challenge and opportunity by integrating the wealth of disparate information found in scattered papers and research. Modern Electric, Hybrid Electric, and Fuel Cell Vehicles focuses on the fundamentals, theory, and design of conventional cars with internal combustion engines (ICE), electric vehicles (EV), hybrid electric vehicles (HEV), and fuel cell vehicles (FCV). It presents vehicle performance, configuration, control strategy, design methodology, modeling, and simulation for different conventional and modern vehicles based on the mathematical equations. Modern Electric, Hybrid Electric, and Fuel Cell Vehicles is the most complete book available on these radical automobiles. Written in an easy-to-understand style with nearly 300 illustrations, the authors emphasize the overall drive train system as well as specific components and describe the design methodology step by step, with design examples and simulation results. This in-depth source and reference in modern automotive systems is ideal for engineers, practitioners, graduate and senior undergraduate students, researchers, managers who are working in the automotive industry, and government agencies.
In the light of European energy efficiency and clean air regulations, as well as an ambitious electric mobility goal of the German government, we examine consumer preferences for alternative fuel vehicles (AFVs) based on a Germany-wide discrete choice experiment among 711 potential car buyers. We estimate consumers’ willingness-to-pay and compensating variation (CV) for improvements in vehicle attributes, also taking taste differences in the population into account by applying a latent class model with 6 distinct consumer segments. Our results indicate that about 1/3 of the consumers are oriented towards at least one AFV option, with almost half of them being AFV-affine, showing a high probability of choosing AFVs despite their current shortcomings. Our results suggest that German car buyers’ willingness-to-pay for improvements of the various vehicle attributes varies considerably across consumer groups and that the vehicle features have to meet some minimum requirements for considering AFVs. The CV values show that decision-makers in the administration and industry should focus on the most promising consumer group of ‘AFV aficionados’ and their needs. It also shows that some vehicle attribute improvements could increase the demand for AFVs cost-effectively, and that consumers would accept surcharges for some vehicle attributes at a level which could enable their private provision and economic operation (e.g. fast-charging infrastructure). Improvement of other attributes will need governmental subsidies to compensate for insufficient consumer valuation (e.g. battery capacity).
Random coefficient models such as mixed logit are increasingly being used to allow for random heterogeneity in willingness to pay (WTP) measures. In the most commonly used specifications, the distribution of WTP for an attribute is derived from the distribution of the ratio of individual coefficients. Since the cost coefficient enters the denominator, its distribution plays a major role in the distribution of WTP. Depending on the choice of distribution for the cost coefficient, and its implied range, the distribution of WTP may or may not have finite moments. In this paper, we identify a criterion to determine whether, with a given distribution for the cost coefficient, the distribution of WTP has finite moments. Using this criterion, we show that some popular distributions used for the cost coefficient in random coefficient models, including normal, truncated normal, uniform and triangular, imply infinite moments for the distribution of WTP, even if truncated or bounded at zero. We also point out that relying on simulation approaches to obtain moments of WTP from the estimated distribution of the cost and attribute coefficients can mask the issue by giving finite moments when the true ones are infinite.
Motorized individual transport strongly contributes to global CO2 emissions, due to its intensive usage of fossil fuels. Current political efforts addressing this issue (i.e. emission performance standards in the EU) are directed towards car manufacturers. This paper focuses on the demand side. It examines whether CO2 emissions per kilometer is a relevant attribute in car choices. Based on a choice experiment among potential car buyers from Germany, a mixed logit specification is estimated. In addition, distributions of willingness-to-pay measures for an abatement of CO2 emissions are obtained. The results suggest that the emissions performance of a car matters substantially, but its consideration varies heavily across the sampled population. In particular, some evidence on gender, age and education effects on climate concerns is provided.
In the light of European energy efficiency and clean air legislations, as well as an ambitious electric mobility goal of the German government, we examine consumer preferences for alternative fuel vehicles (AFVs), based on a Germany-wide discrete choice experiment among 711 potential car buyers. We estimate consumers’ willingness-to-pay (WTP) and contingent variation (CV) for improvements in vehicle purchase price, fuel cost, driving range, refueling infrastructure, CO2 emissions and governmental monetary and non-monetary incentives, hereby accounting for diminishing marginal returns for some of the attributes and taking taste differences in the population into account by applying a latent class model with 6 distinct consumer segments. Our results indicate that almost 36% of the consumers are open-minded towards at least one AFV option, with 15% being AFV-affine insomuch that they show a high probability of choosing AFVs despite their current shortcomings. Our results suggest that German car buyers’ WTP for improvements of the various vehicle attributes varies considerably across consumer segments and that the vehicle features have to meet some minimum requirements so that AFVs are shortlisted. Furthermore, the CV values show that decision-makers in the administration and industry should focus on the most promising consumer group of ‘AFV aficionados’ and their needs, that some vehicle attribute improvements could increase AFV demand rather cost-effectively, and that consumers would accept surcharges for some vehicle attributes at a level, which could enable their economic provision and operation (e.g. fast-charging infrastructure), while others might need governmental subsidies to substitute the insufficient consumer WTP (e.g. battery capacity).
The most comprehensive and applied discussion of stated choice experiment constructions available. The Construction of Optimal Stated Choice Experiments provides an accessible introduction to the construction methods needed to create the best possible designs for use in modeling decision-making. Many aspects of the design of a generic stated choice experiment are independent of its area of application, and until now there has been no single book describing these constructions. This book begins with a brief description of the various areas where stated choice experiments are applicable, including marketing and health economics, transportation, environmental resource economics, and public welfare analysis. The authors focus on recent research results on the construction of optimal and near-optimal choice experiments and conclude with guidelines and insight on how to properly implement these results. Features of the book include: Construction of generic stated choice experiments for the estimation of main effects only, as well as experiments for the estimation of main effects plus two-factor interactions. Constructions for choice sets of any size and for attributes with any number of levels. A discussion of designs that contain a none option or a common base option. Practical techniques for the implementation of the constructions. Class-tested material that presents theoretical discussion of optimal design. Complete and extensive references to the mathematical and statistical literature for the constructions. Exercise sets in most chapters, which reinforce the understanding of the presented material. The Construction of Optimal Stated Choice Experiments serves as an invaluable reference guide for applied statisticians and practitioners in the areas of marketing, health economics, transport, and environmental evaluation. It is also ideal as a supplemental text for courses in the design of experiments, decision support systems, and choice models. A companion web site is available for readers to access web-based software that can be used to implement the constructions described in the book.