Energy Policy 119 (2018) 473–486

Contents lists available at ScienceDirect

Energy Policy
journal homepage: www.elsevier.com/locate/enpol

The impacts of government policies on green utilization diffusion and social T

benefits – A case study of electric motorcycles in Taiwan
⁎
Shihping Kevin Huanga, , Lopin Kuob, Kuei-Lan Choua,c
a
Institute of Management of Technology, National Chiao Tung University, 1001, University Rd, Hsinchu 300, Taiwan, ROC
b
Department of Accounting, Tamkang University, 151, Yingzhuan Rd. Tamsui Dist., New Taipei City 25137, Taiwan, ROC
c
Green Energy and Environment Research Laboratories, Industrial Technology Research Institute, 195, Sec. 4, Chung Hsing Rd., Chutung, Hsinchu 31040, Taiwan, ROC

A R T I C LE I N FO A B S T R A C T

Keywords: Diffusion of green utilization is a big challenge. A growing body of literature has suggested that governments
Green utilization play a core role in promotion of green utilization. The extant research of green utilization behavior is focused
Diffusion curve primarily on consumer preferences. So far, in-depth quantitative research of the government’s role in the con-
Investment decision nection between green utilization and production is absent. Unlike previous research of green utilization be-
Government policies
havior, this study established a dynamic adjustment mechanism for the government's subsidy policy for electric
Social benefits
vehicles through the case study of Taiwan's electric motorcycle, so that the government's finite budget can be
effectively allocated. Results indicate that emerging energy technologies reach the market competition condi-
tions of economies of scale through government subsidy, subsidy policies should gradually phase out the market
and transfer finite budgets to other emerging energy technologies that need to break through economies of scale.
Therefore, the incentives phase-out mechanism of the electric motorcycle in this study will be helpful in the
decision-making thinking of government and quantitative evaluation for the promotion of emerging energy
technologies in the future.

1. Introduction of Taiwan's electric motorcycle, so that the government's limited budget

can be effectively allocated. When emerging energy technologies reach
The motorcycle is an important short-distance transportation tool in the market competition conditions of economies of scale through gov-
densely populated metropolitan areas. The diffusion of the electric ernment subsidy policies, subsidy policies should gradually phase out
motorcycle has an important influence on energy conservation, carbon the market and transfer finite budgets to other emerging energy tech-
reduction and air pollution reduction in metropolitan areas. At present, nologies that need to break through economies of scale. Therefore, the
the major cities in the Japan, Europe and United States have actively phase-out mechanism of the incentives for the electric motorcycle
developed the short-distance transportation of electric motorcycle. For policy in this study will help provide the government with the decision-
example, Japan planned to switch 85,000 postal motorcycles from ga- making thinking and quantitative evaluation model for the promotion
soline to electric modes in 2018, and set up 25,000 battery exchange of emerging energy technologies in the future.
stations around country’s post office. Some cities are actively devel- With the rise of living standards, people are paying more attention
oping electric motorcycle sharing lease services, such as Paris (cityscoot to environmental sustainability. A pro-environment utilization and
1000), San Francisco (scoot networks 400), Barcelona (Cooltra/Yugo production system should be created. Green utilization behavior is af-
561), and Berlin (Coup 1000). fected by a variety of factors. Consumers’ willingness and ability to
From 2017 to 2020, the annual compound growth rate of global engage in green utilization are most critical (Peattie, 2010). Diffusing
electric motorcycle is 12.3%, which will reach 845,000 units in 2020 green utilization is a big challenge. A growing body of literature has
(as shown in Fig. 1). In the first half of 2017, the market share of suggested that governments should play a core role in promotion of
Taiwan's electric motorcycle accounts for 5.3% of global sales (as green utilization (Norberg-Bohn, 1999; Stevens, 2010; Young et al.,
shown in Fig. 2). 2010; Tran, 2012; Tseng et al., 2013; Lorek and Spangenberg, 2014;
This study established a dynamic adjustment mechanism for the Frederiks et al., 2015; Testa et al., 2015). Hence, unlike previous re-
government's subsidy policy for electric vehicles through the case study search which concentrates its focus on consumer preferences, we will

⁎
Corresponding author.
E-mail addresses: ksph@nctu.edu.tw (S.K. Huang), lopinkuo@gmail.com (L. Kuo), Kelly@itri.org.tw (K.-L. Chou).

https://doi.org/10.1016/j.enpol.2018.04.061
Received 11 December 2017; Received in revised form 25 April 2018; Accepted 27 April 2018
0301-4215/ © 2018 Elsevier Ltd. All rights reserved.
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Fig. 1. The sales of E-scooter in the world in 2009–2020.

Source: Materialsnet (2017).

how to dynamically adjust the subsidies to avoid inappropriate allo-

cation of public resources; (3) assess the benefits of diffusion of green
utilization in the aspects of energy saving and pollution reduction.
This paper is structured as follows: Section 2 reviews literature on
definitions of green utilization, the effects of financial benefits on green
utilization, and the effects of government's green policies on diffusion of
green utilization. Section 3 describes the methodology, including the
diffusion model and the energy system engineering model, the data
collection, and assumptions. Section 4 presents and explains the results
of model estimation. Section 5 concludes the results and discusses their
policy implications.

2. Literature review
Fig. 2. The market share of E-scooter in the world in the first half of 2017.
Source: TaiwanJobs (2017). The objective of green utilization research is to explore the major
factors affecting green utilization and provide an important reference
build a quantitative model from the perspective of consumers’ invest- on design and promotion of policies. In this section, we will first review
ment decision to explore factors that promote diffusion of electric literature that address the definition of green utilization, the effects of
motorcycles in Taiwan and forecast the diffusion. The factors include financial benefits on green utilization, and the effects of government
government’s incentives, emission standards for vehicles, oil price, and policies on green utilization.
battery cost. In addition, we will also develop a dynamic adjustment
mechanism for government policies on green utilization and evaluate 2.1. Definitions of green utilization
whether the social benefits from these policies can contribute to the
national goals of development. The results presented a forecast on the The concept of green utilization was first introduced in marketing
market growth of green utilization. The policy implications in the study research during 1970s. Generally, it refers to the decision-making for
ensures fair allocation of public resources. purchase, use or disposal of environmentally friendly products (Peattie,
It is recognized that electric vehicles are a new and important 2010; Cherian and Jacob, 2012; Awan and Raza, 2014; Cheng et al.,
carbon reduction technology for the transportation sector across the 2014). Many similar concepts have been proposed in related literature.
world. Statistics released by Environmental Protection Administration These concepts include sustainable utilization behavior (Ek and
of Taiwan show that the amounts of carbon monoxide, hydrocarbons, Söderholm, 2008; Young et al., 2010; Lorek and Spangenberg, 2014;
and particulate matters generated by traditional motorcycles take up Biswas and Roy, 2015), pro-environmental behavior (Peattie, 2010;
27%, 48%, and 23% of the total amounts generated by all road vehicles. Kaufmann et al., 2012; Cheng et al., 2014), and green consumer choice
Motorcycles have become a major source of urban air pollution in (Tseng et al., 2013; He et al., 2014). Green utilization behavior differs
Taiwan. In order to achieve the target of Sustainable Energy Policy from general consumption behavior. While general consumption be-
Convention, the government has implemented carbon emission controls havior is often driven by an instantly made evaluation of costs and
in the transportation sector and set allowable energy performance benefits, green utilization behavior is driven by a more far-sighted
standards and air pollution emission standards for motorcycles. evaluation of the social benefits from the decision.
Moreover, the government has launched numerous measures aimed at Green utilization behavior can be classified into green curtailment
promoting electric motorcycles. These measures include exemption of behavior and green purchase behavior. Green curtailment behavior is a
excise tax and fuel tax for electric motorcycles and price subsidy for habitual behavior. Reducing energy utilization or vehicle use, recycling,
buyers. and reusing are behavior of this kind. As expenses will be curtailed, this
In this study, we focus on diffusion of electric motorcycles in kind of behavior may cause discomfort to people who are used to their
Taiwan. From a perspective of investment benefits for consumers, we existing living styles. In contrast, green purchase behavior is an in-
will build a green utilization diffusion model to (1) explore how the vestment behavior. It is a non-recurring decision to adopt an innova-
government can use social norms and financial incentives to influence tion. It requires investment of additional money but will not disturb the
consumer perception of the economic benefits of green products and existing living quality (Jansson et al., 2010; Peattie, 2010; Cheng et al.,
further motivate them to have green utilization behavior; (2) evaluate 2014; Testa et al., 2015). Despite the difference, both kinds of green

474
S.K. Huang et al. Energy Policy 119 (2018) 473–486

utilization behaviors contribute to creation of a better and sustainable confirmed that financial benefits are determinants of green utilization.
environment. Hence, governments should take into account the cost-effectiveness and
Based on the above-mentioned definition and classification, we will maximum return on investment when promoting a green technology or
concentrate our focus on green purchase behavior, especially the be- product, especially among early adopters. These two factors are critical
havior of buying high-priced, durable, and innovative technologies or to diffusion of green utilization and closing of the values-action gap. In
products that are designed to replace existing technologies or products, this study, we will add a variable of financial benefits, called payback
to analyze diffusion of the behavior. Green products of this type are period, into the diffusion function to capture the dynamic effects of
durable goods. Consumers’ decision-making for buying this type of financial benefits on diffusion of green utilization.
products can be deemed an investment behavior. We will analyze dif-
fusion of electric motorcycles in Taiwan from an investment decision 2.3. The effects of government policies on diffusion of green utilization
perspective.
To achieve the goal of creating a sustainable environment, radical
2.2. The effects of financial benefits on green utilization technological innovations are essential. In promotion of commercial
green technologies, governments should provide stronger financial in-
Most academic analyses of green utilization behavior are based on centives and clear environmental requirements to drive development of
two views, namely the rational choice in economics and behaviorism in innovative technologies (Norberg-Bohn, 1999; Tseng et al., 2013). The
psychology (Peattie, 2010; Loock et al., 2013; Frederiks et al., 2015). survey conducted by Sang and Bekhet (2015) shows that adoption of
The former suggests that when given sufficient information, consumers electric vehicles in Malaysia is significantly affected by government
will seek to make a rational, utility-maximizing decision (Chen et al., policies. In a study of the correlation between government policies and
2012; Zhang et al., 2013; Wu, 2015). The latter proposes that an in- development of electric vehicles in Japan, Åhman (2006) highlights
dividual’s consumption behavior is affected by a multitude of factors, that government’s support is pivotal to the success of diffusion of
including internal factors (e.g. attitude, value, and demographic vari- electric vehicles in the early stage of market development.
ables) and external factors (e.g. norms, social comparison, incentives, The concept of sustainable consumption and production (SCP) was
and promotional advertisement) (Kim and Choi, 2005; Kaufmann et al., proposed in 2002 at World Summit on Sustainable Development
2012; Cheng et al., 2014; Delafrooz et al., 2014; Ludhiyani et al., 2012; (WSSD). There exist some gaps between sustainable consumption and
Sarumathi, 2014; Ritter et al., 2015). sustainable production, and the government is responsible for closing
Many studies have mentioned the gap between attitude and beha- the gaps. In this role, the government has two major tasks. One is to
vior toward green utilization. This gap exists when one shows a pro- solve market failures by introducing external environmental costs into
environmental protection attitude but hesitates when buying a green market economy. The other is to solve system failures by bridging the
product. This phenomenon is generally referred to as the “attitude-be- information gap between sustainable consumption and production,
havior gap” or the “values-action gap” (Peattie, 2010; Awan et al., which can be carried out by the voluntary or mandated related to en-
2011; Frederiks et al., 2015; Cherian and Jacob, 2012; Lin and Huang, vironmental labelling and production certification (Stevens, 2010).
2012; Sarumathi, 2014). So far, many theories have been applied to Some studies suggest that the government reinforce factors that can
analyze green utilization behavior. These theories include theory of narrow the behavioral gap in green utilization. These factors include
reasoned action (TRA), theory of planned behavior (TPB), norm-acti- consumers’ green value, purchase experience, environmental knowl-
vation model (NAM), value-belief-norm (VNB), and theory of con- edge, access information of green products, and affordable financial
sumption values (Biswas and Roy, 2015; Testa et al., 2015). These cost (Young et al., 2010; Sinnappan and Rahman, 2011; Zhao et al.,
theories are employed to identify factors affecting green utilization 2014; Testa et al., 2015).
decisions and further explain or predict green utilization behavior. The Generally, governments promote sustainable utilization and pro-
key factors affecting green utilization decisions are useful for the gov- duction through two policies. One is called the command-and-control
ernment and businesses when formulating marketing strategies for policy, which is to directly mandate adoption of green production
green products. controls. For instance, setting fuel efficiency or pollution emission
It is documented in many papers that financial benefits from lower standards and levying an external cost tax can compel consumers or
operational cost, improved fuel efficiency or other complementary ex- producers to change their behavior. The other is called the incentive-
penditures are determinants of consumers’ willingness to pay more for based policy, which is to provide consumers a subsidy or tax exemption
green energy or green vehicles (Mourato et al., 2004; Adamson, 2005; to encourage green utilization. With the increase of green utilization,
Potoglou and Kanaroglou, 2007; PwC, 2013; Wu and Chen, 2014). green production will also increase. However, this kind of policy is
Frederiks et al. (2015) indicate that a values-action gap is prevalent effective only when the financial incentive is powerful enough to in-
among energy-saving behavior in the household sector. They suggest fluence consumers’ decisions (Santos et al., 2010; Stevens, 2010).
that green consumers dislike losses and higher value risks. They tend to Tax exemption or reduction can substantially stimulate utilization
evaluate a product’s effectiveness based on relative values and are ea- of green energy or green vehicles (Bomb et al., 2007; Potoglou and
sily influenced by group behaviors. However, these value perceptions Kanaroglou, 2007; Haan et al., 2007; Chandra et al., 2010). Tax credits
are usually ignored by policy makers in promotion of energy efficiency. can not only promote diffusion of high-efficiency clear vehicles but also
To promote an innovative vehicle among early adopters, the cost-ef- bring expectable social benefits (Skerlos and Winebrake, 2010; Rogan
fectiveness of the vehicle should be stressed (Lane and Potter, 2007). In et al., 2011). The evidence in Gallagher and Muehlegger (2011) reveals
the choice between low-carbon vehicles, consumers will consider the that sales tax waivers are more effective than income tax credits in
relative competitiveness of fuel price for different vehicles (Bomb et al., inducing consumer purchase of hybrid electric vehicles (HEV). In ad-
2007). dition, the sales of HEV increases with the rise of gasoline prices, sug-
Our review of literature shows that economic rationality and in- gesting that consumers expect a fuel cost reduction from the high fuel
dividual perceptions have been identified as having a significant rela- economy of HEV. Brand et al. (2013) indicate that in order to reduce
tion with green utilization in a larger number of studies. In other words, GHG emissions by the transportation sector, governments should design
consumers are most concerned about these two factors in the choice of incentive schemes with strong up-front price signals that reward low-
green products, show as Table 1. carbon vehicles and penalize high-carbon ones.
Obviously, factors affecting green utilization vary by demographic From the above-mentioned literature on the correlation between
characteristics and regional cultures. A factor may be significant in one government policies and diffusion of green utilization, we can find that
country but be insignificant in another. However, previous research has governments indeed play the role of a promoter in adoption and

475
 S.K. Huang et al.

Table 1
Literature review of green consumption behavior with statistical analysis.
Driving factors Economic Demographic Environmental Personal Perceived consumer Life style Social Media Information
Author Published time country Green products rationality consciousness perception effectiveness and habit norm spread acquirement

* * *
Kim and Choi 2005 U.S.A. General
* * *
Ek and 2008 Sweden Electricity
Söderholm
* * *
Young et al. 2010 UK General
* * *
Awan et al. 2011 Pakistan Energy
* * *
Lin and Huang 2012 Taiwan General
*
Chen et al. 2012 Taiwan Hydrogen Electric
motorcycle
*
Loock et al. 2013 Switzerland Energy
* * *
Zhang et al. 2013 China New energy vehicle
* *
Wu and Chen 2014 Taiwan General
*

476
Delafrooz et al. 2014 Iran / cluster General
* *
Awan et al. 2014 Sweden Energy
* * *
Zhao et al. 2014 China General
*
He et al. 2014 U.S.A. Hybrid vehicle
* * * *
Ritter et al. 2015 Brazil General
* *
Wu 2015 Taiwan/ cluster General
* *
Biswas and Roy 2015 India General
* *
Testa et al. 2015 Italy Energy

• Economic rationality: Price, Quality, Risk

• Demographic: Gender, Age, Occupation, Education, Income
• Personal perception: Attitude, Believe, Value
• Perceived consumer effectiveness: Personal ability to affect the outcome in a positive way
• Social norm: Social network or collective conscious
• Information acquirement: Advertisement, Knowledge, Education
* means statistical significance.
Energy Policy 119 (2018) 473–486
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Fig. 3. An associative analysis framework of green technology diffusion and social effects.

diffusion of innovative green vehicles. Besides, empirical evidence also derived from the market diffusion model into an energy engineering
confirms that governments’ financial incentives also have a positive system model to calculate the social benefits under different market
effect on utilization and production of green vehicles. However, most conditions. The results can contribute to the implementation of gov-
extant research of the relation between government policies and green ernment policies and the planning of target markets for the industry.
utilization relies on survey data or a qualitative approach to analyze the The market diffusion model, the energy engineering system model,
consumers’ perception on the functional role of government. Research simulated scenarios, and the data collection are respectively explained
dedicated to a quantitative evaluation of values or addressing dynamic as follows.
adjustment of government policies is relatively rare. Unlike previous
research, we will conduct an integrated evaluation of market diffusion 3.1. Market diffusion model
and social benefits of green vehicles.
The development of innovations follows an s-shaped curve and
3. Method and data collection usually consists of four phases, including the embryonic phase, the
growing phase, the mature phase, and the aging phase. The well-known
In extant research of environmentally friendly production and uti- diffusion model is a mathematical function of these four phases.
lization systems, little effort has been made to in-depth quantitatively General forecasting of diffusion is based on the best fit of the empirical
analyze the roles and functions of the government (Stevens, 2010). data of time to the model, and trend extrapolation is applied outside the
Therefore, as shown in Fig. 3, we build a relational structure of green range of the empirical period (Jaakkola et al., 1998). In the extant lit-
utilization diffusion and social benefits by three parties, including the erature, three diffusion models of innovations are documented, in-
government, consumers, and producers. The government uses subsidies cluding the logistic model (also known as Pearl model), Gompertz
and pollution regulations to promote diffusion of green technologies. model, and Bass model. All of these models have been extensively ap-
These policies can reduce the cost of technology learning for producers, plied in various areas.
improve the competitiveness of the industry, and help the government In the energy research area, some studies consider different ex-
attain expected carbon reduction goals. In this cycle of diffusion and planatory variables in addition to time to improve the explaining power
social benefits, the government plays the trigger for the diffusion and of the diffusion model (Won et al., 2009; Virginie, 2010; Zhou et al.,
crossing the chasm of technological lifecycle curve. 2011; Kong and Bi, 2014). The U.S. NEMS model considers the variable
The above literature review reveals that consumers mostly concern of payback period when applied to estimation of the logistic diffusion
about financial benefits and economic rationality when buying green function of new energy technologies. This model can be used to explore
products. The relational loop in Fig. 3 represents the interaction be- consumers’ perception of risk associated with investment in emerging
tween green utilization and production. To form this relational loop, technologies and the effect of timing on diffusion (US EIA, 2011).
diffusion of green utilization must be activated. We adopt the market This logic of calculation for building a diffusion model of emerging
penetration function of U.S. National Energy Modeling System (NEMS) energy technologies is interpreted as follow. Firstly, the technology
model to analyze the relationship between the government and con- learning effect triggers cost reduction under the promotion of external
sumers in diffusion of green utilization. From the perspective of cost- factors. When the incremental investment for the emerging technology
benefit investment on green technologies and products, we build a or product drops below the Net Present Value of future aggregative
market diffusion model for electric motorcycles in Taiwan. This model benefits in energy saving or pollution reduction, the break-even point of
consists of factors affecting market diffusion, including government investment is reached. Meanwhile, the diffusion of this emerging en-
incentives, oil price, and battery replacement cost. ergy technology or product will follow an s-shaped curve toward
Furthermore, we also evaluate the social benefits generated by commercialization.
government policies on green utilization, including energy saving and Lund’s (2006) market diffusion model is a logistic time curve as
pollution reduction. We input the market share of electric motorcycles follows:

477
S.K. Huang et al. Energy Policy 119 (2018) 473–486

1 service demand in the downstream. The energy service demand com-

f (t ) =
1+a0 e−β (t − t0) prise of four sectors, including industrial, residential, service, and
transportation. In this study, we focus on replacement of gasoline-
The U.S. NEMS model includes payback period, a criterion for in-
powered motorcycles for electric motorcycles in the transportation
vestment, into the logistic time curve:
sector to examine change in energy utilization and potential reductions
1 in carbon dioxide and air pollution after the replacement.
Pen y = MaxPen y − 1
MaxPeny
+ e[β *(y − Histyear − PayBack ) In Taiwan’s TIMES model, vehicles used in the transportation sector
are classified by specifications. Among the vehicles, motorcycles are
β is a parameter of diffusion, estimated by regression with historical classified by engine displacement into four categories, including 50cc or
data. Since the historical data about the development of pure electric less, 51–100cc, 101–150cc, and 151–250cc. The inventory for each
vehicles are insufficient for regression analysis, and HEV is a pioneer in category of vehicles is estimated based on the energy balance sheets
the development of electric vehicles, we collect the global growth data published by Bureau of Energy and the statistics released by Ministry of
of HEV (2004–2013) in “Tracking Clean Energy Progress” released by Transportation and Communications with 2014 as the baseline year.
International Energy Agency (IEA) in 2014. Using the data, we take the The trends of transportation efficiency and cost of each category of
logarithm of Lund’s (2006) logistic function and perform linear re- vehicles are derived from the Annual Energy Outlook released by U.S.
gression to obtain β = 2. Energy Information Administration (EIA) and expert evaluations of-
Payback denotes the payback period of investment. It is estimated fered by Industrial Technology Research Institute (ITRI) in Taiwan.
by the future cash flow of consumers’ investment. According to the five
major banks in Taiwan, This study refers to the average consumer loan 3.3. The assumptions of scenarios
interest rate from five public banks in Taiwan in the last decade
(2005–2014) which is around 2.9%. Compared to the interest rate of Motorcycles and automobiles have different uses in densely popu-
8.5% in 1995, the average interest rate in the last ten years has dropped lated metropolitan areas. Motorcycles are mostly used for short-dis-
significantly. The interest rate used in this study is the nominal interest tance transport, while automobiles are mainly used for long-distance
rate, regardless of the inflation rate. In general, the assessment of the transport. This study is aimed at replacing traditional gasoline motor-
discount rate for the investment of emerging energy technologies that cycles with electric motorcycles under the same short-distance trans-
are not yet mature in the market should take into account the market port conditions as a measure of social benefits. Therefore, the replacing
entry risk of immature emerging energy technologies, which means effects of different uses are not within the scope of this study. Besides,
discount rate should consider the cost of risk, called hurdle rate, and motorcycles and automobiles have different considerations for the
add it to the loan interest rate. The market for electric motorcycles is safety. Limited by the scope of modeling, this study only explores the
currently in the growth stage in Taiwan. Taiwan has a complete pro- alternative relationship and the relative direct costs of purchases and
duction technology and industrial supply chain for motorcycle. The maintenance between the electric motorcycle and gasoline motorcycle
charging mode of lithium batteries for the electric motorcycle is por- with the same short-distance use. Therefore, the external costs with
table battery-exchanged which overcomes the obstacle for lacking different uses for motorcycles and automobiles, including the safety,
sufficient charging infrastructure around the country. The market entry traffic congestion, commuting, and noise, are outside the scope of this
barriers for electric motorcycles, relative to that for electric auto- study.
mobiles, are lower, so the discount rate for the electric motorcycle in In Taiwan, motorcycles with engine displacement smaller than 50cc
Taiwan does not need to consider the hurdle rate of risk cost. take up a market share of only 2%, motorcycles with 51–100cc engine
Other items to consider in estimation of cash flow include initial displacement account for 42%, and the remaining 56% market is
capital outlay, battery replacement cost, oil price, electricity price, tax dominated by large motorcycles (with 101cc or more). Table 2 provides
exemption, and subsidy. The details on these items will be provided in a comparison on specifications, purchasing cost, and performance be-
the section about scenario settings. tween electric motorcycles and gasoline-powered motorcycles. An
MaxPeny denotes maximum market share. We use the current analysis of motorcycle users’ behavior conducted by Ministry of
market shares of gasoline-powered motorcycles, by different engine Transportation and Communications (2011) shows that motorcycles
displacement level, as the maximum market shares for electric motor- travel an average distance between 10 and 27.2 km per day for the
cycles. According to government-released statistics, the maximum following purposes: commuting (56.8%), shopping (18.7%), business
market shares of small and medium-sized motorcycles are 2% and 42% (9.3%), picking up family members (8.9%), leisure purposes (4.7%),
respectively. and others (1.6%). Most of today’s electric motorcycles are able to
travel a distance between 40 and 62 km on full charge. One full charge
3.2. Energy engineering system model per day is enough for basic purposes. With the development toward
lighter batteries, models that use portable batteries have been devel-
The integrated MARKAL-EFOM System (TIMES) is an optimization oped. They can be charged without being bound to a fixed location or
model of energy systems developed by IEA in the Energy Technology the first floor of buildings. Given the time-of-use rates of electricity in
System Analysis Program (IEA-ETSAP). Integrated with engineering Taiwan, most rational users of electric motorcycles charge their ve-
methods and economic methods, this model is mainly applied to ana- hicles during off-peak hours to prepare for the next day’s transportation
lysis of long-term energy and environmental scenario simulation. need.
TIMES consists of technology-rich network linking between the supply In the current market of electric motorcycles, small and medium-
and demand for energy system (as shown in Fig. 4). Through dynamic sized models are considered to be more mature. Hence, our subjects are
linear programming, it can optimize energy and technologies portfolio electric motorcycles targeted at replacing gasoline-powered motor-
that meets the energy service demand at least cost and evaluate energy cycles with engine displacement less than 100cc. We set three scenarios
and environmental impacts of the system. with the following variables: (1) government subsidies, (2) oil price
The development of the energy engineering system model in Taiwan trend, and (3) battery replacement cost. All scenarios are presented as
began in 1994 when MARKAL (the abbreviation of MARKet ALlocation) Table 3.
model was introduced. MARKAL was upgraded to TIMES in 2009.
Taiwan’s TIMES model consists of the following components: energy 3.3.1. The assumption of government subsidies
generation in the upstream, including power generation and oil-re- The government in Taiwan began to promote electric motorcycles in
fining; energy transmission and distribution in the midstream; energy 2002 and launched a series of incentive measures. In 2014, these

478
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Fig. 4. The energy system reference of Taiwan TIMES Model.

Source: Huang et al. (2017)

measures include exemption of excise duty for three years (US$125 per lifetime will significantly decrease, and there can be greater diffusion of
vehicle), exemption of fuel taxes (US$9.4–14 per vehicle), and provi- electric motorcycles. In this scenario, the government’s subsidies should
sion of national purchasing subsidies (US$226 for each small-sized also be properly adjusted.
vehicle and US$313 for each medium-sized vehicle). The effective
period of the subsidies was scheduled to end in 2017 but might be 3.3.4. The decreasing trend of vehicle cost
adjusted according to actual conditions. To meet the varying demands With market diffusion, the technology learning effect can trigger a
of users, four types of battery charging facilities have been developed, decline in the cost of electric motorcycles. Based on opinions of experts
including free charging stations set up by dealers, battery charging and in Industrial Technology Research Institute (ITRI) in Taiwan, we set the
monitoring stations where services are paid by EASYCARD, battery relative trends of prices of gasoline-powered motorcycle and electric
exchange stations, and convenience stores with the charging service. As motorcycles as shown in Fig. 6. We use the price of motorcycles in 2014
of 2013, the government has subsidized a total of 2056 battery charging as the baseline to normalize the future costs of the two types of mo-
facilities in 1152 locations nationwide. torcycles. The technology of gasoline-powered motorcycles should be
According to the economic rationality principle, this kind of gov- unceasingly improved so as to comply with new air pollution standards,
ernment’s subsidies should be adjusted over time to avoid unfair allo- so the cost of gasoline-powered motorcycles will continuously rise. For
cation of public resources. When the payback period of investment in electric motorcycles, the technology learning effect in an economy of
electric motorcycles is less than 10 years, the incremental investment scale will not emerge before 2019. Besides, the battery technology is
cost in an electric motorcycle will pay back within its lifetime. In this constantly advancing (from lead-acid battery to lithium battery), so the
scenario, electric motorcycles will pass the initial embryonic phase of cost of electric motorcycles will slightly rise in the innovation stage.
diffusion. It is necessary to progressively phase out the subsidies to let However, after the technology becomes mature and the market begins
diffusion of electric motorcycles follow an s-shaped curve without to expand, the cost of this new innovation will gradually decline under
policy intervention toward the saturation point. the technology learning effect. Trend projection based on expert ex-
periences is usually subject to a period of no longer than 10 years. We
3.3.2. Estimation of gasoline and electricity price trends will use fixed values for the trend outside the 10-year period.
The gasoline and electricity price trends are estimated by regression
analysis with the international historical prices of crude oil, coal, and 4. Results
natural gases. After obtaining the regression coefficients, we use U.S.
EIA projections of energy price in the reference and high oil price cases 4.1. Market diffusion curve and dynamic adjustment of government
(U.S. EIA, 2015) to obtain the estimates of future gasoline and elec- subsidies
tricity prices in Taiwan. The results are as shown in Fig. 5. As high oil
prices offer a niche competitive advantage for electric motorcycles, we We design two scenarios, one without government subsidies and the
will explore the diffusion effects of electric motorcycles and the dyna- other with government subsidies, to evaluate the market diffusion curve
mically adjustment of government subsidies under different oil price and the effect of government subsidies for small-sized and mid-sized
levels. electric motorcycles.

3.3.3. The assumption of battery replacement cost 4.1.1. Small-sized electric motorcycles
Nykvist et al. (2015) indicate that the cost of batteries for electric Small-sized electric motorcycles are aimed at replacing gasoline-
vehicles has nosedived during 2007–2014. The average cost, which powered motorcycles with engine displacement of 50cc. So far, this
remained higher than US$1000/kWh in 2007, dropped to US$410/kWh type of vehicle has a market share of only 2%. Development of electric
in 2014, with the average annual rate of decline reaching 14%. It is motorcycles in Taiwan began in 1992 when the government collabo-
estimated that the cost of batteries for leading vehicles in the market rated with related businesses to form an Alliance for Developing Key
will fall below US$300/kWh in 2014, meaning that the IEA goal for Components and Technologies of Electric Motorcycles. Each year from
2020 can be achieved 6 years earlier. The latest goal of IEA (2015) is to 2011 to 2014, about 8000–10,000 vehicles were sold under the gov-
bring down the battery cost to US$150/kWh by 2022. Electric vehicles ernment’s subsidies. This number is close to the average number of
are expensive than engine-powered vehicles mainly because of the high newly registered small-sized gasoline-powered motorcycles
cost of batteries. Hence, we assume that if the battery cost is down to US (8800–11,500 units). With the advancement of technology and the
$150/kWh, the battery replacement cost for electric motorcycles within promotion of the technology learning effect, small-sized electric

479
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Table 2 Table 3
Comparison between electric motorcycle and gasoline-powered motorcycle in Description for scenarios in this study.
Taiwan.
Scenarios Without With High Battery
Electric motorcycle Gasoline-powered government government oil replacement cost
motorcycle subsidies(BAU) subsidies price decreasing

Medium-size Small-size Medium- Small-size S_A •

size S_B •
S_C •
Battery Capacity/ 20Ah above 20Ah below 51–100cc 50cc or S_D •
Displacement less M_A •
Motor power rated 1～3.7 kW 1 kW or less – – M_B •
Average price (US 2364 1868 2018 1932 M_C •
$/vehicle) M_D •
Subsidy (US$/vehicle) 313 226 0 0 M_B+C • •
Excise duty exemption (US 125 125 0 0 M_B+D • •
$/vehicle)
Fuel tax exemption(US 14 9.4 0 0 S: Small-sized motorcycle.
$/vehicle-year) M: Mid-sized motorcycle.
Battery replacement cost 1098 499 0 0 A~D: codes for different scenarios.
(US$/vehicle)a
Net purchase cost (US 3023 2015 2018 1932
$/vehicle) magnify diffusion of small-sized electric motorcycles. In particular,
Cruising endurability 57–62 40–45 231 287 decreasing battery replacement cost (S_D) has the greatest effect on
(kilometer)b acceleration of market diffusion toward saturation.
Energy used electricity electricity gasoline gasoline
Fuel economy(kcal/km)d – 33.97 – 143.52
Emission factor of CO 0 0 2 below 2 below 4.1.1.2. The scenario with government subsidies. The latest government
air pollutionc HC 0 0 0.8 below 0.8 below subsidy for electric motorcycles is effective between 2014 and 2017 and
(g/km) NOx 0 0 0.15 below 0.15 may be adjusted depending on actual conditions. Because small-sized
below
electric motorcycles are expected to gain market competitiveness by
CO2e – 20.6 – 41.6
2018, we assume that the subsidy can be terminated after its planned
a
Supposed that motorcycle lifetime of 10 years, the lithium battery lifetime expiration in 2017, and the resources should be allocated to promotion
of 5 years, therefore, it need to be replaced twice during motorcycle lifetime. of larger electric motorcycles. In the scenario where the subsidy is
The present value was estimated with discount rate of 3%. expected to expire in 2017 (S_B), a negative growth of the rate of
b
Cruising endurability was measured on a constant speed of 30 km per hour. change in diffusion will appear in 2018 (Fig. 8). Starting in 2024,
c
Air pollution of vehicle is exhaust emissions, so electric motorcycle has no diffusion will grow annually by a rate similar to that appears in the high
exhaust emissions. Gasoline motorcycle is based on Taiwan EPA emissions oil price scenario (S_C). Although the subsidy is absent after 2018, a
standards. certain degree of technology learning effect will be formed. By that
d
In the laboratory's standard mode test, the fuel economy of the motorcycle time, early adopters will cross the chasm to become the early majority
with 1000 W is 30 Wh/km. The electric motorcycle studied in this study does (Rogers, 1962; Moor, 2014). This suggests that the influence of
not consider the power generation efficiency and transmission and distribution government subsidies on market diffusion is considerable.
efficiency of the power plant. The fuel economy of electric motorcycle is Fig. 8 shows the rates of change of diffusion in all scenarios. After
39.5 Wh per km (= 30 Wh/km/95%/80%) while considered the power con- 2025, the diffusion will enter the mature phase of S curve, with a
sumption efficiency (battery charge and discharge efficiency is 95% and gradual and stable decline in the rate of change, year by year. Lower
charger efficiency is 80%), which converts it to calorific value unit is equivalent rate of change indicates sooner market saturation.
to 33.97 kcal/km (= 39.5 Wh/km/1000 Wh/kWh*860 kcal/kWh). Taking the
50cc gasoline motorcycle as a benchmark, the fuel economy of a gasoline
4.1.2. Mid-sized electric motorcycles
motorcycle is 0.0184 L/km, which converts it to calorific value unit is equiva-
Mid-sized electric motorcycles are designed to replace 51–100cc
lent to 143.52 kcal/km (= 0.0184 L/km*7800 kcal/L). Therefore, we infer the
gasoline-powered motorcycles, which take up a market share of 42% in
ratio of the fuel economy (km/kcal) of a 1000 W electric motorcycle relative to
Taiwan. With the diffusion of small-sized electric motorcycles and the
a 50cc gasoline locomotive is 4 times [(1/33.97 km/kcal)/(1/143.52 km/kcal)].
e
The carbon dioxide emitted by an electric motorcycle is the fuel economy
promotion of the technology learning effect, the focus in promotion of
multiplied by the average electricity generation emission coefficient of na- electric vehicles has been shifted onto mid-sized and large electric
tional, equaled to 20.6 g/km (= 39.5 Wh/km/1000 Wh/kWh*0.521 kgCO2/ motorcycles with a much larger market share. However, the higher
kWh*1000 g/kg). Taking the 50cc gasoline motorcycle as a benchmark, the CO2 costs of mid-sized and large electric motorcycles (including the battery
emission of a gasoline motorcycle is the fuel economy multiplied by the gaso- replacement cost), compared to traditional gasoline-powered motor-
line emission factor equal to 41.6 g/km (= 0.0184 L/km*2.26kgCO2/ cycles, are the main competitive obstacle in the market.
L*1000 g/kg). Therefore, we infer that a 1000 W electric motorcycle can reduce
the CO2 emissions by 50% comparing with gasoline motorcycle. 4.1.2.1. The scenario without government subsidies. Without any subsidy
policy (M_A), mid-sized electric motorcycles cannot reach the break-
motorcycles have gained competiveness in the market of small-sized even point until 2048. In this scenario, their market diffusion is not
gasoline-powered motorcycles (50cc). likely to follow an s-shaped curve (Fig. 9). In the high oil price (M_C)
scenario and the decreasing battery replacement cost (M_D) scenario,
4.1.1.1. The scenario without government subsidies. Fig. 7 illustrates the despite the absence of government subsidies, mid-sized electric
trends of market diffusion of small-sized electric motorcycles in motorcycles still reach the break-even point during 2024–2026, and
different scenarios. These trends show that without government their market diffusion can follow an s-shaped curve afterwards (Fig. 9).
subsidies (S_A), small-sized electric motorcycles reach the break-even Obviously, the influence of high oil price is greater than that of
point in 2018 and follow an s-shaped curve of market diffusion toward decreasing battery replacement cost. This finding is opposite to that
saturation. Even without government subsidies, high oil price (S_C) and found among small-sized electric motorcycles. A plausible explanation
decreasing battery replacement cost (S_D) can also significantly is that on average, mid-sized motorcycles have a double annual travel

480
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Fig. 5. Price trend estimation of gasoline and electricity in Taiwan.

Source: Huang et al. (2017)

distance than small-sized ones in Taiwan. For users of mid-sized replacement cost. Fig. 10 shows the rate of change of diffusion in the
motorcycles, high oil price means high fuel cost. Hence, mid-sized two scenarios. The diffusion will enter the mature phase of the s-shaped
electric motorcycles offer a strong economic incentive in the case of curve during 2024–2027. Lower rate of change indicates sooner market
high oil price. saturation. These trends confirm that appropriate subsidies can accel-
erate the replacement of traditional gasoline-powered motorcycles in
4.1.2.2. The scenario with government subsidies. Market diffusion of mid- both the high oil price scenario and the decreasing battery replacement
sized electric motorcycles is greater in government subsidies scenario cost scenario.
compared to the high oil price scenario and the decreasing battery Government’s subsidies should be properly adjusted in the case of
replacement cost scenario. This suggests that government’s subsidies high oil price (M_B+C) and in the case of decreasing battery replace-
can accelerate the diffusion of mid-sized electric motorcycles (Fig. 9), ment cost (M_B+D) to avoid generating windfall profits. In both sce-
resulting in rapid emergence of early adopters related to technological narios, the price subsidy can be phased out in 2022 or 2023. Compared
lifecycle of green innovation. Taking into account the principle of fair to M_B case (only government subsidies without high oil price and
allocation of public resources, we assume that the government adjusts decreasing battery replace cost condition), the withdrawal can shorten
the subsidies when the payback period for consumers is shorter than 10 the duration of subsidization by about 20 years. The ideal adjustments
years. The termination of the price subsidy in 2042 and excise duty of government subsidies in the three scenarios are as shown in Fig. 11.
exemption in 2048 will cause a negative rate of change of diffusion
(Fig. 10). However, by that time, mid-sized electric motorcycles will 4.2. Evaluation of social benefits
gain competitiveness in the market and return to normal market
diffusion that follows an s-shape curve. The government offers tax exemptions or price subsidy to electric
We further analyze if the effect of government subsidies multiplies motorcycles for three main goals: promoting diffusion of green utili-
under the conditions of high oil price and decreasing battery zation, driving development of the green industry, and generating

Fig. 6. The change index trend of motorcycle purchase price in Taiwan (base on 2014 = 1). Note：Data estimated by experts by 2024 and keeping on a constant after
2025.

481
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Fig. 7. Diffusion curve in small-size electric motorcycle.

social benefits. Effectiveness evaluation can provide an important re- considered, the total energy saving can reach 0.08 PJ (1015 Joules).
ference on continuous implementation of a policy. Based on the diffu- Despite the absence of government policies, the energy saving benefit in
sion curve of electric motorcycles, we further evaluate the possible the scenario with decreasing lithium battery cost will be more sig-
social benefits generated by replacing gasoline-powered motorcycles nificant. It is estimated that the total energy saving can reach 0.14 PJ,
with electric motorcycles. The benefits include energy saving and pol- about 1.8 times of the amount in the scenario with only government
lution reduction. We use the scenario without government subsidies as subsidies. If government subsidies are also provided in the case of high
the reference case to calculate the social benefits variation from diffu- oil price or in the case of decreasing battery cost, the energy saving
sion of electric motorcycles under various external conditions (gov- benefit can improve by 1.4–2 times compared to the case with only
ernment subsidies, high oil price, and decreasing battery replacement government subsidies. It can be inferred that diffusion of electric mo-
cost). The results are as follows: torcycles can lead to a decline in the transportation sector’s dependency
on imports of crude oil.

4.2.1. Energy saving benefit

The fuel economy (kilometers per unit of energy) of electric mo- 4.2.2. Environmental benefit
torcycles is four times higher than that of gasoline-powered motor- In Taiwan, the CO2 emissions from the industrial sector ranked first,
cycles, as show in Table 2. Hence, the energy saving benefit from re- accounting for 48% of the national emissions, and CO2 emissions from
placing gasoline-powered motorcycles with electric ones varies by the transportation sector accounted for 14% of the national emissions,
condition of market diffusion. Fig. 12 shows that compared to the ranking second. The CO2 emissions from motorcycles are not the main
scenario without government subsidies, the energy saving benefit in the source of the transport sector, compared to other modes of transport.
scenario with government subsidies is similar to that in the scenario However, the diffusion of electric motorcycles is an important effect in
with high oil price. The benefit improves significantly after 2030. In the reduction of air pollution in densely populated metropolitan areas.
2050, the total use of gasoline can reduce by 2.93 million liters, but the Therefore, this study aims at assessing the social benefits of electric
total of use of electricity will increase by 4.22 million kWh. With both motorcycles in place of traditional gasoline motorcycles. In addition to

Fig. 8. The change rate of diffusion in small-size electric motorcycle.

482
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Fig. 9. Diffusion curve in medium-size electric motorcycle.

CO2 emissions reduction, this study also examines the reduction in air being imposed, the reduction in HC will gradually decline before 2025.
pollution (CO, HC, and NOx). The results of the study are shown in After 2025, the emission standards will not be further raised, so the
Fig. 13, Fig. 14, Fig. 15, and Fig. 16. reduction in HC will increase again.
We use the average CO2 emission per kWh of power plants in
Taiwan (0.521 kg CO2/kWh), which refers to the average emission of 5. Discussion of policy implications and conclusions
the country's power system, including coal-fired, gas-fired, and oil-fired
power plants, as the basis to calculate the CO2 emission of electric To achieve diffusion of an emerging energy technology or product,
motorcycles. As shown in Fig. 13, we calculate the net CO2 emission by it is necessary to create reasonable conditions that can make the tech-
combining the increase in electricity use and the decrease in gasoline nology or product more desirable in the consumer market. Diffusion of
use. Results indicate that although electricity is mainly generated from the technology or product can further trigger a technology learning
coal power plants, the increase in CO2 emission of electricity used by effect. In this study, our literature review suggests that financial ben-
electric motorcycles is smaller than the reduction in CO2 emission from efits are one of the important factors affecting green utilization, and
gasoline-powered motorcycles. Hence, electric motorcycles can con- governments play a key role in bridging the gap between green utili-
tribute to mitigation of GHG emissions in Taiwan. Compared to the zation and production. Hence, we build a market diffusion model from
scenario without government subsidies, the reduction in CO2 emissions a perspective of consumers’ investment decision to explore how
in the scenario with government subsidies can reach 5800 tons by 2050. changes in technology brought about by green incentives are useful
If other favorable external conditions are available in addition to gov- initially, but may create conditions where such incentives can be
ernment subsidies, the reduction can be further boosted by 1.7–2.1 phased out. The government's green subsidy phase-out mechanism as-
times compared to the reduction with government subsidies scenario. sessed in this study is based on the decision-making principle of con-
Electric motorcycles can also contribute to reduction of air pollu- sumers' investment benefits. The benefits of the subsidy policy are two
tions. The Environmental Protection Administration has planned to
folds. First, allowing consumers to receive positive investment benefits
gradually lower the maximum tolerable emissions of air pollution from through the purchase of alternative transportation mode at the same
motorcycles during 2017–2021. Compared to existing standards, the
time changing their purchasing behavior from gasoline motorcycle to
new standards will be raised by 50~88%. With stricter standards are electric motorcycle. Second, the diffusion of emerging energy

Fig. 10. The change rate of diffusion in medium-size electric motorcycle.

483
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Fig. 13. The impact of electric motorcycle diffusion on CO2 emission.

Fig. 14. The impact of electric motorcycle diffusion on CO emission.

Fig. 11. The dynamic adjustment mechanism of government subsidies on

medium-size electric vehicle.

Fig. 15. The impact of electric motorcycle diffusion on HC emission.

the market. The government would consider phase out its subsidy
policy gradually, allowing emerging energy technologies to return to
market mechanism.
Results indicate that the government’s green incentives indeed have
a large and positive effect on diffusion of electric motorcycles in
Taiwan. Because the technology lifecycle curves of small and mid-sized
electric motorcycles are located at different phases, the government’s
incentives for these two types of vehicles should be different. It is es-
Fig. 12. The impact of electric motorcycle diffusion on energy saving. timated that small-sized electric motorcycles can gain market compe-
titiveness after 2018, so the green incentives for this type of vehicle
should be gradually phased out to let the diffusion follow a normal
technologies caused by changing of purchasing behavior will create
course. The suspended subsidy can be allocated to promotion of mid or
economies of scale in production which will reduce production costs
large models of electric motorcycles with a higher utility rate. Mid-sized
gradually making emerging energy technologies more competitive in
motorcycles take up 42% of the market in Taiwan. Without green

484
S.K. Huang et al. Energy Policy 119 (2018) 473–486

again because the standards are not further elevated. It can be inferred
that there is a limit to the improvement of traditional motorcycles to-
ward compliance with emission standards. Given this strong niche ad-
vantage and the strict emission standards for motorcycles, there can be
greater diffusion of electric motorcycles.
In Taiwan, electric motorcycles are mainly used for short-distance
commuting, and electric motorcycles have a daily travel distance be-
tween 10 and 27.2 km. Besides, most electric motorcycles can travel
40–62 km on each full charge, so one full charge can accommodate the
basic transportation needs in the next day. Portable batteries can be
charged without being bound to a specific location. Hence, the gov-
ernment in Taiwan has endeavored to build battery charging facilities
of different models to meet the needs of consumers. If the charging is
performed during off-peak hours, the electricity cost can be reduced.
Therefore, we infer that the rate of diffusion of electric motorcycles in
Fig. 16. The impact of electric motorcycle diffusion on NOx emission.
Taiwan will not be constrained by battery charging facilities.
Previous research of green behavior is primarily focused on survey
incentives, it would be hard to form an s-shaped diffusion curve for of consumer preferences. So far, an in-depth quantitative examination
mid-sized electric motorcycles. Our results show that green incentives of the government’s role in the relation between green utilization and
can facilitate diffusion of mid-sized electric motorcycles. Besides, these production is still absent. Unlike previous research of green behavior,
incentives should be constantly adjusted depending on changes in ex- the main contribution of this paper lies in an in-depth exploration of the
ternal conditions, such as high oil price and decreasing battery cost. effects of factors affecting diffusion of green utilization using a quan-
Under promotion of financial incentives, electric motorcycles will ob- titative model developed based on the associations between govern-
tain competitive conditions sooner or later. It is necessary for the ment incentives and consumers’ investment decisions. The factors ex-
government to set up an incentive termination mechanism based on a amined include incentives policy, oil price, battery cost, and emission
certain rate of return on investment or a break-even point to avoid standards for vehicles. Besides, this study also presents an incentive
generating windfall profits for consumers. adjustment mechanism to ensure fair allocation of public resources and
External conditions that affect diffusion of electric motorcycles in- evaluate whether the social benefits from government policies can
clude government incentives, increase in oil price, decrease in battery contribute to the national goals of sustainable development. In this
cost, and pollution control standards. Compared to the uncertainty of study, our results provide a reference on effective implementation of
oil price or battery cost, government’s incentives are more foreseeable green policies and some criteria for investment in green products. For
for consumers. In other words, government’s incentives can reduce the businesses, these results also indirectly provide a forecast on the market
uncertainty in consumers’ green investment decision. Our results reveal growth of green products.
that the effect of green incentives on diffusion of electric motorcycles is However, our model is subject to some constraints. The main dis-
almost comparable to that of high oil price. Besides, the diffusion of advantages of the quantified modeling analysis in this study are as
green utilization will be amplified when subsidies induce the tech- below:
nology learning. The completeness of research results is constrained. The energy
The budget of government’s green incentives come from the tax system involves a huge and complex causal relationship from upstream
revenue collected from all nationals. Hence, allocation of green in- to downstream. It is necessary to clearly define the scope of the study in
centives should be based on the benefits they can bring, such as in- order to avoid losing the focus of the research topic. The research scope
dustry benefits or social benefits. These benefits are the criteria for of this study is limited to assessing the effects of replacement of
evaluating the legitimacy of a policy or the necessity of continuous transport vehicles with the same use. We do not consider the compre-
implementation of the policy. Hence, we further evaluate the potential hensive impact, including the replacement of transport vehicles with
social benefits from the government’s green incentives for electric different use and the effect of energy production on the energy-saving
motorcycles. Our results indicate that due to higher fuel economy, and carbon reduction of energy use.
electric motorcycles offer a significant energy saving benefit, which in The assumptions of the parameter in the future are uncertainty. To
turn, can reduce Taiwan’s dependency on crude oil imports. Our results simulate impacts of policies in the future, it is necessary to set some
confirm that the increase in CO2 emissions caused by use of electric assumptions. In general, sensitivity analysis of parameter assumptions
motorcycles is smaller than the reduction in CO2 emissions from ga- is usually used to resolve uncertainties. Therefore, how to obtain the
soline-powered motorcycles. In the scenario with government subsidies, objective and representative assumptions is another obstacle for mod-
the net reduction of CO2 emissions can reach 5800 tons by 2050. eling study that needs to be overcome.
In addition to tax exemption and price subsidy, the existing gov-
ernment measures for promoting diffusion of green utilization also in- Acknowledgement
clude strict emission standards for air pollution, which also have a
substantial impact on diffusion of electric motorcycles. When the The authors would like to acknowledge the support from the Bureau
emission standards are raised, the cost of research and development of of Energy, ROC (Grant number 555A70134/H455CG4100) and
low-pollution technologies will become higher for gasoline-powered Industrial Technology Research Institute in Taiwan.
motorcycles than for electric motorcycles. This higher cost will in-
directly narrow the price gap between electric motorcycles and gaso- References
line-powered motorcycles, making electric motorcycles more competi-
tive in the market. In the simulation of scenarios, we have considered Adamson, K.A., 2005. Calculating the price trajectory of adoption of fuel cell vehicles. Int.
the rising tendency of the cost of gasoline-powered motorcycles due to J. Hydrogen Energy 30, 341–350.
increasingly stricter emission standards. A stricter allowable tolerable Awan, U., Sarwar, S., Raza, M.A., 2011. Green consumer behavior and environmental
sustainability. Biomass. Convers. Res. Cent. 1.
emissions standards by 2025 will decrease the contribution of reduction Awan, U., Raza, M.A., 2014. Green consumer behavior: empirical study of Swedish
in air pollution by electric motorcycles. After 2025, the benefit of using consumer behavior. Recent Res. Econ. 89–104.
electric motorcycles in the pollution reduction aspect will magnify Åhman, M., 2006. Government policy and the development of electric vehicles in Japan.

485
S.K. Huang et al. Energy Policy 119 (2018) 473–486

Energy Policy 34, 433–443. Moor, G.A., 2014. Crossing the Chasm-Marketing and Selling Disruptive Products to
Biswas, A., Roy, M., 2015. Leveraging factors for sustained green consumption behavior Mainstream Customers, 3rd Edition.
based on consumption value perceptions: testing the structural model. J. Clean. Prod. Nykvist, B., et al., 2015. Rapidly falling costs of battery packs for electric vehicles. Nat.
95, 332–340. Clim. Change. http://dx.doi.org/10.1038/nclimate2564.
Brand, C., Anable, J., Tran, M., 2013. Accelerating the transformation to a low carbon Norberg-Bohn, V., 1999. Stimulating green technological innovation: an analysis of al-
passenger transport system: the role of car purchase taxes, feebates, road taxes and ternative policy mechanisms. Policy Sci. 32, 1338.
scrappage incentives in the UK. Transp. Res. Part A 49, 132–148. Peattie, K., 2010. Green consumption: behavior and norms. Annu. Rev. Environ. Resour.
Bomb, C., McCormick, K., Deurwaarder, E., Kåberger, T., 2007. Biofuels for transport in 35, 195–228.
Europe: lessons from Germany and the UK. Energy Policy 35, 2256–2267. Potoglou, D., Kanaroglou, P.S., 2007. Household demand and willingness to pay for clean
Chandra, A., Gulati, S., Kandlikar, M., 2010. Green drivers or free riders? An analysis of vehicles. Transp. Res. Part D 12, 264–274.
tax rebates for hybrid vehicles. J. Environ. Econ. Manag. 60, 78–93. PwC(PricewaterhouseCoopers), 2013. Charging forward-PwC’s 2012 electric vehicles
Cherian, J., Jacob, J., 2012. Green marketing: a study of consumers' attitude towards survey.
environment friendly products. Asian Soc. Sci. 8, 117–126. Ritter, Á.M., Borchardt, M., Vaccaro, G.L.R., Pereira, G.M., 2015. Motivations for pro-
Chen, H.S., Chen, C.Y., Chen, H.K., Hsieh, T., 2012. A study of relationships among green moting the consumption of green products in an emerging country: exploring atti-
consumption attitude, perceived risk, perceived value toward hydropgen-electric tudes of Brazilian consumers. J. Clean. Prod. 106, 507–520.
motorcycle purchase intention. AASRI Procedia 2, 163–168. Rogers, E.M., 1962. Diffusion of Innovations. Free Press of Glencoe, New York.
Cheng, Q., Wang, J., Yin, L., 2014. A research framework of green consumption behavior Rogan, F., Dennehy, E., Daly, H., Howley, M., Gallachóir, B.P., 2011. Impacts of an
based on value-belief-norm theory. Advanced Mater. Res. 962-965, 1485–1489. emission based private car taxation policy-First year ex-post analysis. Transp. Res.
Delafrooz, N., Taleghani, M., Nouri, B., 2014. Effect of green marketing on consumer Part A 45, 583–597.
purchase behavior. QSci. Connect 5. Sang, Y., Bekhet, H.A., 2015. Modelling electric vehicle usage intentions: an empirical
Ek, K., Söderholm, P., 2008. Norms and economic motivation in the Swedish green study in Malaysia. J. Clean. Prod. 92, 75–83.
electricity market. Ecol. Econ. 68, 169–182. Santos, G., Behrendt, H., Maconi, L., Shirvani, T., Teytelboym, A., 2010. Part I: ex-
Frederiks, E.R., Stenner, K., Hobman, E.V., 2015. Household energy use: applying be- ternalities and economic policies in road transport. Res. Transp. Econ. 28, 2–45.
havioural economics to understand consumer decision-making and behavior. Renew. Sinnappan, P., Rahman, A.A., 2011. Antecedents of green purchasing behavior among
Sustain. Energy Rev. 41, 1385–1394. Malaysian consumers. Int. Bus. Manag. 5, 129–138.
Gallagher, K.S., Muehlegger, E., 2011. Giving green to get green? Incentives and con- Sarumathi, S., 2014. Green purchase behavior – a conceptual framework socially con-
sumer adoption of hybrid vehicle technology. J. Eviron. Econ. Manag. 61, 1–15. scious consumer behavior. Glob. J. Financ. Manag. 6, 777–782.
Haan, P.D., Peters, A., Scholz, R.W., 2007. Reducing energy consumption in road trans- Skerlos, S.J., Winebrake, J.J., 2010. Targeting plug-in hybrid electric vehicle policies to
port through hybrid vehicles: investigation of rebound effects, and possible effects of increase social benefits. Energy Policy 38, 705–708.
tax rebates. J. Clean. Prod. 15, 1076–1084. Stevens, C., 2010. Linking sustainable consumption and production: the government role.
He, L., Wang, M., Chen, Wei, Conzelmann, G., 2014. Incorporating social impact on new Nat. Resour. Forum 34, 16–23.
product adoption in choice modeling: a case study in green vehicles. Transp. Res. Part Tran, M., 2012. Technology-behavioural modelling of energy innovation diffusion in the
D 32, 421–434. UK. Appl. Energy 95, 1–11.
Huang, S.K., Kuo, L., Chou, K.L., 2017. Integrating the induced market diffusion into the Tseng, M.L., Chiu, S.F., Tan, R.R., Anna, B.S.M., 2013. Sustainable consumption and
carbon reduction contribution assessment of emerging technology: a case study of production for Asia: sustainability through green design and practice. J. Clean. Prod.
Taiwan’s electric vehicles development. J. Technol. Manag. 22, 1–28 (in Chinese). 40, 1–5.
IEA, 2015. Global EV Outlook update. Testa, F., Cosic, A., Iraldo, F., 2015. Determining factors of curtailment and purchasing
Jansson, J., Marell, A., Nordlund, A., 2010. Green consumer behavior: determinants of energy related behaviors. J. Clean. Prod. 1–10. http://dx.doi.org/10.1016/j.jclepro.
curtailment and eco-innovation adoption. J. Consum. Mark. 27/4, 358–370. 2015.07.134.
Jaakkola, H., Gabbouj, M., Neuvo, Y., 1998. Fundamentals of technology diffusion and Taiwan Ministry of Transportation and communications, 2011. Survey report on
mobile phone case study. Circuits Syst. Dign. Process. 17, 421–448. Motorcycle in use.
Kaufmann, H.R., Panni, M.F.A.K., Orphanidou, Y., 2012. Factors affecting consumers’ TaiwanJobs, 2017. Industrial overview and technology application trend of electric mo-
green purchasing behaviou: an integrated conceptual framework. Amfiteatru Econ. torcycles and energy storage systems 〈https://www.taiwanjobs.gov.tw/internet/
14, 50–69. jobwanted/docDetail.aspx?Uid=1590&docid=33252〉.
Kim, Y., Choi, S.M., 2005. Antecedents of green purchase behavior: an examination of U.S. Energy Information Administration, 2011. NEMS model documentation: Commercial
collectivism, environmental, concern, and PCE. Adv. Consum. Res. 32, 592–599. sector demand module, pp. 185–186.
Kong, D.Y., Bi, X.H., 2014. Impact of Social Network and Business Model on Innovation U.S. Energy Information Administration, 2015. Annual Energy Outlook.
Diffusion of Electric Vehicles in China. Hindawi Publishing Corporation Virginie, L., 2010. Residential Electricity demand in China-Can Efficiency Reverse the
Mathematical Problems in Engineering. Growth? Lawrence Berkeley National Laboratory (LBNL-2321E).
Lane, B., Potter, S., 2007. The adoption of cleaner vehicles in the UK: exploring the Wu, S.I., 2015. Effect of green consumption perception degree on relationship model of
consumer attitude-action gap. J. Clean. Prod. 15, 1085–1092. green consumption behavior. J. Manag. Strategy 6, 109–123.
Lin, P.C., Huang, Y.H., 2012. The influence factors on choice behavior regarding green Wu, S.I., Chen, J.Y., 2014. A model of green consumption behavior constructed by the
products based on the theory of consumption values. J. Clean. Prod. 22, 11–18. theory of planned behavior. Int. J. Mark. Stud. 6.
Loock, C.M., Staake, T., Thiesse, F., 2013. Motivating energy-efficient behavior with Won, J.R., Yoo, Y.B., Lee, K.J., 2009. Prediction of electricity demand due to PHEVs
green IS: An investigation of goal setting and the role of defaults. MIS Q. 37, (Plug-in hybrid electric vehicles) distribution in Korea by using diffusion model. IEEE
1313–1332. T&D Asia.
Lorek, S., Spangenberg, J.H., 2014. Sustainable consumption within a sustainable Young, W., Hwang, K., McDonald, S., Oates, C.J., 2010. Sustainable consumption: green
economy-beyond green growth and green economies. J. Clean. Prod. 63, 33–44. consumer behavior when purchasing products. Sustain. Dev. 18, 20–31.
Ludhiyani, A., Joshi, S., Pathak, R., 2012. Consumer behavior and strategic finance in Zhao, H.H., Gao, Q., Wu, Y.P., Wang, Y., Zhu, X., 2014. What affects green consumer
green energy. Intellect. Econ. 6, 695–712. behavior in China? A case study from Qingdao. J. Clean. Prod. 63, 143–151.
Lund, P., 2006. Market penetration rates of new energy technologies. Energy Policy 34, Zhang, X., Wang, K., Hao, Y., Fan, J., Wei, T., 2013. The impact of government policy on
33173326. preference for NEVs: the evidence from China. Energy Policy 61, 382–393.
Materialsnet, 2017. New Age and New Business Opportunities for Electric Motorcycle Zhou, N., Fridley, D., McNeil, M., Zheng, N., Letschert, V., KE, J., Saheb, Y., 2011.
〈https://www.materialsnet.com.tw/DocView.aspx?Id=32726〉. Analysis of potential energy saving and CO2 emission reduction of home appliances
Mourato, S., Saynor, B., Hart, D., 2004. Greening London’s black cabs: a study of driver’s and commercial equipments in China. Energy Policy 39, 4541–4550.
preferences for fuel cell taxis. Energy Policy 32, 685–695.

486